From 93c8ebfa780ebd1495095e794731881aef29e7d3 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Daniel=20Sch=C3=BCrmann?= Date: Tue, 17 Sep 2019 13:22:17 +0200 Subject: [PATCH] aco: Initial commit of independent AMD compiler MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit ACO (short for AMD Compiler) is a new compiler backend with the goal to replace LLVM for Radeon hardware for the RADV driver. ACO currently supports only VS, PS and CS on VI and Vega. There are some optimizations missing because of unmerged NIR changes which may decrease performance. Full commit history can be found at https://github.com/daniel-schuermann/mesa/commits/backend Co-authored-by: Daniel Schürmann Co-authored-by: Rhys Perry Co-authored-by: Bas Nieuwenhuizen Co-authored-by: Connor Abbott Co-authored-by: Michael Schellenberger Costa Co-authored-by: Timur Kristóf Acked-by: Samuel Pitoiset Acked-by: Bas Nieuwenhuizen --- src/amd/compiler/README | 87 + src/amd/compiler/aco_assembler.cpp | 497 ++ src/amd/compiler/aco_builder_h.py | 400 + src/amd/compiler/aco_dead_code_analysis.cpp | 102 + src/amd/compiler/aco_dominance.cpp | 93 + src/amd/compiler/aco_insert_NOPs.cpp | 282 + src/amd/compiler/aco_insert_exec_mask.cpp | 1078 +++ src/amd/compiler/aco_insert_waitcnt.cpp | 697 ++ .../compiler/aco_instruction_selection.cpp | 7621 +++++++++++++++++ .../aco_instruction_selection_setup.cpp | 1366 +++ src/amd/compiler/aco_interface.cpp | 166 + src/amd/compiler/aco_interface.h | 45 + src/amd/compiler/aco_ir.h | 1169 +++ src/amd/compiler/aco_live_var_analysis.cpp | 243 + src/amd/compiler/aco_lower_bool_phis.cpp | 241 + src/amd/compiler/aco_lower_to_hw_instr.cpp | 765 ++ src/amd/compiler/aco_opcodes.py | 1552 ++++ src/amd/compiler/aco_opcodes_cpp.py | 74 + src/amd/compiler/aco_opcodes_h.py | 47 + src/amd/compiler/aco_opt_value_numbering.cpp | 327 + src/amd/compiler/aco_optimizer.cpp | 2401 ++++++ src/amd/compiler/aco_print_asm.cpp | 104 + src/amd/compiler/aco_print_ir.cpp | 575 ++ src/amd/compiler/aco_reduce_assign.cpp | 164 + src/amd/compiler/aco_register_allocation.cpp | 1924 +++++ src/amd/compiler/aco_scheduler.cpp | 835 ++ src/amd/compiler/aco_spill.cpp | 1630 ++++ src/amd/compiler/aco_ssa_elimination.cpp | 291 + src/amd/compiler/aco_util.h | 233 + src/amd/compiler/aco_validate.cpp | 460 + src/amd/compiler/meson.build | 103 + 31 files changed, 25572 insertions(+) create mode 100644 src/amd/compiler/README create mode 100644 src/amd/compiler/aco_assembler.cpp create mode 100644 src/amd/compiler/aco_builder_h.py create mode 100644 src/amd/compiler/aco_dead_code_analysis.cpp create mode 100644 src/amd/compiler/aco_dominance.cpp create mode 100644 src/amd/compiler/aco_insert_NOPs.cpp create mode 100644 src/amd/compiler/aco_insert_exec_mask.cpp create mode 100644 src/amd/compiler/aco_insert_waitcnt.cpp create mode 100644 src/amd/compiler/aco_instruction_selection.cpp create mode 100644 src/amd/compiler/aco_instruction_selection_setup.cpp create mode 100644 src/amd/compiler/aco_interface.cpp create mode 100644 src/amd/compiler/aco_interface.h create mode 100644 src/amd/compiler/aco_ir.h create mode 100644 src/amd/compiler/aco_live_var_analysis.cpp create mode 100644 src/amd/compiler/aco_lower_bool_phis.cpp create mode 100644 src/amd/compiler/aco_lower_to_hw_instr.cpp create mode 100644 src/amd/compiler/aco_opcodes.py create mode 100644 src/amd/compiler/aco_opcodes_cpp.py create mode 100644 src/amd/compiler/aco_opcodes_h.py create mode 100644 src/amd/compiler/aco_opt_value_numbering.cpp create mode 100644 src/amd/compiler/aco_optimizer.cpp create mode 100644 src/amd/compiler/aco_print_asm.cpp create mode 100644 src/amd/compiler/aco_print_ir.cpp create mode 100644 src/amd/compiler/aco_reduce_assign.cpp create mode 100644 src/amd/compiler/aco_register_allocation.cpp create mode 100644 src/amd/compiler/aco_scheduler.cpp create mode 100644 src/amd/compiler/aco_spill.cpp create mode 100644 src/amd/compiler/aco_ssa_elimination.cpp create mode 100644 src/amd/compiler/aco_util.h create mode 100644 src/amd/compiler/aco_validate.cpp create mode 100644 src/amd/compiler/meson.build diff --git a/src/amd/compiler/README b/src/amd/compiler/README new file mode 100644 index 00000000000..87d63c07024 --- /dev/null +++ b/src/amd/compiler/README @@ -0,0 +1,87 @@ +# Unofficial GCN/RDNA ISA reference errata + +## v_sad_u32 + +The Vega ISA reference writes it's behaviour as: +``` +D.u = abs(S0.i - S1.i) + S2.u. +``` +This is incorrect. The actual behaviour is what is written in the GCN3 reference +guide: +``` +ABS_DIFF (A,B) = (A>B) ? (A-B) : (B-A) +D.u = ABS_DIFF (S0.u,S1.u) + S2.u +``` +The instruction doesn't subtract the S0 and S1 and use the absolute value (the +_signed_ distance), it uses the _unsigned_ distance between the operands. So +`v_sad_u32(-5, 0, 0)` would return `4294967291` (`-5` interpreted as unsigned), +not `5`. + +## s_bfe_* + +Both the Vega and GCN3 ISA references write that these instructions don't write +SCC. They do. + +## v_bcnt_u32_b32 + +The Vega ISA reference writes it's behaviour as: +``` +D.u = 0; +for i in 0 ... 31 do +D.u += (S0.u[i] == 1 ? 1 : 0); +endfor. +``` +This is incorrect. The actual behaviour (and number of operands) is what +is written in the GCN3 reference guide: +``` +D.u = CountOneBits(S0.u) + S1.u. +``` + +## SMEM stores + +The Vega ISA references doesn't say this (or doesn't make it clear), but +the offset for SMEM stores must be in m0 if IMM == 0. + +The RDNA ISA doesn't mention SMEM stores at all, but they seem to be supported +by the chip and are present in LLVM. AMD devs however highly recommend avoiding +these instructions. + +## SMEM atomics + +RDNA ISA: same as the SMEM stores, the ISA pretends they don't exist, but they +are there in LLVM. + +## VMEM stores + +All reference guides say (under "Vector Memory Instruction Data Dependencies"): +> When a VM instruction is issued, the address is immediately read out of VGPRs +> and sent to the texture cache. Any texture or buffer resources and samplers +> are also sent immediately. However, write-data is not immediately sent to the +> texture cache. +Reading that, one might think that waitcnts need to be added when writing to +the registers used for a VMEM store's data. Experimentation has shown that this +does not seem to be the case on GFX8 and GFX9 (GFX6 and GFX7 are untested). It +also seems unlikely, since NOPs are apparently needed in a subset of these +situations. + +## MIMG opcodes on GFX8/GCN3 + +The `image_atomic_{swap,cmpswap,add,sub}` opcodes in the GCN3 ISA reference +guide are incorrect. The Vega ISA reference guide has the correct ones. + +## Legacy instructions + +Some instructions have a `_LEGACY` variant which implements "DX9 rules", in which +the zero "wins" in multiplications, ie. `0.0*x` is always `0.0`. The VEGA ISA +mentions `V_MAC_LEGACY_F32` but this instruction is not really there on VEGA. + +# Hardware Bugs + +## SMEM corrupts VCCZ on SI/CI + +https://github.com/llvm/llvm-project/blob/acb089e12ae48b82c0b05c42326196a030df9b82/llvm/lib/Target/AMDGPU/SIInsertWaits.cpp#L580-L616 +After issuing a SMEM instructions, we need to wait for the SMEM instructions to +finish and then write to vcc (for example, `s_mov_b64 vcc, vcc`) to correct vccz + +Currently, we don't do this. + diff --git a/src/amd/compiler/aco_assembler.cpp b/src/amd/compiler/aco_assembler.cpp new file mode 100644 index 00000000000..a6bf2a3e0db --- /dev/null +++ b/src/amd/compiler/aco_assembler.cpp @@ -0,0 +1,497 @@ +#include + +#include "aco_ir.h" +#include "common/sid.h" + +namespace aco { + +struct asm_context { + Program *program; + enum chip_class chip_class; + std::map branches; + std::vector constaddrs; + const int16_t* opcode; + // TODO: keep track of branch instructions referring blocks + // and, when emitting the block, correct the offset in instr + asm_context(Program* program) : program(program), chip_class(program->chip_class) { + if (chip_class <= GFX9) + opcode = &instr_info.opcode_gfx9[0]; + } +}; + +void emit_instruction(asm_context& ctx, std::vector& out, Instruction* instr) +{ + uint32_t instr_offset = out.size() * 4u; + + /* lower remaining pseudo-instructions */ + if (instr->opcode == aco_opcode::p_constaddr) { + unsigned dest = instr->definitions[0].physReg(); + unsigned offset = instr->operands[0].constantValue(); + + /* s_getpc_b64 dest[0:1] */ + uint32_t encoding = (0b101111101 << 23); + uint32_t opcode = ctx.opcode[(int)aco_opcode::s_getpc_b64]; + if (opcode >= 55 && ctx.chip_class <= GFX9) { + assert(ctx.chip_class == GFX9 && opcode < 60); + opcode = opcode - 4; + } + encoding |= dest << 16; + encoding |= opcode << 8; + out.push_back(encoding); + + /* s_add_u32 dest[0], dest[0], ... */ + encoding = (0b10 << 30); + encoding |= ctx.opcode[(int)aco_opcode::s_add_u32] << 23; + encoding |= dest << 16; + encoding |= dest; + encoding |= 255 << 8; + out.push_back(encoding); + ctx.constaddrs.push_back(out.size()); + out.push_back(-(instr_offset + 4) + offset); + + /* s_addc_u32 dest[1], dest[1], 0 */ + encoding = (0b10 << 30); + encoding |= ctx.opcode[(int)aco_opcode::s_addc_u32] << 23; + encoding |= (dest + 1) << 16; + encoding |= dest + 1; + encoding |= 128 << 8; + out.push_back(encoding); + return; + } + + uint32_t opcode = ctx.opcode[(int)instr->opcode]; + if (opcode == (uint32_t)-1) { + fprintf(stderr, "Unsupported opcode: "); + aco_print_instr(instr, stderr); + abort(); + } + + switch (instr->format) { + case Format::SOP2: { + uint32_t encoding = (0b10 << 30); + encoding |= opcode << 23; + encoding |= !instr->definitions.empty() ? instr->definitions[0].physReg() << 16 : 0; + encoding |= instr->operands.size() >= 2 ? instr->operands[1].physReg() << 8 : 0; + encoding |= !instr->operands.empty() ? instr->operands[0].physReg() : 0; + out.push_back(encoding); + break; + } + case Format::SOPK: { + uint32_t encoding = (0b1011 << 28); + encoding |= opcode << 23; + encoding |= + !instr->definitions.empty() && !(instr->definitions[0].physReg() == scc) ? + instr->definitions[0].physReg() << 16 : + !instr->operands.empty() && !(instr->operands[0].physReg() == scc) ? + instr->operands[0].physReg() << 16 : 0; + encoding |= static_cast(instr)->imm; + out.push_back(encoding); + break; + } + case Format::SOP1: { + uint32_t encoding = (0b101111101 << 23); + if (opcode >= 55 && ctx.chip_class <= GFX9) { + assert(ctx.chip_class == GFX9 && opcode < 60); + opcode = opcode - 4; + } + encoding |= !instr->definitions.empty() ? instr->definitions[0].physReg() << 16 : 0; + encoding |= opcode << 8; + encoding |= !instr->operands.empty() ? instr->operands[0].physReg() : 0; + out.push_back(encoding); + break; + } + case Format::SOPC: { + uint32_t encoding = (0b101111110 << 23); + encoding |= opcode << 16; + encoding |= instr->operands.size() == 2 ? instr->operands[1].physReg() << 8 : 0; + encoding |= !instr->operands.empty() ? instr->operands[0].physReg() : 0; + out.push_back(encoding); + break; + } + case Format::SOPP: { + SOPP_instruction* sopp = static_cast(instr); + uint32_t encoding = (0b101111111 << 23); + encoding |= opcode << 16; + encoding |= (uint16_t) sopp->imm; + if (sopp->block != -1) + ctx.branches.insert({out.size(), sopp}); + out.push_back(encoding); + break; + } + case Format::SMEM: { + SMEM_instruction* smem = static_cast(instr); + uint32_t encoding = (0b110000 << 26); + encoding |= opcode << 18; + if (instr->operands.size() >= 2) + encoding |= instr->operands[1].isConstant() ? 1 << 17 : 0; + bool soe = instr->operands.size() >= (!instr->definitions.empty() ? 3 : 4); + assert(!soe || ctx.chip_class >= GFX9); + encoding |= soe ? 1 << 14 : 0; + encoding |= smem->glc ? 1 << 16 : 0; + if (!instr->definitions.empty() || instr->operands.size() >= 3) + encoding |= (!instr->definitions.empty() ? instr->definitions[0].physReg() : instr->operands[2].physReg().reg) << 6; + if (instr->operands.size() >= 1) + encoding |= instr->operands[0].physReg() >> 1; + out.push_back(encoding); + encoding = 0; + if (instr->operands.size() >= 2) + encoding |= instr->operands[1].isConstant() ? instr->operands[1].constantValue() : instr->operands[1].physReg().reg; + encoding |= soe ? instr->operands.back().physReg() << 25 : 0; + out.push_back(encoding); + return; + } + case Format::VOP2: { + uint32_t encoding = 0; + encoding |= opcode << 25; + encoding |= (0xFF & instr->definitions[0].physReg().reg) << 17; + encoding |= (0xFF & instr->operands[1].physReg().reg) << 9; + encoding |= instr->operands[0].physReg().reg; + out.push_back(encoding); + break; + } + case Format::VOP1: { + uint32_t encoding = (0b0111111 << 25); + encoding |= (0xFF & instr->definitions[0].physReg().reg) << 17; + encoding |= opcode << 9; + encoding |= instr->operands[0].physReg().reg; + out.push_back(encoding); + break; + } + case Format::VOPC: { + uint32_t encoding = (0b0111110 << 25); + encoding |= opcode << 17; + encoding |= (0xFF & instr->operands[1].physReg().reg) << 9; + encoding |= instr->operands[0].physReg().reg; + out.push_back(encoding); + break; + } + case Format::VINTRP: { + Interp_instruction* interp = static_cast(instr); + uint32_t encoding = (0b110101 << 26); + encoding |= (0xFF & instr->definitions[0].physReg().reg) << 18; + encoding |= opcode << 16; + encoding |= interp->attribute << 10; + encoding |= interp->component << 8; + if (instr->opcode == aco_opcode::v_interp_mov_f32) + encoding |= (0x3 & instr->operands[0].constantValue()); + else + encoding |= (0xFF & instr->operands[0].physReg().reg); + out.push_back(encoding); + break; + } + case Format::DS: { + DS_instruction* ds = static_cast(instr); + uint32_t encoding = (0b110110 << 26); + encoding |= opcode << 17; + encoding |= (ds->gds ? 1 : 0) << 16; + encoding |= ((0xFF & ds->offset1) << 8); + encoding |= (0xFFFF & ds->offset0); + out.push_back(encoding); + encoding = 0; + unsigned reg = !instr->definitions.empty() ? instr->definitions[0].physReg() : 0; + encoding |= (0xFF & reg) << 24; + reg = instr->operands.size() >= 3 && !(instr->operands[2].physReg() == m0) ? instr->operands[2].physReg() : 0; + encoding |= (0xFF & reg) << 16; + reg = instr->operands.size() >= 2 && !(instr->operands[1].physReg() == m0) ? instr->operands[1].physReg() : 0; + encoding |= (0xFF & reg) << 8; + encoding |= (0xFF & instr->operands[0].physReg().reg); + out.push_back(encoding); + break; + } + case Format::MUBUF: { + MUBUF_instruction* mubuf = static_cast(instr); + uint32_t encoding = (0b111000 << 26); + encoding |= opcode << 18; + encoding |= (mubuf->slc ? 1 : 0) << 17; + encoding |= (mubuf->lds ? 1 : 0) << 16; + encoding |= (mubuf->glc ? 1 : 0) << 14; + encoding |= (mubuf->idxen ? 1 : 0) << 13; + encoding |= (mubuf->offen ? 1 : 0) << 12; + encoding |= 0x0FFF & mubuf->offset; + out.push_back(encoding); + encoding = 0; + encoding |= instr->operands[2].physReg() << 24; + encoding |= (mubuf->tfe ? 1 : 0) << 23; + encoding |= (instr->operands[1].physReg() >> 2) << 16; + unsigned reg = instr->operands.size() > 3 ? instr->operands[3].physReg() : instr->definitions[0].physReg().reg; + encoding |= (0xFF & reg) << 8; + encoding |= (0xFF & instr->operands[0].physReg().reg); + out.push_back(encoding); + break; + } + case Format::MTBUF: { + MTBUF_instruction* mtbuf = static_cast(instr); + uint32_t encoding = (0b111010 << 26); + encoding |= opcode << 15; + encoding |= (mtbuf->glc ? 1 : 0) << 14; + encoding |= (mtbuf->idxen ? 1 : 0) << 13; + encoding |= (mtbuf->offen ? 1 : 0) << 12; + encoding |= 0x0FFF & mtbuf->offset; + encoding |= (0xF & mtbuf->dfmt) << 19; + encoding |= (0x7 & mtbuf->nfmt) << 23; + out.push_back(encoding); + encoding = 0; + encoding |= instr->operands[2].physReg().reg << 24; + encoding |= (mtbuf->tfe ? 1 : 0) << 23; + encoding |= (mtbuf->slc ? 1 : 0) << 22; + encoding |= (instr->operands[1].physReg().reg >> 2) << 16; + unsigned reg = instr->operands.size() > 3 ? instr->operands[3].physReg().reg : instr->definitions[0].physReg().reg; + encoding |= (0xFF & reg) << 8; + encoding |= (0xFF & instr->operands[0].physReg().reg); + out.push_back(encoding); + break; + } + case Format::MIMG: { + MIMG_instruction* mimg = static_cast(instr); + uint32_t encoding = (0b111100 << 26); + encoding |= mimg->slc ? 1 << 25 : 0; + encoding |= opcode << 18; + encoding |= mimg->lwe ? 1 << 17 : 0; + encoding |= mimg->tfe ? 1 << 16 : 0; + encoding |= mimg->r128 ? 1 << 15 : 0; + encoding |= mimg->da ? 1 << 14 : 0; + encoding |= mimg->glc ? 1 << 13 : 0; + encoding |= mimg->unrm ? 1 << 12 : 0; + encoding |= (0xF & mimg->dmask) << 8; + out.push_back(encoding); + encoding = (0xFF & instr->operands[0].physReg().reg); /* VADDR */ + if (!instr->definitions.empty()) { + encoding |= (0xFF & instr->definitions[0].physReg().reg) << 8; /* VDATA */ + } else if (instr->operands.size() == 4) { + encoding |= (0xFF & instr->operands[3].physReg().reg) << 8; /* VDATA */ + } + encoding |= (0x1F & (instr->operands[1].physReg() >> 2)) << 16; /* T# (resource) */ + if (instr->operands.size() > 2) + encoding |= (0x1F & (instr->operands[2].physReg() >> 2)) << 21; /* sampler */ + // TODO VEGA: D16 + out.push_back(encoding); + break; + } + case Format::FLAT: + case Format::SCRATCH: + case Format::GLOBAL: { + FLAT_instruction *flat = static_cast(instr); + uint32_t encoding = (0b110111 << 26); + encoding |= opcode << 18; + encoding |= flat->offset & 0x1fff; + if (instr->format == Format::SCRATCH) + encoding |= 1 << 14; + else if (instr->format == Format::GLOBAL) + encoding |= 2 << 14; + encoding |= flat->lds ? 1 << 13 : 0; + encoding |= flat->glc ? 1 << 13 : 0; + encoding |= flat->slc ? 1 << 13 : 0; + out.push_back(encoding); + encoding = (0xFF & instr->operands[0].physReg().reg); + if (!instr->definitions.empty()) + encoding |= (0xFF & instr->definitions[0].physReg().reg) << 24; + else + encoding |= (0xFF & instr->operands[2].physReg().reg) << 8; + if (!instr->operands[1].isUndefined()) { + assert(instr->operands[1].physReg() != 0x7f); + assert(instr->format != Format::FLAT); + encoding |= instr->operands[1].physReg() << 16; + } else if (instr->format != Format::FLAT) { + encoding |= 0x7F << 16; + } + encoding |= flat->nv ? 1 << 23 : 0; + out.push_back(encoding); + break; + } + case Format::EXP: { + Export_instruction* exp = static_cast(instr); + uint32_t encoding = (0b110001 << 26); + encoding |= exp->valid_mask ? 0b1 << 12 : 0; + encoding |= exp->done ? 0b1 << 11 : 0; + encoding |= exp->compressed ? 0b1 << 10 : 0; + encoding |= exp->dest << 4; + encoding |= exp->enabled_mask; + out.push_back(encoding); + encoding = 0xFF & exp->operands[0].physReg().reg; + encoding |= (0xFF & exp->operands[1].physReg().reg) << 8; + encoding |= (0xFF & exp->operands[2].physReg().reg) << 16; + encoding |= (0xFF & exp->operands[3].physReg().reg) << 24; + out.push_back(encoding); + break; + } + case Format::PSEUDO: + case Format::PSEUDO_BARRIER: + unreachable("Pseudo instructions should be lowered before assembly."); + default: + if ((uint16_t) instr->format & (uint16_t) Format::VOP3A) { + VOP3A_instruction* vop3 = static_cast(instr); + + if ((uint16_t) instr->format & (uint16_t) Format::VOP2) + opcode = opcode + 0x100; + else if ((uint16_t) instr->format & (uint16_t) Format::VOP1) + opcode = opcode + 0x140; + else if ((uint16_t) instr->format & (uint16_t) Format::VOPC) + opcode = opcode + 0x0; + else if ((uint16_t) instr->format & (uint16_t) Format::VINTRP) + opcode = opcode + 0x270; + + // TODO: op_sel + uint32_t encoding = (0b110100 << 26); + encoding |= opcode << 16; + encoding |= (vop3->clamp ? 1 : 0) << 15; + for (unsigned i = 0; i < 3; i++) + encoding |= vop3->abs[i] << (8+i); + if (instr->definitions.size() == 2) + encoding |= instr->definitions[1].physReg() << 8; + encoding |= (0xFF & instr->definitions[0].physReg().reg); + out.push_back(encoding); + encoding = 0; + if (instr->opcode == aco_opcode::v_interp_mov_f32) { + encoding = 0x3 & instr->operands[0].constantValue(); + } else { + for (unsigned i = 0; i < instr->operands.size(); i++) + encoding |= instr->operands[i].physReg() << (i * 9); + } + encoding |= vop3->omod << 27; + for (unsigned i = 0; i < 3; i++) + encoding |= vop3->neg[i] << (29+i); + out.push_back(encoding); + return; + + } else if (instr->isDPP()){ + /* first emit the instruction without the DPP operand */ + Operand dpp_op = instr->operands[0]; + instr->operands[0] = Operand(PhysReg{250}, v1); + instr->format = (Format) ((uint32_t) instr->format & ~(1 << 14)); + emit_instruction(ctx, out, instr); + DPP_instruction* dpp = static_cast(instr); + uint32_t encoding = (0xF & dpp->row_mask) << 28; + encoding |= (0xF & dpp->bank_mask) << 24; + encoding |= dpp->abs[1] << 23; + encoding |= dpp->neg[1] << 22; + encoding |= dpp->abs[0] << 21; + encoding |= dpp->neg[0] << 20; + encoding |= dpp->bound_ctrl << 19; + encoding |= dpp->dpp_ctrl << 8; + encoding |= (0xFF) & dpp_op.physReg().reg; + out.push_back(encoding); + return; + } else { + unreachable("unimplemented instruction format"); + } + } + + /* append literal dword */ + for (const Operand& op : instr->operands) { + if (op.isLiteral()) { + out.push_back(op.constantValue()); + break; + } + } +} + +void emit_block(asm_context& ctx, std::vector& out, Block& block) +{ + for (aco_ptr& instr : block.instructions) { +#if 0 + int start_idx = out.size(); + std::cerr << "Encoding:\t" << std::endl; + aco_print_instr(&*instr, stderr); + std::cerr << std::endl; +#endif + emit_instruction(ctx, out, instr.get()); +#if 0 + for (int i = start_idx; i < out.size(); i++) + std::cerr << "encoding: " << "0x" << std::setfill('0') << std::setw(8) << std::hex << out[i] << std::endl; +#endif + } +} + +void fix_exports(asm_context& ctx, std::vector& out, Program* program) +{ + for (int idx = program->blocks.size() - 1; idx >= 0; idx--) { + Block& block = program->blocks[idx]; + std::vector>::reverse_iterator it = block.instructions.rbegin(); + bool endBlock = false; + bool exported = false; + while ( it != block.instructions.rend()) + { + if ((*it)->format == Format::EXP && endBlock) { + Export_instruction* exp = static_cast((*it).get()); + if (program->stage & hw_vs) { + if (exp->dest >= V_008DFC_SQ_EXP_POS && exp->dest <= (V_008DFC_SQ_EXP_POS + 3)) { + exp->done = true; + exported = true; + break; + } + } else { + exp->done = true; + exp->valid_mask = true; + exported = true; + break; + } + } else if ((*it)->definitions.size() && (*it)->definitions[0].physReg() == exec) + break; + else if ((*it)->opcode == aco_opcode::s_endpgm) { + if (endBlock) + break; + endBlock = true; + } + ++it; + } + if (!endBlock || exported) + continue; + /* we didn't find an Export instruction and have to insert a null export */ + aco_ptr exp{create_instruction(aco_opcode::exp, Format::EXP, 4, 0)}; + for (unsigned i = 0; i < 4; i++) + exp->operands[i] = Operand(v1); + exp->enabled_mask = 0; + exp->compressed = false; + exp->done = true; + exp->valid_mask = program->stage & hw_fs; + if (program->stage & hw_fs) + exp->dest = 9; /* NULL */ + else + exp->dest = V_008DFC_SQ_EXP_POS; + /* insert the null export 1 instruction before endpgm */ + block.instructions.insert(block.instructions.end() - 1, std::move(exp)); + } +} + +void fix_branches(asm_context& ctx, std::vector& out) +{ + for (std::pair branch : ctx.branches) + { + int offset = (int)ctx.program->blocks[branch.second->block].offset - branch.first - 1; + out[branch.first] |= (uint16_t) offset; + } +} + +void fix_constaddrs(asm_context& ctx, std::vector& out) +{ + for (unsigned addr : ctx.constaddrs) + out[addr] += out.size() * 4u; +} + +unsigned emit_program(Program* program, + std::vector& code) +{ + asm_context ctx(program); + + if (program->stage & (hw_vs | hw_fs)) + fix_exports(ctx, code, program); + + for (Block& block : program->blocks) { + block.offset = code.size(); + emit_block(ctx, code, block); + } + + fix_branches(ctx, code); + fix_constaddrs(ctx, code); + + unsigned constant_data_offset = code.size() * sizeof(uint32_t); + while (program->constant_data.size() % 4u) + program->constant_data.push_back(0); + /* Copy constant data */ + code.insert(code.end(), (uint32_t*)program->constant_data.data(), + (uint32_t*)(program->constant_data.data() + program->constant_data.size())); + + return constant_data_offset; +} + +} diff --git a/src/amd/compiler/aco_builder_h.py b/src/amd/compiler/aco_builder_h.py new file mode 100644 index 00000000000..7be3a664c4a --- /dev/null +++ b/src/amd/compiler/aco_builder_h.py @@ -0,0 +1,400 @@ + +template = """\ +/* + * Copyright (c) 2019 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * This file was generated by aco_builder_h.py + */ + +#ifndef _ACO_BUILDER_ +#define _ACO_BUILDER_ + +#include "aco_ir.h" +#include "util/u_math.h" +#include "util/bitscan.h" + +namespace aco { +enum dpp_ctrl { + _dpp_quad_perm = 0x000, + _dpp_row_sl = 0x100, + _dpp_row_sr = 0x110, + _dpp_row_rr = 0x120, + dpp_wf_sl1 = 0x130, + dpp_wf_rl1 = 0x134, + dpp_wf_sr1 = 0x138, + dpp_wf_rr1 = 0x13C, + dpp_row_mirror = 0x140, + dpp_row_half_mirror = 0x141, + dpp_row_bcast15 = 0x142, + dpp_row_bcast31 = 0x143 +}; + +inline dpp_ctrl +dpp_quad_perm(unsigned lane0, unsigned lane1, unsigned lane2, unsigned lane3) +{ + assert(lane0 < 4 && lane1 < 4 && lane2 < 4 && lane3 < 4); + return (dpp_ctrl)(lane0 | (lane1 << 2) | (lane2 << 4) | (lane3 << 6)); +} + +inline dpp_ctrl +dpp_row_sl(unsigned amount) +{ + assert(amount > 0 && amount < 16); + return (dpp_ctrl)(((unsigned) _dpp_row_sl) | amount); +} + +inline dpp_ctrl +dpp_row_sr(unsigned amount) +{ + assert(amount > 0 && amount < 16); + return (dpp_ctrl)(((unsigned) _dpp_row_sr) | amount); +} + +inline unsigned +ds_pattern_bitmode(unsigned and_mask, unsigned or_mask, unsigned xor_mask) +{ + assert(and_mask < 32 && or_mask < 32 && xor_mask < 32); + return and_mask | (or_mask << 5) | (xor_mask << 10); +} + +aco_ptr create_s_mov(Definition dst, Operand src); + +class Builder { +public: + struct Result { + Instruction *instr; + + Result(Instruction *instr) : instr(instr) {} + + operator Instruction *() const { + return instr; + } + + operator Temp() const { + return instr->definitions[0].getTemp(); + } + + operator Operand() const { + return Operand((Temp)*this); + } + + Definition& def(unsigned index) const { + return instr->definitions[index]; + } + + aco_ptr get_ptr() const { + return aco_ptr(instr); + } + }; + + struct Op { + Operand op; + Op(Temp tmp) : op(tmp) {} + Op(Operand op_) : op(op_) {} + Op(Result res) : op((Temp)res) {} + }; + + Program *program; + bool use_iterator; + union { + bool forwards; //when use_iterator == true + bool start; //when use_iterator == false + }; + std::vector> *instructions; + std::vector>::iterator it; + + Builder(Program *pgm) : program(pgm), use_iterator(false), start(false), instructions(NULL) {} + Builder(Program *pgm, Block *block) : program(pgm), use_iterator(false), start(false), instructions(&block->instructions) {} + Builder(Program *pgm, std::vector> *instrs) : program(pgm), use_iterator(false), start(false), instructions(instrs) {} + + void moveEnd(Block *block) { + instructions = &block->instructions; + } + + void reset() { + use_iterator = false; + start = false; + instructions = NULL; + } + + void reset(Block *block) { + use_iterator = false; + start = false; + instructions = &block->instructions; + } + + void reset(std::vector> *instrs) { + use_iterator = false; + start = false; + instructions = instrs; + } + + Result insert(aco_ptr instr) { + Instruction *instr_ptr = instr.get(); + if (instructions) { + if (use_iterator) { + it = instructions->emplace(it, std::move(instr)); + if (forwards) + it = std::next(it); + } else if (!start) { + instructions->emplace_back(std::move(instr)); + } else { + instructions->emplace(instructions->begin(), std::move(instr)); + } + } + return Result(instr_ptr); + } + + Result insert(Instruction* instr) { + if (instructions) { + if (use_iterator) { + it = instructions->emplace(it, aco_ptr(instr)); + if (forwards) + it = std::next(it); + } else if (!start) { + instructions->emplace_back(aco_ptr(instr)); + } else { + instructions->emplace(instructions->begin(), aco_ptr(instr)); + } + } + return Result(instr); + } + + Temp tmp(RegClass rc) { + return (Temp){program->allocateId(), rc}; + } + + Temp tmp(RegType type, unsigned size) { + return (Temp){program->allocateId(), RegClass(type, size)}; + } + + Definition def(RegClass rc) { + return Definition((Temp){program->allocateId(), rc}); + } + + Definition def(RegType type, unsigned size) { + return Definition((Temp){program->allocateId(), RegClass(type, size)}); + } + + Definition def(RegClass rc, PhysReg reg) { + return Definition(program->allocateId(), reg, rc); + } + +% for fixed in ['m0', 'vcc', 'exec', 'scc']: + Operand ${fixed}(Temp tmp) { + Operand op(tmp); + op.setFixed(aco::${fixed}); + return op; + } + + Definition ${fixed}(Definition def) { + def.setFixed(aco::${fixed}); + return def; + } + + Definition hint_${fixed}(Definition def) { + def.setHint(aco::${fixed}); + return def; + } + +% endfor + /* hand-written helpers */ + Temp as_uniform(Op op) + { + assert(op.op.isTemp()); + if (op.op.getTemp().type() == RegType::vgpr) + return pseudo(aco_opcode::p_as_uniform, def(RegType::sgpr, op.op.size()), op); + else + return op.op.getTemp(); + } + + Result v_mul_imm(Definition dst, Temp tmp, uint32_t imm, bool bits24=false) + { + assert(tmp.type() == RegType::vgpr); + if (imm == 0) { + return vop1(aco_opcode::v_mov_b32, dst, Operand(0u)); + } else if (imm == 1) { + return copy(dst, Operand(tmp)); + } else if (util_is_power_of_two_or_zero(imm)) { + return vop2(aco_opcode::v_lshlrev_b32, dst, Operand((uint32_t)ffs(imm) - 1u), tmp); + } else if (bits24) { + return vop2(aco_opcode::v_mul_u32_u24, dst, Operand(imm), tmp); + } else { + Temp imm_tmp = copy(def(v1), Operand(imm)); + return vop3(aco_opcode::v_mul_lo_u32, dst, imm_tmp, tmp); + } + } + + Result v_mul24_imm(Definition dst, Temp tmp, uint32_t imm) + { + return v_mul_imm(dst, tmp, imm, true); + } + + Result copy(Definition dst, Op op_) { + Operand op = op_.op; + if (dst.regClass() == s1 && op.size() == 1 && op.isLiteral()) { + uint32_t imm = op.constantValue(); + if (imm >= 0xffff8000 || imm <= 0x7fff) { + return sopk(aco_opcode::s_movk_i32, dst, imm & 0xFFFFu); + } else if (util_bitreverse(imm) <= 64 || util_bitreverse(imm) >= 0xFFFFFFF0) { + uint32_t rev = util_bitreverse(imm); + return dst.regClass() == v1 ? + vop1(aco_opcode::v_bfrev_b32, dst, Operand(rev)) : + sop1(aco_opcode::s_brev_b32, dst, Operand(rev)); + } else if (imm != 0) { + unsigned start = (ffs(imm) - 1) & 0x1f; + unsigned size = util_bitcount(imm) & 0x1f; + if ((((1u << size) - 1u) << start) == imm) + return sop2(aco_opcode::s_bfm_b32, dst, Operand(size), Operand(start)); + } + } + + if (dst.regClass() == s2) { + return sop1(aco_opcode::s_mov_b64, dst, op); + } else if (op.size() > 1) { + return pseudo(aco_opcode::p_create_vector, dst, op); + } else if (dst.regClass() == v1 || dst.regClass() == v1.as_linear()) { + return vop1(aco_opcode::v_mov_b32, dst, op); + } else { + assert(dst.regClass() == s1); + return sop1(aco_opcode::s_mov_b32, dst, op); + } + } + + Result vadd32(Definition dst, Op a, Op b, bool carry_out=false, Op carry_in=Op(Operand(s2))) { + if (!b.op.isTemp() || b.op.regClass().type() != RegType::vgpr) + std::swap(a, b); + assert(b.op.isTemp() && b.op.regClass().type() == RegType::vgpr); + + if (!carry_in.op.isUndefined()) + return vop2(aco_opcode::v_addc_co_u32, Definition(dst), hint_vcc(def(s2)), a, b, carry_in); + else if (program->chip_class < GFX9 || carry_out) + return vop2(aco_opcode::v_add_co_u32, Definition(dst), hint_vcc(def(s2)), a, b); + else + return vop2(aco_opcode::v_add_u32, Definition(dst), a, b); + } + + Result vsub32(Definition dst, Op a, Op b, bool carry_out=false, Op borrow=Op(Operand(s2))) + { + if (!borrow.op.isUndefined() || program->chip_class < GFX9) + carry_out = true; + + bool reverse = !b.op.isTemp() || b.op.regClass().type() != RegType::vgpr; + if (reverse) + std::swap(a, b); + assert(b.op.isTemp() && b.op.regClass().type() == RegType::vgpr); + + aco_opcode op; + Temp carry; + if (carry_out) { + carry = tmp(s2); + if (borrow.op.isUndefined()) + op = reverse ? aco_opcode::v_subrev_co_u32 : aco_opcode::v_sub_co_u32; + else + op = reverse ? aco_opcode::v_subbrev_co_u32 : aco_opcode::v_subb_co_u32; + } else { + op = reverse ? aco_opcode::v_subrev_u32 : aco_opcode::v_sub_u32; + } + + int num_ops = borrow.op.isUndefined() ? 2 : 3; + int num_defs = carry_out ? 2 : 1; + aco_ptr sub{create_instruction(op, Format::VOP2, num_ops, num_defs)}; + sub->operands[0] = a.op; + sub->operands[1] = b.op; + if (!borrow.op.isUndefined()) + sub->operands[2] = borrow.op; + sub->definitions[0] = dst; + if (carry_out) { + sub->definitions[1] = Definition(carry); + sub->definitions[1].setHint(aco::vcc); + } + return insert(std::move(sub)); + } +<% +import itertools +formats = [("pseudo", [Format.PSEUDO], 'Pseudo_instruction', list(itertools.product(range(5), range(5))) + [(8, 1), (1, 8)]), + ("sop1", [Format.SOP1], 'SOP1_instruction', [(1, 1), (2, 1), (3, 2)]), + ("sop2", [Format.SOP2], 'SOP2_instruction', itertools.product([1, 2], [2, 3])), + ("sopk", [Format.SOPK], 'SOPK_instruction', itertools.product([0, 1, 2], [0, 1])), + ("sopp", [Format.SOPP], 'SOPP_instruction', [(0, 0), (0, 1)]), + ("sopc", [Format.SOPC], 'SOPC_instruction', [(1, 2)]), + ("smem", [Format.SMEM], 'SMEM_instruction', [(0, 4), (0, 3), (1, 0), (1, 3), (1, 2), (0, 0)]), + ("ds", [Format.DS], 'DS_instruction', [(1, 1), (1, 2), (0, 3), (0, 4)]), + ("mubuf", [Format.MUBUF], 'MUBUF_instruction', [(0, 4), (1, 3)]), + ("mimg", [Format.MIMG], 'MIMG_instruction', [(0, 4), (1, 3), (0, 3), (1, 2)]), #TODO(pendingchaos): less shapes? + ("exp", [Format.EXP], 'Export_instruction', [(0, 4)]), + ("branch", [Format.PSEUDO_BRANCH], 'Pseudo_branch_instruction', itertools.product([0], [0, 1])), + ("barrier", [Format.PSEUDO_BARRIER], 'Pseudo_barrier_instruction', [(0, 0)]), + ("reduction", [Format.PSEUDO_REDUCTION], 'Pseudo_reduction_instruction', [(3, 2)]), + ("vop1", [Format.VOP1], 'VOP1_instruction', [(1, 1), (2, 2)]), + ("vop2", [Format.VOP2], 'VOP2_instruction', itertools.product([1, 2], [2, 3])), + ("vopc", [Format.VOPC], 'VOPC_instruction', itertools.product([1, 2], [2])), + ("vop3", [Format.VOP3A], 'VOP3A_instruction', [(1, 3), (1, 2), (1, 1), (2, 2)]), + ("vintrp", [Format.VINTRP], 'Interp_instruction', [(1, 2), (1, 3)]), + ("vop1_dpp", [Format.VOP1, Format.DPP], 'DPP_instruction', [(1, 1)]), + ("vop2_dpp", [Format.VOP2, Format.DPP], 'DPP_instruction', itertools.product([1, 2], [2, 3])), + ("vopc_dpp", [Format.VOPC, Format.DPP], 'DPP_instruction', itertools.product([1, 2], [2])), + ("vop1_e64", [Format.VOP1, Format.VOP3A], 'VOP3A_instruction', itertools.product([1], [1])), + ("vop2_e64", [Format.VOP2, Format.VOP3A], 'VOP3A_instruction', itertools.product([1, 2], [2, 3])), + ("vopc_e64", [Format.VOPC, Format.VOP3A], 'VOP3A_instruction', itertools.product([1, 2], [2])), + ("flat", [Format.FLAT], 'FLAT_instruction', [(0, 3), (1, 2)]), + ("global", [Format.GLOBAL], 'FLAT_instruction', [(0, 3), (1, 2)])] +%>\\ +% for name, formats, struct, shapes in formats: + % for num_definitions, num_operands in shapes: + <% + args = ['aco_opcode opcode'] + for i in range(num_definitions): + args.append('Definition def%d' % i) + for i in range(num_operands): + args.append('Op op%d' % i) + for f in formats: + args += f.get_builder_field_decls() + %>\\ + + Result ${name}(${', '.join(args)}) + { + ${struct} *instr = create_instruction<${struct}>(opcode, (Format)(${'|'.join('(int)Format::%s' % f.name for f in formats)}), ${num_operands}, ${num_definitions}); + % for i in range(num_definitions): + instr->definitions[${i}] = def${i}; + % endfor + % for i in range(num_operands): + instr->operands[${i}] = op${i}.op; + % endfor + % for f in formats: + % for dest, field_name in zip(f.get_builder_field_dests(), f.get_builder_field_names()): + instr->${dest} = ${field_name}; + % endfor + % endfor + return insert(instr); + } + % endfor +% endfor +}; + +} +#endif /* _ACO_BUILDER_ */""" + +from aco_opcodes import opcodes, Format +from mako.template import Template + +print(Template(template).render(opcodes=opcodes, Format=Format)) diff --git a/src/amd/compiler/aco_dead_code_analysis.cpp b/src/amd/compiler/aco_dead_code_analysis.cpp new file mode 100644 index 00000000000..f56718f0479 --- /dev/null +++ b/src/amd/compiler/aco_dead_code_analysis.cpp @@ -0,0 +1,102 @@ +/* + * Copyright © 2019 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" + +#include + +/* + * Implements an analysis pass to determine the number of uses + * for each SSA-definition. + */ + +namespace aco { +namespace { + +struct dce_ctx { + int current_block; + std::vector uses; + std::vector> live; + + dce_ctx(Program* program) : current_block(program->blocks.size() - 1), uses(program->peekAllocationId()) + { + live.reserve(program->blocks.size()); + for (Block& block : program->blocks) + live.emplace_back(block.instructions.size()); + } +}; + +void process_block(dce_ctx& ctx, Block& block) +{ + std::vector& live = ctx.live[block.index]; + assert(live.size() == block.instructions.size()); + bool process_predecessors = false; + for (int idx = block.instructions.size() - 1; idx >= 0; idx--) { + if (live[idx]) + continue; + + aco_ptr& instr = block.instructions[idx]; + const bool is_live = instr->definitions.empty() || + std::any_of(instr->definitions.begin(), instr->definitions.end(), + [&ctx] (const Definition& def) { return !def.isTemp() || ctx.uses[def.tempId()];}); + + if (is_live) { + for (const Operand& op : instr->operands) { + if (op.isTemp()) { + if (ctx.uses[op.tempId()] == 0) + process_predecessors = true; + ctx.uses[op.tempId()]++; + } + } + live[idx] = true; + } + } + + if (process_predecessors) { + for (unsigned pred_idx : block.linear_preds) + ctx.current_block = std::max(ctx.current_block, (int) pred_idx); + } +} + +} /* end namespace */ + +std::vector dead_code_analysis(Program *program) { + + dce_ctx ctx(program); + + while (ctx.current_block >= 0) { + unsigned next_block = ctx.current_block--; + process_block(ctx, program->blocks[next_block]); + } + + /* add one use to exec to prevent startpgm from being removed */ + aco_ptr& startpgm = program->blocks[0].instructions[0]; + assert(startpgm->opcode == aco_opcode::p_startpgm); + ctx.uses[startpgm->definitions.back().tempId()]++; + + return ctx.uses; +} + +} + diff --git a/src/amd/compiler/aco_dominance.cpp b/src/amd/compiler/aco_dominance.cpp new file mode 100644 index 00000000000..de5549eec1a --- /dev/null +++ b/src/amd/compiler/aco_dominance.cpp @@ -0,0 +1,93 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Daniel Schürmann (daniel.schuermann@campus.tu-berlin.de) + * + */ + +#ifndef ACO_DOMINANCE_CPP +#define ACO_DOMINANCE_CPP + +#include "aco_ir.h" + +/* + * Implements the algorithms for computing the dominator tree from + * "A Simple, Fast Dominance Algorithm" by Cooper, Harvey, and Kennedy. + * + * Different from the paper, our CFG allows to compute the dominator tree + * in a single pass as it is guaranteed that the dominating predecessors + * are processed before the current block. + */ + +namespace aco { + +void dominator_tree(Program* program) +{ + program->blocks[0].logical_idom = 0; + program->blocks[0].linear_idom = 0; + + for (unsigned i = 1; i < program->blocks.size(); i++) { + Block& block = program->blocks[i]; + int new_logical_idom = -1; + int new_linear_idom = -1; + for (unsigned pred_idx : block.logical_preds) { + if ((int) program->blocks[pred_idx].logical_idom == -1) + continue; + + if (new_logical_idom == -1) { + new_logical_idom = pred_idx; + continue; + } + + while ((int) pred_idx != new_logical_idom) { + if ((int) pred_idx > new_logical_idom) + pred_idx = program->blocks[pred_idx].logical_idom; + if ((int) pred_idx < new_logical_idom) + new_logical_idom = program->blocks[new_logical_idom].logical_idom; + } + } + + for (unsigned pred_idx : block.linear_preds) { + if ((int) program->blocks[pred_idx].linear_idom == -1) + continue; + + if (new_linear_idom == -1) { + new_linear_idom = pred_idx; + continue; + } + + while ((int) pred_idx != new_linear_idom) { + if ((int) pred_idx > new_linear_idom) + pred_idx = program->blocks[pred_idx].linear_idom; + if ((int) pred_idx < new_linear_idom) + new_linear_idom = program->blocks[new_linear_idom].linear_idom; + } + } + + block.logical_idom = new_logical_idom; + block.linear_idom = new_linear_idom; + } +} + +} +#endif diff --git a/src/amd/compiler/aco_insert_NOPs.cpp b/src/amd/compiler/aco_insert_NOPs.cpp new file mode 100644 index 00000000000..fea1364072e --- /dev/null +++ b/src/amd/compiler/aco_insert_NOPs.cpp @@ -0,0 +1,282 @@ +/* + * Copyright © 2019 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" + +namespace aco { +namespace { + +struct NOP_ctx { + /* just initialize these with something less than max NOPs */ + int VALU_wrexec = -10; + int VALU_wrvcc = -10; + int VALU_wrsgpr = -10; + enum chip_class chip_class; + unsigned vcc_physical; + NOP_ctx(Program* program) : chip_class(program->chip_class) { + vcc_physical = program->config->num_sgprs - 2; + } +}; + +bool VALU_writes_sgpr(aco_ptr& instr) +{ + if ((uint32_t) instr->format & (uint32_t) Format::VOPC) + return true; + if (instr->isVOP3() && instr->definitions.size() == 2) + return true; + if (instr->opcode == aco_opcode::v_readfirstlane_b32 || instr->opcode == aco_opcode::v_readlane_b32) + return true; + return false; +} + +bool regs_intersect(PhysReg a_reg, unsigned a_size, PhysReg b_reg, unsigned b_size) +{ + return a_reg > b_reg ? + (a_reg - b_reg < b_size) : + (b_reg - a_reg < a_size); +} + +int handle_instruction(NOP_ctx& ctx, aco_ptr& instr, + std::vector>& old_instructions, + std::vector>& new_instructions) +{ + int new_idx = new_instructions.size(); + + // TODO: setreg / getreg / m0 writes + // TODO: try to schedule the NOP-causing instruction up to reduce the number of stall cycles + + /* break off from prevous SMEM clause if needed */ + if (instr->format == Format::SMEM && ctx.chip_class >= GFX8) { + const bool is_store = instr->definitions.empty(); + for (int pred_idx = new_idx - 1; pred_idx >= 0; pred_idx--) { + aco_ptr& pred = new_instructions[pred_idx]; + if (pred->format != Format::SMEM) + break; + + /* Don't allow clauses with store instructions since the clause's + * instructions may use the same address. */ + if (is_store || pred->definitions.empty()) + return 1; + + Definition& instr_def = instr->definitions[0]; + Definition& pred_def = pred->definitions[0]; + + /* ISA reference doesn't say anything about this, but best to be safe */ + if (regs_intersect(instr_def.physReg(), instr_def.size(), pred_def.physReg(), pred_def.size())) + return 1; + + for (const Operand& op : pred->operands) { + if (op.isConstant() || !op.isFixed()) + continue; + if (regs_intersect(instr_def.physReg(), instr_def.size(), op.physReg(), op.size())) + return 1; + } + for (const Operand& op : instr->operands) { + if (op.isConstant() || !op.isFixed()) + continue; + if (regs_intersect(pred_def.physReg(), pred_def.size(), op.physReg(), op.size())) + return 1; + } + } + } else if (instr->isVALU() || instr->format == Format::VINTRP) { + int NOPs = 0; + + if (instr->isDPP()) { + /* VALU does not forward EXEC to DPP. */ + if (ctx.VALU_wrexec + 5 >= new_idx) + NOPs = 5 + ctx.VALU_wrexec - new_idx + 1; + + /* VALU DPP reads VGPR written by VALU */ + for (int pred_idx = new_idx - 1; pred_idx >= 0 && pred_idx >= new_idx - 2; pred_idx--) { + aco_ptr& pred = new_instructions[pred_idx]; + if ((pred->isVALU() || pred->format == Format::VINTRP) && + !pred->definitions.empty() && + pred->definitions[0].physReg() == instr->operands[0].physReg()) { + NOPs = std::max(NOPs, 2 + pred_idx - new_idx + 1); + break; + } + } + } + + /* SALU writes M0 */ + if (instr->format == Format::VINTRP && new_idx > 0 && ctx.chip_class >= GFX9) { + aco_ptr& pred = new_instructions.back(); + if (pred->isSALU() && + !pred->definitions.empty() && + pred->definitions[0].physReg() == m0) + NOPs = std::max(NOPs, 1); + } + + for (const Operand& op : instr->operands) { + /* VALU which uses VCCZ */ + if (op.physReg() == PhysReg{251} && + ctx.VALU_wrvcc + 5 >= new_idx) + NOPs = std::max(NOPs, 5 + ctx.VALU_wrvcc - new_idx + 1); + + /* VALU which uses EXECZ */ + if (op.physReg() == PhysReg{252} && + ctx.VALU_wrexec + 5 >= new_idx) + NOPs = std::max(NOPs, 5 + ctx.VALU_wrexec - new_idx + 1); + + /* VALU which reads VCC as a constant */ + if (ctx.VALU_wrvcc + 1 >= new_idx) { + for (unsigned k = 0; k < op.size(); k++) { + unsigned reg = op.physReg() + k; + if (reg == ctx.vcc_physical || reg == ctx.vcc_physical + 1) + NOPs = std::max(NOPs, 1); + } + } + } + + switch (instr->opcode) { + case aco_opcode::v_readlane_b32: + case aco_opcode::v_writelane_b32: { + if (ctx.VALU_wrsgpr + 4 < new_idx) + break; + PhysReg reg = instr->operands[1].physReg(); + for (int pred_idx = new_idx - 1; pred_idx >= 0 && pred_idx >= new_idx - 4; pred_idx--) { + aco_ptr& pred = new_instructions[pred_idx]; + if (!pred->isVALU() || !VALU_writes_sgpr(pred)) + continue; + for (const Definition& def : pred->definitions) { + if (def.physReg() == reg) + NOPs = std::max(NOPs, 4 + pred_idx - new_idx + 1); + } + } + break; + } + case aco_opcode::v_div_fmas_f32: + case aco_opcode::v_div_fmas_f64: { + if (ctx.VALU_wrvcc + 4 >= new_idx) + NOPs = std::max(NOPs, 4 + ctx.VALU_wrvcc - new_idx + 1); + break; + } + default: + break; + } + + /* Write VGPRs holding writedata > 64 bit from MIMG/MUBUF instructions */ + // FIXME: handle case if the last instruction of a block without branch is such store + // TODO: confirm that DS instructions cannot cause WAR hazards here + if (new_idx > 0) { + aco_ptr& pred = new_instructions.back(); + if (pred->isVMEM() && + pred->operands.size() == 4 && + pred->operands[3].size() > 2 && + pred->operands[1].size() != 8 && + (pred->format != Format::MUBUF || pred->operands[2].physReg() >= 102)) { + /* Ops that use a 256-bit T# do not need a wait state. + * BUFFER_STORE_* operations that use an SGPR for "offset" + * do not require any wait states. */ + PhysReg wrdata = pred->operands[3].physReg(); + unsigned size = pred->operands[3].size(); + assert(wrdata >= 256); + for (const Definition& def : instr->definitions) { + if (regs_intersect(def.physReg(), def.size(), wrdata, size)) + NOPs = std::max(NOPs, 1); + } + } + } + + if (VALU_writes_sgpr(instr)) { + for (const Definition& def : instr->definitions) { + if (def.physReg() == vcc) + ctx.VALU_wrvcc = NOPs ? new_idx : new_idx + 1; + else if (def.physReg() == exec) + ctx.VALU_wrexec = NOPs ? new_idx : new_idx + 1; + else if (def.physReg() <= 102) + ctx.VALU_wrsgpr = NOPs ? new_idx : new_idx + 1; + } + } + return NOPs; + } else if (instr->isVMEM() && ctx.VALU_wrsgpr + 5 >= new_idx) { + /* If the VALU writes the SGPR that is used by a VMEM, the user must add five wait states. */ + for (int pred_idx = new_idx - 1; pred_idx >= 0 && pred_idx >= new_idx - 5; pred_idx--) { + aco_ptr& pred = new_instructions[pred_idx]; + if (!(pred->isVALU() && VALU_writes_sgpr(pred))) + continue; + + for (const Definition& def : pred->definitions) { + if (def.physReg() > 102) + continue; + + if (instr->operands.size() > 1 && + regs_intersect(instr->operands[1].physReg(), instr->operands[1].size(), + def.physReg(), def.size())) { + return 5 + pred_idx - new_idx + 1; + } + + if (instr->operands.size() > 2 && + regs_intersect(instr->operands[2].physReg(), instr->operands[2].size(), + def.physReg(), def.size())) { + return 5 + pred_idx - new_idx + 1; + } + } + } + } + + return 0; +} + + +void handle_block(NOP_ctx& ctx, Block& block) +{ + std::vector> instructions; + instructions.reserve(block.instructions.size()); + for (unsigned i = 0; i < block.instructions.size(); i++) { + aco_ptr& instr = block.instructions[i]; + unsigned NOPs = handle_instruction(ctx, instr, block.instructions, instructions); + if (NOPs) { + // TODO: try to move the instruction down + /* create NOP */ + aco_ptr nop{create_instruction(aco_opcode::s_nop, Format::SOPP, 0, 0)}; + nop->imm = NOPs - 1; + nop->block = -1; + instructions.emplace_back(std::move(nop)); + } + + instructions.emplace_back(std::move(instr)); + } + + ctx.VALU_wrvcc -= instructions.size(); + ctx.VALU_wrexec -= instructions.size(); + ctx.VALU_wrsgpr -= instructions.size(); + block.instructions = std::move(instructions); +} + +} /* end namespace */ + + +void insert_NOPs(Program* program) +{ + NOP_ctx ctx(program); + for (Block& block : program->blocks) { + if (block.instructions.empty()) + continue; + + handle_block(ctx, block); + } +} + +} diff --git a/src/amd/compiler/aco_insert_exec_mask.cpp b/src/amd/compiler/aco_insert_exec_mask.cpp new file mode 100644 index 00000000000..7886a4c77e2 --- /dev/null +++ b/src/amd/compiler/aco_insert_exec_mask.cpp @@ -0,0 +1,1078 @@ +/* + * Copyright © 2019 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" +#include "aco_builder.h" + +namespace aco { + +namespace { + +enum WQMState : uint8_t { + Unspecified = 0, + Exact = 1 << 0, + WQM = 1 << 1, /* with control flow applied */ + Preserve_WQM = 1 << 2, + Exact_Branch = 1 << 3, +}; + +enum mask_type : uint8_t { + mask_type_global = 1 << 0, + mask_type_exact = 1 << 1, + mask_type_wqm = 1 << 2, + mask_type_loop = 1 << 3, /* active lanes of a loop */ + mask_type_initial = 1 << 4, /* initially active lanes */ +}; + +struct wqm_ctx { + Program* program; + /* state for WQM propagation */ + std::set worklist; + std::vector defined_in; + std::vector needs_wqm; + std::vector branch_wqm; /* true if the branch condition in this block should be in wqm */ + bool loop; + bool wqm; + wqm_ctx(Program* program) : program(program), + defined_in(program->peekAllocationId(), 0xFFFF), + needs_wqm(program->peekAllocationId()), + branch_wqm(program->blocks.size()), + loop(false), + wqm(false) + { + for (unsigned i = 0; i < program->blocks.size(); i++) + worklist.insert(i); + } +}; + +struct loop_info { + Block* loop_header; + uint16_t num_exec_masks; + uint8_t needs; + bool has_divergent_break; + bool has_divergent_continue; + bool has_discard; + loop_info(Block* b, uint16_t num, uint8_t needs, bool breaks, bool cont, bool discard) : + loop_header(b), num_exec_masks(num), needs(needs), has_divergent_break(breaks), + has_divergent_continue(cont), has_discard(discard) {} +}; + +struct block_info { + std::vector> exec; + std::vector instr_needs; + uint8_t block_needs; + uint8_t ever_again_needs; + /* more... */ +}; + +struct exec_ctx { + Program *program; + std::vector info; + std::vector loop; + bool handle_wqm = false; + exec_ctx(Program *program) : program(program), info(program->blocks.size()) {} +}; + +bool pred_by_exec_mask(aco_ptr& instr) { + if (instr->format == Format::SMEM || instr->isSALU()) + return false; + if (instr->format == Format::PSEUDO_BARRIER) + return false; + + if (instr->format == Format::PSEUDO) { + switch (instr->opcode) { + case aco_opcode::p_create_vector: + return instr->definitions[0].getTemp().type() == RegType::vgpr; + case aco_opcode::p_extract_vector: + case aco_opcode::p_split_vector: + return instr->operands[0].getTemp().type() == RegType::vgpr; + case aco_opcode::p_spill: + case aco_opcode::p_reload: + return false; + default: + break; + } + } + + if (instr->opcode == aco_opcode::v_readlane_b32 || + instr->opcode == aco_opcode::v_writelane_b32) + return false; + + return true; +} + +bool needs_exact(aco_ptr& instr) { + if (instr->format == Format::MUBUF) { + MUBUF_instruction *mubuf = static_cast(instr.get()); + return mubuf->disable_wqm; + } else if (instr->format == Format::MTBUF) { + MTBUF_instruction *mtbuf = static_cast(instr.get()); + return mtbuf->disable_wqm; + } else if (instr->format == Format::MIMG) { + MIMG_instruction *mimg = static_cast(instr.get()); + return mimg->disable_wqm; + } else { + return instr->format == Format::EXP || instr->opcode == aco_opcode::p_fs_buffer_store_smem; + } +} + +void set_needs_wqm(wqm_ctx &ctx, Temp tmp) +{ + if (!ctx.needs_wqm[tmp.id()]) { + ctx.needs_wqm[tmp.id()] = true; + if (ctx.defined_in[tmp.id()] != 0xFFFF) + ctx.worklist.insert(ctx.defined_in[tmp.id()]); + } +} + +void mark_block_wqm(wqm_ctx &ctx, unsigned block_idx) +{ + if (ctx.branch_wqm[block_idx]) + return; + + ctx.branch_wqm[block_idx] = true; + Block& block = ctx.program->blocks[block_idx]; + aco_ptr& branch = block.instructions.back(); + + if (branch->opcode != aco_opcode::p_branch) { + assert(!branch->operands.empty() && branch->operands[0].isTemp()); + set_needs_wqm(ctx, branch->operands[0].getTemp()); + } + + /* TODO: this sets more branch conditions to WQM than it needs to + * it should be enough to stop at the "exec mask top level" */ + if (block.kind & block_kind_top_level) + return; + + for (unsigned pred_idx : block.logical_preds) + mark_block_wqm(ctx, pred_idx); +} + +void get_block_needs(wqm_ctx &ctx, exec_ctx &exec_ctx, Block* block) +{ + block_info& info = exec_ctx.info[block->index]; + + std::vector instr_needs(block->instructions.size()); + + if (block->kind & block_kind_top_level) { + if (ctx.loop && ctx.wqm) { + /* mark all break conditions as WQM */ + unsigned block_idx = block->index + 1; + while (!(ctx.program->blocks[block_idx].kind & block_kind_top_level)) { + if (ctx.program->blocks[block_idx].kind & block_kind_break) + mark_block_wqm(ctx, block_idx); + block_idx++; + } + } else if (ctx.loop && !ctx.wqm) { + /* Ensure a branch never results in an exec mask with only helper + * invocations (which can cause a loop to repeat infinitively if it's + * break branches are done in exact). */ + unsigned block_idx = block->index; + do { + if ((ctx.program->blocks[block_idx].kind & block_kind_branch)) + exec_ctx.info[block_idx].block_needs |= Exact_Branch; + block_idx++; + } while (!(ctx.program->blocks[block_idx].kind & block_kind_top_level)); + } + + ctx.loop = false; + ctx.wqm = false; + } + + for (int i = block->instructions.size() - 1; i >= 0; --i) + { + aco_ptr& instr = block->instructions[i]; + + WQMState needs = needs_exact(instr) ? Exact : Unspecified; + bool propagate_wqm = instr->opcode == aco_opcode::p_wqm; + bool preserve_wqm = instr->opcode == aco_opcode::p_discard_if; + bool pred_by_exec = pred_by_exec_mask(instr); + for (const Definition& definition : instr->definitions) { + if (!definition.isTemp()) + continue; + const unsigned def = definition.tempId(); + ctx.defined_in[def] = block->index; + if (needs == Unspecified && ctx.needs_wqm[def]) { + needs = pred_by_exec ? WQM : Unspecified; + propagate_wqm = true; + } + } + + if (propagate_wqm) { + for (const Operand& op : instr->operands) { + if (op.isTemp()) { + set_needs_wqm(ctx, op.getTemp()); + } + } + } else if (preserve_wqm && info.block_needs & WQM) { + needs = Preserve_WQM; + } + + /* ensure the condition controlling the control flow for this phi is in WQM */ + if (needs == WQM && instr->opcode == aco_opcode::p_phi) { + for (unsigned pred_idx : block->logical_preds) + mark_block_wqm(ctx, pred_idx); + } + + instr_needs[i] = needs; + info.block_needs |= needs; + } + + info.instr_needs = instr_needs; + + /* for "if () " or "while () ", + * should be computed in WQM */ + if (info.block_needs & WQM && !(block->kind & block_kind_top_level)) { + for (unsigned pred_idx : block->logical_preds) + mark_block_wqm(ctx, pred_idx); + ctx.wqm = true; + } + if (block->kind & block_kind_loop_header) + ctx.loop = true; +} + +void calculate_wqm_needs(exec_ctx& exec_ctx) +{ + wqm_ctx ctx(exec_ctx.program); + + while (!ctx.worklist.empty()) { + unsigned block_index = *std::prev(ctx.worklist.end()); + ctx.worklist.erase(std::prev(ctx.worklist.end())); + + get_block_needs(ctx, exec_ctx, &exec_ctx.program->blocks[block_index]); + } + + uint8_t ever_again_needs = 0; + for (int i = exec_ctx.program->blocks.size() - 1; i >= 0; i--) { + exec_ctx.info[i].ever_again_needs = ever_again_needs; + Block& block = exec_ctx.program->blocks[i]; + + if (block.kind & block_kind_needs_lowering) + exec_ctx.info[i].block_needs |= Exact; + + /* if discard is used somewhere in nested CF, we need to preserve the WQM mask */ + if ((block.kind & block_kind_discard || + block.kind & block_kind_uses_discard_if) && + ever_again_needs & WQM) + exec_ctx.info[i].block_needs |= Preserve_WQM; + + ever_again_needs |= exec_ctx.info[i].block_needs & ~Exact_Branch; + if (block.kind & block_kind_discard || + block.kind & block_kind_uses_discard_if) + ever_again_needs |= Exact; + + /* don't propagate WQM preservation further than the next top_level block */ + if (block.kind & block_kind_top_level) + ever_again_needs &= ~Preserve_WQM; + else + exec_ctx.info[i].block_needs &= ~Preserve_WQM; + } + exec_ctx.handle_wqm = true; +} + +void transition_to_WQM(exec_ctx& ctx, Builder bld, unsigned idx) +{ + if (ctx.info[idx].exec.back().second & mask_type_wqm) + return; + if (ctx.info[idx].exec.back().second & mask_type_global) { + Temp exec_mask = ctx.info[idx].exec.back().first; + exec_mask = bld.sop1(aco_opcode::s_wqm_b64, bld.def(s2, exec), bld.def(s1, scc), exec_mask); + ctx.info[idx].exec.emplace_back(exec_mask, mask_type_global | mask_type_wqm); + return; + } + /* otherwise, the WQM mask should be one below the current mask */ + ctx.info[idx].exec.pop_back(); + assert(ctx.info[idx].exec.back().second & mask_type_wqm); + ctx.info[idx].exec.back().first = bld.pseudo(aco_opcode::p_parallelcopy, bld.def(s2, exec), + ctx.info[idx].exec.back().first); +} + +void transition_to_Exact(exec_ctx& ctx, Builder bld, unsigned idx) +{ + if (ctx.info[idx].exec.back().second & mask_type_exact) + return; + if (ctx.info[idx].exec.back().second & mask_type_global) { + ctx.info[idx].exec.pop_back(); + assert(ctx.info[idx].exec.back().second & mask_type_exact); + ctx.info[idx].exec.back().first = bld.pseudo(aco_opcode::p_parallelcopy, bld.def(s2, exec), + ctx.info[idx].exec.back().first); + return; + } + /* otherwise, we create an exact mask and push to the stack */ + Temp wqm = ctx.info[idx].exec.back().first; + Temp exact = bld.tmp(s2); + wqm = bld.sop1(aco_opcode::s_and_saveexec_b64, bld.def(s2), bld.def(s1, scc), + bld.exec(Definition(exact)), ctx.info[idx].exec[0].first, bld.exec(wqm)); + ctx.info[idx].exec.back().first = wqm; + ctx.info[idx].exec.emplace_back(exact, mask_type_exact); +} + +unsigned add_coupling_code(exec_ctx& ctx, Block* block, + std::vector>& instructions) +{ + unsigned idx = block->index; + Builder bld(ctx.program, &instructions); + std::vector& preds = block->linear_preds; + + /* start block */ + if (idx == 0) { + aco_ptr& startpgm = block->instructions[0]; + assert(startpgm->opcode == aco_opcode::p_startpgm); + Temp exec_mask = startpgm->definitions.back().getTemp(); + bld.insert(std::move(startpgm)); + + if (ctx.handle_wqm) { + ctx.info[0].exec.emplace_back(exec_mask, mask_type_global | mask_type_exact | mask_type_initial); + /* if this block only needs WQM, initialize already */ + if (ctx.info[0].block_needs == WQM) + transition_to_WQM(ctx, bld, 0); + } else { + uint8_t mask = mask_type_global; + if (ctx.program->needs_wqm) { + exec_mask = bld.sop1(aco_opcode::s_wqm_b64, bld.def(s2, exec), bld.def(s1, scc), bld.exec(exec_mask)); + mask |= mask_type_wqm; + } else { + mask |= mask_type_exact; + } + ctx.info[0].exec.emplace_back(exec_mask, mask); + } + + return 1; + } + + /* loop entry block */ + if (block->kind & block_kind_loop_header) { + assert(preds[0] == idx - 1); + ctx.info[idx].exec = ctx.info[idx - 1].exec; + loop_info& info = ctx.loop.back(); + while (ctx.info[idx].exec.size() > info.num_exec_masks) + ctx.info[idx].exec.pop_back(); + + /* create ssa names for outer exec masks */ + if (info.has_discard) { + aco_ptr phi; + for (int i = 0; i < info.num_exec_masks - 1; i++) { + phi.reset(create_instruction(aco_opcode::p_linear_phi, Format::PSEUDO, preds.size(), 1)); + phi->definitions[0] = bld.def(s2); + phi->operands[0] = Operand(ctx.info[preds[0]].exec[i].first); + ctx.info[idx].exec[i].first = bld.insert(std::move(phi)); + } + } + + /* create ssa name for restore mask */ + if (info.has_divergent_break) { + /* this phi might be trivial but ensures a parallelcopy on the loop header */ + aco_ptr phi{create_instruction(aco_opcode::p_linear_phi, Format::PSEUDO, preds.size(), 1)}; + phi->definitions[0] = bld.def(s2); + phi->operands[0] = Operand(ctx.info[preds[0]].exec[info.num_exec_masks - 1].first); + ctx.info[idx].exec.back().first = bld.insert(std::move(phi)); + } + + /* create ssa name for loop active mask */ + aco_ptr phi{create_instruction(aco_opcode::p_linear_phi, Format::PSEUDO, preds.size(), 1)}; + if (info.has_divergent_continue) + phi->definitions[0] = bld.def(s2); + else + phi->definitions[0] = bld.def(s2, exec); + phi->operands[0] = Operand(ctx.info[preds[0]].exec.back().first); + Temp loop_active = bld.insert(std::move(phi)); + + if (info.has_divergent_break) { + uint8_t mask_type = (ctx.info[idx].exec.back().second & (mask_type_wqm | mask_type_exact)) | mask_type_loop; + ctx.info[idx].exec.emplace_back(loop_active, mask_type); + } else { + ctx.info[idx].exec.back().first = loop_active; + ctx.info[idx].exec.back().second |= mask_type_loop; + } + + /* create a parallelcopy to move the active mask to exec */ + unsigned i = 0; + if (info.has_divergent_continue) { + while (block->instructions[i]->opcode != aco_opcode::p_logical_start) { + bld.insert(std::move(block->instructions[i])); + i++; + } + uint8_t mask_type = ctx.info[idx].exec.back().second & (mask_type_wqm | mask_type_exact); + ctx.info[idx].exec.emplace_back(bld.pseudo(aco_opcode::p_parallelcopy, bld.def(s2, exec), + ctx.info[idx].exec.back().first), mask_type); + } + + return i; + } + + /* loop exit block */ + if (block->kind & block_kind_loop_exit) { + Block* header = ctx.loop.back().loop_header; + loop_info& info = ctx.loop.back(); + + for (ASSERTED unsigned pred : preds) + assert(ctx.info[pred].exec.size() >= info.num_exec_masks); + + /* fill the loop header phis */ + std::vector& header_preds = header->linear_preds; + int k = 0; + if (info.has_discard) { + while (k < info.num_exec_masks - 1) { + aco_ptr& phi = header->instructions[k]; + assert(phi->opcode == aco_opcode::p_linear_phi); + for (unsigned i = 1; i < phi->operands.size(); i++) + phi->operands[i] = Operand(ctx.info[header_preds[i]].exec[k].first); + k++; + } + } + aco_ptr& phi = header->instructions[k++]; + assert(phi->opcode == aco_opcode::p_linear_phi); + for (unsigned i = 1; i < phi->operands.size(); i++) + phi->operands[i] = Operand(ctx.info[header_preds[i]].exec[info.num_exec_masks - 1].first); + + if (info.has_divergent_break) { + aco_ptr& phi = header->instructions[k]; + assert(phi->opcode == aco_opcode::p_linear_phi); + for (unsigned i = 1; i < phi->operands.size(); i++) + phi->operands[i] = Operand(ctx.info[header_preds[i]].exec[info.num_exec_masks].first); + } + + assert(!(block->kind & block_kind_top_level) || info.num_exec_masks <= 2); + + /* create the loop exit phis if not trivial */ + for (unsigned k = 0; k < info.num_exec_masks; k++) { + Temp same = ctx.info[preds[0]].exec[k].first; + uint8_t type = ctx.info[header_preds[0]].exec[k].second; + bool trivial = true; + + for (unsigned i = 1; i < preds.size() && trivial; i++) { + if (ctx.info[preds[i]].exec[k].first != same) + trivial = false; + } + + if (trivial) { + ctx.info[idx].exec.emplace_back(same, type); + } else { + /* create phi for loop footer */ + aco_ptr phi{create_instruction(aco_opcode::p_linear_phi, Format::PSEUDO, preds.size(), 1)}; + phi->definitions[0] = bld.def(s2); + for (unsigned i = 0; i < phi->operands.size(); i++) + phi->operands[i] = Operand(ctx.info[preds[i]].exec[k].first); + ctx.info[idx].exec.emplace_back(bld.insert(std::move(phi)), type); + } + } + assert(ctx.info[idx].exec.size() == info.num_exec_masks); + + /* create a parallelcopy to move the live mask to exec */ + unsigned i = 0; + while (block->instructions[i]->opcode != aco_opcode::p_logical_start) { + bld.insert(std::move(block->instructions[i])); + i++; + } + + if (ctx.handle_wqm) { + if (block->kind & block_kind_top_level && ctx.info[idx].exec.size() == 2) { + if ((ctx.info[idx].block_needs | ctx.info[idx].ever_again_needs) == 0 || + (ctx.info[idx].block_needs | ctx.info[idx].ever_again_needs) == Exact) { + ctx.info[idx].exec.back().second |= mask_type_global; + transition_to_Exact(ctx, bld, idx); + ctx.handle_wqm = false; + } + } + if (ctx.info[idx].block_needs == WQM) + transition_to_WQM(ctx, bld, idx); + else if (ctx.info[idx].block_needs == Exact) + transition_to_Exact(ctx, bld, idx); + } + + ctx.info[idx].exec.back().first = bld.pseudo(aco_opcode::p_parallelcopy, bld.def(s2, exec), + ctx.info[idx].exec.back().first); + + ctx.loop.pop_back(); + return i; + } + + if (preds.size() == 1) { + ctx.info[idx].exec = ctx.info[preds[0]].exec; + } else { + assert(preds.size() == 2); + /* if one of the predecessors ends in exact mask, we pop it from stack */ + unsigned num_exec_masks = std::min(ctx.info[preds[0]].exec.size(), + ctx.info[preds[1]].exec.size()); + if (block->kind & block_kind_top_level && !(block->kind & block_kind_merge)) + num_exec_masks = std::min(num_exec_masks, 2u); + + /* create phis for diverged exec masks */ + for (unsigned i = 0; i < num_exec_masks; i++) { + bool in_exec = i == num_exec_masks - 1 && !(block->kind & block_kind_merge); + if (!in_exec && ctx.info[preds[0]].exec[i].first == ctx.info[preds[1]].exec[i].first) { + assert(ctx.info[preds[0]].exec[i].second == ctx.info[preds[1]].exec[i].second); + ctx.info[idx].exec.emplace_back(ctx.info[preds[0]].exec[i]); + continue; + } + + Temp phi = bld.pseudo(aco_opcode::p_linear_phi, in_exec ? bld.def(s2, exec) : bld.def(s2), + ctx.info[preds[0]].exec[i].first, + ctx.info[preds[1]].exec[i].first); + uint8_t mask_type = ctx.info[preds[0]].exec[i].second & ctx.info[preds[1]].exec[i].second; + ctx.info[idx].exec.emplace_back(phi, mask_type); + } + } + + unsigned i = 0; + while (block->instructions[i]->opcode == aco_opcode::p_phi || + block->instructions[i]->opcode == aco_opcode::p_linear_phi) { + bld.insert(std::move(block->instructions[i])); + i++; + } + + if (block->kind & block_kind_merge) + ctx.info[idx].exec.pop_back(); + + if (block->kind & block_kind_top_level && ctx.info[idx].exec.size() == 3) { + assert(ctx.info[idx].exec.back().second == mask_type_exact); + assert(block->kind & block_kind_merge); + ctx.info[idx].exec.pop_back(); + } + + /* try to satisfy the block's needs */ + if (ctx.handle_wqm) { + if (block->kind & block_kind_top_level && ctx.info[idx].exec.size() == 2) { + if ((ctx.info[idx].block_needs | ctx.info[idx].ever_again_needs) == 0 || + (ctx.info[idx].block_needs | ctx.info[idx].ever_again_needs) == Exact) { + ctx.info[idx].exec.back().second |= mask_type_global; + transition_to_Exact(ctx, bld, idx); + ctx.handle_wqm = false; + } + } + if (ctx.info[idx].block_needs == WQM) + transition_to_WQM(ctx, bld, idx); + else if (ctx.info[idx].block_needs == Exact) + transition_to_Exact(ctx, bld, idx); + } + + if (block->kind & block_kind_merge) { + Temp restore = ctx.info[idx].exec.back().first; + ctx.info[idx].exec.back().first = bld.pseudo(aco_opcode::p_parallelcopy, bld.def(s2, exec), restore); + } + + return i; +} + +void lower_fs_buffer_store_smem(Builder& bld, bool need_check, aco_ptr& instr, Temp cur_exec) +{ + Operand offset = instr->operands[1]; + if (need_check) { + /* if exec is zero, then use UINT32_MAX as an offset and make this store a no-op */ + Temp nonempty = bld.sopc(aco_opcode::s_cmp_lg_u64, bld.def(s1, scc), cur_exec, Operand(0u)); + + if (offset.isLiteral()) + offset = bld.sop1(aco_opcode::s_mov_b32, bld.def(s1), offset); + + offset = bld.sop2(aco_opcode::s_cselect_b32, bld.hint_m0(bld.def(s1)), + offset, Operand(UINT32_MAX), bld.scc(nonempty)); + } else if (offset.isConstant() && offset.constantValue() > 0xFFFFF) { + offset = bld.sop1(aco_opcode::s_mov_b32, bld.hint_m0(bld.def(s1)), offset); + } + if (!offset.isConstant()) + offset.setFixed(m0); + + switch (instr->operands[2].size()) { + case 1: + instr->opcode = aco_opcode::s_buffer_store_dword; + break; + case 2: + instr->opcode = aco_opcode::s_buffer_store_dwordx2; + break; + case 4: + instr->opcode = aco_opcode::s_buffer_store_dwordx4; + break; + default: + unreachable("Invalid SMEM buffer store size"); + } + instr->operands[1] = offset; + /* as_uniform() needs to be done here so it's done in exact mode and helper + * lanes don't contribute. */ + instr->operands[2] = Operand(bld.as_uniform(instr->operands[2])); +} + +void process_instructions(exec_ctx& ctx, Block* block, + std::vector>& instructions, + unsigned idx) +{ + WQMState state; + if (ctx.info[block->index].exec.back().second & mask_type_wqm) + state = WQM; + else { + assert(!ctx.handle_wqm || ctx.info[block->index].exec.back().second & mask_type_exact); + state = Exact; + } + + /* if the block doesn't need both, WQM and Exact, we can skip processing the instructions */ + bool process = (ctx.handle_wqm && + (ctx.info[block->index].block_needs & state) != + (ctx.info[block->index].block_needs & (WQM | Exact))) || + block->kind & block_kind_uses_discard_if || + block->kind & block_kind_needs_lowering; + if (!process) { + std::vector>::iterator it = std::next(block->instructions.begin(), idx); + instructions.insert(instructions.end(), + std::move_iterator>::iterator>(it), + std::move_iterator>::iterator>(block->instructions.end())); + return; + } + + Builder bld(ctx.program, &instructions); + + for (; idx < block->instructions.size(); idx++) { + aco_ptr instr = std::move(block->instructions[idx]); + + WQMState needs = ctx.handle_wqm ? ctx.info[block->index].instr_needs[idx] : Unspecified; + + if (instr->opcode == aco_opcode::p_discard_if) { + if (ctx.info[block->index].block_needs & Preserve_WQM) { + assert(block->kind & block_kind_top_level); + transition_to_WQM(ctx, bld, block->index); + ctx.info[block->index].exec.back().second &= ~mask_type_global; + } + unsigned num = ctx.info[block->index].exec.size(); + assert(num); + Operand cond = instr->operands[0]; + instr.reset(create_instruction(aco_opcode::p_discard_if, Format::PSEUDO, num + 1, num + 1)); + for (unsigned i = 0; i < num; i++) { + instr->operands[i] = Operand(ctx.info[block->index].exec[i].first); + if (i == num - 1) + instr->operands[i].setFixed(exec); + Temp new_mask = bld.tmp(s2); + instr->definitions[i] = Definition(new_mask); + ctx.info[block->index].exec[i].first = new_mask; + } + assert((ctx.info[block->index].exec[0].second & mask_type_wqm) == 0); + instr->definitions[num - 1].setFixed(exec); + instr->operands[num] = cond; + instr->definitions[num] = bld.def(s1, scc); + + } else if (needs == WQM && state != WQM) { + transition_to_WQM(ctx, bld, block->index); + state = WQM; + } else if (needs == Exact && state != Exact) { + transition_to_Exact(ctx, bld, block->index); + state = Exact; + } + + if (instr->opcode == aco_opcode::p_is_helper || instr->opcode == aco_opcode::p_load_helper) { + Definition dst = instr->definitions[0]; + if (state == Exact) { + instr.reset(create_instruction(aco_opcode::s_mov_b64, Format::SOP1, 1, 1)); + instr->operands[0] = Operand(0u); + instr->definitions[0] = dst; + } else { + std::pair& exact_mask = ctx.info[block->index].exec[0]; + if (instr->opcode == aco_opcode::p_load_helper && + !(ctx.info[block->index].exec[0].second & mask_type_initial)) { + /* find last initial exact mask */ + for (int i = block->index; i >= 0; i--) { + if (ctx.program->blocks[i].kind & block_kind_top_level && + ctx.info[i].exec[0].second & mask_type_initial) { + exact_mask = ctx.info[i].exec[0]; + break; + } + } + } + + assert(instr->opcode == aco_opcode::p_is_helper || exact_mask.second & mask_type_initial); + assert(exact_mask.second & mask_type_exact); + + instr.reset(create_instruction(aco_opcode::s_andn2_b64, Format::SOP2, 2, 2)); + instr->operands[0] = Operand(ctx.info[block->index].exec.back().first); /* current exec */ + instr->operands[1] = Operand(exact_mask.first); + instr->definitions[0] = dst; + instr->definitions[1] = bld.def(s1, scc); + } + } else if (instr->opcode == aco_opcode::p_demote_to_helper) { + /* turn demote into discard_if with only exact masks */ + assert((ctx.info[block->index].exec[0].second & (mask_type_exact | mask_type_global)) == (mask_type_exact | mask_type_global)); + ctx.info[block->index].exec[0].second &= ~mask_type_initial; + + int num = 0; + Temp cond; + if (instr->operands.empty()) { + /* transition to exact and set exec to zero */ + Temp old_exec = ctx.info[block->index].exec.back().first; + Temp new_exec = bld.tmp(s2); + cond = bld.sop1(aco_opcode::s_and_saveexec_b64, bld.def(s2), bld.def(s1, scc), + bld.exec(Definition(new_exec)), Operand(0u), bld.exec(old_exec)); + if (ctx.info[block->index].exec.back().second & mask_type_exact) { + ctx.info[block->index].exec.back().first = new_exec; + } else { + ctx.info[block->index].exec.back().first = cond; + ctx.info[block->index].exec.emplace_back(new_exec, mask_type_exact); + } + } else { + /* demote_if: transition to exact */ + transition_to_Exact(ctx, bld, block->index); + assert(instr->operands[0].isTemp()); + cond = instr->operands[0].getTemp(); + num = 1; + } + + for (unsigned i = 0; i < ctx.info[block->index].exec.size() - 1; i++) + num += ctx.info[block->index].exec[i].second & mask_type_exact ? 1 : 0; + instr.reset(create_instruction(aco_opcode::p_discard_if, Format::PSEUDO, num + 1, num + 1)); + int k = 0; + for (unsigned i = 0; k < num; i++) { + if (ctx.info[block->index].exec[i].second & mask_type_exact) { + instr->operands[k] = Operand(ctx.info[block->index].exec[i].first); + Temp new_mask = bld.tmp(s2); + instr->definitions[k] = Definition(new_mask); + if (i == ctx.info[block->index].exec.size() - 1) + instr->definitions[k].setFixed(exec); + k++; + ctx.info[block->index].exec[i].first = new_mask; + } + } + assert(k == num); + instr->definitions[num] = bld.def(s1, scc); + instr->operands[num] = Operand(cond); + state = Exact; + + } else if (instr->opcode == aco_opcode::p_fs_buffer_store_smem) { + bool need_check = ctx.info[block->index].exec.size() != 1 && + !(ctx.info[block->index].exec[ctx.info[block->index].exec.size() - 2].second & Exact); + lower_fs_buffer_store_smem(bld, need_check, instr, ctx.info[block->index].exec.back().first); + } + + bld.insert(std::move(instr)); + } +} + +void add_branch_code(exec_ctx& ctx, Block* block) +{ + unsigned idx = block->index; + Builder bld(ctx.program, block); + + if (idx == ctx.program->blocks.size() - 1) + return; + + /* try to disable wqm handling */ + if (ctx.handle_wqm && block->kind & block_kind_top_level) { + if (ctx.info[idx].exec.size() == 3) { + assert(ctx.info[idx].exec[1].second == mask_type_wqm); + ctx.info[idx].exec.pop_back(); + } + assert(ctx.info[idx].exec.size() <= 2); + + if (ctx.info[idx].ever_again_needs == 0 || + ctx.info[idx].ever_again_needs == Exact) { + /* transition to Exact */ + aco_ptr branch = std::move(block->instructions.back()); + block->instructions.pop_back(); + ctx.info[idx].exec.back().second |= mask_type_global; + transition_to_Exact(ctx, bld, idx); + bld.insert(std::move(branch)); + ctx.handle_wqm = false; + + } else if (ctx.info[idx].block_needs & Preserve_WQM) { + /* transition to WQM and remove global flag */ + aco_ptr branch = std::move(block->instructions.back()); + block->instructions.pop_back(); + transition_to_WQM(ctx, bld, idx); + ctx.info[idx].exec.back().second &= ~mask_type_global; + bld.insert(std::move(branch)); + } + } + + if (block->kind & block_kind_loop_preheader) { + /* collect information about the succeeding loop */ + bool has_divergent_break = false; + bool has_divergent_continue = false; + bool has_discard = false; + uint8_t needs = 0; + unsigned loop_nest_depth = ctx.program->blocks[idx + 1].loop_nest_depth; + + for (unsigned i = idx + 1; ctx.program->blocks[i].loop_nest_depth >= loop_nest_depth; i++) { + Block& loop_block = ctx.program->blocks[i]; + needs |= ctx.info[i].block_needs; + + if (loop_block.kind & block_kind_uses_discard_if || + loop_block.kind & block_kind_discard) + has_discard = true; + if (loop_block.loop_nest_depth != loop_nest_depth) + continue; + + if (loop_block.kind & block_kind_uniform) + continue; + else if (loop_block.kind & block_kind_break) + has_divergent_break = true; + else if (loop_block.kind & block_kind_continue) + has_divergent_continue = true; + } + + if (ctx.handle_wqm) { + if (needs & WQM) { + aco_ptr branch = std::move(block->instructions.back()); + block->instructions.pop_back(); + transition_to_WQM(ctx, bld, idx); + bld.insert(std::move(branch)); + } else { + aco_ptr branch = std::move(block->instructions.back()); + block->instructions.pop_back(); + transition_to_Exact(ctx, bld, idx); + bld.insert(std::move(branch)); + } + } + + unsigned num_exec_masks = ctx.info[idx].exec.size(); + if (block->kind & block_kind_top_level) + num_exec_masks = std::min(num_exec_masks, 2u); + + ctx.loop.emplace_back(&ctx.program->blocks[block->linear_succs[0]], + num_exec_masks, + needs, + has_divergent_break, + has_divergent_continue, + has_discard); + } + + if (block->kind & block_kind_discard) { + + assert(block->instructions.back()->format == Format::PSEUDO_BRANCH); + aco_ptr branch = std::move(block->instructions.back()); + block->instructions.pop_back(); + + /* create a discard_if() instruction with the exec mask as condition */ + unsigned num = 0; + if (ctx.loop.size()) { + /* if we're in a loop, only discard from the outer exec masks */ + num = ctx.loop.back().num_exec_masks; + } else { + num = ctx.info[idx].exec.size() - 1; + } + + Temp old_exec = ctx.info[idx].exec.back().first; + Temp new_exec = bld.tmp(s2); + Temp cond = bld.sop1(aco_opcode::s_and_saveexec_b64, bld.def(s2), bld.def(s1, scc), + bld.exec(Definition(new_exec)), Operand(0u), bld.exec(old_exec)); + ctx.info[idx].exec.back().first = new_exec; + + aco_ptr discard{create_instruction(aco_opcode::p_discard_if, Format::PSEUDO, num + 1, num + 1)}; + for (unsigned i = 0; i < num; i++) { + discard->operands[i] = Operand(ctx.info[block->index].exec[i].first); + Temp new_mask = bld.tmp(s2); + discard->definitions[i] = Definition(new_mask); + ctx.info[block->index].exec[i].first = new_mask; + } + assert(!ctx.handle_wqm || (ctx.info[block->index].exec[0].second & mask_type_wqm) == 0); + discard->operands[num] = Operand(cond); + discard->definitions[num] = bld.def(s1, scc); + + bld.insert(std::move(discard)); + if ((block->kind & (block_kind_break | block_kind_uniform)) == block_kind_break) + ctx.info[idx].exec.back().first = cond; + bld.insert(std::move(branch)); + /* no return here as it can be followed by a divergent break */ + } + + if (block->kind & block_kind_continue_or_break) { + assert(block->instructions.back()->opcode == aco_opcode::p_branch); + block->instructions.pop_back(); + + /* because of how linear_succs is created, this needs to be swapped */ + std::swap(block->linear_succs[0], block->linear_succs[1]); + + assert(ctx.program->blocks[block->linear_succs[1]].kind & block_kind_loop_header); + assert(ctx.program->blocks[ctx.program->blocks[block->linear_succs[0]].linear_succs[0]].kind & block_kind_loop_exit); + + if (ctx.info[idx].exec.back().second & mask_type_loop) { + bld.branch(aco_opcode::p_cbranch_nz, bld.exec(ctx.info[idx].exec.back().first), block->linear_succs[1], block->linear_succs[0]); + } else { + Temp cond = Temp(); + for (int exec_idx = ctx.info[idx].exec.size() - 1; exec_idx >= 0; exec_idx--) { + if (ctx.info[idx].exec[exec_idx].second & mask_type_loop) { + cond = bld.sopc(aco_opcode::s_cmp_lg_u64, bld.def(s1, scc), ctx.info[idx].exec[exec_idx].first, Operand(0u)); + break; + } + } + assert(cond != Temp()); + + bld.branch(aco_opcode::p_cbranch_nz, bld.scc(cond), block->linear_succs[1], block->linear_succs[0]); + } + return; + } + + if (block->kind & block_kind_uniform) { + Pseudo_branch_instruction* branch = static_cast(block->instructions.back().get()); + if (branch->opcode == aco_opcode::p_branch) { + branch->target[0] = block->linear_succs[0]; + } else { + branch->target[0] = block->linear_succs[1]; + branch->target[1] = block->linear_succs[0]; + } + return; + } + + if (block->kind & block_kind_branch) { + + if (ctx.handle_wqm && + ctx.info[idx].exec.size() >= 2 && + ctx.info[idx].exec.back().second == mask_type_exact && + !(ctx.info[idx].block_needs & Exact_Branch) && + ctx.info[idx].exec[ctx.info[idx].exec.size() - 2].second & mask_type_wqm) { + /* return to wqm before branching */ + ctx.info[idx].exec.pop_back(); + } + + // orig = s_and_saveexec_b64 + assert(block->linear_succs.size() == 2); + assert(block->instructions.back()->opcode == aco_opcode::p_cbranch_z); + Temp cond = block->instructions.back()->operands[0].getTemp(); + block->instructions.pop_back(); + + if (ctx.info[idx].block_needs & Exact_Branch) + transition_to_Exact(ctx, bld, idx); + + Temp current_exec = ctx.info[idx].exec.back().first; + uint8_t mask_type = ctx.info[idx].exec.back().second & (mask_type_wqm | mask_type_exact); + + Temp then_mask = bld.tmp(s2); + Temp old_exec = bld.sop1(aco_opcode::s_and_saveexec_b64, bld.def(s2), bld.def(s1, scc), + bld.exec(Definition(then_mask)), cond, bld.exec(current_exec)); + + ctx.info[idx].exec.back().first = old_exec; + + /* add next current exec to the stack */ + ctx.info[idx].exec.emplace_back(then_mask, mask_type); + + bld.branch(aco_opcode::p_cbranch_z, bld.exec(then_mask), block->linear_succs[1], block->linear_succs[0]); + return; + } + + if (block->kind & block_kind_invert) { + // exec = s_andn2_b64 (original_exec, exec) + assert(block->instructions.back()->opcode == aco_opcode::p_cbranch_nz); + block->instructions.pop_back(); + Temp then_mask = ctx.info[idx].exec.back().first; + uint8_t mask_type = ctx.info[idx].exec.back().second; + ctx.info[idx].exec.pop_back(); + Temp orig_exec = ctx.info[idx].exec.back().first; + Temp else_mask = bld.sop2(aco_opcode::s_andn2_b64, bld.def(s2, exec), + bld.def(s1, scc), orig_exec, bld.exec(then_mask)); + + /* add next current exec to the stack */ + ctx.info[idx].exec.emplace_back(else_mask, mask_type); + + bld.branch(aco_opcode::p_cbranch_z, bld.exec(else_mask), block->linear_succs[1], block->linear_succs[0]); + return; + } + + if (block->kind & block_kind_break) { + // loop_mask = s_andn2_b64 (loop_mask, exec) + assert(block->instructions.back()->opcode == aco_opcode::p_branch); + block->instructions.pop_back(); + + Temp current_exec = ctx.info[idx].exec.back().first; + Temp cond = Temp(); + for (int exec_idx = ctx.info[idx].exec.size() - 2; exec_idx >= 0; exec_idx--) { + cond = bld.tmp(s1); + Temp exec_mask = ctx.info[idx].exec[exec_idx].first; + exec_mask = bld.sop2(aco_opcode::s_andn2_b64, bld.def(s2), bld.scc(Definition(cond)), + exec_mask, current_exec); + ctx.info[idx].exec[exec_idx].first = exec_mask; + if (ctx.info[idx].exec[exec_idx].second & mask_type_loop) + break; + } + + /* check if the successor is the merge block, otherwise set exec to 0 */ + // TODO: this could be done better by directly branching to the merge block + unsigned succ_idx = ctx.program->blocks[block->linear_succs[1]].linear_succs[0]; + Block& succ = ctx.program->blocks[succ_idx]; + if (!(succ.kind & block_kind_invert || succ.kind & block_kind_merge)) { + ctx.info[idx].exec.back().first = bld.sop1(aco_opcode::s_mov_b64, bld.def(s2, exec), Operand(0u)); + } + + bld.branch(aco_opcode::p_cbranch_nz, bld.scc(cond), block->linear_succs[1], block->linear_succs[0]); + return; + } + + if (block->kind & block_kind_continue) { + assert(block->instructions.back()->opcode == aco_opcode::p_branch); + block->instructions.pop_back(); + + Temp current_exec = ctx.info[idx].exec.back().first; + Temp cond = Temp(); + for (int exec_idx = ctx.info[idx].exec.size() - 2; exec_idx >= 0; exec_idx--) { + if (ctx.info[idx].exec[exec_idx].second & mask_type_loop) + break; + cond = bld.tmp(s1); + Temp exec_mask = ctx.info[idx].exec[exec_idx].first; + exec_mask = bld.sop2(aco_opcode::s_andn2_b64, bld.def(s2), bld.scc(Definition(cond)), + exec_mask, bld.exec(current_exec)); + ctx.info[idx].exec[exec_idx].first = exec_mask; + } + assert(cond != Temp()); + + /* check if the successor is the merge block, otherwise set exec to 0 */ + // TODO: this could be done better by directly branching to the merge block + unsigned succ_idx = ctx.program->blocks[block->linear_succs[1]].linear_succs[0]; + Block& succ = ctx.program->blocks[succ_idx]; + if (!(succ.kind & block_kind_invert || succ.kind & block_kind_merge)) { + ctx.info[idx].exec.back().first = bld.sop1(aco_opcode::s_mov_b64, bld.def(s2, exec), Operand(0u)); + } + + bld.branch(aco_opcode::p_cbranch_nz, bld.scc(cond), block->linear_succs[1], block->linear_succs[0]); + return; + } +} + +void process_block(exec_ctx& ctx, Block* block) +{ + std::vector> instructions; + instructions.reserve(block->instructions.size()); + + unsigned idx = add_coupling_code(ctx, block, instructions); + + assert(block->index != ctx.program->blocks.size() - 1 || + ctx.info[block->index].exec.size() <= 2); + + process_instructions(ctx, block, instructions, idx); + + block->instructions = std::move(instructions); + + add_branch_code(ctx, block); + + block->live_out_exec = ctx.info[block->index].exec.back().first; +} + +} /* end namespace */ + + +void insert_exec_mask(Program *program) +{ + exec_ctx ctx(program); + + if (program->needs_wqm && program->needs_exact) + calculate_wqm_needs(ctx); + + for (Block& block : program->blocks) + process_block(ctx, &block); + +} + +} + diff --git a/src/amd/compiler/aco_insert_waitcnt.cpp b/src/amd/compiler/aco_insert_waitcnt.cpp new file mode 100644 index 00000000000..d19fdadadea --- /dev/null +++ b/src/amd/compiler/aco_insert_waitcnt.cpp @@ -0,0 +1,697 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include +#include + +#include "aco_ir.h" +#include "vulkan/radv_shader.h" + +namespace aco { + +namespace { + +/** + * The general idea of this pass is: + * The CFG is traversed in reverse postorder (forward). + * Per BB one wait_ctx is maintained. + * The in-context is the joined out-contexts of the predecessors. + * The context contains a map: gpr -> wait_entry + * consisting of the information about the cnt values to be waited for. + * Note: After merge-nodes, it might occur that for the same register + * multiple cnt values are to be waited for. + * + * The values are updated according to the encountered instructions: + * - additional events increment the counter of waits of the same type + * - or erase gprs with counters higher than to be waited for. + */ + +// TODO: do a more clever insertion of wait_cnt (lgkm_cnt) when there is a load followed by a use of a previous load + +/* Instructions of the same event will finish in-order except for smem + * and maybe flat. Instructions of different events may not finish in-order. */ +enum wait_event : uint16_t { + event_smem = 1 << 0, + event_lds = 1 << 1, + event_gds = 1 << 2, + event_vmem = 1 << 3, + event_vmem_store = 1 << 4, /* GFX10+ */ + event_flat = 1 << 5, + event_exp_pos = 1 << 6, + event_exp_param = 1 << 7, + event_exp_mrt_null = 1 << 8, + event_gds_gpr_lock = 1 << 9, + event_vmem_gpr_lock = 1 << 10, +}; + +enum counter_type : uint8_t { + counter_exp = 1 << 0, + counter_lgkm = 1 << 1, + counter_vm = 1 << 2, + counter_vs = 1 << 3, +}; + +static const uint16_t exp_events = event_exp_pos | event_exp_param | event_exp_mrt_null | event_gds_gpr_lock | event_vmem_gpr_lock; +static const uint16_t lgkm_events = event_smem | event_lds | event_gds | event_flat; +static const uint16_t vm_events = event_vmem | event_flat; +static const uint16_t vs_events = event_vmem_store; + +uint8_t get_counters_for_event(wait_event ev) +{ + switch (ev) { + case event_smem: + case event_lds: + case event_gds: + return counter_lgkm; + case event_vmem: + return counter_vm; + case event_vmem_store: + return counter_vs; + case event_flat: + return counter_vm | counter_lgkm; + case event_exp_pos: + case event_exp_param: + case event_exp_mrt_null: + case event_gds_gpr_lock: + case event_vmem_gpr_lock: + return counter_exp; + default: + return 0; + } +} + +struct wait_imm { + static const uint8_t unset_counter = 0xff; + + uint8_t vm; + uint8_t exp; + uint8_t lgkm; + uint8_t vs; + + wait_imm() : + vm(unset_counter), exp(unset_counter), lgkm(unset_counter), vs(unset_counter) {} + wait_imm(uint16_t vm_, uint16_t exp_, uint16_t lgkm_, uint16_t vs_) : + vm(vm_), exp(exp_), lgkm(lgkm_), vs(vs_) {} + + uint16_t pack(enum chip_class chip) const + { + uint16_t imm = 0; + assert(exp == unset_counter || exp <= 0x7); + switch (chip) { + case GFX10: + assert(lgkm == unset_counter || lgkm <= 0x3f); + assert(vm == unset_counter || vm <= 0x3f); + imm = ((vm & 0x30) << 10) | ((lgkm & 0x3f) << 8) | ((exp & 0x7) << 4) | (vm & 0xf); + break; + case GFX9: + assert(lgkm == unset_counter || lgkm <= 0xf); + assert(vm == unset_counter || vm <= 0x3f); + imm = ((vm & 0x30) << 10) | ((lgkm & 0xf) << 8) | ((exp & 0x7) << 4) | (vm & 0xf); + break; + default: + assert(lgkm == unset_counter || lgkm <= 0xf); + assert(vm == unset_counter || vm <= 0xf); + imm = ((lgkm & 0xf) << 8) | ((exp & 0x7) << 4) | (vm & 0xf); + break; + } + if (chip < GFX9 && vm == wait_imm::unset_counter) + imm |= 0xc000; /* should have no effect on pre-GFX9 and now we won't have to worry about the architecture when interpreting the immediate */ + if (chip < GFX10 && lgkm == wait_imm::unset_counter) + imm |= 0x3000; /* should have no effect on pre-GFX10 and now we won't have to worry about the architecture when interpreting the immediate */ + return imm; + } + + void combine(const wait_imm& other) + { + vm = std::min(vm, other.vm); + exp = std::min(exp, other.exp); + lgkm = std::min(lgkm, other.lgkm); + vs = std::min(vs, other.vs); + } + + bool empty() const + { + return vm == unset_counter && exp == unset_counter && + lgkm == unset_counter && vs == unset_counter; + } +}; + +struct wait_entry { + wait_imm imm; + uint16_t events; /* use wait_event notion */ + uint8_t counters; /* use counter_type notion */ + bool wait_on_read:1; + bool logical:1; + + wait_entry(wait_event event, wait_imm imm, bool logical, bool wait_on_read) + : imm(imm), events(event), counters(get_counters_for_event(event)), + wait_on_read(wait_on_read), logical(logical) {} + + void join(const wait_entry& other) + { + events |= other.events; + counters |= other.counters; + imm.combine(other.imm); + wait_on_read = wait_on_read || other.wait_on_read; + assert(logical == other.logical); + } + + void remove_counter(counter_type counter) + { + counters &= ~counter; + + if (counter == counter_lgkm) { + imm.lgkm = wait_imm::unset_counter; + events &= ~(event_smem | event_lds | event_gds); + } + + if (counter == counter_vm) { + imm.vm = wait_imm::unset_counter; + events &= ~event_vmem; + } + + if (counter == counter_exp) { + imm.exp = wait_imm::unset_counter; + events &= ~(event_exp_pos | event_exp_param | event_exp_mrt_null | event_gds_gpr_lock | event_vmem_gpr_lock); + } + + if (counter == counter_vs) { + imm.vs = wait_imm::unset_counter; + events &= ~event_vmem_store; + } + + if (!(counters & counter_lgkm) && !(counters & counter_vm)) + events &= ~event_flat; + } +}; + +struct wait_ctx { + Program *program; + enum chip_class chip_class; + uint16_t max_vm_cnt; + uint16_t max_exp_cnt; + uint16_t max_lgkm_cnt; + uint16_t max_vs_cnt; + uint16_t unordered_events = event_smem | event_flat; + + uint8_t vm_cnt = 0; + uint8_t exp_cnt = 0; + uint8_t lgkm_cnt = 0; + uint8_t vs_cnt = 0; + bool pending_flat_lgkm = false; + bool pending_flat_vm = false; + + wait_imm barrier_imm[barrier_count]; + + std::map gpr_map; + + wait_ctx() {} + wait_ctx(Program *program_) + : program(program_), + chip_class(program_->chip_class), + max_vm_cnt(program_->chip_class >= GFX9 ? 62 : 14), + max_exp_cnt(6), + max_lgkm_cnt(program_->chip_class >= GFX10 ? 62 : 14), + max_vs_cnt(program_->chip_class >= GFX10 ? 62 : 0), + unordered_events(event_smem | (program_->chip_class < GFX10 ? event_flat : 0)) {} + + void join(const wait_ctx* other, bool logical) + { + exp_cnt = std::max(exp_cnt, other->exp_cnt); + vm_cnt = std::max(vm_cnt, other->vm_cnt); + lgkm_cnt = std::max(lgkm_cnt, other->lgkm_cnt); + vs_cnt = std::max(vs_cnt, other->vs_cnt); + pending_flat_lgkm |= other->pending_flat_lgkm; + pending_flat_vm |= other->pending_flat_vm; + + for (std::pair entry : other->gpr_map) + { + std::map::iterator it = gpr_map.find(entry.first); + if (entry.second.logical != logical) + continue; + + if (it != gpr_map.end()) + it->second.join(entry.second); + else + gpr_map.insert(entry); + } + + for (unsigned i = 0; i < barrier_count; i++) + barrier_imm[i].combine(other->barrier_imm[i]); + } +}; + +wait_imm check_instr(Instruction* instr, wait_ctx& ctx) +{ + wait_imm wait; + + for (const Operand op : instr->operands) { + if (op.isConstant() || op.isUndefined()) + continue; + + /* check consecutively read gprs */ + for (unsigned j = 0; j < op.size(); j++) { + PhysReg reg{op.physReg() + j}; + std::map::iterator it = ctx.gpr_map.find(reg); + if (it == ctx.gpr_map.end() || !it->second.wait_on_read) + continue; + + wait.combine(it->second.imm); + } + } + + for (const Definition& def : instr->definitions) { + /* check consecutively written gprs */ + for (unsigned j = 0; j < def.getTemp().size(); j++) + { + PhysReg reg{def.physReg() + j}; + + std::map::iterator it = ctx.gpr_map.find(reg); + if (it == ctx.gpr_map.end()) + continue; + + /* Vector Memory reads and writes return in the order they were issued */ + if (instr->isVMEM() && ((it->second.events & vm_events) == event_vmem)) { + it->second.remove_counter(counter_vm); + if (!it->second.counters) + it = ctx.gpr_map.erase(it); + continue; + } + + /* LDS reads and writes return in the order they were issued. same for GDS */ + if (instr->format == Format::DS) { + bool gds = static_cast(instr)->gds; + if ((it->second.events & lgkm_events) == (gds ? event_gds : event_lds)) { + it->second.remove_counter(counter_lgkm); + if (!it->second.counters) + it = ctx.gpr_map.erase(it); + continue; + } + } + + wait.combine(it->second.imm); + } + } + + return wait; +} + +wait_imm kill(Instruction* instr, wait_ctx& ctx) +{ + wait_imm imm; + if (ctx.exp_cnt || ctx.vm_cnt || ctx.lgkm_cnt) + imm.combine(check_instr(instr, ctx)); + + if (instr->format == Format::PSEUDO_BARRIER) { + unsigned* bsize = ctx.program->info->cs.block_size; + unsigned workgroup_size = bsize[0] * bsize[1] * bsize[2]; + switch (instr->opcode) { + case aco_opcode::p_memory_barrier_all: + for (unsigned i = 0; i < barrier_count; i++) { + if ((1 << i) == barrier_shared && workgroup_size <= 64) + continue; + imm.combine(ctx.barrier_imm[i]); + } + break; + case aco_opcode::p_memory_barrier_atomic: + imm.combine(ctx.barrier_imm[ffs(barrier_atomic) - 1]); + break; + /* see comment in aco_scheduler.cpp's can_move_instr() on why these barriers are merged */ + case aco_opcode::p_memory_barrier_buffer: + case aco_opcode::p_memory_barrier_image: + imm.combine(ctx.barrier_imm[ffs(barrier_buffer) - 1]); + imm.combine(ctx.barrier_imm[ffs(barrier_image) - 1]); + break; + case aco_opcode::p_memory_barrier_shared: + if (workgroup_size > 64) + imm.combine(ctx.barrier_imm[ffs(barrier_shared) - 1]); + break; + default: + assert(false); + break; + } + } + + if (!imm.empty()) { + if (ctx.pending_flat_vm && imm.vm != wait_imm::unset_counter) + imm.vm = 0; + if (ctx.pending_flat_lgkm && imm.lgkm != wait_imm::unset_counter) + imm.lgkm = 0; + + /* reset counters */ + ctx.exp_cnt = std::min(ctx.exp_cnt, imm.exp); + ctx.vm_cnt = std::min(ctx.vm_cnt, imm.vm); + ctx.lgkm_cnt = std::min(ctx.lgkm_cnt, imm.lgkm); + ctx.vs_cnt = std::min(ctx.vs_cnt, imm.vs); + + /* update barrier wait imms */ + for (unsigned i = 0; i < barrier_count; i++) { + wait_imm& bar = ctx.barrier_imm[i]; + if (bar.exp != wait_imm::unset_counter && imm.exp <= bar.exp) + bar.exp = wait_imm::unset_counter; + if (bar.vm != wait_imm::unset_counter && imm.vm <= bar.vm) + bar.vm = wait_imm::unset_counter; + if (bar.lgkm != wait_imm::unset_counter && imm.lgkm <= bar.lgkm) + bar.lgkm = wait_imm::unset_counter; + if (bar.vs != wait_imm::unset_counter && imm.vs <= bar.vs) + bar.vs = wait_imm::unset_counter; + } + + /* remove all vgprs with higher counter from map */ + std::map::iterator it = ctx.gpr_map.begin(); + while (it != ctx.gpr_map.end()) + { + if (imm.exp != wait_imm::unset_counter && imm.exp <= it->second.imm.exp) + it->second.remove_counter(counter_exp); + if (imm.vm != wait_imm::unset_counter && imm.vm <= it->second.imm.vm) + it->second.remove_counter(counter_vm); + if (imm.lgkm != wait_imm::unset_counter && imm.lgkm <= it->second.imm.lgkm) + it->second.remove_counter(counter_lgkm); + if (imm.lgkm != wait_imm::unset_counter && imm.vs <= it->second.imm.vs) + it->second.remove_counter(counter_vs); + if (!it->second.counters) + it = ctx.gpr_map.erase(it); + else + it++; + } + } + + if (imm.vm == 0) + ctx.pending_flat_vm = false; + if (imm.lgkm == 0) + ctx.pending_flat_lgkm = false; + + return imm; +} + +void update_barrier_imm(wait_ctx& ctx, uint8_t counters, barrier_interaction barrier) +{ + unsigned barrier_index = ffs(barrier) - 1; + for (unsigned i = 0; i < barrier_count; i++) { + wait_imm& bar = ctx.barrier_imm[i]; + if (i == barrier_index) { + if (counters & counter_lgkm) + bar.lgkm = 0; + if (counters & counter_vm) + bar.vm = 0; + if (counters & counter_exp) + bar.exp = 0; + if (counters & counter_vs) + bar.vs = 0; + } else { + if (counters & counter_lgkm && bar.lgkm != wait_imm::unset_counter && bar.lgkm < ctx.max_lgkm_cnt) + bar.lgkm++; + if (counters & counter_vm && bar.vm != wait_imm::unset_counter && bar.vm < ctx.max_vm_cnt) + bar.vm++; + if (counters & counter_exp && bar.exp != wait_imm::unset_counter && bar.exp < ctx.max_exp_cnt) + bar.exp++; + if (counters & counter_vs && bar.vs != wait_imm::unset_counter && bar.vs < ctx.max_vs_cnt) + bar.vs++; + } + } +} + +void update_counters(wait_ctx& ctx, wait_event event, barrier_interaction barrier=barrier_none) +{ + uint8_t counters = get_counters_for_event(event); + + if (counters & counter_lgkm && ctx.lgkm_cnt <= ctx.max_lgkm_cnt) + ctx.lgkm_cnt++; + if (counters & counter_vm && ctx.vm_cnt <= ctx.max_vm_cnt) + ctx.vm_cnt++; + if (counters & counter_exp && ctx.exp_cnt <= ctx.max_exp_cnt) + ctx.exp_cnt++; + if (counters & counter_vs && ctx.vs_cnt <= ctx.max_vs_cnt) + ctx.vs_cnt++; + + update_barrier_imm(ctx, counters, barrier); + + if (ctx.unordered_events & event) + return; + + if (ctx.pending_flat_lgkm) + counters &= ~counter_lgkm; + if (ctx.pending_flat_vm) + counters &= ~counter_vm; + + for (std::pair& e : ctx.gpr_map) { + wait_entry& entry = e.second; + + if (entry.events & ctx.unordered_events) + continue; + + assert(entry.events); + + if ((counters & counter_exp) && (entry.events & exp_events) == event && entry.imm.exp < ctx.max_exp_cnt) + entry.imm.exp++; + if ((counters & counter_lgkm) && (entry.events & lgkm_events) == event && entry.imm.lgkm < ctx.max_lgkm_cnt) + entry.imm.lgkm++; + if ((counters & counter_vm) && (entry.events & vm_events) == event && entry.imm.vm < ctx.max_vm_cnt) + entry.imm.vm++; + if ((counters & counter_vs) && (entry.events & vs_events) == event && entry.imm.vs < ctx.max_vs_cnt) + entry.imm.vs++; + } +} + +void update_counters_for_flat_load(wait_ctx& ctx, barrier_interaction barrier=barrier_none) +{ + assert(ctx.chip_class < GFX10); + + if (ctx.lgkm_cnt <= ctx.max_lgkm_cnt) + ctx.lgkm_cnt++; + if (ctx.lgkm_cnt <= ctx.max_vm_cnt) + ctx.vm_cnt++; + + update_barrier_imm(ctx, counter_vm | counter_lgkm, barrier); + + for (std::pair e : ctx.gpr_map) + { + if (e.second.counters & counter_vm) + e.second.imm.vm = 0; + if (e.second.counters & counter_lgkm) + e.second.imm.lgkm = 0; + } + ctx.pending_flat_lgkm = true; + ctx.pending_flat_vm = true; +} + +void insert_wait_entry(wait_ctx& ctx, PhysReg reg, RegClass rc, wait_event event, bool wait_on_read) +{ + uint16_t counters = get_counters_for_event(event); + wait_imm imm; + if (counters & counter_lgkm) + imm.lgkm = 0; + if (counters & counter_vm) + imm.vm = 0; + if (counters & counter_exp) + imm.exp = 0; + if (counters & counter_vs) + imm.vs = 0; + + wait_entry new_entry(event, imm, !rc.is_linear(), wait_on_read); + + for (unsigned i = 0; i < rc.size(); i++) { + auto it = ctx.gpr_map.emplace(PhysReg{reg.reg+i}, new_entry); + if (!it.second) + it.first->second.join(new_entry); + } +} + +void insert_wait_entry(wait_ctx& ctx, Operand op, wait_event event) +{ + if (!op.isConstant() && !op.isUndefined()) + insert_wait_entry(ctx, op.physReg(), op.regClass(), event, false); +} + +void insert_wait_entry(wait_ctx& ctx, Definition def, wait_event event) +{ + insert_wait_entry(ctx, def.physReg(), def.regClass(), event, true); +} + +void gen(Instruction* instr, wait_ctx& ctx) +{ + switch (instr->format) { + case Format::EXP: { + Export_instruction* exp_instr = static_cast(instr); + + wait_event ev; + if (exp_instr->dest <= 9) + ev = event_exp_mrt_null; + else if (exp_instr->dest <= 15) + ev = event_exp_pos; + else + ev = event_exp_param; + update_counters(ctx, ev); + + /* insert new entries for exported vgprs */ + for (unsigned i = 0; i < 4; i++) + { + if (exp_instr->enabled_mask & (1 << i)) { + unsigned idx = exp_instr->compressed ? i >> 1 : i; + assert(idx < exp_instr->operands.size()); + insert_wait_entry(ctx, exp_instr->operands[idx], ev); + + } + } + insert_wait_entry(ctx, exec, s2, ev, false); + break; + } + case Format::FLAT: { + if (ctx.chip_class < GFX10 && !instr->definitions.empty()) + update_counters_for_flat_load(ctx, barrier_buffer); + else + update_counters(ctx, event_flat, barrier_buffer); + + if (!instr->definitions.empty()) + insert_wait_entry(ctx, instr->definitions[0], event_flat); + break; + } + case Format::SMEM: { + update_counters(ctx, event_smem, static_cast(instr)->barrier); + + if (!instr->definitions.empty()) + insert_wait_entry(ctx, instr->definitions[0], event_smem); + break; + } + case Format::DS: { + bool gds = static_cast(instr)->gds; + update_counters(ctx, gds ? event_gds : event_lds, gds ? barrier_none : barrier_shared); + if (gds) + update_counters(ctx, event_gds_gpr_lock); + + if (!instr->definitions.empty()) + insert_wait_entry(ctx, instr->definitions[0], gds ? event_gds : event_lds); + + if (gds) { + for (const Operand& op : instr->operands) + insert_wait_entry(ctx, op, event_gds_gpr_lock); + insert_wait_entry(ctx, exec, s2, event_gds_gpr_lock, false); + } + break; + } + case Format::MUBUF: + case Format::MTBUF: + case Format::MIMG: + case Format::GLOBAL: { + wait_event ev = !instr->definitions.empty() || ctx.chip_class < GFX10 ? event_vmem : event_vmem_store; + update_counters(ctx, ev, get_barrier_interaction(instr)); + + if (!instr->definitions.empty()) + insert_wait_entry(ctx, instr->definitions[0], ev); + + if (instr->operands.size() == 4 && ctx.chip_class == GFX6) { + ctx.exp_cnt++; + update_counters(ctx, event_vmem_gpr_lock); + insert_wait_entry(ctx, instr->operands[3], event_vmem_gpr_lock); + } + break; + } + default: + break; + } +} + +void emit_waitcnt(wait_ctx& ctx, std::vector>& instructions, wait_imm imm) +{ + if (imm.vs != wait_imm::unset_counter) { + assert(ctx.chip_class >= GFX10); + SOPK_instruction* waitcnt_vs = create_instruction(aco_opcode::s_waitcnt_vscnt, Format::SOPK, 0, 0); + waitcnt_vs->imm = imm.vs; + instructions.emplace_back(waitcnt_vs); + imm.vs = wait_imm::unset_counter; + } + if (!imm.empty()) { + SOPP_instruction* waitcnt = create_instruction(aco_opcode::s_waitcnt, Format::SOPP, 0, 0); + waitcnt->imm = imm.pack(ctx.chip_class); + waitcnt->block = -1; + instructions.emplace_back(waitcnt); + } +} + +void handle_block(Program *program, Block& block, wait_ctx& ctx) +{ + std::vector> new_instructions; + + for (aco_ptr& instr : block.instructions) { + wait_imm imm = kill(instr.get(), ctx); + + if (!imm.empty()) + emit_waitcnt(ctx, new_instructions, imm); + + gen(instr.get(), ctx); + + if (instr->format != Format::PSEUDO_BARRIER) + new_instructions.emplace_back(std::move(instr)); + } + + /* check if this block is at the end of a loop */ + for (unsigned succ_idx : block.linear_succs) { + /* eliminate any remaining counters */ + if (succ_idx <= block.index && (ctx.vm_cnt || ctx.exp_cnt || ctx.lgkm_cnt || ctx.vs_cnt)) { + // TODO: we could do better if we only wait if the regs between the block and other predecessors differ + + aco_ptr branch = std::move(new_instructions.back()); + new_instructions.pop_back(); + + wait_imm imm(ctx.vm_cnt ? 0 : wait_imm::unset_counter, + ctx.exp_cnt ? 0 : wait_imm::unset_counter, + ctx.lgkm_cnt ? 0 : wait_imm::unset_counter, + ctx.vs_cnt ? 0 : wait_imm::unset_counter); + emit_waitcnt(ctx, new_instructions, imm); + + new_instructions.push_back(std::move(branch)); + + ctx = wait_ctx(program); + break; + } + } + block.instructions.swap(new_instructions); +} + +} /* end namespace */ + +void insert_wait_states(Program* program) +{ + wait_ctx out_ctx[program->blocks.size()]; /* per BB ctx */ + for (unsigned i = 0; i < program->blocks.size(); i++) + out_ctx[i] = wait_ctx(program); + + for (unsigned i = 0; i < program->blocks.size(); i++) { + Block& current = program->blocks[i]; + wait_ctx& in = out_ctx[current.index]; + + for (unsigned b : current.linear_preds) + in.join(&out_ctx[b], false); + for (unsigned b : current.logical_preds) + in.join(&out_ctx[b], true); + + if (current.instructions.empty()) + continue; + + handle_block(program, current, in); + } +} + +} + diff --git a/src/amd/compiler/aco_instruction_selection.cpp b/src/amd/compiler/aco_instruction_selection.cpp new file mode 100644 index 00000000000..d52043f3c0d --- /dev/null +++ b/src/amd/compiler/aco_instruction_selection.cpp @@ -0,0 +1,7621 @@ +/* + * Copyright © 2018 Valve Corporation + * Copyright © 2018 Google + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include +#include + +#include "aco_ir.h" +#include "aco_builder.h" +#include "aco_interface.h" +#include "aco_instruction_selection_setup.cpp" +#include "util/fast_idiv_by_const.h" + +namespace aco { +namespace { + +class loop_info_RAII { + isel_context* ctx; + unsigned header_idx_old; + Block* exit_old; + bool divergent_cont_old; + bool divergent_branch_old; + bool divergent_if_old; + +public: + loop_info_RAII(isel_context* ctx, unsigned loop_header_idx, Block* loop_exit) + : ctx(ctx), + header_idx_old(ctx->cf_info.parent_loop.header_idx), exit_old(ctx->cf_info.parent_loop.exit), + divergent_cont_old(ctx->cf_info.parent_loop.has_divergent_continue), + divergent_branch_old(ctx->cf_info.parent_loop.has_divergent_branch), + divergent_if_old(ctx->cf_info.parent_if.is_divergent) + { + ctx->cf_info.parent_loop.header_idx = loop_header_idx; + ctx->cf_info.parent_loop.exit = loop_exit; + ctx->cf_info.parent_loop.has_divergent_continue = false; + ctx->cf_info.parent_loop.has_divergent_branch = false; + ctx->cf_info.parent_if.is_divergent = false; + ctx->cf_info.loop_nest_depth = ctx->cf_info.loop_nest_depth + 1; + } + + ~loop_info_RAII() + { + ctx->cf_info.parent_loop.header_idx = header_idx_old; + ctx->cf_info.parent_loop.exit = exit_old; + ctx->cf_info.parent_loop.has_divergent_continue = divergent_cont_old; + ctx->cf_info.parent_loop.has_divergent_branch = divergent_branch_old; + ctx->cf_info.parent_if.is_divergent = divergent_if_old; + ctx->cf_info.loop_nest_depth = ctx->cf_info.loop_nest_depth - 1; + if (!ctx->cf_info.loop_nest_depth && !ctx->cf_info.parent_if.is_divergent) + ctx->cf_info.exec_potentially_empty = false; + } +}; + +struct if_context { + Temp cond; + + bool divergent_old; + bool exec_potentially_empty_old; + + unsigned BB_if_idx; + unsigned invert_idx; + bool then_branch_divergent; + Block BB_invert; + Block BB_endif; +}; + +static void visit_cf_list(struct isel_context *ctx, + struct exec_list *list); + +static void add_logical_edge(unsigned pred_idx, Block *succ) +{ + succ->logical_preds.emplace_back(pred_idx); +} + + +static void add_linear_edge(unsigned pred_idx, Block *succ) +{ + succ->linear_preds.emplace_back(pred_idx); +} + +static void add_edge(unsigned pred_idx, Block *succ) +{ + add_logical_edge(pred_idx, succ); + add_linear_edge(pred_idx, succ); +} + +static void append_logical_start(Block *b) +{ + Builder(NULL, b).pseudo(aco_opcode::p_logical_start); +} + +static void append_logical_end(Block *b) +{ + Builder(NULL, b).pseudo(aco_opcode::p_logical_end); +} + +Temp get_ssa_temp(struct isel_context *ctx, nir_ssa_def *def) +{ + assert(ctx->allocated[def->index].id()); + return ctx->allocated[def->index]; +} + +Temp emit_wqm(isel_context *ctx, Temp src, Temp dst=Temp(0, s1), bool program_needs_wqm = false) +{ + Builder bld(ctx->program, ctx->block); + + if (!dst.id()) + dst = bld.tmp(src.regClass()); + + if (ctx->stage != fragment_fs) { + if (!dst.id()) + return src; + + if (src.type() == RegType::vgpr || src.size() > 1) + bld.copy(Definition(dst), src); + else + bld.sop1(aco_opcode::s_mov_b32, Definition(dst), src); + return dst; + } + + bld.pseudo(aco_opcode::p_wqm, Definition(dst), src); + ctx->program->needs_wqm |= program_needs_wqm; + return dst; +} + +Temp as_vgpr(isel_context *ctx, Temp val) +{ + if (val.type() == RegType::sgpr) { + Builder bld(ctx->program, ctx->block); + return bld.copy(bld.def(RegType::vgpr, val.size()), val); + } + assert(val.type() == RegType::vgpr); + return val; +} + +//assumes a != 0xffffffff +void emit_v_div_u32(isel_context *ctx, Temp dst, Temp a, uint32_t b) +{ + assert(b != 0); + Builder bld(ctx->program, ctx->block); + + if (util_is_power_of_two_or_zero(b)) { + bld.vop2(aco_opcode::v_lshrrev_b32, Definition(dst), Operand((uint32_t)util_logbase2(b)), a); + return; + } + + util_fast_udiv_info info = util_compute_fast_udiv_info(b, 32, 32); + + assert(info.multiplier <= 0xffffffff); + + bool pre_shift = info.pre_shift != 0; + bool increment = info.increment != 0; + bool multiply = true; + bool post_shift = info.post_shift != 0; + + if (!pre_shift && !increment && !multiply && !post_shift) { + bld.vop1(aco_opcode::v_mov_b32, Definition(dst), a); + return; + } + + Temp pre_shift_dst = a; + if (pre_shift) { + pre_shift_dst = (increment || multiply || post_shift) ? bld.tmp(v1) : dst; + bld.vop2(aco_opcode::v_lshrrev_b32, Definition(pre_shift_dst), Operand((uint32_t)info.pre_shift), a); + } + + Temp increment_dst = pre_shift_dst; + if (increment) { + increment_dst = (post_shift || multiply) ? bld.tmp(v1) : dst; + bld.vadd32(Definition(increment_dst), Operand((uint32_t) info.increment), pre_shift_dst); + } + + Temp multiply_dst = increment_dst; + if (multiply) { + multiply_dst = post_shift ? bld.tmp(v1) : dst; + bld.vop3(aco_opcode::v_mul_hi_u32, Definition(multiply_dst), increment_dst, + bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand((uint32_t)info.multiplier))); + } + + if (post_shift) { + bld.vop2(aco_opcode::v_lshrrev_b32, Definition(dst), Operand((uint32_t)info.post_shift), multiply_dst); + } +} + +void emit_extract_vector(isel_context* ctx, Temp src, uint32_t idx, Temp dst) +{ + Builder bld(ctx->program, ctx->block); + bld.pseudo(aco_opcode::p_extract_vector, Definition(dst), src, Operand(idx)); +} + + +Temp emit_extract_vector(isel_context* ctx, Temp src, uint32_t idx, RegClass dst_rc) +{ + /* no need to extract the whole vector */ + if (src.regClass() == dst_rc) { + assert(idx == 0); + return src; + } + assert(src.size() > idx); + Builder bld(ctx->program, ctx->block); + auto it = ctx->allocated_vec.find(src.id()); + /* the size check needs to be early because elements other than 0 may be garbage */ + if (it != ctx->allocated_vec.end() && it->second[0].size() == dst_rc.size()) { + if (it->second[idx].regClass() == dst_rc) { + return it->second[idx]; + } else { + assert(dst_rc.size() == it->second[idx].regClass().size()); + assert(dst_rc.type() == RegType::vgpr && it->second[idx].type() == RegType::sgpr); + return bld.copy(bld.def(dst_rc), it->second[idx]); + } + } + + if (src.size() == dst_rc.size()) { + assert(idx == 0); + return bld.copy(bld.def(dst_rc), src); + } else { + Temp dst = bld.tmp(dst_rc); + emit_extract_vector(ctx, src, idx, dst); + return dst; + } +} + +void emit_split_vector(isel_context* ctx, Temp vec_src, unsigned num_components) +{ + if (num_components == 1) + return; + if (ctx->allocated_vec.find(vec_src.id()) != ctx->allocated_vec.end()) + return; + aco_ptr split{create_instruction(aco_opcode::p_split_vector, Format::PSEUDO, 1, num_components)}; + split->operands[0] = Operand(vec_src); + std::array elems; + for (unsigned i = 0; i < num_components; i++) { + elems[i] = {ctx->program->allocateId(), RegClass(vec_src.type(), vec_src.size() / num_components)}; + split->definitions[i] = Definition(elems[i]); + } + ctx->block->instructions.emplace_back(std::move(split)); + ctx->allocated_vec.emplace(vec_src.id(), elems); +} + +/* This vector expansion uses a mask to determine which elements in the new vector + * come from the original vector. The other elements are undefined. */ +void expand_vector(isel_context* ctx, Temp vec_src, Temp dst, unsigned num_components, unsigned mask) +{ + emit_split_vector(ctx, vec_src, util_bitcount(mask)); + + if (vec_src == dst) + return; + + Builder bld(ctx->program, ctx->block); + if (num_components == 1) { + if (dst.type() == RegType::sgpr) + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), vec_src); + else + bld.copy(Definition(dst), vec_src); + return; + } + + unsigned component_size = dst.size() / num_components; + std::array elems; + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, num_components, 1)}; + vec->definitions[0] = Definition(dst); + unsigned k = 0; + for (unsigned i = 0; i < num_components; i++) { + if (mask & (1 << i)) { + Temp src = emit_extract_vector(ctx, vec_src, k++, RegClass(vec_src.type(), component_size)); + if (dst.type() == RegType::sgpr) + src = bld.as_uniform(src); + vec->operands[i] = Operand(src); + } else { + vec->operands[i] = Operand(0u); + } + elems[i] = vec->operands[i].getTemp(); + } + ctx->block->instructions.emplace_back(std::move(vec)); + ctx->allocated_vec.emplace(dst.id(), elems); +} + +Temp as_divergent_bool(isel_context *ctx, Temp val, bool vcc_hint) +{ + if (val.regClass() == s2) { + return val; + } else { + assert(val.regClass() == s1); + Builder bld(ctx->program, ctx->block); + Definition& def = bld.sop2(aco_opcode::s_cselect_b64, bld.def(s2), + Operand((uint32_t) -1), Operand(0u), bld.scc(val)).def(0); + if (vcc_hint) + def.setHint(vcc); + return def.getTemp(); + } +} + +Temp as_uniform_bool(isel_context *ctx, Temp val) +{ + if (val.regClass() == s1) { + return val; + } else { + assert(val.regClass() == s2); + Builder bld(ctx->program, ctx->block); + return bld.sopc(aco_opcode::s_cmp_lg_u64, bld.def(s1, scc), Operand(0u), Operand(val)); + } +} + +Temp get_alu_src(struct isel_context *ctx, nir_alu_src src, unsigned size=1) +{ + if (src.src.ssa->num_components == 1 && src.swizzle[0] == 0 && size == 1) + return get_ssa_temp(ctx, src.src.ssa); + + if (src.src.ssa->num_components == size) { + bool identity_swizzle = true; + for (unsigned i = 0; identity_swizzle && i < size; i++) { + if (src.swizzle[i] != i) + identity_swizzle = false; + } + if (identity_swizzle) + return get_ssa_temp(ctx, src.src.ssa); + } + + Temp vec = get_ssa_temp(ctx, src.src.ssa); + unsigned elem_size = vec.size() / src.src.ssa->num_components; + assert(elem_size > 0); /* TODO: 8 and 16-bit vectors not supported */ + assert(vec.size() % elem_size == 0); + + RegClass elem_rc = RegClass(vec.type(), elem_size); + if (size == 1) { + return emit_extract_vector(ctx, vec, src.swizzle[0], elem_rc); + } else { + assert(size <= 4); + std::array elems; + aco_ptr vec_instr{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, size, 1)}; + for (unsigned i = 0; i < size; ++i) { + elems[i] = emit_extract_vector(ctx, vec, src.swizzle[i], elem_rc); + vec_instr->operands[i] = Operand{elems[i]}; + } + Temp dst{ctx->program->allocateId(), RegClass(vec.type(), elem_size * size)}; + vec_instr->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec_instr)); + ctx->allocated_vec.emplace(dst.id(), elems); + return dst; + } +} + +Temp convert_pointer_to_64_bit(isel_context *ctx, Temp ptr) +{ + if (ptr.size() == 2) + return ptr; + Builder bld(ctx->program, ctx->block); + return bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), + ptr, Operand((unsigned)ctx->options->address32_hi)); +} + +void emit_sop2_instruction(isel_context *ctx, nir_alu_instr *instr, aco_opcode op, Temp dst, bool writes_scc) +{ + aco_ptr sop2{create_instruction(op, Format::SOP2, 2, writes_scc ? 2 : 1)}; + sop2->operands[0] = Operand(get_alu_src(ctx, instr->src[0])); + sop2->operands[1] = Operand(get_alu_src(ctx, instr->src[1])); + sop2->definitions[0] = Definition(dst); + if (writes_scc) + sop2->definitions[1] = Definition(ctx->program->allocateId(), scc, s1); + ctx->block->instructions.emplace_back(std::move(sop2)); +} + +void emit_vop2_instruction(isel_context *ctx, nir_alu_instr *instr, aco_opcode op, Temp dst, bool commutative, bool swap_srcs=false) +{ + Builder bld(ctx->program, ctx->block); + Temp src0 = get_alu_src(ctx, instr->src[swap_srcs ? 1 : 0]); + Temp src1 = get_alu_src(ctx, instr->src[swap_srcs ? 0 : 1]); + if (src1.type() == RegType::sgpr) { + if (commutative && src0.type() == RegType::vgpr) { + Temp t = src0; + src0 = src1; + src1 = t; + } else if (src0.type() == RegType::vgpr && + op != aco_opcode::v_madmk_f32 && + op != aco_opcode::v_madak_f32 && + op != aco_opcode::v_madmk_f16 && + op != aco_opcode::v_madak_f16) { + /* If the instruction is not commutative, we emit a VOP3A instruction */ + bld.vop2_e64(op, Definition(dst), src0, src1); + return; + } else { + src1 = bld.copy(bld.def(RegType::vgpr, src1.size()), src1); //TODO: as_vgpr + } + } + bld.vop2(op, Definition(dst), src0, src1); +} + +void emit_vop3a_instruction(isel_context *ctx, nir_alu_instr *instr, aco_opcode op, Temp dst) +{ + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + Temp src2 = get_alu_src(ctx, instr->src[2]); + + /* ensure that the instruction has at most 1 sgpr operand + * The optimizer will inline constants for us */ + if (src0.type() == RegType::sgpr && src1.type() == RegType::sgpr) + src0 = as_vgpr(ctx, src0); + if (src1.type() == RegType::sgpr && src2.type() == RegType::sgpr) + src1 = as_vgpr(ctx, src1); + if (src2.type() == RegType::sgpr && src0.type() == RegType::sgpr) + src2 = as_vgpr(ctx, src2); + + Builder bld(ctx->program, ctx->block); + bld.vop3(op, Definition(dst), src0, src1, src2); +} + +void emit_vop1_instruction(isel_context *ctx, nir_alu_instr *instr, aco_opcode op, Temp dst) +{ + Builder bld(ctx->program, ctx->block); + bld.vop1(op, Definition(dst), get_alu_src(ctx, instr->src[0])); +} + +void emit_vopc_instruction(isel_context *ctx, nir_alu_instr *instr, aco_opcode op, Temp dst) +{ + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + aco_ptr vopc; + if (src1.type() == RegType::sgpr) { + if (src0.type() == RegType::vgpr) { + /* to swap the operands, we might also have to change the opcode */ + switch (op) { + case aco_opcode::v_cmp_lt_f32: + op = aco_opcode::v_cmp_gt_f32; + break; + case aco_opcode::v_cmp_ge_f32: + op = aco_opcode::v_cmp_le_f32; + break; + case aco_opcode::v_cmp_lt_i32: + op = aco_opcode::v_cmp_gt_i32; + break; + case aco_opcode::v_cmp_ge_i32: + op = aco_opcode::v_cmp_le_i32; + break; + case aco_opcode::v_cmp_lt_u32: + op = aco_opcode::v_cmp_gt_u32; + break; + case aco_opcode::v_cmp_ge_u32: + op = aco_opcode::v_cmp_le_u32; + break; + case aco_opcode::v_cmp_lt_f64: + op = aco_opcode::v_cmp_gt_f64; + break; + case aco_opcode::v_cmp_ge_f64: + op = aco_opcode::v_cmp_le_f64; + break; + case aco_opcode::v_cmp_lt_i64: + op = aco_opcode::v_cmp_gt_i64; + break; + case aco_opcode::v_cmp_ge_i64: + op = aco_opcode::v_cmp_le_i64; + break; + case aco_opcode::v_cmp_lt_u64: + op = aco_opcode::v_cmp_gt_u64; + break; + case aco_opcode::v_cmp_ge_u64: + op = aco_opcode::v_cmp_le_u64; + break; + default: /* eq and ne are commutative */ + break; + } + Temp t = src0; + src0 = src1; + src1 = t; + } else { + src1 = as_vgpr(ctx, src1); + } + } + Builder bld(ctx->program, ctx->block); + bld.vopc(op, Definition(dst), src0, src1).def(0).setHint(vcc); +} + +void emit_comparison(isel_context *ctx, nir_alu_instr *instr, aco_opcode op, Temp dst) +{ + if (dst.regClass() == s2) { + emit_vopc_instruction(ctx, instr, op, dst); + if (!ctx->divergent_vals[instr->dest.dest.ssa.index]) + emit_split_vector(ctx, dst, 2); + } else if (dst.regClass() == s1) { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + assert(src0.type() == RegType::sgpr && src1.type() == RegType::sgpr); + + Builder bld(ctx->program, ctx->block); + bld.sopc(op, bld.scc(Definition(dst)), src0, src1); + + } else { + assert(false); + } +} + +void emit_boolean_logic(isel_context *ctx, nir_alu_instr *instr, aco_opcode op32, aco_opcode op64, Temp dst) +{ + Builder bld(ctx->program, ctx->block); + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + if (dst.regClass() == s2) { + bld.sop2(op64, Definition(dst), bld.def(s1, scc), + as_divergent_bool(ctx, src0, false), as_divergent_bool(ctx, src1, false)); + } else { + assert(dst.regClass() == s1); + bld.sop2(op32, bld.def(s1), bld.scc(Definition(dst)), + as_uniform_bool(ctx, src0), as_uniform_bool(ctx, src1)); + } +} + + +void emit_bcsel(isel_context *ctx, nir_alu_instr *instr, Temp dst) +{ + Builder bld(ctx->program, ctx->block); + Temp cond = get_alu_src(ctx, instr->src[0]); + Temp then = get_alu_src(ctx, instr->src[1]); + Temp els = get_alu_src(ctx, instr->src[2]); + + if (dst.type() == RegType::vgpr) { + cond = as_divergent_bool(ctx, cond, true); + + aco_ptr bcsel; + if (dst.size() == 1) { + then = as_vgpr(ctx, then); + els = as_vgpr(ctx, els); + + bld.vop2(aco_opcode::v_cndmask_b32, Definition(dst), els, then, cond); + } else if (dst.size() == 2) { + Temp then_lo = bld.tmp(v1), then_hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(then_lo), Definition(then_hi), then); + Temp else_lo = bld.tmp(v1), else_hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(else_lo), Definition(else_hi), els); + + Temp dst0 = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), else_lo, then_lo, cond); + Temp dst1 = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), else_hi, then_hi, cond); + + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), dst0, dst1); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + return; + } + + if (instr->dest.dest.ssa.bit_size != 1) { /* uniform condition and values in sgpr */ + if (dst.regClass() == s1 || dst.regClass() == s2) { + assert((then.regClass() == s1 || then.regClass() == s2) && els.regClass() == then.regClass()); + aco_opcode op = dst.regClass() == s1 ? aco_opcode::s_cselect_b32 : aco_opcode::s_cselect_b64; + bld.sop2(op, Definition(dst), then, els, bld.scc(as_uniform_bool(ctx, cond))); + } else { + fprintf(stderr, "Unimplemented uniform bcsel bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + return; + } + + /* boolean bcsel */ + assert(instr->dest.dest.ssa.bit_size == 1); + + if (dst.regClass() == s1) + cond = as_uniform_bool(ctx, cond); + + if (cond.regClass() == s1) { /* uniform selection */ + aco_opcode op; + if (dst.regClass() == s2) { + op = aco_opcode::s_cselect_b64; + then = as_divergent_bool(ctx, then, false); + els = as_divergent_bool(ctx, els, false); + } else { + assert(dst.regClass() == s1); + op = aco_opcode::s_cselect_b32; + then = as_uniform_bool(ctx, then); + els = as_uniform_bool(ctx, els); + } + bld.sop2(op, Definition(dst), then, els, bld.scc(cond)); + return; + } + + /* divergent boolean bcsel + * this implements bcsel on bools: dst = s0 ? s1 : s2 + * are going to be: dst = (s0 & s1) | (~s0 & s2) */ + assert (dst.regClass() == s2); + then = as_divergent_bool(ctx, then, false); + els = as_divergent_bool(ctx, els, false); + + if (cond.id() != then.id()) + then = bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), cond, then); + + if (cond.id() == els.id()) + bld.sop1(aco_opcode::s_mov_b64, Definition(dst), then); + else + bld.sop2(aco_opcode::s_or_b64, Definition(dst), bld.def(s1, scc), then, + bld.sop2(aco_opcode::s_andn2_b64, bld.def(s2), bld.def(s1, scc), els, cond)); +} + +void visit_alu_instr(isel_context *ctx, nir_alu_instr *instr) +{ + if (!instr->dest.dest.is_ssa) { + fprintf(stderr, "nir alu dst not in ssa: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + abort(); + } + Builder bld(ctx->program, ctx->block); + Temp dst = get_ssa_temp(ctx, &instr->dest.dest.ssa); + switch(instr->op) { + case nir_op_vec2: + case nir_op_vec3: + case nir_op_vec4: { + std::array elems; + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, instr->dest.dest.ssa.num_components, 1)}; + for (unsigned i = 0; i < instr->dest.dest.ssa.num_components; ++i) { + elems[i] = get_alu_src(ctx, instr->src[i]); + vec->operands[i] = Operand{elems[i]}; + } + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + ctx->allocated_vec.emplace(dst.id(), elems); + break; + } + case nir_op_mov: { + Temp src = get_alu_src(ctx, instr->src[0]); + aco_ptr mov; + if (dst.type() == RegType::sgpr) { + if (src.type() == RegType::vgpr) + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), src); + else if (src.regClass() == s1) + bld.sop1(aco_opcode::s_mov_b32, Definition(dst), src); + else if (src.regClass() == s2) + bld.sop1(aco_opcode::s_mov_b64, Definition(dst), src); + else + unreachable("wrong src register class for nir_op_imov"); + } else if (dst.regClass() == v1) { + bld.vop1(aco_opcode::v_mov_b32, Definition(dst), src); + } else if (dst.regClass() == v2) { + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), src); + } else { + nir_print_instr(&instr->instr, stderr); + unreachable("Should have been lowered to scalar."); + } + break; + } + case nir_op_inot: { + Temp src = get_alu_src(ctx, instr->src[0]); + /* uniform booleans */ + if (instr->dest.dest.ssa.bit_size == 1 && dst.regClass() == s1) { + if (src.regClass() == s1) { + /* in this case, src is either 1 or 0 */ + bld.sop2(aco_opcode::s_xor_b32, bld.def(s1), bld.scc(Definition(dst)), Operand(1u), src); + } else { + /* src is either exec_mask or 0 */ + assert(src.regClass() == s2); + bld.sopc(aco_opcode::s_cmp_eq_u64, bld.scc(Definition(dst)), Operand(0u), src); + } + } else if (dst.regClass() == v1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_not_b32, dst); + } else if (dst.type() == RegType::sgpr) { + aco_opcode opcode = dst.size() == 1 ? aco_opcode::s_not_b32 : aco_opcode::s_not_b64; + bld.sop1(opcode, Definition(dst), bld.def(s1, scc), src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ineg: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.regClass() == v1) { + bld.vsub32(Definition(dst), Operand(0u), Operand(src)); + } else if (dst.regClass() == s1) { + bld.sop2(aco_opcode::s_mul_i32, Definition(dst), Operand((uint32_t) -1), src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_iabs: { + if (dst.regClass() == s1) { + bld.sop1(aco_opcode::s_abs_i32, Definition(dst), bld.def(s1, scc), get_alu_src(ctx, instr->src[0])); + } else if (dst.regClass() == v1) { + Temp src = get_alu_src(ctx, instr->src[0]); + bld.vop2(aco_opcode::v_max_i32, Definition(dst), src, bld.vsub32(bld.def(v1), Operand(0u), src)); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_isign: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.regClass() == s1) { + Temp tmp = bld.sop2(aco_opcode::s_ashr_i32, bld.def(s1), bld.def(s1, scc), src, Operand(31u)); + Temp gtz = bld.sopc(aco_opcode::s_cmp_gt_i32, bld.def(s1, scc), src, Operand(0u)); + bld.sop2(aco_opcode::s_add_i32, Definition(dst), bld.def(s1, scc), gtz, tmp); + } else if (dst.regClass() == s2) { + Temp neg = bld.sop2(aco_opcode::s_ashr_i64, bld.def(s2), bld.def(s1, scc), src, Operand(63u)); + Temp neqz = bld.sopc(aco_opcode::s_cmp_lg_u64, bld.def(s1, scc), src, Operand(0u)); + bld.sop2(aco_opcode::s_or_b64, Definition(dst), bld.def(s1, scc), neg, neqz); + } else if (dst.regClass() == v1) { + Temp tmp = bld.vop2(aco_opcode::v_ashrrev_i32, bld.def(v1), Operand(31u), src); + Temp gtz = bld.vopc(aco_opcode::v_cmp_ge_i32, bld.hint_vcc(bld.def(s2)), Operand(0u), src); + bld.vop2(aco_opcode::v_cndmask_b32, Definition(dst), Operand(1u), tmp, gtz); + } else if (dst.regClass() == v2) { + Temp upper = emit_extract_vector(ctx, src, 1, v1); + Temp neg = bld.vop2(aco_opcode::v_ashrrev_i32, bld.def(v1), Operand(31u), upper); + Temp gtz = bld.vopc(aco_opcode::v_cmp_ge_i64, bld.hint_vcc(bld.def(s2)), Operand(0u), src); + Temp lower = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(1u), neg, gtz); + upper = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0u), neg, gtz); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_imax: { + if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_max_i32, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_max_i32, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_umax: { + if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_max_u32, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_max_u32, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_imin: { + if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_min_i32, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_min_i32, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_umin: { + if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_min_u32, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_min_u32, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ior: { + if (instr->dest.dest.ssa.bit_size == 1) { + emit_boolean_logic(ctx, instr, aco_opcode::s_or_b32, aco_opcode::s_or_b64, dst); + } else if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_or_b32, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_or_b32, dst, true); + } else if (dst.regClass() == s2) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_or_b64, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_iand: { + if (instr->dest.dest.ssa.bit_size == 1) { + emit_boolean_logic(ctx, instr, aco_opcode::s_and_b32, aco_opcode::s_and_b64, dst); + } else if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_and_b32, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_and_b32, dst, true); + } else if (dst.regClass() == s2) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_and_b64, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ixor: { + if (instr->dest.dest.ssa.bit_size == 1) { + emit_boolean_logic(ctx, instr, aco_opcode::s_xor_b32, aco_opcode::s_xor_b64, dst); + } else if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_xor_b32, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_xor_b32, dst, true); + } else if (dst.regClass() == s2) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_xor_b64, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ushr: { + if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_lshrrev_b32, dst, false, true); + } else if (dst.regClass() == v2) { + bld.vop3(aco_opcode::v_lshrrev_b64, Definition(dst), + get_alu_src(ctx, instr->src[1]), get_alu_src(ctx, instr->src[0])); + } else if (dst.regClass() == s2) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_lshr_b64, dst, true); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_lshr_b32, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ishl: { + if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_lshlrev_b32, dst, false, true); + } else if (dst.regClass() == v2) { + bld.vop3(aco_opcode::v_lshlrev_b64, Definition(dst), + get_alu_src(ctx, instr->src[1]), get_alu_src(ctx, instr->src[0])); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_lshl_b32, dst, true); + } else if (dst.regClass() == s2) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_lshl_b64, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ishr: { + if (dst.regClass() == v1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_ashrrev_i32, dst, false, true); + } else if (dst.regClass() == v2) { + bld.vop3(aco_opcode::v_ashrrev_i64, Definition(dst), + get_alu_src(ctx, instr->src[1]), get_alu_src(ctx, instr->src[0])); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_ashr_i32, dst, true); + } else if (dst.regClass() == s2) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_ashr_i64, dst, true); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_find_lsb: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (src.regClass() == s1) { + bld.sop1(aco_opcode::s_ff1_i32_b32, Definition(dst), src); + } else if (src.regClass() == v1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_ffbl_b32, dst); + } else if (src.regClass() == s2) { + bld.sop1(aco_opcode::s_ff1_i32_b64, Definition(dst), src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ufind_msb: + case nir_op_ifind_msb: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (src.regClass() == s1 || src.regClass() == s2) { + aco_opcode op = src.regClass() == s2 ? + (instr->op == nir_op_ufind_msb ? aco_opcode::s_flbit_i32_b64 : aco_opcode::s_flbit_i32_i64) : + (instr->op == nir_op_ufind_msb ? aco_opcode::s_flbit_i32_b32 : aco_opcode::s_flbit_i32); + Temp msb_rev = bld.sop1(op, bld.def(s1), src); + + Builder::Result sub = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.def(s1, scc), + Operand(src.size() * 32u - 1u), msb_rev); + Temp msb = sub.def(0).getTemp(); + Temp carry = sub.def(1).getTemp(); + + bld.sop2(aco_opcode::s_cselect_b32, Definition(dst), Operand((uint32_t)-1), msb, carry); + } else if (src.regClass() == v1) { + aco_opcode op = instr->op == nir_op_ufind_msb ? aco_opcode::v_ffbh_u32 : aco_opcode::v_ffbh_i32; + Temp msb_rev = bld.tmp(v1); + emit_vop1_instruction(ctx, instr, op, msb_rev); + Temp msb = bld.tmp(v1); + Temp carry = bld.vsub32(Definition(msb), Operand(31u), Operand(msb_rev), true).def(1).getTemp(); + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(dst), msb, Operand((uint32_t)-1), carry); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_bitfield_reverse: { + if (dst.regClass() == s1) { + bld.sop1(aco_opcode::s_brev_b32, Definition(dst), get_alu_src(ctx, instr->src[0])); + } else if (dst.regClass() == v1) { + bld.vop1(aco_opcode::v_bfrev_b32, Definition(dst), get_alu_src(ctx, instr->src[0])); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_iadd: { + if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_add_u32, dst, true); + break; + } + + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + if (dst.regClass() == v1) { + bld.vadd32(Definition(dst), Operand(src0), Operand(src1)); + break; + } + + assert(src0.size() == 2 && src1.size() == 2); + Temp src00 = bld.tmp(src0.type(), 1); + Temp src01 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src00), Definition(src01), src0); + Temp src10 = bld.tmp(src1.type(), 1); + Temp src11 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src10), Definition(src11), src1); + + if (dst.regClass() == s2) { + Temp carry = bld.tmp(s1); + Temp dst0 = bld.sop2(aco_opcode::s_add_u32, bld.def(s1), bld.scc(Definition(carry)), src00, src10); + Temp dst1 = bld.sop2(aco_opcode::s_addc_u32, bld.def(s1), bld.def(s1, scc), src01, src11, bld.scc(carry)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), dst0, dst1); + } else if (dst.regClass() == v2) { + Temp dst0 = bld.tmp(v1); + Temp carry = bld.vadd32(Definition(dst0), src00, src10, true).def(1).getTemp(); + Temp dst1 = bld.vadd32(bld.def(v1), src01, src11, false, carry); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), dst0, dst1); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_uadd_sat: { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + if (dst.regClass() == s1) { + Temp tmp = bld.tmp(s1), carry = bld.tmp(s1); + bld.sop2(aco_opcode::s_add_u32, Definition(tmp), bld.scc(Definition(carry)), + src0, src1); + bld.sop2(aco_opcode::s_cselect_b32, Definition(dst), Operand((uint32_t) -1), tmp, bld.scc(carry)); + } else if (dst.regClass() == v1) { + if (ctx->options->chip_class >= GFX9) { + aco_ptr add{create_instruction(aco_opcode::v_add_u32, asVOP3(Format::VOP2), 2, 1)}; + add->operands[0] = Operand(src0); + add->operands[1] = Operand(src1); + add->definitions[0] = Definition(dst); + add->clamp = 1; + ctx->block->instructions.emplace_back(std::move(add)); + } else { + if (src1.regClass() != v1) + std::swap(src0, src1); + assert(src1.regClass() == v1); + Temp tmp = bld.tmp(v1); + Temp carry = bld.vadd32(Definition(tmp), src0, src1, true).def(1).getTemp(); + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(dst), tmp, Operand((uint32_t) -1), carry); + } + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_uadd_carry: { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + if (dst.regClass() == s1) { + bld.sop2(aco_opcode::s_add_u32, bld.def(s1), bld.scc(Definition(dst)), src0, src1); + break; + } + if (dst.regClass() == v1) { + Temp carry = bld.vadd32(bld.def(v1), src0, src1, true).def(1).getTemp(); + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(dst), Operand(0u), Operand(1u), carry); + break; + } + + Temp src00 = bld.tmp(src0.type(), 1); + Temp src01 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src00), Definition(src01), src0); + Temp src10 = bld.tmp(src1.type(), 1); + Temp src11 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src10), Definition(src11), src1); + if (dst.regClass() == s2) { + Temp carry = bld.tmp(s1); + bld.sop2(aco_opcode::s_add_u32, bld.def(s1), bld.scc(Definition(carry)), src00, src10); + carry = bld.sop2(aco_opcode::s_addc_u32, bld.def(s1), bld.scc(bld.def(s1)), src01, src11, bld.scc(carry)).def(1).getTemp(); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), carry, Operand(0u)); + } else if (dst.regClass() == v2) { + Temp carry = bld.vadd32(bld.def(v1), src00, src10, true).def(1).getTemp(); + carry = bld.vadd32(bld.def(v1), src01, src11, true, carry).def(1).getTemp(); + carry = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0u), Operand(1u), carry); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), carry, Operand(0u)); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_isub: { + if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_sub_i32, dst, true); + break; + } + + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + if (dst.regClass() == v1) { + bld.vsub32(Definition(dst), src0, src1); + break; + } + + Temp src00 = bld.tmp(src0.type(), 1); + Temp src01 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src00), Definition(src01), src0); + Temp src10 = bld.tmp(src1.type(), 1); + Temp src11 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src10), Definition(src11), src1); + if (dst.regClass() == s2) { + Temp carry = bld.tmp(s1); + Temp dst0 = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.scc(Definition(carry)), src00, src10); + Temp dst1 = bld.sop2(aco_opcode::s_subb_u32, bld.def(s1), bld.def(s1, scc), src01, src11, carry); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), dst0, dst1); + } else if (dst.regClass() == v2) { + Temp lower = bld.tmp(v1); + Temp borrow = bld.vsub32(Definition(lower), src00, src10, true).def(1).getTemp(); + Temp upper = bld.vsub32(bld.def(v1), src01, src11, false, borrow); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_usub_borrow: { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + if (dst.regClass() == s1) { + bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.scc(Definition(dst)), src0, src1); + break; + } else if (dst.regClass() == v1) { + Temp borrow = bld.vsub32(bld.def(v1), src0, src1, true).def(1).getTemp(); + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(dst), Operand(0u), Operand(1u), borrow); + break; + } + + Temp src00 = bld.tmp(src0.type(), 1); + Temp src01 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src00), Definition(src01), src0); + Temp src10 = bld.tmp(src1.type(), 1); + Temp src11 = bld.tmp(dst.type(), 1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src10), Definition(src11), src1); + if (dst.regClass() == s2) { + Temp borrow = bld.tmp(s1); + bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.scc(Definition(borrow)), src00, src10); + borrow = bld.sop2(aco_opcode::s_subb_u32, bld.def(s1), bld.scc(bld.def(s1)), src01, src11, bld.scc(borrow)).def(1).getTemp(); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), borrow, Operand(0u)); + } else if (dst.regClass() == v2) { + Temp borrow = bld.vsub32(bld.def(v1), src00, src10, true).def(1).getTemp(); + borrow = bld.vsub32(bld.def(v1), src01, src11, true, Operand(borrow)).def(1).getTemp(); + borrow = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0u), Operand(1u), borrow); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), borrow, Operand(0u)); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_imul: { + if (dst.regClass() == v1) { + bld.vop3(aco_opcode::v_mul_lo_u32, Definition(dst), + get_alu_src(ctx, instr->src[0]), get_alu_src(ctx, instr->src[1])); + } else if (dst.regClass() == s1) { + emit_sop2_instruction(ctx, instr, aco_opcode::s_mul_i32, dst, false); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_umul_high: { + if (dst.regClass() == v1) { + bld.vop3(aco_opcode::v_mul_hi_u32, Definition(dst), get_alu_src(ctx, instr->src[0]), get_alu_src(ctx, instr->src[1])); + } else if (dst.regClass() == s1 && ctx->options->chip_class >= GFX9) { + bld.sop2(aco_opcode::s_mul_hi_u32, Definition(dst), get_alu_src(ctx, instr->src[0]), get_alu_src(ctx, instr->src[1])); + } else if (dst.regClass() == s1) { + Temp tmp = bld.vop3(aco_opcode::v_mul_hi_u32, bld.def(v1), get_alu_src(ctx, instr->src[0]), + as_vgpr(ctx, get_alu_src(ctx, instr->src[1]))); + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), tmp); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_imul_high: { + if (dst.regClass() == v1) { + bld.vop3(aco_opcode::v_mul_hi_i32, Definition(dst), get_alu_src(ctx, instr->src[0]), get_alu_src(ctx, instr->src[1])); + } else if (dst.regClass() == s1 && ctx->options->chip_class >= GFX9) { + bld.sop2(aco_opcode::s_mul_hi_i32, Definition(dst), get_alu_src(ctx, instr->src[0]), get_alu_src(ctx, instr->src[1])); + } else if (dst.regClass() == s1) { + Temp tmp = bld.vop3(aco_opcode::v_mul_hi_i32, bld.def(v1), get_alu_src(ctx, instr->src[0]), + as_vgpr(ctx, get_alu_src(ctx, instr->src[1]))); + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), tmp); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fmul: { + if (dst.size() == 1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_mul_f32, dst, true); + } else if (dst.size() == 2) { + bld.vop3(aco_opcode::v_mul_f64, Definition(dst), get_alu_src(ctx, instr->src[0]), + as_vgpr(ctx, get_alu_src(ctx, instr->src[1]))); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fadd: { + if (dst.size() == 1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_add_f32, dst, true); + } else if (dst.size() == 2) { + bld.vop3(aco_opcode::v_add_f64, Definition(dst), get_alu_src(ctx, instr->src[0]), + as_vgpr(ctx, get_alu_src(ctx, instr->src[1]))); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fsub: { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + if (dst.size() == 1) { + if (src1.type() == RegType::vgpr || src0.type() != RegType::vgpr) + emit_vop2_instruction(ctx, instr, aco_opcode::v_sub_f32, dst, false); + else + emit_vop2_instruction(ctx, instr, aco_opcode::v_subrev_f32, dst, true); + } else if (dst.size() == 2) { + Instruction* add = bld.vop3(aco_opcode::v_add_f64, Definition(dst), + get_alu_src(ctx, instr->src[0]), + as_vgpr(ctx, get_alu_src(ctx, instr->src[1]))); + VOP3A_instruction* sub = static_cast(add); + sub->neg[1] = true; + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fmod: + case nir_op_frem: { + if (dst.size() == 1) { + Temp rcp = bld.vop1(aco_opcode::v_rcp_f32, bld.def(v1), get_alu_src(ctx, instr->src[1])); + Temp mul = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), get_alu_src(ctx, instr->src[0]), rcp); + + aco_opcode op = instr->op == nir_op_fmod ? aco_opcode::v_floor_f32 : aco_opcode::v_trunc_f32; + Temp floor = bld.vop1(op, bld.def(v1), mul); + + mul = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), get_alu_src(ctx, instr->src[1]), floor); + bld.vop2(aco_opcode::v_sub_f32, Definition(dst), get_alu_src(ctx, instr->src[0]), mul); + } else if (dst.size() == 2) { + Temp rcp = bld.vop1(aco_opcode::v_rcp_f64, bld.def(v2), get_alu_src(ctx, instr->src[1])); + Temp mul = bld.vop3(aco_opcode::v_mul_f64, bld.def(v2), get_alu_src(ctx, instr->src[0]), rcp); + + aco_opcode op = instr->op == nir_op_fmod ? aco_opcode::v_floor_f64 : aco_opcode::v_trunc_f64; + Temp floor = bld.vop1(op, bld.def(v1), mul); + + mul = bld.vop3(aco_opcode::v_mul_f64, bld.def(v2), get_alu_src(ctx, instr->src[1]), floor); + Instruction* add = bld.vop3(aco_opcode::v_add_f64, Definition(dst), get_alu_src(ctx, instr->src[0]), mul); + VOP3A_instruction* sub = static_cast(add); + sub->neg[1] = true; + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fmax: { + if (dst.size() == 1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_max_f32, dst, true); + } else if (dst.size() == 2) { + bld.vop3(aco_opcode::v_max_f64, Definition(dst), + get_alu_src(ctx, instr->src[0]), + as_vgpr(ctx, get_alu_src(ctx, instr->src[1]))); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fmin: { + if (dst.size() == 1) { + emit_vop2_instruction(ctx, instr, aco_opcode::v_min_f32, dst, true); + } else if (dst.size() == 2) { + bld.vop3(aco_opcode::v_min_f64, Definition(dst), + get_alu_src(ctx, instr->src[0]), + as_vgpr(ctx, get_alu_src(ctx, instr->src[1]))); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fmax3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_max3_f32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fmin3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_min3_f32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fmed3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_med3_f32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_umax3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_max3_u32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_umin3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_min3_u32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_umed3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_med3_u32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_imax3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_max3_i32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_imin3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_min3_i32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_imed3: { + if (dst.size() == 1) { + emit_vop3a_instruction(ctx, instr, aco_opcode::v_med3_i32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_cube_face_coord: { + Temp in = get_alu_src(ctx, instr->src[0], 3); + Temp src[3] = { emit_extract_vector(ctx, in, 0, v1), + emit_extract_vector(ctx, in, 1, v1), + emit_extract_vector(ctx, in, 2, v1) }; + Temp ma = bld.vop3(aco_opcode::v_cubema_f32, bld.def(v1), src[0], src[1], src[2]); + ma = bld.vop1(aco_opcode::v_rcp_f32, bld.def(v1), ma); + Temp sc = bld.vop3(aco_opcode::v_cubesc_f32, bld.def(v1), src[0], src[1], src[2]); + Temp tc = bld.vop3(aco_opcode::v_cubetc_f32, bld.def(v1), src[0], src[1], src[2]); + sc = bld.vop2(aco_opcode::v_madak_f32, bld.def(v1), sc, ma, Operand(0x3f000000u/*0.5*/)); + tc = bld.vop2(aco_opcode::v_madak_f32, bld.def(v1), tc, ma, Operand(0x3f000000u/*0.5*/)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), sc, tc); + break; + } + case nir_op_cube_face_index: { + Temp in = get_alu_src(ctx, instr->src[0], 3); + Temp src[3] = { emit_extract_vector(ctx, in, 0, v1), + emit_extract_vector(ctx, in, 1, v1), + emit_extract_vector(ctx, in, 2, v1) }; + bld.vop3(aco_opcode::v_cubeid_f32, Definition(dst), src[0], src[1], src[2]); + break; + } + case nir_op_bcsel: { + emit_bcsel(ctx, instr, dst); + break; + } + case nir_op_frsq: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_rsq_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_rsq_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fneg: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.size() == 1) { + bld.vop2(aco_opcode::v_xor_b32, Definition(dst), Operand(0x80000000u), as_vgpr(ctx, src)); + } else if (dst.size() == 2) { + Temp upper = bld.tmp(v1), lower = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), src); + upper = bld.vop2(aco_opcode::v_xor_b32, bld.def(v1), Operand(0x80000000u), upper); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fabs: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.size() == 1) { + bld.vop2(aco_opcode::v_and_b32, Definition(dst), Operand(0x7FFFFFFFu), as_vgpr(ctx, src)); + } else if (dst.size() == 2) { + Temp upper = bld.tmp(v1), lower = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), src); + upper = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(0x7FFFFFFFu), upper); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fsat: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.size() == 1) { + bld.vop3(aco_opcode::v_med3_f32, Definition(dst), Operand(0u), Operand(0x3f800000u), src); + } else if (dst.size() == 2) { + Instruction* add = bld.vop3(aco_opcode::v_add_f64, Definition(dst), src, Operand(0u)); + VOP3A_instruction* vop3 = static_cast(add); + vop3->clamp = true; + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_flog2: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_log_f32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_frcp: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_rcp_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_rcp_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fexp2: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_exp_f32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fsqrt: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_sqrt_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_sqrt_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ffract: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_fract_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_fract_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ffloor: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_floor_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_floor_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fceil: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_ceil_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_ceil_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ftrunc: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_trunc_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_trunc_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fround_even: { + if (dst.size() == 1) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_rndne_f32, dst); + } else if (dst.size() == 2) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_rndne_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fsin: + case nir_op_fcos: { + Temp src = get_alu_src(ctx, instr->src[0]); + aco_ptr norm; + if (dst.size() == 1) { + Temp tmp; + Operand half_pi(0x3e22f983u); + if (src.type() == RegType::sgpr) + tmp = bld.vop2_e64(aco_opcode::v_mul_f32, bld.def(v1), half_pi, src); + else + tmp = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), half_pi, src); + + /* before GFX9, v_sin_f32 and v_cos_f32 had a valid input domain of [-256, +256] */ + if (ctx->options->chip_class < GFX9) + tmp = bld.vop1(aco_opcode::v_fract_f32, bld.def(v1), tmp); + + aco_opcode opcode = instr->op == nir_op_fsin ? aco_opcode::v_sin_f32 : aco_opcode::v_cos_f32; + bld.vop1(opcode, Definition(dst), tmp); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ldexp: { + if (dst.size() == 1) { + bld.vop3(aco_opcode::v_ldexp_f32, Definition(dst), + as_vgpr(ctx, get_alu_src(ctx, instr->src[0])), + get_alu_src(ctx, instr->src[1])); + } else if (dst.size() == 2) { + bld.vop3(aco_opcode::v_ldexp_f64, Definition(dst), + as_vgpr(ctx, get_alu_src(ctx, instr->src[0])), + get_alu_src(ctx, instr->src[1])); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_frexp_sig: { + if (dst.size() == 1) { + bld.vop1(aco_opcode::v_frexp_mant_f32, Definition(dst), + get_alu_src(ctx, instr->src[0])); + } else if (dst.size() == 2) { + bld.vop1(aco_opcode::v_frexp_mant_f64, Definition(dst), + get_alu_src(ctx, instr->src[0])); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_frexp_exp: { + if (instr->src[0].src.ssa->bit_size == 32) { + bld.vop1(aco_opcode::v_frexp_exp_i32_f32, Definition(dst), + get_alu_src(ctx, instr->src[0])); + } else if (instr->src[0].src.ssa->bit_size == 64) { + bld.vop1(aco_opcode::v_frexp_exp_i32_f64, Definition(dst), + get_alu_src(ctx, instr->src[0])); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fsign: { + Temp src = as_vgpr(ctx, get_alu_src(ctx, instr->src[0])); + if (dst.size() == 1) { + Temp cond = bld.vopc(aco_opcode::v_cmp_nlt_f32, bld.hint_vcc(bld.def(s2)), Operand(0u), src); + src = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0x3f800000u), src, cond); + cond = bld.vopc(aco_opcode::v_cmp_le_f32, bld.hint_vcc(bld.def(s2)), Operand(0u), src); + bld.vop2(aco_opcode::v_cndmask_b32, Definition(dst), Operand(0xbf800000u), src, cond); + } else if (dst.size() == 2) { + Temp cond = bld.vopc(aco_opcode::v_cmp_nlt_f64, bld.hint_vcc(bld.def(s2)), Operand(0u), src); + Temp tmp = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand(0x3FF00000u)); + Temp upper = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), tmp, src, cond); + + cond = bld.vopc(aco_opcode::v_cmp_le_f64, bld.hint_vcc(bld.def(s2)), Operand(0u), src); + tmp = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand(0xBFF00000u)); + upper = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), tmp, upper, cond); + + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), Operand(0u), upper); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_f2f32: { + if (instr->src[0].src.ssa->bit_size == 64) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_cvt_f32_f64, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_f2f64: { + if (instr->src[0].src.ssa->bit_size == 32) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_cvt_f64_f32, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_i2f32: { + assert(dst.size() == 1); + emit_vop1_instruction(ctx, instr, aco_opcode::v_cvt_f32_i32, dst); + break; + } + case nir_op_i2f64: { + if (instr->src[0].src.ssa->bit_size == 32) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_cvt_f64_i32, dst); + } else if (instr->src[0].src.ssa->bit_size == 64) { + Temp src = get_alu_src(ctx, instr->src[0]); + RegClass rc = RegClass(src.type(), 1); + Temp lower = bld.tmp(rc), upper = bld.tmp(rc); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), src); + lower = bld.vop1(aco_opcode::v_cvt_f64_u32, bld.def(v2), lower); + upper = bld.vop1(aco_opcode::v_cvt_f64_i32, bld.def(v2), upper); + upper = bld.vop3(aco_opcode::v_ldexp_f64, bld.def(v2), upper, Operand(32u)); + bld.vop3(aco_opcode::v_add_f64, Definition(dst), lower, upper); + + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_u2f32: { + assert(dst.size() == 1); + emit_vop1_instruction(ctx, instr, aco_opcode::v_cvt_f32_u32, dst); + break; + } + case nir_op_u2f64: { + if (instr->src[0].src.ssa->bit_size == 32) { + emit_vop1_instruction(ctx, instr, aco_opcode::v_cvt_f64_u32, dst); + } else if (instr->src[0].src.ssa->bit_size == 64) { + Temp src = get_alu_src(ctx, instr->src[0]); + RegClass rc = RegClass(src.type(), 1); + Temp lower = bld.tmp(rc), upper = bld.tmp(rc); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), src); + lower = bld.vop1(aco_opcode::v_cvt_f64_u32, bld.def(v2), lower); + upper = bld.vop1(aco_opcode::v_cvt_f64_u32, bld.def(v2), upper); + upper = bld.vop3(aco_opcode::v_ldexp_f64, bld.def(v2), upper, Operand(32u)); + bld.vop3(aco_opcode::v_add_f64, Definition(dst), lower, upper); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_f2i32: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 32) { + if (dst.type() == RegType::vgpr) + bld.vop1(aco_opcode::v_cvt_i32_f32, Definition(dst), src); + else + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), + bld.vop1(aco_opcode::v_cvt_i32_f32, bld.def(v1), src)); + + } else if (instr->src[0].src.ssa->bit_size == 64) { + if (dst.type() == RegType::vgpr) + bld.vop1(aco_opcode::v_cvt_i32_f64, Definition(dst), src); + else + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), + bld.vop1(aco_opcode::v_cvt_i32_f64, bld.def(v1), src)); + + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_f2u32: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 32) { + if (dst.type() == RegType::vgpr) + bld.vop1(aco_opcode::v_cvt_u32_f32, Definition(dst), src); + else + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), + bld.vop1(aco_opcode::v_cvt_u32_f32, bld.def(v1), src)); + + } else if (instr->src[0].src.ssa->bit_size == 64) { + if (dst.type() == RegType::vgpr) + bld.vop1(aco_opcode::v_cvt_u32_f64, Definition(dst), src); + else + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), + bld.vop1(aco_opcode::v_cvt_u32_f64, bld.def(v1), src)); + + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_f2i64: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 32 && dst.type() == RegType::vgpr) { + Temp exponent = bld.vop1(aco_opcode::v_frexp_exp_i32_f32, bld.def(v1), src); + exponent = bld.vop3(aco_opcode::v_med3_i32, bld.def(v1), Operand(0x0u), exponent, Operand(64u)); + Temp mantissa = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(0x7fffffu), src); + Temp sign = bld.vop2(aco_opcode::v_ashrrev_i32, bld.def(v1), Operand(31u), src); + mantissa = bld.vop2(aco_opcode::v_or_b32, bld.def(v1), Operand(0x800000u), mantissa); + mantissa = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(7u), mantissa); + mantissa = bld.pseudo(aco_opcode::p_create_vector, bld.def(v2), Operand(0u), mantissa); + Temp new_exponent = bld.tmp(v1); + Temp borrow = bld.vsub32(Definition(new_exponent), Operand(63u), exponent, true).def(1).getTemp(); + mantissa = bld.vop3(aco_opcode::v_lshrrev_b64, bld.def(v2), new_exponent, mantissa); + Temp saturate = bld.vop1(aco_opcode::v_bfrev_b32, bld.def(v1), Operand(0xfffffffeu)); + Temp lower = bld.tmp(v1), upper = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), mantissa); + lower = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), lower, Operand(0xffffffffu), borrow); + upper = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), upper, saturate, borrow); + lower = bld.vop2(aco_opcode::v_xor_b32, bld.def(v1), sign, lower); + upper = bld.vop2(aco_opcode::v_xor_b32, bld.def(v1), sign, upper); + Temp new_lower = bld.tmp(v1); + borrow = bld.vsub32(Definition(new_lower), lower, sign, true).def(1).getTemp(); + Temp new_upper = bld.vsub32(bld.def(v1), upper, sign, false, borrow); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), new_lower, new_upper); + + } else if (instr->src[0].src.ssa->bit_size == 32 && dst.type() == RegType::sgpr) { + if (src.type() == RegType::vgpr) + src = bld.as_uniform(src); + Temp exponent = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), src, Operand(0x80017u)); + exponent = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.def(s1, scc), exponent, Operand(126u)); + exponent = bld.sop2(aco_opcode::s_max_u32, bld.def(s1), bld.def(s1, scc), Operand(0u), exponent); + exponent = bld.sop2(aco_opcode::s_min_u32, bld.def(s1), bld.def(s1, scc), Operand(64u), exponent); + Temp mantissa = bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), Operand(0x7fffffu), src); + Temp sign = bld.sop2(aco_opcode::s_ashr_i32, bld.def(s1), bld.def(s1, scc), src, Operand(31u)); + mantissa = bld.sop2(aco_opcode::s_or_b32, bld.def(s1), bld.def(s1, scc), Operand(0x800000u), mantissa); + mantissa = bld.sop2(aco_opcode::s_lshl_b32, bld.def(s1), bld.def(s1, scc), mantissa, Operand(7u)); + mantissa = bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), Operand(0u), mantissa); + exponent = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.def(s1, scc), Operand(63u), exponent); + mantissa = bld.sop2(aco_opcode::s_lshr_b64, bld.def(s2), bld.def(s1, scc), mantissa, exponent); + Temp cond = bld.sopc(aco_opcode::s_cmp_eq_u32, bld.def(s1, scc), exponent, Operand(0xffffffffu)); // exp >= 64 + Temp saturate = bld.sop1(aco_opcode::s_brev_b64, bld.def(s2), Operand(0xfffffffeu)); + mantissa = bld.sop2(aco_opcode::s_cselect_b64, bld.def(s2), saturate, mantissa, cond); + Temp lower = bld.tmp(s1), upper = bld.tmp(s1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), mantissa); + lower = bld.sop2(aco_opcode::s_xor_b32, bld.def(s1), bld.def(s1, scc), sign, lower); + upper = bld.sop2(aco_opcode::s_xor_b32, bld.def(s1), bld.def(s1, scc), sign, upper); + Temp borrow = bld.tmp(s1); + lower = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.scc(Definition(borrow)), lower, sign); + upper = bld.sop2(aco_opcode::s_subb_u32, bld.def(s1), bld.def(s1, scc), upper, sign, borrow); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + + } else if (instr->src[0].src.ssa->bit_size == 64) { + Temp vec = bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), Operand(0u), Operand(0x3df00000u)); + Temp trunc = bld.vop1(aco_opcode::v_trunc_f64, bld.def(v2), src); + Temp mul = bld.vop3(aco_opcode::v_mul_f64, bld.def(v2), trunc, vec); + vec = bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), Operand(0u), Operand(0xc1f00000u)); + Temp floor = bld.vop1(aco_opcode::v_floor_f64, bld.def(v2), mul); + Temp fma = bld.vop3(aco_opcode::v_fma_f64, bld.def(v2), floor, vec, trunc); + Temp lower = bld.vop1(aco_opcode::v_cvt_u32_f64, bld.def(v1), fma); + Temp upper = bld.vop1(aco_opcode::v_cvt_i32_f64, bld.def(v1), floor); + if (dst.type() == RegType::sgpr) { + lower = bld.as_uniform(lower); + upper = bld.as_uniform(upper); + } + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_f2u64: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 32 && dst.type() == RegType::vgpr) { + Temp exponent = bld.vop1(aco_opcode::v_frexp_exp_i32_f32, bld.def(v1), src); + Temp exponent_in_range = bld.vopc(aco_opcode::v_cmp_ge_i32, bld.hint_vcc(bld.def(s2)), Operand(64u), exponent); + exponent = bld.vop2(aco_opcode::v_max_i32, bld.def(v1), Operand(0x0u), exponent); + Temp mantissa = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(0x7fffffu), src); + mantissa = bld.vop2(aco_opcode::v_or_b32, bld.def(v1), Operand(0x800000u), mantissa); + Temp exponent_small = bld.vsub32(bld.def(v1), Operand(24u), exponent); + Temp small = bld.vop2(aco_opcode::v_lshrrev_b32, bld.def(v1), exponent_small, mantissa); + mantissa = bld.pseudo(aco_opcode::p_create_vector, bld.def(v2), Operand(0u), mantissa); + Temp new_exponent = bld.tmp(v1); + Temp cond_small = bld.vsub32(Definition(new_exponent), exponent, Operand(24u), true).def(1).getTemp(); + mantissa = bld.vop3(aco_opcode::v_lshlrev_b64, bld.def(v2), new_exponent, mantissa); + Temp lower = bld.tmp(v1), upper = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), mantissa); + lower = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), lower, small, cond_small); + upper = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), upper, Operand(0u), cond_small); + lower = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0xffffffffu), lower, exponent_in_range); + upper = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0xffffffffu), upper, exponent_in_range); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + + } else if (instr->src[0].src.ssa->bit_size == 32 && dst.type() == RegType::sgpr) { + if (src.type() == RegType::vgpr) + src = bld.as_uniform(src); + Temp exponent = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), src, Operand(0x80017u)); + exponent = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.def(s1, scc), exponent, Operand(126u)); + exponent = bld.sop2(aco_opcode::s_max_u32, bld.def(s1), bld.def(s1, scc), Operand(0u), exponent); + Temp mantissa = bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), Operand(0x7fffffu), src); + mantissa = bld.sop2(aco_opcode::s_or_b32, bld.def(s1), bld.def(s1, scc), Operand(0x800000u), mantissa); + Temp exponent_small = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.def(s1, scc), Operand(24u), exponent); + Temp small = bld.sop2(aco_opcode::s_lshr_b32, bld.def(s1), bld.def(s1, scc), mantissa, exponent_small); + mantissa = bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), Operand(0u), mantissa); + Temp exponent_large = bld.sop2(aco_opcode::s_sub_u32, bld.def(s1), bld.def(s1, scc), exponent, Operand(24u)); + mantissa = bld.sop2(aco_opcode::s_lshl_b64, bld.def(s2), bld.def(s1, scc), mantissa, exponent_large); + Temp cond = bld.sopc(aco_opcode::s_cmp_ge_i32, bld.def(s1, scc), Operand(64u), exponent); + mantissa = bld.sop2(aco_opcode::s_cselect_b64, bld.def(s2), mantissa, Operand(0xffffffffu), cond); + Temp lower = bld.tmp(s1), upper = bld.tmp(s1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lower), Definition(upper), mantissa); + Temp cond_small = bld.sopc(aco_opcode::s_cmp_le_i32, bld.def(s1, scc), exponent, Operand(24u)); + lower = bld.sop2(aco_opcode::s_cselect_b32, bld.def(s1), small, lower, cond_small); + upper = bld.sop2(aco_opcode::s_cselect_b32, bld.def(s1), Operand(0u), upper, cond_small); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + + } else if (instr->src[0].src.ssa->bit_size == 64) { + Temp vec = bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), Operand(0u), Operand(0x3df00000u)); + Temp trunc = bld.vop1(aco_opcode::v_trunc_f64, bld.def(v2), src); + Temp mul = bld.vop3(aco_opcode::v_mul_f64, bld.def(v2), trunc, vec); + vec = bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), Operand(0u), Operand(0xc1f00000u)); + Temp floor = bld.vop1(aco_opcode::v_floor_f64, bld.def(v2), mul); + Temp fma = bld.vop3(aco_opcode::v_fma_f64, bld.def(v2), floor, vec, trunc); + Temp lower = bld.vop1(aco_opcode::v_cvt_u32_f64, bld.def(v1), fma); + Temp upper = bld.vop1(aco_opcode::v_cvt_u32_f64, bld.def(v1), floor); + if (dst.type() == RegType::sgpr) { + lower = bld.as_uniform(lower); + upper = bld.as_uniform(upper); + } + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lower, upper); + + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_b2f32: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.regClass() == s1) { + src = as_uniform_bool(ctx, src); + bld.sop2(aco_opcode::s_mul_i32, Definition(dst), Operand(0x3f800000u), src); + } else if (dst.regClass() == v1) { + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(dst), Operand(0u), Operand(0x3f800000u), + as_divergent_bool(ctx, src, true)); + } else { + unreachable("Wrong destination register class for nir_op_b2f32."); + } + break; + } + case nir_op_b2f64: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.regClass() == s2) { + src = as_uniform_bool(ctx, src); + bld.sop2(aco_opcode::s_cselect_b64, Definition(dst), Operand(0x3f800000u), Operand(0u), bld.scc(src)); + } else if (dst.regClass() == v2) { + Temp one = bld.vop1(aco_opcode::v_mov_b32, bld.def(v2), Operand(0x3FF00000u)); + Temp upper = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0u), one, + as_divergent_bool(ctx, src, true)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), Operand(0u), upper); + } else { + unreachable("Wrong destination register class for nir_op_b2f64."); + } + break; + } + case nir_op_i2i32: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 64) { + /* we can actually just say dst = src, as it would map the lower register */ + emit_extract_vector(ctx, src, 0, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_u2u32: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 16) { + if (dst.regClass() == s1) { + bld.sop2(aco_opcode::s_and_b32, Definition(dst), bld.def(s1, scc), Operand(0xFFFFu), src); + } else { + // TODO: do better with SDWA + bld.vop2(aco_opcode::v_and_b32, Definition(dst), Operand(0xFFFFu), src); + } + } else if (instr->src[0].src.ssa->bit_size == 64) { + /* we can actually just say dst = src, as it would map the lower register */ + emit_extract_vector(ctx, src, 0, dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_i2i64: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 32) { + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), src, Operand(0u)); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_u2u64: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (instr->src[0].src.ssa->bit_size == 32) { + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), src, Operand(0u)); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_b2i32: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.regClass() == s1) { + if (src.regClass() == s1) { + bld.copy(Definition(dst), src); + } else { + // TODO: in a post-RA optimization, we can check if src is in VCC, and directly use VCCNZ + assert(src.regClass() == s2); + bld.sopc(aco_opcode::s_cmp_lg_u64, bld.scc(Definition(dst)), Operand(0u), src); + } + } else { + assert(dst.regClass() == v1 && src.regClass() == s2); + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(dst), Operand(0u), Operand(1u), src); + } + break; + } + case nir_op_i2b1: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (dst.regClass() == s2) { + assert(src.regClass() == v1 || src.regClass() == v2); + bld.vopc(src.size() == 2 ? aco_opcode::v_cmp_lg_u64 : aco_opcode::v_cmp_lg_u32, + Definition(dst), Operand(0u), src).def(0).setHint(vcc); + } else { + assert(src.regClass() == s1 && dst.regClass() == s1); + bld.sopc(aco_opcode::s_cmp_lg_u32, bld.scc(Definition(dst)), Operand(0u), src); + } + break; + } + case nir_op_pack_64_2x32_split: { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), src0, src1); + break; + } + case nir_op_unpack_64_2x32_split_x: + bld.pseudo(aco_opcode::p_split_vector, Definition(dst), bld.def(dst.regClass()), get_alu_src(ctx, instr->src[0])); + break; + case nir_op_unpack_64_2x32_split_y: + bld.pseudo(aco_opcode::p_split_vector, bld.def(dst.regClass()), Definition(dst), get_alu_src(ctx, instr->src[0])); + break; + case nir_op_pack_half_2x16: { + Temp src = get_alu_src(ctx, instr->src[0], 2); + + if (dst.regClass() == v1) { + Temp src0 = bld.tmp(v1); + Temp src1 = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src0), Definition(src1), src); + bld.vop3(aco_opcode::v_cvt_pkrtz_f16_f32, Definition(dst), src0, src1); + + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_unpack_half_2x16_split_x: { + if (dst.regClass() == v1) { + Builder bld(ctx->program, ctx->block); + bld.vop1(aco_opcode::v_cvt_f32_f16, Definition(dst), get_alu_src(ctx, instr->src[0])); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_unpack_half_2x16_split_y: { + if (dst.regClass() == v1) { + Builder bld(ctx->program, ctx->block); + /* TODO: use SDWA here */ + bld.vop1(aco_opcode::v_cvt_f32_f16, Definition(dst), + bld.vop2(aco_opcode::v_lshrrev_b32, bld.def(v1), Operand(16u), as_vgpr(ctx, get_alu_src(ctx, instr->src[0])))); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_fquantize2f16: { + Temp f16 = bld.vop1(aco_opcode::v_cvt_f16_f32, bld.def(v1), get_alu_src(ctx, instr->src[0])); + + Temp mask = bld.copy(bld.def(s1), Operand(0x36Fu)); /* value is NOT negative/positive denormal value */ + + Temp cmp_res = bld.tmp(s2); + bld.vopc_e64(aco_opcode::v_cmp_class_f16, Definition(cmp_res), f16, mask).def(0).setHint(vcc); + + Temp f32 = bld.vop1(aco_opcode::v_cvt_f32_f16, bld.def(v1), f16); + + bld.vop2(aco_opcode::v_cndmask_b32, Definition(dst), Operand(0u), f32, cmp_res); + break; + } + case nir_op_bfm: { + Temp bits = get_alu_src(ctx, instr->src[0]); + Temp offset = get_alu_src(ctx, instr->src[1]); + + if (dst.regClass() == s1) { + bld.sop2(aco_opcode::s_bfm_b32, Definition(dst), bits, offset); + } else if (dst.regClass() == v1) { + bld.vop3(aco_opcode::v_bfm_b32, Definition(dst), bits, offset); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_bitfield_select: { + /* (mask & insert) | (~mask & base) */ + Temp bitmask = get_alu_src(ctx, instr->src[0]); + Temp insert = get_alu_src(ctx, instr->src[1]); + Temp base = get_alu_src(ctx, instr->src[2]); + + /* dst = (insert & bitmask) | (base & ~bitmask) */ + if (dst.regClass() == s1) { + aco_ptr sop2; + nir_const_value* const_bitmask = nir_src_as_const_value(instr->src[0].src); + nir_const_value* const_insert = nir_src_as_const_value(instr->src[1].src); + Operand lhs; + if (const_insert && const_bitmask) { + lhs = Operand(const_insert->u32 & const_bitmask->u32); + } else { + insert = bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), insert, bitmask); + lhs = Operand(insert); + } + + Operand rhs; + nir_const_value* const_base = nir_src_as_const_value(instr->src[2].src); + if (const_base && const_bitmask) { + rhs = Operand(const_base->u32 & ~const_bitmask->u32); + } else { + base = bld.sop2(aco_opcode::s_andn2_b32, bld.def(s1), bld.def(s1, scc), base, bitmask); + rhs = Operand(base); + } + + bld.sop2(aco_opcode::s_or_b32, Definition(dst), bld.def(s1, scc), rhs, lhs); + + } else if (dst.regClass() == v1) { + if (base.type() == RegType::sgpr && (bitmask.type() == RegType::sgpr || (insert.type() == RegType::sgpr))) + base = as_vgpr(ctx, base); + if (insert.type() == RegType::sgpr && bitmask.type() == RegType::sgpr) + insert = as_vgpr(ctx, insert); + + bld.vop3(aco_opcode::v_bfi_b32, Definition(dst), bitmask, insert, base); + + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ubfe: + case nir_op_ibfe: { + Temp base = get_alu_src(ctx, instr->src[0]); + Temp offset = get_alu_src(ctx, instr->src[1]); + Temp bits = get_alu_src(ctx, instr->src[2]); + + if (dst.type() == RegType::sgpr) { + Operand extract; + nir_const_value* const_offset = nir_src_as_const_value(instr->src[1].src); + nir_const_value* const_bits = nir_src_as_const_value(instr->src[2].src); + if (const_offset && const_bits) { + uint32_t const_extract = (const_bits->u32 << 16) | const_offset->u32; + extract = Operand(const_extract); + } else { + Operand width; + if (const_bits) { + width = Operand(const_bits->u32 << 16); + } else { + width = bld.sop2(aco_opcode::s_lshl_b32, bld.def(s1), bld.def(s1, scc), bits, Operand(16u)); + } + extract = bld.sop2(aco_opcode::s_or_b32, bld.def(s1), bld.def(s1, scc), offset, width); + } + + aco_opcode opcode; + if (dst.regClass() == s1) { + if (instr->op == nir_op_ubfe) + opcode = aco_opcode::s_bfe_u32; + else + opcode = aco_opcode::s_bfe_i32; + } else if (dst.regClass() == s2) { + if (instr->op == nir_op_ubfe) + opcode = aco_opcode::s_bfe_u64; + else + opcode = aco_opcode::s_bfe_i64; + } else { + unreachable("Unsupported BFE bit size"); + } + + bld.sop2(opcode, Definition(dst), bld.def(s1, scc), base, extract); + + } else { + aco_opcode opcode; + if (dst.regClass() == v1) { + if (instr->op == nir_op_ubfe) + opcode = aco_opcode::v_bfe_u32; + else + opcode = aco_opcode::v_bfe_i32; + } else { + unreachable("Unsupported BFE bit size"); + } + + emit_vop3a_instruction(ctx, instr, opcode, dst); + } + break; + } + case nir_op_bit_count: { + Temp src = get_alu_src(ctx, instr->src[0]); + if (src.regClass() == s1) { + bld.sop1(aco_opcode::s_bcnt1_i32_b32, Definition(dst), bld.def(s1, scc), src); + } else if (src.regClass() == v1) { + bld.vop3(aco_opcode::v_bcnt_u32_b32, Definition(dst), src, Operand(0u)); + } else if (src.regClass() == v2) { + bld.vop3(aco_opcode::v_bcnt_u32_b32, Definition(dst), + emit_extract_vector(ctx, src, 1, v1), + bld.vop3(aco_opcode::v_bcnt_u32_b32, bld.def(v1), + emit_extract_vector(ctx, src, 0, v1), Operand(0u))); + } else if (src.regClass() == s2) { + bld.sop1(aco_opcode::s_bcnt1_i32_b64, Definition(dst), bld.def(s1, scc), src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_flt: { + if (instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_lt_f32, dst); + else if (instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_lt_f64, dst); + break; + } + case nir_op_fge: { + if (instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_ge_f32, dst); + else if (instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_ge_f64, dst); + break; + } + case nir_op_feq: { + if (instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_eq_f32, dst); + else if (instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_eq_f64, dst); + break; + } + case nir_op_fne: { + if (instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_neq_f32, dst); + else if (instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_neq_f64, dst); + break; + } + case nir_op_ilt: { + if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_lt_i32, dst); + else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::s_cmp_lt_i32, dst); + else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_lt_i64, dst); + break; + } + case nir_op_ige: { + if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_ge_i32, dst); + else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::s_cmp_ge_i32, dst); + else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_ge_i64, dst); + break; + } + case nir_op_ieq: { + if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 32) { + emit_comparison(ctx, instr, aco_opcode::v_cmp_eq_i32, dst); + } else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 32) { + emit_comparison(ctx, instr, aco_opcode::s_cmp_eq_i32, dst); + } else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 64) { + emit_comparison(ctx, instr, aco_opcode::v_cmp_eq_i64, dst); + } else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 64) { + emit_comparison(ctx, instr, aco_opcode::s_cmp_eq_u64, dst); + } else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 1) { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + bld.sopc(aco_opcode::s_cmp_eq_i32, bld.scc(Definition(dst)), + as_uniform_bool(ctx, src0), as_uniform_bool(ctx, src1)); + } else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 1) { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + bld.sop2(aco_opcode::s_xnor_b64, Definition(dst), bld.def(s1, scc), + as_divergent_bool(ctx, src0, false), as_divergent_bool(ctx, src1, false)); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ine: { + if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 32) { + emit_comparison(ctx, instr, aco_opcode::v_cmp_lg_i32, dst); + } else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 64) { + emit_comparison(ctx, instr, aco_opcode::v_cmp_lg_i64, dst); + } else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 32) { + emit_comparison(ctx, instr, aco_opcode::s_cmp_lg_i32, dst); + } else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 64) { + emit_comparison(ctx, instr, aco_opcode::s_cmp_lg_u64, dst); + } else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 1) { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + bld.sopc(aco_opcode::s_cmp_lg_i32, bld.scc(Definition(dst)), + as_uniform_bool(ctx, src0), as_uniform_bool(ctx, src1)); + } else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 1) { + Temp src0 = get_alu_src(ctx, instr->src[0]); + Temp src1 = get_alu_src(ctx, instr->src[1]); + bld.sop2(aco_opcode::s_xor_b64, Definition(dst), bld.def(s1, scc), + as_divergent_bool(ctx, src0, false), as_divergent_bool(ctx, src1, false)); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_op_ult: { + if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_lt_u32, dst); + else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::s_cmp_lt_u32, dst); + else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_lt_u64, dst); + break; + } + case nir_op_uge: { + if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::v_cmp_ge_u32, dst); + else if (dst.regClass() == s1 && instr->src[0].src.ssa->bit_size == 32) + emit_comparison(ctx, instr, aco_opcode::s_cmp_ge_u32, dst); + else if (dst.regClass() == s2 && instr->src[0].src.ssa->bit_size == 64) + emit_comparison(ctx, instr, aco_opcode::v_cmp_ge_u64, dst); + break; + } + case nir_op_fddx: + case nir_op_fddy: + case nir_op_fddx_fine: + case nir_op_fddy_fine: + case nir_op_fddx_coarse: + case nir_op_fddy_coarse: { + Definition tl = bld.def(v1); + uint16_t dpp_ctrl; + if (instr->op == nir_op_fddx_fine) { + bld.vop1_dpp(aco_opcode::v_mov_b32, tl, get_alu_src(ctx, instr->src[0]), dpp_quad_perm(0, 0, 2, 2)); + dpp_ctrl = dpp_quad_perm(1, 1, 3, 3); + } else if (instr->op == nir_op_fddy_fine) { + bld.vop1_dpp(aco_opcode::v_mov_b32, tl, get_alu_src(ctx, instr->src[0]), dpp_quad_perm(0, 1, 0, 1)); + dpp_ctrl = dpp_quad_perm(2, 3, 2, 3); + } else { + bld.vop1_dpp(aco_opcode::v_mov_b32, tl, get_alu_src(ctx, instr->src[0]), dpp_quad_perm(0, 0, 0, 0)); + if (instr->op == nir_op_fddx || instr->op == nir_op_fddx_coarse) + dpp_ctrl = dpp_quad_perm(1, 1, 1, 1); + else + dpp_ctrl = dpp_quad_perm(2, 2, 2, 2); + } + + Definition tmp = bld.def(v1); + bld.vop2_dpp(aco_opcode::v_sub_f32, tmp, get_alu_src(ctx, instr->src[0]), tl.getTemp(), dpp_ctrl); + emit_wqm(ctx, tmp.getTemp(), dst, true); + break; + } + default: + fprintf(stderr, "Unknown NIR ALU instr: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } +} + +void visit_load_const(isel_context *ctx, nir_load_const_instr *instr) +{ + Temp dst = get_ssa_temp(ctx, &instr->def); + + // TODO: we really want to have the resulting type as this would allow for 64bit literals + // which get truncated the lsb if double and msb if int + // for now, we only use s_mov_b64 with 64bit inline constants + assert(instr->def.num_components == 1 && "Vector load_const should be lowered to scalar."); + assert(dst.type() == RegType::sgpr); + + if (dst.size() == 1) + { + Builder(ctx->program, ctx->block).copy(Definition(dst), Operand(instr->value[0].u32)); + } else { + assert(dst.size() != 1); + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, dst.size(), 1)}; + if (instr->def.bit_size == 64) + for (unsigned i = 0; i < dst.size(); i++) + vec->operands[i] = Operand{(uint32_t)(instr->value[0].u64 >> i * 32)}; + else { + for (unsigned i = 0; i < dst.size(); i++) + vec->operands[i] = Operand{instr->value[i].u32}; + } + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + } +} + +uint32_t widen_mask(uint32_t mask, unsigned multiplier) +{ + uint32_t new_mask = 0; + for(unsigned i = 0; i < 32 && (1u << i) <= mask; ++i) + if (mask & (1u << i)) + new_mask |= ((1u << multiplier) - 1u) << (i * multiplier); + return new_mask; +} + +void visit_store_vs_output(isel_context *ctx, nir_intrinsic_instr *instr) +{ + /* This wouldn't work inside control flow or with indirect offsets but + * that doesn't happen because of nir_lower_io_to_temporaries(). */ + + unsigned write_mask = nir_intrinsic_write_mask(instr); + unsigned component = nir_intrinsic_component(instr); + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + unsigned idx = nir_intrinsic_base(instr) + component; + + nir_instr *off_instr = instr->src[1].ssa->parent_instr; + if (off_instr->type != nir_instr_type_load_const) { + fprintf(stderr, "Unimplemented nir_intrinsic_load_input offset\n"); + nir_print_instr(off_instr, stderr); + fprintf(stderr, "\n"); + } + idx += nir_instr_as_load_const(off_instr)->value[0].u32 * 4u; + + if (instr->src[0].ssa->bit_size == 64) + write_mask = widen_mask(write_mask, 2); + + for (unsigned i = 0; i < 8; ++i) { + if (write_mask & (1 << i)) { + ctx->vs_output.mask[idx / 4u] |= 1 << (idx % 4u); + ctx->vs_output.outputs[idx / 4u][idx % 4u] = emit_extract_vector(ctx, src, i, v1); + } + idx++; + } +} + +void visit_store_fs_output(isel_context *ctx, nir_intrinsic_instr *instr) +{ + unsigned write_mask = nir_intrinsic_write_mask(instr); + Operand values[4]; + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + for (unsigned i = 0; i < 4; ++i) { + if (write_mask & (1 << i)) { + Temp tmp = emit_extract_vector(ctx, src, i, v1); + values[i] = Operand(tmp); + } else { + values[i] = Operand(v1); + } + } + + unsigned index = nir_intrinsic_base(instr) / 4; + unsigned target, col_format; + unsigned enabled_channels = 0xF; + aco_opcode compr_op = (aco_opcode)0; + + nir_const_value* offset = nir_src_as_const_value(instr->src[1]); + assert(offset && "Non-const offsets on exports not yet supported"); + index += offset->u32; + + assert(index != FRAG_RESULT_COLOR); + + /* Unlike vertex shader exports, it's fine to use multiple exports to + * export separate channels of one target. So shaders which export both + * FRAG_RESULT_SAMPLE_MASK and FRAG_RESULT_DEPTH should work fine. + * TODO: combine the exports in those cases and create better code + */ + + if (index == FRAG_RESULT_SAMPLE_MASK) { + + if (ctx->program->info->ps.writes_z) { + target = V_008DFC_SQ_EXP_MRTZ; + enabled_channels = 0x4; + col_format = (unsigned) -1; + + values[2] = values[0]; + values[0] = Operand(v1); + } else { + aco_ptr exp{create_instruction(aco_opcode::exp, Format::EXP, 4, 0)}; + exp->valid_mask = false; + exp->done = false; + exp->compressed = true; + exp->dest = V_008DFC_SQ_EXP_MRTZ; + exp->enabled_mask = 0xc; + for (int i = 0; i < 4; i++) + exp->operands[i] = Operand(v1); + exp->operands[1] = Operand(values[0]); + ctx->block->instructions.emplace_back(std::move(exp)); + return; + } + + } else if (index == FRAG_RESULT_DEPTH) { + + target = V_008DFC_SQ_EXP_MRTZ; + enabled_channels = 0x1; + col_format = (unsigned) -1; + + } else if (index == FRAG_RESULT_STENCIL) { + + if (ctx->program->info->ps.writes_z) { + target = V_008DFC_SQ_EXP_MRTZ; + enabled_channels = 0x2; + col_format = (unsigned) -1; + + values[1] = values[0]; + values[0] = Operand(v1); + } else { + aco_ptr shift{create_instruction(aco_opcode::v_lshlrev_b32, Format::VOP2, 2, 1)}; + shift->operands[0] = Operand((uint32_t) 16); + shift->operands[1] = values[0]; + Temp tmp = {ctx->program->allocateId(), v1}; + shift->definitions[0] = Definition(tmp); + ctx->block->instructions.emplace_back(std::move(shift)); + + aco_ptr exp{create_instruction(aco_opcode::exp, Format::EXP, 4, 0)}; + exp->valid_mask = false; + exp->done = false; + exp->compressed = true; + exp->dest = V_008DFC_SQ_EXP_MRTZ; + exp->enabled_mask = 0x3; + exp->operands[0] = Operand(tmp); + for (int i = 1; i < 4; i++) + exp->operands[i] = Operand(v1); + ctx->block->instructions.emplace_back(std::move(exp)); + return; + } + + } else { + index -= FRAG_RESULT_DATA0; + target = V_008DFC_SQ_EXP_MRT + index; + col_format = (ctx->options->key.fs.col_format >> (4 * index)) & 0xf; + } + ASSERTED bool is_int8 = (ctx->options->key.fs.is_int8 >> index) & 1; + ASSERTED bool is_int10 = (ctx->options->key.fs.is_int10 >> index) & 1; + assert(!is_int8 && !is_int10); + + switch (col_format) + { + case V_028714_SPI_SHADER_ZERO: + enabled_channels = 0; /* writemask */ + target = V_008DFC_SQ_EXP_NULL; + break; + + case V_028714_SPI_SHADER_32_R: + enabled_channels = 1; + break; + + case V_028714_SPI_SHADER_32_GR: + enabled_channels = 0x3; + break; + + case V_028714_SPI_SHADER_32_AR: + enabled_channels = 0x9; + break; + + case V_028714_SPI_SHADER_FP16_ABGR: + enabled_channels = 0x5; + compr_op = aco_opcode::v_cvt_pkrtz_f16_f32; + break; + + case V_028714_SPI_SHADER_UNORM16_ABGR: + enabled_channels = 0x5; + compr_op = aco_opcode::v_cvt_pknorm_u16_f32; + break; + + case V_028714_SPI_SHADER_SNORM16_ABGR: + enabled_channels = 0x5; + compr_op = aco_opcode::v_cvt_pknorm_i16_f32; + break; + + case V_028714_SPI_SHADER_UINT16_ABGR: + enabled_channels = 0x5; + compr_op = aco_opcode::v_cvt_pk_u16_u32; + break; + + case V_028714_SPI_SHADER_SINT16_ABGR: + enabled_channels = 0x5; + compr_op = aco_opcode::v_cvt_pk_i16_i32; + break; + + case V_028714_SPI_SHADER_32_ABGR: + enabled_channels = 0xF; + break; + + default: + break; + } + + if (target == V_008DFC_SQ_EXP_NULL) + return; + + if ((bool)compr_op) + { + for (int i = 0; i < 2; i++) + { + /* check if at least one of the values to be compressed is enabled */ + unsigned enabled = (write_mask >> (i*2) | write_mask >> (i*2+1)) & 0x1; + if (enabled) { + enabled_channels |= enabled << (i*2); + aco_ptr compr{create_instruction(compr_op, Format::VOP3A, 2, 1)}; + Temp tmp{ctx->program->allocateId(), v1}; + compr->operands[0] = values[i*2].isUndefined() ? Operand(0u) : values[i*2]; + compr->operands[1] = values[i*2+1].isUndefined() ? Operand(0u): values[i*2+1]; + compr->definitions[0] = Definition(tmp); + values[i] = Operand(tmp); + ctx->block->instructions.emplace_back(std::move(compr)); + } else { + values[i] = Operand(v1); + } + } + } + + aco_ptr exp{create_instruction(aco_opcode::exp, Format::EXP, 4, 0)}; + exp->valid_mask = false; + exp->done = false; + exp->compressed = (bool) compr_op; + exp->dest = target; + exp->enabled_mask = enabled_channels; + if ((bool) compr_op) { + for (int i = 0; i < 2; i++) + exp->operands[i] = enabled_channels & (3 << (i * 2)) ? values[i] : Operand(v1); + exp->operands[2] = Operand(v1); + exp->operands[3] = Operand(v1); + } else { + for (int i = 0; i < 4; i++) + exp->operands[i] = enabled_channels & (1 << i) ? values[i] : Operand(v1); + } + + ctx->block->instructions.emplace_back(std::move(exp)); +} + +void visit_store_output(isel_context *ctx, nir_intrinsic_instr *instr) +{ + if (ctx->stage == vertex_vs) { + visit_store_vs_output(ctx, instr); + } else if (ctx->stage == fragment_fs) { + visit_store_fs_output(ctx, instr); + } else { + unreachable("Shader stage not implemented"); + } +} + +void emit_interp_instr(isel_context *ctx, unsigned idx, unsigned component, Temp src, Temp dst, Temp prim_mask) +{ + Temp coord1 = emit_extract_vector(ctx, src, 0, v1); + Temp coord2 = emit_extract_vector(ctx, src, 1, v1); + + Builder bld(ctx->program, ctx->block); + Temp tmp = bld.vintrp(aco_opcode::v_interp_p1_f32, bld.def(v1), coord1, bld.m0(prim_mask), idx, component); + bld.vintrp(aco_opcode::v_interp_p2_f32, Definition(dst), coord2, bld.m0(prim_mask), tmp, idx, component); +} + +void emit_load_frag_coord(isel_context *ctx, Temp dst, unsigned num_components) +{ + aco_ptr vec(create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, num_components, 1)); + for (unsigned i = 0; i < num_components; i++) + vec->operands[i] = Operand(ctx->fs_inputs[fs_input::frag_pos_0 + i]); + + if (ctx->fs_vgpr_args[fs_input::frag_pos_3]) { + assert(num_components == 4); + Builder bld(ctx->program, ctx->block); + vec->operands[3] = bld.vop1(aco_opcode::v_rcp_f32, bld.def(v1), ctx->fs_inputs[fs_input::frag_pos_3]); + } + + for (Operand& op : vec->operands) + op = op.isUndefined() ? Operand(0u) : op; + + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + emit_split_vector(ctx, dst, num_components); + return; +} + +void visit_load_interpolated_input(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + Temp coords = get_ssa_temp(ctx, instr->src[0].ssa); + unsigned idx = nir_intrinsic_base(instr); + unsigned component = nir_intrinsic_component(instr); + Temp prim_mask = ctx->prim_mask; + + nir_const_value* offset = nir_src_as_const_value(instr->src[1]); + if (offset) { + assert(offset->u32 == 0); + } else { + /* the lower 15bit of the prim_mask contain the offset into LDS + * while the upper bits contain the number of prims */ + Temp offset_src = get_ssa_temp(ctx, instr->src[1].ssa); + assert(offset_src.regClass() == s1 && "TODO: divergent offsets..."); + Builder bld(ctx->program, ctx->block); + Temp stride = bld.sop2(aco_opcode::s_lshr_b32, bld.def(s1), bld.def(s1, scc), prim_mask, Operand(16u)); + stride = bld.sop1(aco_opcode::s_bcnt1_i32_b32, bld.def(s1), bld.def(s1, scc), stride); + stride = bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), stride, Operand(48u)); + offset_src = bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), stride, offset_src); + prim_mask = bld.sop2(aco_opcode::s_add_i32, bld.def(s1, m0), bld.def(s1, scc), offset_src, prim_mask); + } + + if (instr->dest.ssa.num_components == 1) { + emit_interp_instr(ctx, idx, component, coords, dst, prim_mask); + } else { + aco_ptr vec(create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, instr->dest.ssa.num_components, 1)); + for (unsigned i = 0; i < instr->dest.ssa.num_components; i++) + { + Temp tmp = {ctx->program->allocateId(), v1}; + emit_interp_instr(ctx, idx, component+i, coords, tmp, prim_mask); + vec->operands[i] = Operand(tmp); + } + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + } +} + +unsigned get_num_channels_from_data_format(unsigned data_format) +{ + switch (data_format) { + case V_008F0C_BUF_DATA_FORMAT_8: + case V_008F0C_BUF_DATA_FORMAT_16: + case V_008F0C_BUF_DATA_FORMAT_32: + return 1; + case V_008F0C_BUF_DATA_FORMAT_8_8: + case V_008F0C_BUF_DATA_FORMAT_16_16: + case V_008F0C_BUF_DATA_FORMAT_32_32: + return 2; + case V_008F0C_BUF_DATA_FORMAT_10_11_11: + case V_008F0C_BUF_DATA_FORMAT_11_11_10: + case V_008F0C_BUF_DATA_FORMAT_32_32_32: + return 3; + case V_008F0C_BUF_DATA_FORMAT_8_8_8_8: + case V_008F0C_BUF_DATA_FORMAT_10_10_10_2: + case V_008F0C_BUF_DATA_FORMAT_2_10_10_10: + case V_008F0C_BUF_DATA_FORMAT_16_16_16_16: + case V_008F0C_BUF_DATA_FORMAT_32_32_32_32: + return 4; + default: + break; + } + + return 4; +} + +/* For 2_10_10_10 formats the alpha is handled as unsigned by pre-vega HW. + * so we may need to fix it up. */ +Temp adjust_vertex_fetch_alpha(isel_context *ctx, unsigned adjustment, Temp alpha) +{ + Builder bld(ctx->program, ctx->block); + + if (adjustment == RADV_ALPHA_ADJUST_SSCALED) + alpha = bld.vop1(aco_opcode::v_cvt_u32_f32, bld.def(v1), alpha); + + /* For the integer-like cases, do a natural sign extension. + * + * For the SNORM case, the values are 0.0, 0.333, 0.666, 1.0 + * and happen to contain 0, 1, 2, 3 as the two LSBs of the + * exponent. + */ + alpha = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(adjustment == RADV_ALPHA_ADJUST_SNORM ? 7u : 30u), alpha); + alpha = bld.vop2(aco_opcode::v_ashrrev_i32, bld.def(v1), Operand(30u), alpha); + + /* Convert back to the right type. */ + if (adjustment == RADV_ALPHA_ADJUST_SNORM) { + alpha = bld.vop1(aco_opcode::v_cvt_f32_i32, bld.def(v1), alpha); + Temp clamp = bld.vopc(aco_opcode::v_cmp_le_f32, bld.hint_vcc(bld.def(s2)), Operand(0xbf800000u), alpha); + alpha = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0xbf800000u), alpha, clamp); + } else if (adjustment == RADV_ALPHA_ADJUST_SSCALED) { + alpha = bld.vop1(aco_opcode::v_cvt_f32_i32, bld.def(v1), alpha); + } + + return alpha; +} + +void visit_load_input(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + if (ctx->stage & sw_vs) { + + nir_instr *off_instr = instr->src[0].ssa->parent_instr; + if (off_instr->type != nir_instr_type_load_const) { + fprintf(stderr, "Unimplemented nir_intrinsic_load_input offset\n"); + nir_print_instr(off_instr, stderr); + fprintf(stderr, "\n"); + } + uint32_t offset = nir_instr_as_load_const(off_instr)->value[0].u32; + + Temp vertex_buffers = convert_pointer_to_64_bit(ctx, ctx->vertex_buffers); + + unsigned location = nir_intrinsic_base(instr) / 4 - VERT_ATTRIB_GENERIC0 + offset; + unsigned component = nir_intrinsic_component(instr); + unsigned attrib_binding = ctx->options->key.vs.vertex_attribute_bindings[location]; + uint32_t attrib_offset = ctx->options->key.vs.vertex_attribute_offsets[location]; + uint32_t attrib_stride = ctx->options->key.vs.vertex_attribute_strides[location]; + unsigned attrib_format = ctx->options->key.vs.vertex_attribute_formats[location]; + + unsigned dfmt = attrib_format & 0xf; + + unsigned nfmt = (attrib_format >> 4) & 0x7; + unsigned num_dfmt_channels = get_num_channels_from_data_format(dfmt); + unsigned mask = nir_ssa_def_components_read(&instr->dest.ssa) << component; + unsigned num_channels = MIN2(util_last_bit(mask), num_dfmt_channels); + unsigned alpha_adjust = (ctx->options->key.vs.alpha_adjust >> (location * 2)) & 3; + bool post_shuffle = ctx->options->key.vs.post_shuffle & (1 << location); + if (post_shuffle) + num_channels = MAX2(num_channels, 3); + + Temp list = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), vertex_buffers, Operand(attrib_binding * 16u)); + + Temp index; + if (ctx->options->key.vs.instance_rate_inputs & (1u << location)) { + uint32_t divisor = ctx->options->key.vs.instance_rate_divisors[location]; + if (divisor) { + ctx->needs_instance_id = true; + + if (divisor != 1) { + Temp divided = bld.tmp(v1); + emit_v_div_u32(ctx, divided, as_vgpr(ctx, ctx->instance_id), divisor); + index = bld.vadd32(bld.def(v1), ctx->start_instance, divided); + } else { + index = bld.vadd32(bld.def(v1), ctx->start_instance, ctx->instance_id); + } + } else { + index = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), ctx->start_instance); + } + } else { + index = bld.vadd32(bld.def(v1), ctx->base_vertex, ctx->vertex_id); + } + + if (attrib_stride != 0 && attrib_offset > attrib_stride) { + index = bld.vadd32(bld.def(v1), Operand(attrib_offset / attrib_stride), index); + attrib_offset = attrib_offset % attrib_stride; + } + + Operand soffset(0u); + if (attrib_offset >= 4096) { + soffset = bld.copy(bld.def(s1), Operand(attrib_offset)); + attrib_offset = 0; + } + + aco_opcode opcode; + switch (num_channels) { + case 1: + opcode = aco_opcode::tbuffer_load_format_x; + break; + case 2: + opcode = aco_opcode::tbuffer_load_format_xy; + break; + case 3: + opcode = aco_opcode::tbuffer_load_format_xyz; + break; + case 4: + opcode = aco_opcode::tbuffer_load_format_xyzw; + break; + default: + unreachable("Unimplemented load_input vector size"); + } + + Temp tmp = post_shuffle || num_channels != dst.size() || alpha_adjust != RADV_ALPHA_ADJUST_NONE || component ? bld.tmp(RegType::vgpr, num_channels) : dst; + + aco_ptr mubuf{create_instruction(opcode, Format::MTBUF, 3, 1)}; + mubuf->operands[0] = Operand(index); + mubuf->operands[1] = Operand(list); + mubuf->operands[2] = soffset; + mubuf->definitions[0] = Definition(tmp); + mubuf->idxen = true; + mubuf->can_reorder = true; + mubuf->dfmt = dfmt; + mubuf->nfmt = nfmt; + assert(attrib_offset < 4096); + mubuf->offset = attrib_offset; + ctx->block->instructions.emplace_back(std::move(mubuf)); + + emit_split_vector(ctx, tmp, tmp.size()); + + if (tmp.id() != dst.id()) { + bool is_float = nfmt != V_008F0C_BUF_NUM_FORMAT_UINT && + nfmt != V_008F0C_BUF_NUM_FORMAT_SINT; + + static const unsigned swizzle_normal[4] = {0, 1, 2, 3}; + static const unsigned swizzle_post_shuffle[4] = {2, 1, 0, 3}; + const unsigned *swizzle = post_shuffle ? swizzle_post_shuffle : swizzle_normal; + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, dst.size(), 1)}; + for (unsigned i = 0; i < dst.size(); i++) { + unsigned idx = i + component; + if (idx == 3 && alpha_adjust != RADV_ALPHA_ADJUST_NONE && num_channels >= 4) { + Temp alpha = emit_extract_vector(ctx, tmp, swizzle[3], v1); + vec->operands[3] = Operand(adjust_vertex_fetch_alpha(ctx, alpha_adjust, alpha)); + } else if (idx < num_channels) { + vec->operands[i] = Operand(emit_extract_vector(ctx, tmp, swizzle[idx], v1)); + } else if (is_float && idx == 3) { + vec->operands[i] = Operand(0x3f800000u); + } else if (!is_float && idx == 3) { + vec->operands[i] = Operand(1u); + } else { + vec->operands[i] = Operand(0u); + } + } + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + emit_split_vector(ctx, dst, dst.size()); + } + + } else if (ctx->stage == fragment_fs) { + nir_instr *off_instr = instr->src[0].ssa->parent_instr; + if (off_instr->type != nir_instr_type_load_const || + nir_instr_as_load_const(off_instr)->value[0].u32 != 0) { + fprintf(stderr, "Unimplemented nir_intrinsic_load_input offset\n"); + nir_print_instr(off_instr, stderr); + fprintf(stderr, "\n"); + } + + Temp prim_mask = ctx->prim_mask; + nir_const_value* offset = nir_src_as_const_value(instr->src[0]); + if (offset) { + assert(offset->u32 == 0); + } else { + /* the lower 15bit of the prim_mask contain the offset into LDS + * while the upper bits contain the number of prims */ + Temp offset_src = get_ssa_temp(ctx, instr->src[0].ssa); + assert(offset_src.regClass() == s1 && "TODO: divergent offsets..."); + Builder bld(ctx->program, ctx->block); + Temp stride = bld.sop2(aco_opcode::s_lshr_b32, bld.def(s1), bld.def(s1, scc), prim_mask, Operand(16u)); + stride = bld.sop1(aco_opcode::s_bcnt1_i32_b32, bld.def(s1), bld.def(s1, scc), stride); + stride = bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), stride, Operand(48u)); + offset_src = bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), stride, offset_src); + prim_mask = bld.sop2(aco_opcode::s_add_i32, bld.def(s1, m0), bld.def(s1, scc), offset_src, prim_mask); + } + + unsigned idx = nir_intrinsic_base(instr); + unsigned component = nir_intrinsic_component(instr); + + if (dst.size() == 1) { + bld.vintrp(aco_opcode::v_interp_mov_f32, Definition(dst), Operand(2u), bld.m0(prim_mask), idx, component); + } else { + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, dst.size(), 1)}; + for (unsigned i = 0; i < dst.size(); i++) + vec->operands[i] = bld.vintrp(aco_opcode::v_interp_mov_f32, bld.def(v1), Operand(2u), bld.m0(prim_mask), idx, component + i); + vec->definitions[0] = Definition(dst); + bld.insert(std::move(vec)); + } + + } else { + unreachable("Shader stage not implemented"); + } +} + +Temp load_desc_ptr(isel_context *ctx, unsigned desc_set) +{ + if (ctx->program->info->need_indirect_descriptor_sets) { + Builder bld(ctx->program, ctx->block); + Temp ptr64 = convert_pointer_to_64_bit(ctx, ctx->descriptor_sets[0]); + return bld.smem(aco_opcode::s_load_dword, bld.def(s1), ptr64, Operand(desc_set << 2));//, false, false, false); + } + + return ctx->descriptor_sets[desc_set]; +} + + +void visit_load_resource(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + Temp index = bld.as_uniform(get_ssa_temp(ctx, instr->src[0].ssa)); + unsigned desc_set = nir_intrinsic_desc_set(instr); + unsigned binding = nir_intrinsic_binding(instr); + + Temp desc_ptr; + radv_pipeline_layout *pipeline_layout = ctx->options->layout; + radv_descriptor_set_layout *layout = pipeline_layout->set[desc_set].layout; + unsigned offset = layout->binding[binding].offset; + unsigned stride; + if (layout->binding[binding].type == VK_DESCRIPTOR_TYPE_UNIFORM_BUFFER_DYNAMIC || + layout->binding[binding].type == VK_DESCRIPTOR_TYPE_STORAGE_BUFFER_DYNAMIC) { + unsigned idx = pipeline_layout->set[desc_set].dynamic_offset_start + layout->binding[binding].dynamic_offset_offset; + desc_ptr = ctx->push_constants; + offset = pipeline_layout->push_constant_size + 16 * idx; + stride = 16; + } else { + desc_ptr = load_desc_ptr(ctx, desc_set); + stride = layout->binding[binding].size; + } + + nir_const_value* nir_const_index = nir_src_as_const_value(instr->src[0]); + unsigned const_index = nir_const_index ? nir_const_index->u32 : 0; + if (stride != 1) { + if (nir_const_index) { + const_index = const_index * stride; + } else { + index = bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), Operand(stride), Operand(index)); + } + } + if (offset) { + if (nir_const_index) { + const_index = const_index + offset; + } else { + index = bld.sop2(aco_opcode::s_add_i32, bld.def(s1), bld.def(s1, scc), Operand(offset), Operand(index)); + } + } + + if (nir_const_index && const_index == 0) { + index = desc_ptr; + } else { + index = bld.sop2(aco_opcode::s_add_i32, bld.def(s1), bld.def(s1, scc), + nir_const_index ? Operand(const_index) : Operand(index), + Operand(desc_ptr)); + } + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.sop1(aco_opcode::s_mov_b32, Definition(dst), index); +} + +void load_buffer(isel_context *ctx, unsigned num_components, Temp dst, Temp rsrc, Temp offset, bool glc=false) +{ + Builder bld(ctx->program, ctx->block); + + unsigned num_bytes = dst.size() * 4; + + aco_opcode op; + if (dst.type() == RegType::vgpr || (glc && ctx->options->chip_class < GFX8)) { + if (ctx->options->chip_class < GFX8) + offset = as_vgpr(ctx, offset); + + Operand vaddr = offset.type() == RegType::vgpr ? Operand(offset) : Operand(v1); + Operand soffset = offset.type() == RegType::sgpr ? Operand(offset) : Operand((uint32_t) 0); + unsigned const_offset = 0; + + Temp lower = Temp(); + if (num_bytes > 16) { + assert(num_components == 3 || num_components == 4); + op = aco_opcode::buffer_load_dwordx4; + lower = bld.tmp(v4); + aco_ptr mubuf{create_instruction(op, Format::MUBUF, 3, 1)}; + mubuf->definitions[0] = Definition(lower); + mubuf->operands[0] = vaddr; + mubuf->operands[1] = Operand(rsrc); + mubuf->operands[2] = soffset; + mubuf->offen = (offset.type() == RegType::vgpr); + mubuf->glc = glc; + mubuf->barrier = barrier_buffer; + bld.insert(std::move(mubuf)); + emit_split_vector(ctx, lower, 2); + num_bytes -= 16; + const_offset = 16; + } + + switch (num_bytes) { + case 4: + op = aco_opcode::buffer_load_dword; + break; + case 8: + op = aco_opcode::buffer_load_dwordx2; + break; + case 12: + op = aco_opcode::buffer_load_dwordx3; + break; + case 16: + op = aco_opcode::buffer_load_dwordx4; + break; + default: + unreachable("Load SSBO not implemented for this size."); + } + aco_ptr mubuf{create_instruction(op, Format::MUBUF, 3, 1)}; + mubuf->operands[0] = vaddr; + mubuf->operands[1] = Operand(rsrc); + mubuf->operands[2] = soffset; + mubuf->offen = (offset.type() == RegType::vgpr); + mubuf->glc = glc; + mubuf->barrier = barrier_buffer; + mubuf->offset = const_offset; + aco_ptr instr = std::move(mubuf); + + if (dst.size() > 4) { + assert(lower != Temp()); + Temp upper = bld.tmp(RegType::vgpr, dst.size() - lower.size()); + instr->definitions[0] = Definition(upper); + bld.insert(std::move(instr)); + if (dst.size() == 8) + emit_split_vector(ctx, upper, 2); + instr.reset(create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, dst.size() / 2, 1)); + instr->operands[0] = Operand(emit_extract_vector(ctx, lower, 0, v2)); + instr->operands[1] = Operand(emit_extract_vector(ctx, lower, 1, v2)); + instr->operands[2] = Operand(emit_extract_vector(ctx, upper, 0, v2)); + if (dst.size() == 8) + instr->operands[3] = Operand(emit_extract_vector(ctx, upper, 1, v2)); + } + + if (dst.type() == RegType::sgpr) { + Temp vec = bld.tmp(RegType::vgpr, dst.size()); + instr->definitions[0] = Definition(vec); + bld.insert(std::move(instr)); + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), vec); + } else { + instr->definitions[0] = Definition(dst); + bld.insert(std::move(instr)); + } + } else { + switch (num_bytes) { + case 4: + op = aco_opcode::s_buffer_load_dword; + break; + case 8: + op = aco_opcode::s_buffer_load_dwordx2; + break; + case 12: + case 16: + op = aco_opcode::s_buffer_load_dwordx4; + break; + case 24: + case 32: + op = aco_opcode::s_buffer_load_dwordx8; + break; + default: + unreachable("Load SSBO not implemented for this size."); + } + aco_ptr load{create_instruction(op, Format::SMEM, 2, 1)}; + load->operands[0] = Operand(rsrc); + load->operands[1] = Operand(bld.as_uniform(offset)); + assert(load->operands[1].getTemp().type() == RegType::sgpr); + load->definitions[0] = Definition(dst); + load->glc = glc; + load->barrier = barrier_buffer; + assert(ctx->options->chip_class >= GFX8 || !glc); + + /* trim vector */ + if (dst.size() == 3) { + Temp vec = bld.tmp(s4); + load->definitions[0] = Definition(vec); + bld.insert(std::move(load)); + emit_split_vector(ctx, vec, 4); + + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + emit_extract_vector(ctx, vec, 0, s1), + emit_extract_vector(ctx, vec, 1, s1), + emit_extract_vector(ctx, vec, 2, s1)); + } else if (dst.size() == 6) { + Temp vec = bld.tmp(s8); + load->definitions[0] = Definition(vec); + bld.insert(std::move(load)); + emit_split_vector(ctx, vec, 4); + + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + emit_extract_vector(ctx, vec, 0, s2), + emit_extract_vector(ctx, vec, 1, s2), + emit_extract_vector(ctx, vec, 2, s2)); + } else { + bld.insert(std::move(load)); + } + + } + emit_split_vector(ctx, dst, num_components); +} + +void visit_load_ubo(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + Temp rsrc = get_ssa_temp(ctx, instr->src[0].ssa); + + Builder bld(ctx->program, ctx->block); + + nir_intrinsic_instr* idx_instr = nir_instr_as_intrinsic(instr->src[0].ssa->parent_instr); + unsigned desc_set = nir_intrinsic_desc_set(idx_instr); + unsigned binding = nir_intrinsic_binding(idx_instr); + radv_descriptor_set_layout *layout = ctx->options->layout->set[desc_set].layout; + + if (layout->binding[binding].type == VK_DESCRIPTOR_TYPE_INLINE_UNIFORM_BLOCK_EXT) { + uint32_t desc_type = S_008F0C_DST_SEL_X(V_008F0C_SQ_SEL_X) | + S_008F0C_DST_SEL_Y(V_008F0C_SQ_SEL_Y) | + S_008F0C_DST_SEL_Z(V_008F0C_SQ_SEL_Z) | + S_008F0C_DST_SEL_W(V_008F0C_SQ_SEL_W) | + S_008F0C_NUM_FORMAT(V_008F0C_BUF_NUM_FORMAT_FLOAT) | + S_008F0C_DATA_FORMAT(V_008F0C_BUF_DATA_FORMAT_32); + Temp upper_dwords = bld.pseudo(aco_opcode::p_create_vector, bld.def(s3), + Operand(S_008F04_BASE_ADDRESS_HI(ctx->options->address32_hi)), + Operand(0xFFFFFFFFu), + Operand(desc_type)); + rsrc = bld.pseudo(aco_opcode::p_create_vector, bld.def(s4), + rsrc, upper_dwords); + } else { + rsrc = convert_pointer_to_64_bit(ctx, rsrc); + rsrc = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), rsrc, Operand(0u)); + } + + load_buffer(ctx, instr->num_components, dst, rsrc, get_ssa_temp(ctx, instr->src[1].ssa)); +} + +void visit_load_push_constant(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + unsigned offset = nir_intrinsic_base(instr); + nir_const_value *index_cv = nir_src_as_const_value(instr->src[0]); + if (index_cv && instr->dest.ssa.bit_size == 32) { + + unsigned count = instr->dest.ssa.num_components; + unsigned start = (offset + index_cv->u32) / 4u; + start -= ctx->base_inline_push_consts; + if (start + count <= ctx->num_inline_push_consts) { + std::array elems; + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, count, 1)}; + for (unsigned i = 0; i < count; ++i) { + elems[i] = ctx->inline_push_consts[start + i]; + vec->operands[i] = Operand{elems[i]}; + } + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + ctx->allocated_vec.emplace(dst.id(), elems); + return; + } + } + + Temp index = bld.as_uniform(get_ssa_temp(ctx, instr->src[0].ssa)); + if (offset != 0) // TODO check if index != 0 as well + index = bld.sop2(aco_opcode::s_add_i32, bld.def(s1), bld.def(s1, scc), Operand(offset), index); + Temp ptr = convert_pointer_to_64_bit(ctx, ctx->push_constants); + Temp vec = dst; + bool trim = false; + aco_opcode op; + + switch (dst.size()) { + case 1: + op = aco_opcode::s_load_dword; + break; + case 2: + op = aco_opcode::s_load_dwordx2; + break; + case 3: + vec = bld.tmp(s4); + trim = true; + case 4: + op = aco_opcode::s_load_dwordx4; + break; + case 6: + vec = bld.tmp(s8); + trim = true; + case 8: + op = aco_opcode::s_load_dwordx8; + break; + default: + unreachable("unimplemented or forbidden load_push_constant."); + } + + bld.smem(op, Definition(vec), ptr, index); + + if (trim) { + emit_split_vector(ctx, vec, 4); + RegClass rc = dst.size() == 3 ? s1 : s2; + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + emit_extract_vector(ctx, vec, 0, rc), + emit_extract_vector(ctx, vec, 1, rc), + emit_extract_vector(ctx, vec, 2, rc)); + + } + emit_split_vector(ctx, dst, instr->dest.ssa.num_components); +} + +void visit_load_constant(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + Builder bld(ctx->program, ctx->block); + + uint32_t desc_type = S_008F0C_DST_SEL_X(V_008F0C_SQ_SEL_X) | + S_008F0C_DST_SEL_Y(V_008F0C_SQ_SEL_Y) | + S_008F0C_DST_SEL_Z(V_008F0C_SQ_SEL_Z) | + S_008F0C_DST_SEL_W(V_008F0C_SQ_SEL_W); + if (ctx->options->chip_class >= GFX10) { + desc_type |= S_008F0C_FORMAT(V_008F0C_IMG_FORMAT_32_FLOAT) | + S_008F0C_OOB_SELECT(3) | + S_008F0C_RESOURCE_LEVEL(1); + } else { + desc_type |= S_008F0C_NUM_FORMAT(V_008F0C_BUF_NUM_FORMAT_FLOAT) | + S_008F0C_DATA_FORMAT(V_008F0C_BUF_DATA_FORMAT_32); + } + + unsigned base = nir_intrinsic_base(instr) + ctx->constant_data_offset; + unsigned range = nir_intrinsic_range(instr); + + Temp offset = get_ssa_temp(ctx, instr->src[0].ssa); + if (base && offset.type() == RegType::sgpr) + offset = bld.sop2(aco_opcode::s_add_u32, bld.def(s1), bld.def(s1, scc), offset, Operand(base)); + else if (base && offset.type() == RegType::vgpr) + offset = bld.vadd32(bld.def(v1), Operand(base), offset); + + Temp rsrc = bld.pseudo(aco_opcode::p_create_vector, bld.def(s4), + bld.sop1(aco_opcode::p_constaddr, bld.def(s2), bld.def(s1, scc), Operand(0u)), + Operand(MIN2(range, ctx->shader->constant_data_size - nir_intrinsic_base(instr))), + Operand(desc_type)); + + load_buffer(ctx, instr->num_components, dst, rsrc, offset); +} + +void visit_discard_if(isel_context *ctx, nir_intrinsic_instr *instr) +{ + if (ctx->cf_info.loop_nest_depth || ctx->cf_info.parent_if.is_divergent) + ctx->cf_info.exec_potentially_empty = true; + + ctx->program->needs_exact = true; + + Builder bld(ctx->program, ctx->block); + Temp src = as_divergent_bool(ctx, get_ssa_temp(ctx, instr->src[0].ssa), false); + src = bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)); + bld.pseudo(aco_opcode::p_discard_if, src); + ctx->block->kind |= block_kind_uses_discard_if; + return; +} + +void visit_discard(isel_context* ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + + if (ctx->cf_info.loop_nest_depth || ctx->cf_info.parent_if.is_divergent) + ctx->cf_info.exec_potentially_empty = true; + + bool divergent = ctx->cf_info.parent_if.is_divergent || + ctx->cf_info.parent_loop.has_divergent_continue; + + if (ctx->block->loop_nest_depth && + ((nir_instr_is_last(&instr->instr) && !divergent) || divergent)) { + /* we handle discards the same way as jump instructions */ + append_logical_end(ctx->block); + + /* in loops, discard behaves like break */ + Block *linear_target = ctx->cf_info.parent_loop.exit; + ctx->block->kind |= block_kind_discard; + + if (!divergent) { + /* uniform discard - loop ends here */ + assert(nir_instr_is_last(&instr->instr)); + ctx->block->kind |= block_kind_uniform; + ctx->cf_info.has_branch = true; + bld.branch(aco_opcode::p_branch); + add_linear_edge(ctx->block->index, linear_target); + return; + } + + /* we add a break right behind the discard() instructions */ + ctx->block->kind |= block_kind_break; + unsigned idx = ctx->block->index; + + /* remove critical edges from linear CFG */ + bld.branch(aco_opcode::p_branch); + Block* break_block = ctx->program->create_and_insert_block(); + break_block->loop_nest_depth = ctx->cf_info.loop_nest_depth; + break_block->kind |= block_kind_uniform; + add_linear_edge(idx, break_block); + add_linear_edge(break_block->index, linear_target); + bld.reset(break_block); + bld.branch(aco_opcode::p_branch); + + Block* continue_block = ctx->program->create_and_insert_block(); + continue_block->loop_nest_depth = ctx->cf_info.loop_nest_depth; + add_linear_edge(idx, continue_block); + append_logical_start(continue_block); + ctx->block = continue_block; + + return; + } + + /* it can currently happen that NIR doesn't remove the unreachable code */ + if (!nir_instr_is_last(&instr->instr)) { + ctx->program->needs_exact = true; + /* save exec somewhere temporarily so that it doesn't get + * overwritten before the discard from outer exec masks */ + Temp cond = bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), Operand(0xFFFFFFFF), Operand(exec, s2)); + bld.pseudo(aco_opcode::p_discard_if, cond); + ctx->block->kind |= block_kind_uses_discard_if; + return; + } + + /* This condition is incorrect for uniformly branched discards in a loop + * predicated by a divergent condition, but the above code catches that case + * and the discard would end up turning into a discard_if. + * For example: + * if (divergent) { + * while (...) { + * if (uniform) { + * discard; + * } + * } + * } + */ + if (!ctx->cf_info.parent_if.is_divergent) { + /* program just ends here */ + ctx->block->kind |= block_kind_uniform; + bld.exp(aco_opcode::exp, Operand(v1), Operand(v1), Operand(v1), Operand(v1), + 0 /* enabled mask */, 9 /* dest */, + false /* compressed */, true/* done */, true /* valid mask */); + bld.sopp(aco_opcode::s_endpgm); + // TODO: it will potentially be followed by a branch which is dead code to sanitize NIR phis + } else { + ctx->block->kind |= block_kind_discard; + /* branch and linear edge is added by visit_if() */ + } +} + +enum aco_descriptor_type { + ACO_DESC_IMAGE, + ACO_DESC_FMASK, + ACO_DESC_SAMPLER, + ACO_DESC_BUFFER, + ACO_DESC_PLANE_0, + ACO_DESC_PLANE_1, + ACO_DESC_PLANE_2, +}; + +enum aco_image_dim { + aco_image_1d, + aco_image_2d, + aco_image_3d, + aco_image_cube, // includes cube arrays + aco_image_1darray, + aco_image_2darray, + aco_image_2dmsaa, + aco_image_2darraymsaa, +}; + +static enum aco_image_dim +get_sampler_dim(isel_context *ctx, enum glsl_sampler_dim dim, bool is_array) +{ + switch (dim) { + case GLSL_SAMPLER_DIM_1D: + if (ctx->options->chip_class >= GFX9) + return is_array ? aco_image_2darray : aco_image_2d; + return is_array ? aco_image_1darray : aco_image_1d; + case GLSL_SAMPLER_DIM_2D: + case GLSL_SAMPLER_DIM_RECT: + case GLSL_SAMPLER_DIM_EXTERNAL: + return is_array ? aco_image_2darray : aco_image_2d; + case GLSL_SAMPLER_DIM_3D: + return aco_image_3d; + case GLSL_SAMPLER_DIM_CUBE: + return aco_image_cube; + case GLSL_SAMPLER_DIM_MS: + return is_array ? aco_image_2darraymsaa : aco_image_2dmsaa; + case GLSL_SAMPLER_DIM_SUBPASS: + return aco_image_2darray; + case GLSL_SAMPLER_DIM_SUBPASS_MS: + return aco_image_2darraymsaa; + default: + unreachable("bad sampler dim"); + } +} + +static bool +should_declare_array(isel_context *ctx, enum glsl_sampler_dim sampler_dim, bool is_array) { + if (sampler_dim == GLSL_SAMPLER_DIM_BUF) + return false; + aco_image_dim dim = get_sampler_dim(ctx, sampler_dim, is_array); + return dim == aco_image_cube || + dim == aco_image_1darray || + dim == aco_image_2darray || + dim == aco_image_2darraymsaa; +} + +Temp get_sampler_desc(isel_context *ctx, nir_deref_instr *deref_instr, + enum aco_descriptor_type desc_type, + const nir_tex_instr *tex_instr, bool image, bool write) +{ +/* FIXME: we should lower the deref with some new nir_intrinsic_load_desc + std::unordered_map::iterator it = ctx->tex_desc.find((uint64_t) desc_type << 32 | deref_instr->dest.ssa.index); + if (it != ctx->tex_desc.end()) + return it->second; +*/ + Temp index = Temp(); + bool index_set = false; + unsigned constant_index = 0; + unsigned descriptor_set; + unsigned base_index; + Builder bld(ctx->program, ctx->block); + + if (!deref_instr) { + assert(tex_instr && !image); + descriptor_set = 0; + base_index = tex_instr->sampler_index; + } else { + while(deref_instr->deref_type != nir_deref_type_var) { + unsigned array_size = glsl_get_aoa_size(deref_instr->type); + if (!array_size) + array_size = 1; + + assert(deref_instr->deref_type == nir_deref_type_array); + nir_const_value *const_value = nir_src_as_const_value(deref_instr->arr.index); + if (const_value) { + constant_index += array_size * const_value->u32; + } else { + Temp indirect = bld.as_uniform(get_ssa_temp(ctx, deref_instr->arr.index.ssa)); + + if (array_size != 1) + indirect = bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), Operand(array_size), indirect); + + if (!index_set) { + index = indirect; + index_set = true; + } else { + index = bld.sop2(aco_opcode::s_add_i32, bld.def(s1), bld.def(s1, scc), index, indirect); + } + } + + deref_instr = nir_src_as_deref(deref_instr->parent); + } + descriptor_set = deref_instr->var->data.descriptor_set; + base_index = deref_instr->var->data.binding; + } + + Temp list = load_desc_ptr(ctx, descriptor_set); + list = convert_pointer_to_64_bit(ctx, list); + + struct radv_descriptor_set_layout *layout = ctx->options->layout->set[descriptor_set].layout; + struct radv_descriptor_set_binding_layout *binding = layout->binding + base_index; + unsigned offset = binding->offset; + unsigned stride = binding->size; + aco_opcode opcode; + RegClass type; + + assert(base_index < layout->binding_count); + + switch (desc_type) { + case ACO_DESC_IMAGE: + type = s8; + opcode = aco_opcode::s_load_dwordx8; + break; + case ACO_DESC_FMASK: + type = s8; + opcode = aco_opcode::s_load_dwordx8; + offset += 32; + break; + case ACO_DESC_SAMPLER: + type = s4; + opcode = aco_opcode::s_load_dwordx4; + if (binding->type == VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER) + offset += radv_combined_image_descriptor_sampler_offset(binding); + break; + case ACO_DESC_BUFFER: + type = s4; + opcode = aco_opcode::s_load_dwordx4; + break; + case ACO_DESC_PLANE_0: + case ACO_DESC_PLANE_1: + type = s8; + opcode = aco_opcode::s_load_dwordx8; + offset += 32 * (desc_type - ACO_DESC_PLANE_0); + break; + case ACO_DESC_PLANE_2: + type = s4; + opcode = aco_opcode::s_load_dwordx4; + offset += 64; + break; + default: + unreachable("invalid desc_type\n"); + } + + offset += constant_index * stride; + + if (desc_type == ACO_DESC_SAMPLER && binding->immutable_samplers_offset && + (!index_set || binding->immutable_samplers_equal)) { + if (binding->immutable_samplers_equal) + constant_index = 0; + + const uint32_t *samplers = radv_immutable_samplers(layout, binding); + return bld.pseudo(aco_opcode::p_create_vector, bld.def(s4), + Operand(samplers[constant_index * 4 + 0]), + Operand(samplers[constant_index * 4 + 1]), + Operand(samplers[constant_index * 4 + 2]), + Operand(samplers[constant_index * 4 + 3])); + } + + Operand off; + if (!index_set) { + off = Operand(offset); + } else { + off = Operand((Temp)bld.sop2(aco_opcode::s_add_i32, bld.def(s1), bld.def(s1, scc), Operand(offset), + bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), Operand(stride), index))); + } + + Temp res = bld.smem(opcode, bld.def(type), list, off); + + if (desc_type == ACO_DESC_PLANE_2) { + Temp components[8]; + for (unsigned i = 0; i < 8; i++) + components[i] = bld.tmp(s1); + bld.pseudo(aco_opcode::p_split_vector, + Definition(components[0]), + Definition(components[1]), + Definition(components[2]), + Definition(components[3]), + res); + + Temp desc2 = get_sampler_desc(ctx, deref_instr, ACO_DESC_PLANE_1, tex_instr, image, write); + bld.pseudo(aco_opcode::p_split_vector, + bld.def(s1), bld.def(s1), bld.def(s1), bld.def(s1), + Definition(components[4]), + Definition(components[5]), + Definition(components[6]), + Definition(components[7]), + desc2); + + res = bld.pseudo(aco_opcode::p_create_vector, bld.def(s8), + components[0], components[1], components[2], components[3], + components[4], components[5], components[6], components[7]); + } + + return res; +} + +static int image_type_to_components_count(enum glsl_sampler_dim dim, bool array) +{ + switch (dim) { + case GLSL_SAMPLER_DIM_BUF: + return 1; + case GLSL_SAMPLER_DIM_1D: + return array ? 2 : 1; + case GLSL_SAMPLER_DIM_2D: + return array ? 3 : 2; + case GLSL_SAMPLER_DIM_MS: + return array ? 4 : 3; + case GLSL_SAMPLER_DIM_3D: + case GLSL_SAMPLER_DIM_CUBE: + return 3; + case GLSL_SAMPLER_DIM_RECT: + case GLSL_SAMPLER_DIM_SUBPASS: + return 2; + case GLSL_SAMPLER_DIM_SUBPASS_MS: + return 3; + default: + break; + } + return 0; +} + + +/* Adjust the sample index according to FMASK. + * + * For uncompressed MSAA surfaces, FMASK should return 0x76543210, + * which is the identity mapping. Each nibble says which physical sample + * should be fetched to get that sample. + * + * For example, 0x11111100 means there are only 2 samples stored and + * the second sample covers 3/4 of the pixel. When reading samples 0 + * and 1, return physical sample 0 (determined by the first two 0s + * in FMASK), otherwise return physical sample 1. + * + * The sample index should be adjusted as follows: + * sample_index = (fmask >> (sample_index * 4)) & 0xF; + */ +static Temp adjust_sample_index_using_fmask(isel_context *ctx, bool da, Temp coords, Operand sample_index, Temp fmask_desc_ptr) +{ + Builder bld(ctx->program, ctx->block); + Temp fmask = bld.tmp(v1); + + aco_ptr load{create_instruction(aco_opcode::image_load, Format::MIMG, 2, 1)}; + load->operands[0] = Operand(coords); + load->operands[1] = Operand(fmask_desc_ptr); + load->definitions[0] = Definition(fmask); + load->glc = false; + load->dmask = 0x1; + load->unrm = true; + load->da = da; + load->can_reorder = true; /* fmask images shouldn't be modified */ + ctx->block->instructions.emplace_back(std::move(load)); + + Operand sample_index4; + if (sample_index.isConstant() && sample_index.constantValue() < 16) { + sample_index4 = Operand(sample_index.constantValue() << 2); + } else if (sample_index.regClass() == s1) { + sample_index4 = bld.sop2(aco_opcode::s_lshl_b32, bld.def(s1), bld.def(s1, scc), sample_index, Operand(2u)); + } else { + assert(sample_index.regClass() == v1); + sample_index4 = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(2u), sample_index); + } + + Temp final_sample; + if (sample_index4.isConstant() && sample_index4.constantValue() == 0) + final_sample = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(15u), fmask); + else if (sample_index4.isConstant() && sample_index4.constantValue() == 28) + final_sample = bld.vop2(aco_opcode::v_lshrrev_b32, bld.def(v1), Operand(28u), fmask); + else + final_sample = bld.vop3(aco_opcode::v_bfe_u32, bld.def(v1), fmask, sample_index4, Operand(4u)); + + /* Don't rewrite the sample index if WORD1.DATA_FORMAT of the FMASK + * resource descriptor is 0 (invalid), + */ + Temp compare = bld.tmp(s2); + bld.vopc_e64(aco_opcode::v_cmp_lg_u32, Definition(compare), + Operand(0u), emit_extract_vector(ctx, fmask_desc_ptr, 1, s1)).def(0).setHint(vcc); + + Temp sample_index_v = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), sample_index); + + /* Replace the MSAA sample index. */ + return bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), sample_index_v, final_sample, compare); +} + +static Temp get_image_coords(isel_context *ctx, const nir_intrinsic_instr *instr, const struct glsl_type *type) +{ + + Temp src0 = get_ssa_temp(ctx, instr->src[1].ssa); + enum glsl_sampler_dim dim = glsl_get_sampler_dim(type); + bool is_array = glsl_sampler_type_is_array(type); + ASSERTED bool add_frag_pos = (dim == GLSL_SAMPLER_DIM_SUBPASS || dim == GLSL_SAMPLER_DIM_SUBPASS_MS); + assert(!add_frag_pos && "Input attachments should be lowered."); + bool is_ms = (dim == GLSL_SAMPLER_DIM_MS || dim == GLSL_SAMPLER_DIM_SUBPASS_MS); + bool gfx9_1d = ctx->options->chip_class >= GFX9 && dim == GLSL_SAMPLER_DIM_1D; + int count = image_type_to_components_count(dim, is_array); + std::vector coords(count); + + if (is_ms) { + Operand sample_index; + nir_const_value *sample_cv = nir_src_as_const_value(instr->src[2]); + if (sample_cv) + sample_index = Operand(sample_cv->u32); + else + sample_index = Operand(emit_extract_vector(ctx, get_ssa_temp(ctx, instr->src[2].ssa), 0, v1)); + + if (instr->intrinsic == nir_intrinsic_image_deref_load) { + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, is_array ? 3 : 2, 1)}; + for (unsigned i = 0; i < vec->operands.size(); i++) + vec->operands[i] = Operand(emit_extract_vector(ctx, src0, i, v1)); + Temp fmask_load_address = {ctx->program->allocateId(), is_array ? v3 : v2}; + vec->definitions[0] = Definition(fmask_load_address); + ctx->block->instructions.emplace_back(std::move(vec)); + + Temp fmask_desc_ptr = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_FMASK, nullptr, false, false); + sample_index = Operand(adjust_sample_index_using_fmask(ctx, is_array, fmask_load_address, sample_index, fmask_desc_ptr)); + } + count--; + coords[count] = sample_index; + } + + if (count == 1 && !gfx9_1d) + return emit_extract_vector(ctx, src0, 0, v1); + + if (gfx9_1d) { + coords[0] = Operand(emit_extract_vector(ctx, src0, 0, v1)); + coords.resize(coords.size() + 1); + coords[1] = Operand((uint32_t) 0); + if (is_array) + coords[2] = Operand(emit_extract_vector(ctx, src0, 1, v1)); + } else { + for (int i = 0; i < count; i++) + coords[i] = Operand(emit_extract_vector(ctx, src0, i, v1)); + } + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, coords.size(), 1)}; + for (unsigned i = 0; i < coords.size(); i++) + vec->operands[i] = coords[i]; + Temp res = {ctx->program->allocateId(), RegClass(RegType::vgpr, coords.size())}; + vec->definitions[0] = Definition(res); + ctx->block->instructions.emplace_back(std::move(vec)); + return res; +} + + +void visit_image_load(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + const nir_variable *var = nir_deref_instr_get_variable(nir_instr_as_deref(instr->src[0].ssa->parent_instr)); + const struct glsl_type *type = glsl_without_array(var->type); + const enum glsl_sampler_dim dim = glsl_get_sampler_dim(type); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + if (dim == GLSL_SAMPLER_DIM_BUF) { + unsigned mask = nir_ssa_def_components_read(&instr->dest.ssa); + unsigned num_channels = util_last_bit(mask); + Temp rsrc = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_BUFFER, nullptr, true, true); + Temp vindex = emit_extract_vector(ctx, get_ssa_temp(ctx, instr->src[1].ssa), 0, v1); + + aco_opcode opcode; + switch (num_channels) { + case 1: + opcode = aco_opcode::buffer_load_format_x; + break; + case 2: + opcode = aco_opcode::buffer_load_format_xy; + break; + case 3: + opcode = aco_opcode::buffer_load_format_xyz; + break; + case 4: + opcode = aco_opcode::buffer_load_format_xyzw; + break; + default: + unreachable(">4 channel buffer image load"); + } + aco_ptr load{create_instruction(opcode, Format::MUBUF, 3, 1)}; + load->operands[0] = Operand(vindex); + load->operands[1] = Operand(rsrc); + load->operands[2] = Operand((uint32_t) 0); + Temp tmp; + if (num_channels == instr->dest.ssa.num_components && dst.type() == RegType::vgpr) + tmp = dst; + else + tmp = {ctx->program->allocateId(), RegClass(RegType::vgpr, num_channels)}; + load->definitions[0] = Definition(tmp); + load->idxen = true; + load->barrier = barrier_image; + ctx->block->instructions.emplace_back(std::move(load)); + + expand_vector(ctx, tmp, dst, instr->dest.ssa.num_components, (1 << num_channels) - 1); + return; + } + + Temp coords = get_image_coords(ctx, instr, type); + Temp resource = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_IMAGE, nullptr, true, true); + //aco_image_dim img_dim = get_image_dim(ctx, glsl_get_sampler_dim(type), glsl_sampler_type_is_array(type)); + + unsigned dmask = nir_ssa_def_components_read(&instr->dest.ssa); + unsigned num_components = util_bitcount(dmask); + Temp tmp; + if (num_components == instr->dest.ssa.num_components && dst.type() == RegType::vgpr) + tmp = dst; + else + tmp = {ctx->program->allocateId(), RegClass(RegType::vgpr, num_components)}; + + aco_ptr load{create_instruction(aco_opcode::image_load, Format::MIMG, 2, 1)}; + load->operands[0] = Operand(coords); + load->operands[1] = Operand(resource); + load->definitions[0] = Definition(tmp); + load->glc = var->data.image.access & (ACCESS_VOLATILE | ACCESS_COHERENT) ? 1 : 0; + load->dmask = dmask; + load->unrm = true; + load->da = should_declare_array(ctx, dim, glsl_sampler_type_is_array(type)); + load->barrier = barrier_image; + ctx->block->instructions.emplace_back(std::move(load)); + + expand_vector(ctx, tmp, dst, instr->dest.ssa.num_components, dmask); + return; +} + +void visit_image_store(isel_context *ctx, nir_intrinsic_instr *instr) +{ + const nir_variable *var = nir_deref_instr_get_variable(nir_instr_as_deref(instr->src[0].ssa->parent_instr)); + const struct glsl_type *type = glsl_without_array(var->type); + const enum glsl_sampler_dim dim = glsl_get_sampler_dim(type); + Temp data = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[3].ssa)); + + bool glc = ctx->options->chip_class == GFX6 || var->data.image.access & (ACCESS_VOLATILE | ACCESS_COHERENT | ACCESS_NON_READABLE) ? 1 : 0; + + if (dim == GLSL_SAMPLER_DIM_BUF) { + Temp rsrc = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_BUFFER, nullptr, true, true); + Temp vindex = emit_extract_vector(ctx, get_ssa_temp(ctx, instr->src[1].ssa), 0, v1); + aco_opcode opcode; + switch (data.size()) { + case 1: + opcode = aco_opcode::buffer_store_format_x; + break; + case 2: + opcode = aco_opcode::buffer_store_format_xy; + break; + case 3: + opcode = aco_opcode::buffer_store_format_xyz; + break; + case 4: + opcode = aco_opcode::buffer_store_format_xyzw; + break; + default: + unreachable(">4 channel buffer image store"); + } + aco_ptr store{create_instruction(opcode, Format::MUBUF, 4, 0)}; + store->operands[0] = Operand(vindex); + store->operands[1] = Operand(rsrc); + store->operands[2] = Operand((uint32_t) 0); + store->operands[3] = Operand(data); + store->idxen = true; + store->glc = glc; + store->disable_wqm = true; + store->barrier = barrier_image; + ctx->program->needs_exact = true; + ctx->block->instructions.emplace_back(std::move(store)); + return; + } + + assert(data.type() == RegType::vgpr); + Temp coords = get_image_coords(ctx, instr, type); + Temp resource = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_IMAGE, nullptr, true, true); + + aco_ptr store{create_instruction(aco_opcode::image_store, Format::MIMG, 4, 0)}; + store->operands[0] = Operand(coords); + store->operands[1] = Operand(resource); + store->operands[2] = Operand(s4); + store->operands[3] = Operand(data); + store->glc = glc; + store->dmask = (1 << data.size()) - 1; + store->unrm = true; + store->da = should_declare_array(ctx, dim, glsl_sampler_type_is_array(type)); + store->disable_wqm = true; + store->barrier = barrier_image; + ctx->program->needs_exact = true; + ctx->block->instructions.emplace_back(std::move(store)); + return; +} + +void visit_image_atomic(isel_context *ctx, nir_intrinsic_instr *instr) +{ + /* return the previous value if dest is ever used */ + bool return_previous = false; + nir_foreach_use_safe(use_src, &instr->dest.ssa) { + return_previous = true; + break; + } + nir_foreach_if_use_safe(use_src, &instr->dest.ssa) { + return_previous = true; + break; + } + + const nir_variable *var = nir_deref_instr_get_variable(nir_instr_as_deref(instr->src[0].ssa->parent_instr)); + const struct glsl_type *type = glsl_without_array(var->type); + const enum glsl_sampler_dim dim = glsl_get_sampler_dim(type); + Builder bld(ctx->program, ctx->block); + + Temp data = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[3].ssa)); + assert(data.size() == 1 && "64bit ssbo atomics not yet implemented."); + + if (instr->intrinsic == nir_intrinsic_image_deref_atomic_comp_swap) + data = bld.pseudo(aco_opcode::p_create_vector, bld.def(v2), get_ssa_temp(ctx, instr->src[4].ssa), data); + + aco_opcode buf_op, image_op; + switch (instr->intrinsic) { + case nir_intrinsic_image_deref_atomic_add: + buf_op = aco_opcode::buffer_atomic_add; + image_op = aco_opcode::image_atomic_add; + break; + case nir_intrinsic_image_deref_atomic_umin: + buf_op = aco_opcode::buffer_atomic_umin; + image_op = aco_opcode::image_atomic_umin; + break; + case nir_intrinsic_image_deref_atomic_imin: + buf_op = aco_opcode::buffer_atomic_smin; + image_op = aco_opcode::image_atomic_smin; + break; + case nir_intrinsic_image_deref_atomic_umax: + buf_op = aco_opcode::buffer_atomic_umax; + image_op = aco_opcode::image_atomic_umax; + break; + case nir_intrinsic_image_deref_atomic_imax: + buf_op = aco_opcode::buffer_atomic_smax; + image_op = aco_opcode::image_atomic_smax; + break; + case nir_intrinsic_image_deref_atomic_and: + buf_op = aco_opcode::buffer_atomic_and; + image_op = aco_opcode::image_atomic_and; + break; + case nir_intrinsic_image_deref_atomic_or: + buf_op = aco_opcode::buffer_atomic_or; + image_op = aco_opcode::image_atomic_or; + break; + case nir_intrinsic_image_deref_atomic_xor: + buf_op = aco_opcode::buffer_atomic_xor; + image_op = aco_opcode::image_atomic_xor; + break; + case nir_intrinsic_image_deref_atomic_exchange: + buf_op = aco_opcode::buffer_atomic_swap; + image_op = aco_opcode::image_atomic_swap; + break; + case nir_intrinsic_image_deref_atomic_comp_swap: + buf_op = aco_opcode::buffer_atomic_cmpswap; + image_op = aco_opcode::image_atomic_cmpswap; + break; + default: + unreachable("visit_image_atomic should only be called with nir_intrinsic_image_deref_atomic_* instructions."); + } + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + if (dim == GLSL_SAMPLER_DIM_BUF) { + Temp vindex = emit_extract_vector(ctx, get_ssa_temp(ctx, instr->src[1].ssa), 0, v1); + Temp resource = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_BUFFER, nullptr, true, true); + //assert(ctx->options->chip_class < GFX9 && "GFX9 stride size workaround not yet implemented."); + aco_ptr mubuf{create_instruction(buf_op, Format::MUBUF, 4, return_previous ? 1 : 0)}; + mubuf->operands[0] = Operand(vindex); + mubuf->operands[1] = Operand(resource); + mubuf->operands[2] = Operand((uint32_t)0); + mubuf->operands[3] = Operand(data); + if (return_previous) + mubuf->definitions[0] = Definition(dst); + mubuf->offset = 0; + mubuf->idxen = true; + mubuf->glc = return_previous; + mubuf->disable_wqm = true; + mubuf->barrier = barrier_image; + ctx->program->needs_exact = true; + ctx->block->instructions.emplace_back(std::move(mubuf)); + return; + } + + Temp coords = get_image_coords(ctx, instr, type); + Temp resource = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_IMAGE, nullptr, true, true); + aco_ptr mimg{create_instruction(image_op, Format::MIMG, 4, return_previous ? 1 : 0)}; + mimg->operands[0] = Operand(coords); + mimg->operands[1] = Operand(resource); + mimg->operands[2] = Operand(s4); /* no sampler */ + mimg->operands[3] = Operand(data); + if (return_previous) + mimg->definitions[0] = Definition(dst); + mimg->glc = return_previous; + mimg->dmask = (1 << data.size()) - 1; + mimg->unrm = true; + mimg->da = should_declare_array(ctx, dim, glsl_sampler_type_is_array(type)); + mimg->disable_wqm = true; + mimg->barrier = barrier_image; + ctx->program->needs_exact = true; + ctx->block->instructions.emplace_back(std::move(mimg)); + return; +} + +void get_buffer_size(isel_context *ctx, Temp desc, Temp dst, bool in_elements) +{ + if (in_elements && ctx->options->chip_class == GFX8) { + Builder bld(ctx->program, ctx->block); + + Temp stride = emit_extract_vector(ctx, desc, 1, s1); + stride = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), stride, Operand((5u << 16) | 16u)); + stride = bld.vop1(aco_opcode::v_cvt_f32_ubyte0, bld.def(v1), stride); + stride = bld.vop1(aco_opcode::v_rcp_iflag_f32, bld.def(v1), stride); + + Temp size = emit_extract_vector(ctx, desc, 2, s1); + size = bld.vop1(aco_opcode::v_cvt_f32_u32, bld.def(v1), size); + + Temp res = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), size, stride); + res = bld.vop1(aco_opcode::v_cvt_u32_f32, bld.def(v1), res); + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), res); + + // TODO: we can probably calculate this faster on the scalar unit to do: size / stride{1,2,4,8,12,16} + /* idea + * for 1,2,4,8,16, the result is just (stride >> S_FF1_I32_B32) + * in case 12 (or 3?), we have to divide by 3: + * set v_skip in case it's 12 (if we also have to take care of 3, shift first) + * use v_mul_hi_u32 with magic number to divide + * we need some pseudo merge opcode to overwrite the original SALU result with readfirstlane + * disable v_skip + * total: 6 SALU + 2 VALU instructions vs 1 SALU + 6 VALU instructions + */ + + } else { + emit_extract_vector(ctx, desc, 2, dst); + } +} + +void visit_image_size(isel_context *ctx, nir_intrinsic_instr *instr) +{ + const nir_variable *var = nir_deref_instr_get_variable(nir_instr_as_deref(instr->src[0].ssa->parent_instr)); + const struct glsl_type *type = glsl_without_array(var->type); + Builder bld(ctx->program, ctx->block); + + if (glsl_get_sampler_dim(type) == GLSL_SAMPLER_DIM_BUF) { + Temp desc = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_BUFFER, NULL, true, false); + return get_buffer_size(ctx, desc, get_ssa_temp(ctx, &instr->dest.ssa), true); + } + + /* LOD */ + Temp lod = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand(0u)); + + /* Resource */ + Temp resource = get_sampler_desc(ctx, nir_instr_as_deref(instr->src[0].ssa->parent_instr), ACO_DESC_IMAGE, NULL, true, false); + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + aco_ptr mimg{create_instruction(aco_opcode::image_get_resinfo, Format::MIMG, 2, 1)}; + mimg->operands[0] = Operand(lod); + mimg->operands[1] = Operand(resource); + unsigned& dmask = mimg->dmask; + mimg->dmask = (1 << instr->dest.ssa.num_components) - 1; + mimg->da = glsl_sampler_type_is_array(type); + mimg->can_reorder = true; + Definition& def = mimg->definitions[0]; + ctx->block->instructions.emplace_back(std::move(mimg)); + + if (glsl_get_sampler_dim(type) == GLSL_SAMPLER_DIM_CUBE && + glsl_sampler_type_is_array(type)) { + + assert(instr->dest.ssa.num_components == 3); + Temp tmp = {ctx->program->allocateId(), v3}; + def = Definition(tmp); + emit_split_vector(ctx, tmp, 3); + + /* divide 3rd value by 6 by multiplying with magic number */ + Temp c = bld.copy(bld.def(s1), Operand((uint32_t) 0x2AAAAAAB)); + Temp by_6 = bld.vop3(aco_opcode::v_mul_hi_i32, bld.def(v1), emit_extract_vector(ctx, tmp, 2, v1), c); + + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + emit_extract_vector(ctx, tmp, 0, v1), + emit_extract_vector(ctx, tmp, 1, v1), + by_6); + + } else if (ctx->options->chip_class >= GFX9 && + glsl_get_sampler_dim(type) == GLSL_SAMPLER_DIM_1D && + glsl_sampler_type_is_array(type)) { + assert(instr->dest.ssa.num_components == 2); + def = Definition(dst); + dmask = 0x5; + } else { + def = Definition(dst); + } + + emit_split_vector(ctx, dst, instr->dest.ssa.num_components); +} + +void visit_load_ssbo(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + unsigned num_components = instr->num_components; + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + Temp rsrc = convert_pointer_to_64_bit(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + rsrc = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), rsrc, Operand(0u)); + + bool glc = nir_intrinsic_access(instr) & (ACCESS_VOLATILE | ACCESS_COHERENT); + load_buffer(ctx, num_components, dst, rsrc, get_ssa_temp(ctx, instr->src[1].ssa), glc); +} + +void visit_store_ssbo(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + Temp data = get_ssa_temp(ctx, instr->src[0].ssa); + unsigned elem_size_bytes = instr->src[0].ssa->bit_size / 8; + unsigned writemask = nir_intrinsic_write_mask(instr); + + Temp offset; + if (ctx->options->chip_class < GFX8) + offset = as_vgpr(ctx,get_ssa_temp(ctx, instr->src[2].ssa)); + else + offset = get_ssa_temp(ctx, instr->src[2].ssa); + + Temp rsrc = convert_pointer_to_64_bit(ctx, get_ssa_temp(ctx, instr->src[1].ssa)); + rsrc = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), rsrc, Operand(0u)); + + bool smem = !ctx->divergent_vals[instr->src[2].ssa->index] && + ctx->options->chip_class >= GFX8; + if (smem) + offset = bld.as_uniform(offset); + bool smem_nonfs = smem && ctx->stage != fragment_fs; + + while (writemask) { + int start, count; + u_bit_scan_consecutive_range(&writemask, &start, &count); + if (count == 3 && smem) { + writemask |= 1u << (start + 2); + count = 2; + } + int num_bytes = count * elem_size_bytes; + + if (num_bytes > 16) { + assert(elem_size_bytes == 8); + writemask |= (((count - 2) << 1) - 1) << (start + 2); + count = 2; + num_bytes = 16; + } + + // TODO: check alignment of sub-dword stores + // TODO: split 3 bytes. there is no store instruction for that + + Temp write_data; + if (count != instr->num_components) { + emit_split_vector(ctx, data, instr->num_components); + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, count, 1)}; + for (int i = 0; i < count; i++) { + Temp elem = emit_extract_vector(ctx, data, start + i, RegClass(data.type(), elem_size_bytes / 4)); + vec->operands[i] = Operand(smem_nonfs ? bld.as_uniform(elem) : elem); + } + write_data = bld.tmp(smem_nonfs ? RegType::sgpr : data.type(), count * elem_size_bytes / 4); + vec->definitions[0] = Definition(write_data); + ctx->block->instructions.emplace_back(std::move(vec)); + } else if (!smem && data.type() != RegType::vgpr) { + assert(num_bytes % 4 == 0); + write_data = bld.copy(bld.def(RegType::vgpr, num_bytes / 4), data); + } else if (smem_nonfs && data.type() == RegType::vgpr) { + assert(num_bytes % 4 == 0); + write_data = bld.as_uniform(data); + } else { + write_data = data; + } + + aco_opcode vmem_op, smem_op; + switch (num_bytes) { + case 4: + vmem_op = aco_opcode::buffer_store_dword; + smem_op = aco_opcode::s_buffer_store_dword; + break; + case 8: + vmem_op = aco_opcode::buffer_store_dwordx2; + smem_op = aco_opcode::s_buffer_store_dwordx2; + break; + case 12: + vmem_op = aco_opcode::buffer_store_dwordx3; + smem_op = aco_opcode::last_opcode; + assert(!smem); + break; + case 16: + vmem_op = aco_opcode::buffer_store_dwordx4; + smem_op = aco_opcode::s_buffer_store_dwordx4; + break; + default: + unreachable("Store SSBO not implemented for this size."); + } + if (ctx->stage == fragment_fs) + smem_op = aco_opcode::p_fs_buffer_store_smem; + + if (smem) { + aco_ptr store{create_instruction(smem_op, Format::SMEM, 3, 0)}; + store->operands[0] = Operand(rsrc); + if (start) { + Temp off = bld.sop2(aco_opcode::s_add_i32, bld.def(s1), bld.def(s1, scc), + offset, Operand(start * elem_size_bytes)); + store->operands[1] = Operand(off); + } else { + store->operands[1] = Operand(offset); + } + if (smem_op != aco_opcode::p_fs_buffer_store_smem) + store->operands[1].setFixed(m0); + store->operands[2] = Operand(write_data); + store->glc = nir_intrinsic_access(instr) & (ACCESS_VOLATILE | ACCESS_COHERENT | ACCESS_NON_READABLE); + store->disable_wqm = true; + store->barrier = barrier_buffer; + ctx->block->instructions.emplace_back(std::move(store)); + ctx->program->wb_smem_l1_on_end = true; + if (smem_op == aco_opcode::p_fs_buffer_store_smem) { + ctx->block->kind |= block_kind_needs_lowering; + ctx->program->needs_exact = true; + } + } else { + aco_ptr store{create_instruction(vmem_op, Format::MUBUF, 4, 0)}; + store->operands[0] = offset.type() == RegType::vgpr ? Operand(offset) : Operand(v1); + store->operands[1] = Operand(rsrc); + store->operands[2] = offset.type() == RegType::sgpr ? Operand(offset) : Operand((uint32_t) 0); + store->operands[3] = Operand(write_data); + store->offset = start * elem_size_bytes; + store->offen = (offset.type() == RegType::vgpr); + store->glc = nir_intrinsic_access(instr) & (ACCESS_VOLATILE | ACCESS_COHERENT | ACCESS_NON_READABLE); + store->disable_wqm = true; + store->barrier = barrier_buffer; + ctx->program->needs_exact = true; + ctx->block->instructions.emplace_back(std::move(store)); + } + } +} + +void visit_atomic_ssbo(isel_context *ctx, nir_intrinsic_instr *instr) +{ + /* return the previous value if dest is ever used */ + bool return_previous = false; + nir_foreach_use_safe(use_src, &instr->dest.ssa) { + return_previous = true; + break; + } + nir_foreach_if_use_safe(use_src, &instr->dest.ssa) { + return_previous = true; + break; + } + + Builder bld(ctx->program, ctx->block); + Temp data = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[2].ssa)); + + if (instr->intrinsic == nir_intrinsic_ssbo_atomic_comp_swap) + data = bld.pseudo(aco_opcode::p_create_vector, bld.def(RegType::vgpr, data.size() * 2), + get_ssa_temp(ctx, instr->src[3].ssa), data); + + Temp offset; + if (ctx->options->chip_class < GFX8) + offset = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[1].ssa)); + else + offset = get_ssa_temp(ctx, instr->src[1].ssa); + + Temp rsrc = convert_pointer_to_64_bit(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + rsrc = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), rsrc, Operand(0u)); + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + aco_opcode op32, op64; + switch (instr->intrinsic) { + case nir_intrinsic_ssbo_atomic_add: + op32 = aco_opcode::buffer_atomic_add; + op64 = aco_opcode::buffer_atomic_add_x2; + break; + case nir_intrinsic_ssbo_atomic_imin: + op32 = aco_opcode::buffer_atomic_smin; + op64 = aco_opcode::buffer_atomic_smin_x2; + break; + case nir_intrinsic_ssbo_atomic_umin: + op32 = aco_opcode::buffer_atomic_umin; + op64 = aco_opcode::buffer_atomic_umin_x2; + break; + case nir_intrinsic_ssbo_atomic_imax: + op32 = aco_opcode::buffer_atomic_smax; + op64 = aco_opcode::buffer_atomic_smax_x2; + break; + case nir_intrinsic_ssbo_atomic_umax: + op32 = aco_opcode::buffer_atomic_umax; + op64 = aco_opcode::buffer_atomic_umax_x2; + break; + case nir_intrinsic_ssbo_atomic_and: + op32 = aco_opcode::buffer_atomic_and; + op64 = aco_opcode::buffer_atomic_and_x2; + break; + case nir_intrinsic_ssbo_atomic_or: + op32 = aco_opcode::buffer_atomic_or; + op64 = aco_opcode::buffer_atomic_or_x2; + break; + case nir_intrinsic_ssbo_atomic_xor: + op32 = aco_opcode::buffer_atomic_xor; + op64 = aco_opcode::buffer_atomic_xor_x2; + break; + case nir_intrinsic_ssbo_atomic_exchange: + op32 = aco_opcode::buffer_atomic_swap; + op64 = aco_opcode::buffer_atomic_swap_x2; + break; + case nir_intrinsic_ssbo_atomic_comp_swap: + op32 = aco_opcode::buffer_atomic_cmpswap; + op64 = aco_opcode::buffer_atomic_cmpswap_x2; + break; + default: + unreachable("visit_atomic_ssbo should only be called with nir_intrinsic_ssbo_atomic_* instructions."); + } + aco_opcode op = instr->dest.ssa.bit_size == 32 ? op32 : op64; + aco_ptr mubuf{create_instruction(op, Format::MUBUF, 4, return_previous ? 1 : 0)}; + mubuf->operands[0] = offset.type() == RegType::vgpr ? Operand(offset) : Operand(v1); + mubuf->operands[1] = Operand(rsrc); + mubuf->operands[2] = offset.type() == RegType::sgpr ? Operand(offset) : Operand((uint32_t) 0); + mubuf->operands[3] = Operand(data); + if (return_previous) + mubuf->definitions[0] = Definition(dst); + mubuf->offset = 0; + mubuf->offen = (offset.type() == RegType::vgpr); + mubuf->glc = return_previous; + mubuf->disable_wqm = true; + mubuf->barrier = barrier_buffer; + ctx->program->needs_exact = true; + ctx->block->instructions.emplace_back(std::move(mubuf)); +} + +void visit_get_buffer_size(isel_context *ctx, nir_intrinsic_instr *instr) { + + Temp index = convert_pointer_to_64_bit(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + Builder bld(ctx->program, ctx->block); + Temp desc = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), index, Operand(0u)); + get_buffer_size(ctx, desc, get_ssa_temp(ctx, &instr->dest.ssa), false); +} + +void visit_load_global(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + unsigned num_components = instr->num_components; + unsigned num_bytes = num_components * instr->dest.ssa.bit_size / 8; + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + Temp addr = get_ssa_temp(ctx, instr->src[0].ssa); + + bool glc = nir_intrinsic_access(instr) & (ACCESS_VOLATILE | ACCESS_COHERENT); + aco_opcode op; + if (dst.type() == RegType::vgpr || (glc && ctx->options->chip_class < GFX8)) { + bool global = ctx->options->chip_class >= GFX9; + aco_opcode op; + switch (num_bytes) { + case 4: + op = global ? aco_opcode::global_load_dword : aco_opcode::flat_load_dword; + break; + case 8: + op = global ? aco_opcode::global_load_dwordx2 : aco_opcode::flat_load_dwordx2; + break; + case 12: + op = global ? aco_opcode::global_load_dwordx3 : aco_opcode::flat_load_dwordx3; + break; + case 16: + op = global ? aco_opcode::global_load_dwordx4 : aco_opcode::flat_load_dwordx4; + break; + default: + unreachable("load_global not implemented for this size."); + } + aco_ptr flat{create_instruction(op, global ? Format::GLOBAL : Format::FLAT, 2, 1)}; + flat->operands[0] = Operand(addr); + flat->operands[1] = Operand(s1); + flat->glc = glc; + + if (dst.type() == RegType::sgpr) { + Temp vec = bld.tmp(RegType::vgpr, dst.size()); + flat->definitions[0] = Definition(vec); + ctx->block->instructions.emplace_back(std::move(flat)); + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), vec); + } else { + flat->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(flat)); + } + emit_split_vector(ctx, dst, num_components); + } else { + switch (num_bytes) { + case 4: + op = aco_opcode::s_load_dword; + break; + case 8: + op = aco_opcode::s_load_dwordx2; + break; + case 12: + case 16: + op = aco_opcode::s_load_dwordx4; + break; + default: + unreachable("load_global not implemented for this size."); + } + aco_ptr load{create_instruction(op, Format::SMEM, 2, 1)}; + load->operands[0] = Operand(addr); + load->operands[1] = Operand(0u); + load->definitions[0] = Definition(dst); + load->glc = glc; + load->barrier = barrier_buffer; + assert(ctx->options->chip_class >= GFX8 || !glc); + + if (dst.size() == 3) { + /* trim vector */ + Temp vec = bld.tmp(s4); + load->definitions[0] = Definition(vec); + ctx->block->instructions.emplace_back(std::move(load)); + emit_split_vector(ctx, vec, 4); + + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + emit_extract_vector(ctx, vec, 0, s1), + emit_extract_vector(ctx, vec, 1, s1), + emit_extract_vector(ctx, vec, 2, s1)); + } else { + ctx->block->instructions.emplace_back(std::move(load)); + } + } +} + +void visit_store_global(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + unsigned elem_size_bytes = instr->src[0].ssa->bit_size / 8; + + Temp data = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + Temp addr = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[1].ssa)); + + unsigned writemask = nir_intrinsic_write_mask(instr); + while (writemask) { + int start, count; + u_bit_scan_consecutive_range(&writemask, &start, &count); + unsigned num_bytes = count * elem_size_bytes; + + Temp write_data = data; + if (count != instr->num_components) { + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, count, 1)}; + for (int i = 0; i < count; i++) + vec->operands[i] = Operand(emit_extract_vector(ctx, data, start + i, v1)); + write_data = bld.tmp(RegType::vgpr, count); + vec->definitions[0] = Definition(write_data); + ctx->block->instructions.emplace_back(std::move(vec)); + } + + unsigned offset = start * elem_size_bytes; + if (offset > 0 && ctx->options->chip_class < GFX9) { + Temp addr0 = bld.tmp(v1), addr1 = bld.tmp(v1); + Temp new_addr0 = bld.tmp(v1), new_addr1 = bld.tmp(v1); + Temp carry = bld.tmp(s2); + bld.pseudo(aco_opcode::p_split_vector, Definition(addr0), Definition(addr1), addr); + + bld.vop2(aco_opcode::v_add_co_u32, Definition(new_addr0), bld.hint_vcc(Definition(carry)), + Operand(offset), addr0); + bld.vop2(aco_opcode::v_addc_co_u32, Definition(new_addr1), bld.def(s2), + Operand(0u), addr1, + carry).def(1).setHint(vcc); + + addr = bld.pseudo(aco_opcode::p_create_vector, bld.def(v2), new_addr0, new_addr1); + + offset = 0; + } + + bool glc = nir_intrinsic_access(instr) & (ACCESS_VOLATILE | ACCESS_COHERENT | ACCESS_NON_READABLE); + bool global = ctx->options->chip_class >= GFX9; + aco_opcode op; + switch (num_bytes) { + case 4: + op = global ? aco_opcode::global_store_dword : aco_opcode::flat_store_dword; + break; + case 8: + op = global ? aco_opcode::global_store_dwordx2 : aco_opcode::flat_store_dwordx2; + break; + case 12: + op = global ? aco_opcode::global_store_dwordx3 : aco_opcode::flat_store_dwordx3; + break; + case 16: + op = global ? aco_opcode::global_store_dwordx4 : aco_opcode::flat_store_dwordx4; + break; + default: + unreachable("store_global not implemented for this size."); + } + aco_ptr flat{create_instruction(op, global ? Format::GLOBAL : Format::FLAT, 3, 0)}; + flat->operands[0] = Operand(addr); + flat->operands[1] = Operand(s1); + flat->operands[2] = Operand(data); + flat->glc = glc; + flat->offset = offset; + ctx->block->instructions.emplace_back(std::move(flat)); + } +} + +void emit_memory_barrier(isel_context *ctx, nir_intrinsic_instr *instr) { + Builder bld(ctx->program, ctx->block); + switch(instr->intrinsic) { + case nir_intrinsic_group_memory_barrier: + case nir_intrinsic_memory_barrier: + bld.barrier(aco_opcode::p_memory_barrier_all); + break; + case nir_intrinsic_memory_barrier_atomic_counter: + bld.barrier(aco_opcode::p_memory_barrier_atomic); + break; + case nir_intrinsic_memory_barrier_buffer: + bld.barrier(aco_opcode::p_memory_barrier_buffer); + break; + case nir_intrinsic_memory_barrier_image: + bld.barrier(aco_opcode::p_memory_barrier_image); + break; + case nir_intrinsic_memory_barrier_shared: + bld.barrier(aco_opcode::p_memory_barrier_shared); + break; + default: + unreachable("Unimplemented memory barrier intrinsic"); + break; + } +} + +Operand load_lds_size_m0(isel_context *ctx) +{ + /* TODO: m0 does not need to be initialized on GFX9+ */ + Builder bld(ctx->program, ctx->block); + return bld.m0((Temp)bld.sopk(aco_opcode::s_movk_i32, bld.def(s1, m0), 0xffff)); +} + + +void visit_load_shared(isel_context *ctx, nir_intrinsic_instr *instr) +{ + // TODO: implement sparse reads using ds_read2_b32 and nir_ssa_def_components_read() + Operand m = load_lds_size_m0(ctx); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + assert(instr->dest.ssa.bit_size >= 32 && "Bitsize not supported in load_shared."); + Temp address = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + Builder bld(ctx->program, ctx->block); + + unsigned elem_size_bytes = instr->dest.ssa.bit_size / 8; + unsigned bytes_read = 0; + unsigned result_size = 0; + unsigned total_bytes = instr->num_components * elem_size_bytes; + unsigned align = nir_intrinsic_align_mul(instr) ? nir_intrinsic_align(instr) : instr->dest.ssa.bit_size / 8; + std::array result; + + while (bytes_read < total_bytes) { + unsigned todo = total_bytes - bytes_read; + bool aligned8 = bytes_read % 8 == 0 && align % 8 == 0; + bool aligned16 = bytes_read % 16 == 0 && align % 16 == 0; + + aco_opcode op = aco_opcode::last_opcode; + if (todo >= 16 && aligned16) { + op = aco_opcode::ds_read_b128; + todo = 16; + } else if (todo >= 12 && aligned16) { + op = aco_opcode::ds_read_b96; + todo = 12; + } else if (todo >= 8) { + op = aligned8 ? aco_opcode::ds_read_b64 : aco_opcode::ds_read2_b32; + todo = 8; + } else if (todo >= 4) { + op = aco_opcode::ds_read_b32; + todo = 4; + } else { + assert(false); + } + assert(todo % elem_size_bytes == 0); + unsigned num_elements = todo / elem_size_bytes; + unsigned offset = nir_intrinsic_base(instr) + bytes_read; + unsigned max_offset = op == aco_opcode::ds_read2_b32 ? 1019 : 65535; + + Temp address_offset = address; + if (offset > max_offset) { + address_offset = bld.vadd32(bld.def(v1), Operand((uint32_t)nir_intrinsic_base(instr)), address_offset); + offset = bytes_read; + } + assert(offset <= max_offset); /* bytes_read shouldn't be large enough for this to happen */ + + Temp res; + if (instr->num_components == 1 && dst.type() == RegType::vgpr) + res = dst; + else + res = bld.tmp(RegClass(RegType::vgpr, todo / 4)); + + if (op == aco_opcode::ds_read2_b32) + res = bld.ds(op, Definition(res), address_offset, m, offset >> 2, (offset >> 2) + 1); + else + res = bld.ds(op, Definition(res), address_offset, m, offset); + + if (instr->num_components == 1) { + assert(todo == total_bytes); + if (dst.type() == RegType::sgpr) + bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), res); + return; + } + + if (dst.type() == RegType::sgpr) + res = bld.as_uniform(res); + + if (num_elements == 1) { + result[result_size++] = res; + } else { + assert(res != dst && res.size() % num_elements == 0); + aco_ptr split{create_instruction(aco_opcode::p_split_vector, Format::PSEUDO, 1, num_elements)}; + split->operands[0] = Operand(res); + for (unsigned i = 0; i < num_elements; i++) + split->definitions[i] = Definition(result[result_size++] = bld.tmp(res.type(), elem_size_bytes / 4)); + ctx->block->instructions.emplace_back(std::move(split)); + } + + bytes_read += todo; + } + + assert(result_size == instr->num_components && result_size > 1); + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, result_size, 1)}; + for (unsigned i = 0; i < result_size; i++) + vec->operands[i] = Operand(result[i]); + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + ctx->allocated_vec.emplace(dst.id(), result); +} + +void ds_write_helper(isel_context *ctx, Operand m, Temp address, Temp data, unsigned offset0, unsigned offset1, unsigned align) +{ + Builder bld(ctx->program, ctx->block); + unsigned bytes_written = 0; + while (bytes_written < data.size() * 4) { + unsigned todo = data.size() * 4 - bytes_written; + bool aligned8 = bytes_written % 8 == 0 && align % 8 == 0; + bool aligned16 = bytes_written % 16 == 0 && align % 16 == 0; + + aco_opcode op = aco_opcode::last_opcode; + unsigned size = 0; + if (todo >= 16 && aligned16) { + op = aco_opcode::ds_write_b128; + size = 4; + } else if (todo >= 12 && aligned16) { + op = aco_opcode::ds_write_b96; + size = 3; + } else if (todo >= 8) { + op = aligned8 ? aco_opcode::ds_write_b64 : aco_opcode::ds_write2_b32; + size = 2; + } else if (todo >= 4) { + op = aco_opcode::ds_write_b32; + size = 1; + } else { + assert(false); + } + + bool write2 = op == aco_opcode::ds_write2_b32; + unsigned offset = offset0 + offset1 + bytes_written; + unsigned max_offset = write2 ? 1020 : 65535; + Temp address_offset = address; + if (offset > max_offset) { + address_offset = bld.vadd32(bld.def(v1), Operand(offset0), address_offset); + offset = offset1 + bytes_written; + } + assert(offset <= max_offset); /* offset1 shouldn't be large enough for this to happen */ + + if (write2) { + Temp val0 = emit_extract_vector(ctx, data, bytes_written >> 2, v1); + Temp val1 = emit_extract_vector(ctx, data, (bytes_written >> 2) + 1, v1); + bld.ds(op, address_offset, val0, val1, m, offset >> 2, (offset >> 2) + 1); + } else { + Temp val = emit_extract_vector(ctx, data, bytes_written >> 2, RegClass(RegType::vgpr, size)); + bld.ds(op, address_offset, val, m, offset); + } + + bytes_written += size * 4; + } +} + +void visit_store_shared(isel_context *ctx, nir_intrinsic_instr *instr) +{ + unsigned offset = nir_intrinsic_base(instr); + unsigned writemask = nir_intrinsic_write_mask(instr); + Operand m = load_lds_size_m0(ctx); + Temp data = get_ssa_temp(ctx, instr->src[0].ssa); + Temp address = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[1].ssa)); + unsigned elem_size_bytes = instr->src[0].ssa->bit_size / 8; + assert(elem_size_bytes >= 4 && "Only 32bit & 64bit store_shared currently supported."); + + /* we need at most two stores for 32bit variables */ + int start[2], count[2]; + u_bit_scan_consecutive_range(&writemask, &start[0], &count[0]); + u_bit_scan_consecutive_range(&writemask, &start[1], &count[1]); + assert(writemask == 0); + + /* one combined store is sufficient */ + if (count[0] == count[1]) { + Builder bld(ctx->program, ctx->block); + + Temp address_offset = address; + if ((offset >> 2) + start[1] > 255) { + address_offset = bld.vadd32(bld.def(v1), Operand(offset), address_offset); + offset = 0; + } + + assert(count[0] == 1); + Temp val0 = emit_extract_vector(ctx, data, start[0], v1); + Temp val1 = emit_extract_vector(ctx, data, start[1], v1); + aco_opcode op = elem_size_bytes == 4 ? aco_opcode::ds_write2_b32 : aco_opcode::ds_write2_b64; + offset = offset / elem_size_bytes; + bld.ds(op, address_offset, val0, val1, m, + offset + start[0], offset + start[1]); + return; + } + + unsigned align = nir_intrinsic_align_mul(instr) ? nir_intrinsic_align(instr) : elem_size_bytes; + for (unsigned i = 0; i < 2; i++) { + if (count[i] == 0) + continue; + + Temp write_data = emit_extract_vector(ctx, data, start[i], RegClass(RegType::vgpr, count[i] * elem_size_bytes / 4)); + ds_write_helper(ctx, m, address, write_data, offset, start[i] * elem_size_bytes, align); + } + return; +} + +void visit_shared_atomic(isel_context *ctx, nir_intrinsic_instr *instr) +{ + unsigned offset = nir_intrinsic_base(instr); + Operand m = load_lds_size_m0(ctx); + Temp data = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[1].ssa)); + Temp address = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + + unsigned num_operands = 3; + aco_opcode op32, op64, op32_rtn, op64_rtn; + switch(instr->intrinsic) { + case nir_intrinsic_shared_atomic_add: + op32 = aco_opcode::ds_add_u32; + op64 = aco_opcode::ds_add_u64; + op32_rtn = aco_opcode::ds_add_rtn_u32; + op64_rtn = aco_opcode::ds_add_rtn_u64; + break; + case nir_intrinsic_shared_atomic_imin: + op32 = aco_opcode::ds_min_i32; + op64 = aco_opcode::ds_min_i64; + op32_rtn = aco_opcode::ds_min_rtn_i32; + op64_rtn = aco_opcode::ds_min_rtn_i64; + break; + case nir_intrinsic_shared_atomic_umin: + op32 = aco_opcode::ds_min_u32; + op64 = aco_opcode::ds_min_u64; + op32_rtn = aco_opcode::ds_min_rtn_u32; + op64_rtn = aco_opcode::ds_min_rtn_u64; + break; + case nir_intrinsic_shared_atomic_imax: + op32 = aco_opcode::ds_max_i32; + op64 = aco_opcode::ds_max_i64; + op32_rtn = aco_opcode::ds_max_rtn_i32; + op64_rtn = aco_opcode::ds_max_rtn_i64; + break; + case nir_intrinsic_shared_atomic_umax: + op32 = aco_opcode::ds_max_u32; + op64 = aco_opcode::ds_max_u64; + op32_rtn = aco_opcode::ds_max_rtn_u32; + op64_rtn = aco_opcode::ds_max_rtn_u64; + break; + case nir_intrinsic_shared_atomic_and: + op32 = aco_opcode::ds_and_b32; + op64 = aco_opcode::ds_and_b64; + op32_rtn = aco_opcode::ds_and_rtn_b32; + op64_rtn = aco_opcode::ds_and_rtn_b64; + break; + case nir_intrinsic_shared_atomic_or: + op32 = aco_opcode::ds_or_b32; + op64 = aco_opcode::ds_or_b64; + op32_rtn = aco_opcode::ds_or_rtn_b32; + op64_rtn = aco_opcode::ds_or_rtn_b64; + break; + case nir_intrinsic_shared_atomic_xor: + op32 = aco_opcode::ds_xor_b32; + op64 = aco_opcode::ds_xor_b64; + op32_rtn = aco_opcode::ds_xor_rtn_b32; + op64_rtn = aco_opcode::ds_xor_rtn_b64; + break; + case nir_intrinsic_shared_atomic_exchange: + op32 = aco_opcode::ds_write_b32; + op64 = aco_opcode::ds_write_b64; + op32_rtn = aco_opcode::ds_wrxchg_rtn_b32; + op64_rtn = aco_opcode::ds_wrxchg2_rtn_b64; + break; + case nir_intrinsic_shared_atomic_comp_swap: + op32 = aco_opcode::ds_cmpst_b32; + op64 = aco_opcode::ds_cmpst_b64; + op32_rtn = aco_opcode::ds_cmpst_rtn_b32; + op64_rtn = aco_opcode::ds_cmpst_rtn_b64; + num_operands = 4; + break; + default: + unreachable("Unhandled shared atomic intrinsic"); + } + + /* return the previous value if dest is ever used */ + bool return_previous = false; + nir_foreach_use_safe(use_src, &instr->dest.ssa) { + return_previous = true; + break; + } + nir_foreach_if_use_safe(use_src, &instr->dest.ssa) { + return_previous = true; + break; + } + + aco_opcode op; + if (data.size() == 1) { + assert(instr->dest.ssa.bit_size == 32); + op = return_previous ? op32_rtn : op32; + } else { + assert(instr->dest.ssa.bit_size == 64); + op = return_previous ? op64_rtn : op64; + } + + if (offset > 65535) { + Builder bld(ctx->program, ctx->block); + address = bld.vadd32(bld.def(v1), Operand(offset), address); + offset = 0; + } + + aco_ptr ds; + ds.reset(create_instruction(op, Format::DS, num_operands, return_previous ? 1 : 0)); + ds->operands[0] = Operand(address); + ds->operands[1] = Operand(data); + if (num_operands == 4) + ds->operands[2] = Operand(get_ssa_temp(ctx, instr->src[2].ssa)); + ds->operands[num_operands - 1] = m; + ds->offset0 = offset; + if (return_previous) + ds->definitions[0] = Definition(get_ssa_temp(ctx, &instr->dest.ssa)); + ctx->block->instructions.emplace_back(std::move(ds)); +} + +void visit_load_scratch(isel_context *ctx, nir_intrinsic_instr *instr) { + assert(instr->dest.ssa.bit_size == 32 || instr->dest.ssa.bit_size == 64); + Builder bld(ctx->program, ctx->block); + Temp scratch_addr = ctx->private_segment_buffer; + if (ctx->stage != MESA_SHADER_COMPUTE) + scratch_addr = bld.smem(aco_opcode::s_load_dwordx2, bld.def(s2), ctx->private_segment_buffer, Operand(0u)); + uint32_t rsrc_conf; + /* older generations need element size = 16 bytes */ + if (ctx->program->chip_class >= GFX9) + rsrc_conf = 0x00E00000u; + else + rsrc_conf = 0x00F80000u; + /* buffer res = addr + num_records = -1, index_stride = 64, add_tid_enable = true */ + Temp rsrc = bld.pseudo(aco_opcode::p_create_vector, bld.def(s4), scratch_addr, Operand(-1u), Operand(rsrc_conf)); + Temp offset = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + aco_opcode op; + switch (dst.size()) { + case 1: + op = aco_opcode::buffer_load_dword; + break; + case 2: + op = aco_opcode::buffer_load_dwordx2; + break; + case 3: + op = aco_opcode::buffer_load_dwordx3; + break; + case 4: + op = aco_opcode::buffer_load_dwordx4; + break; + case 6: + case 8: { + std::array elems; + Temp lower = bld.mubuf(aco_opcode::buffer_load_dwordx4, + bld.def(v4), offset, rsrc, + ctx->scratch_offset, 0, true); + Temp upper = bld.mubuf(dst.size() == 6 ? aco_opcode::buffer_load_dwordx2 : + aco_opcode::buffer_load_dwordx4, + dst.size() == 6 ? bld.def(v2) : bld.def(v4), + offset, rsrc, ctx->scratch_offset, 16, true); + emit_split_vector(ctx, lower, 2); + elems[0] = emit_extract_vector(ctx, lower, 0, v2); + elems[1] = emit_extract_vector(ctx, lower, 1, v2); + if (dst.size() == 8) { + emit_split_vector(ctx, upper, 2); + elems[2] = emit_extract_vector(ctx, upper, 0, v2); + elems[3] = emit_extract_vector(ctx, upper, 1, v2); + } else { + elems[2] = upper; + } + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, + Format::PSEUDO, dst.size() / 2, 1)}; + for (unsigned i = 0; i < dst.size() / 2; i++) + vec->operands[i] = Operand(elems[i]); + vec->definitions[0] = Definition(dst); + bld.insert(std::move(vec)); + ctx->allocated_vec.emplace(dst.id(), elems); + return; + } + default: + unreachable("Wrong dst size for nir_intrinsic_load_scratch"); + } + + bld.mubuf(op, Definition(dst), offset, rsrc, ctx->scratch_offset, 0, true); + emit_split_vector(ctx, dst, instr->num_components); +} + +void visit_store_scratch(isel_context *ctx, nir_intrinsic_instr *instr) { + assert(instr->src[0].ssa->bit_size == 32 || instr->src[0].ssa->bit_size == 64); + Builder bld(ctx->program, ctx->block); + Temp scratch_addr = ctx->private_segment_buffer; + if (ctx->stage != MESA_SHADER_COMPUTE) + scratch_addr = bld.smem(aco_opcode::s_load_dwordx2, bld.def(s2), ctx->private_segment_buffer, Operand(0u)); + uint32_t rsrc_conf; + /* older generations need element size = 16 bytes */ + if (ctx->program->chip_class >= GFX9) + rsrc_conf = 0x00E00000u; + else + rsrc_conf = 0x00F80000u; + /* buffer res = addr + num_records = -1, index_stride = 64, add_tid_enable = true */ + Temp rsrc = bld.pseudo(aco_opcode::p_create_vector, bld.def(s4), scratch_addr, Operand(-1u), Operand(rsrc_conf)); + Temp data = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + Temp offset = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[1].ssa)); + + unsigned elem_size_bytes = instr->src[0].ssa->bit_size / 8; + unsigned writemask = nir_intrinsic_write_mask(instr); + + while (writemask) { + int start, count; + u_bit_scan_consecutive_range(&writemask, &start, &count); + int num_bytes = count * elem_size_bytes; + + if (num_bytes > 16) { + assert(elem_size_bytes == 8); + writemask |= (((count - 2) << 1) - 1) << (start + 2); + count = 2; + num_bytes = 16; + } + + // TODO: check alignment of sub-dword stores + // TODO: split 3 bytes. there is no store instruction for that + + Temp write_data; + if (count != instr->num_components) { + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, count, 1)}; + for (int i = 0; i < count; i++) { + Temp elem = emit_extract_vector(ctx, data, start + i, RegClass(RegType::vgpr, elem_size_bytes / 4)); + vec->operands[i] = Operand(elem); + } + write_data = bld.tmp(RegClass(RegType::vgpr, count * elem_size_bytes / 4)); + vec->definitions[0] = Definition(write_data); + ctx->block->instructions.emplace_back(std::move(vec)); + } else { + write_data = data; + } + + aco_opcode op; + switch (num_bytes) { + case 4: + op = aco_opcode::buffer_store_dword; + break; + case 8: + op = aco_opcode::buffer_store_dwordx2; + break; + case 12: + op = aco_opcode::buffer_store_dwordx3; + break; + case 16: + op = aco_opcode::buffer_store_dwordx4; + break; + default: + unreachable("Invalid data size for nir_intrinsic_store_scratch."); + } + + bld.mubuf(op, offset, rsrc, ctx->scratch_offset, write_data, start * elem_size_bytes, true); + } +} + +void visit_load_sample_mask_in(isel_context *ctx, nir_intrinsic_instr *instr) { + uint8_t log2_ps_iter_samples; + if (ctx->program->info->ps.force_persample) { + log2_ps_iter_samples = + util_logbase2(ctx->options->key.fs.num_samples); + } else { + log2_ps_iter_samples = ctx->options->key.fs.log2_ps_iter_samples; + } + + /* The bit pattern matches that used by fixed function fragment + * processing. */ + static const unsigned ps_iter_masks[] = { + 0xffff, /* not used */ + 0x5555, + 0x1111, + 0x0101, + 0x0001, + }; + assert(log2_ps_iter_samples < ARRAY_SIZE(ps_iter_masks)); + + Builder bld(ctx->program, ctx->block); + + Temp sample_id = bld.vop3(aco_opcode::v_bfe_u32, bld.def(v1), ctx->fs_inputs[fs_input::ancillary], Operand(8u), Operand(4u)); + Temp ps_iter_mask = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand(ps_iter_masks[log2_ps_iter_samples])); + Temp mask = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), sample_id, ps_iter_mask); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.vop2(aco_opcode::v_and_b32, Definition(dst), mask, ctx->fs_inputs[fs_input::sample_coverage]); +} + +Temp emit_boolean_reduce(isel_context *ctx, nir_op op, unsigned cluster_size, Temp src) +{ + Builder bld(ctx->program, ctx->block); + + if (cluster_size == 1) { + return src; + } if (op == nir_op_iand && cluster_size == 4) { + //subgroupClusteredAnd(val, 4) -> ~wqm(exec & ~val) + Temp tmp = bld.sop2(aco_opcode::s_andn2_b64, bld.def(s2), bld.def(s1, scc), Operand(exec, s2), src); + return bld.sop1(aco_opcode::s_not_b64, bld.def(s2), bld.def(s1, scc), + bld.sop1(aco_opcode::s_wqm_b64, bld.def(s2), bld.def(s1, scc), tmp)); + } else if (op == nir_op_ior && cluster_size == 4) { + //subgroupClusteredOr(val, 4) -> wqm(val & exec) + return bld.sop1(aco_opcode::s_wqm_b64, bld.def(s2), bld.def(s1, scc), + bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2))); + } else if (op == nir_op_iand && cluster_size == 64) { + //subgroupAnd(val) -> (exec & ~val) == 0 + Temp tmp = bld.sop2(aco_opcode::s_andn2_b64, bld.def(s2), bld.def(s1, scc), Operand(exec, s2), src).def(1).getTemp(); + return bld.sopc(aco_opcode::s_cmp_eq_u32, bld.def(s1, scc), tmp, Operand(0u)); + } else if (op == nir_op_ior && cluster_size == 64) { + //subgroupOr(val) -> (val & exec) != 0 + return bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)).def(1).getTemp(); + } else if (op == nir_op_ixor && cluster_size == 64) { + //subgroupXor(val) -> s_bcnt1_i32_b64(val & exec) & 1 + Temp tmp = bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)); + tmp = bld.sop1(aco_opcode::s_bcnt1_i32_b64, bld.def(s2), bld.def(s1, scc), tmp); + return bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), tmp, Operand(1u)).def(1).getTemp(); + } else { + //subgroupClustered{And,Or,Xor}(val, n) -> + //lane_id = v_mbcnt_hi_u32_b32(-1, v_mbcnt_lo_u32_b32(-1, 0)) + //cluster_offset = ~(n - 1) & lane_id + //cluster_mask = ((1 << n) - 1) + //subgroupClusteredAnd(): + // return ((val | ~exec) >> cluster_offset) & cluster_mask == cluster_mask + //subgroupClusteredOr(): + // return ((val & exec) >> cluster_offset) & cluster_mask != 0 + //subgroupClusteredXor(): + // return v_bnt_u32_b32(((val & exec) >> cluster_offset) & cluster_mask, 0) & 1 != 0 + Temp lane_id = bld.vop3(aco_opcode::v_mbcnt_hi_u32_b32, bld.def(v1), Operand((uint32_t) -1), + bld.vop3(aco_opcode::v_mbcnt_lo_u32_b32, bld.def(v1), Operand((uint32_t) -1), Operand(0u))); + Temp cluster_offset = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(~uint32_t(cluster_size - 1)), lane_id); + + Temp tmp; + if (op == nir_op_iand) + tmp = bld.sop2(aco_opcode::s_orn2_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)); + else + tmp = bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)); + + uint32_t cluster_mask = cluster_size == 32 ? -1 : (1u << cluster_size) - 1u; + tmp = bld.vop3(aco_opcode::v_lshrrev_b64, bld.def(v2), cluster_offset, tmp); + tmp = emit_extract_vector(ctx, tmp, 0, v1); + if (cluster_mask != 0xffffffff) + tmp = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(cluster_mask), tmp); + + Definition cmp_def = Definition(); + if (op == nir_op_iand) { + cmp_def = bld.vopc(aco_opcode::v_cmp_eq_u32, bld.def(s2), Operand(cluster_mask), tmp).def(0); + } else if (op == nir_op_ior) { + cmp_def = bld.vopc(aco_opcode::v_cmp_lg_u32, bld.def(s2), Operand(0u), tmp).def(0); + } else if (op == nir_op_ixor) { + tmp = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(1u), + bld.vop3(aco_opcode::v_bcnt_u32_b32, bld.def(v1), tmp, Operand(0u))); + cmp_def = bld.vopc(aco_opcode::v_cmp_lg_u32, bld.def(s2), Operand(0u), tmp).def(0); + } + cmp_def.setHint(vcc); + return cmp_def.getTemp(); + } +} + +Temp emit_boolean_exclusive_scan(isel_context *ctx, nir_op op, Temp src) +{ + Builder bld(ctx->program, ctx->block); + + //subgroupExclusiveAnd(val) -> mbcnt(exec & ~val) == 0 + //subgroupExclusiveOr(val) -> mbcnt(val & exec) != 0 + //subgroupExclusiveXor(val) -> mbcnt(val & exec) & 1 != 0 + Temp tmp; + if (op == nir_op_iand) + tmp = bld.sop2(aco_opcode::s_andn2_b64, bld.def(s2), bld.def(s1, scc), Operand(exec, s2), src); + else + tmp = bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)); + + Builder::Result lohi = bld.pseudo(aco_opcode::p_split_vector, bld.def(s1), bld.def(s1), tmp); + Temp lo = lohi.def(0).getTemp(); + Temp hi = lohi.def(1).getTemp(); + Temp mbcnt = bld.vop3(aco_opcode::v_mbcnt_hi_u32_b32, bld.def(v1), hi, + bld.vop3(aco_opcode::v_mbcnt_lo_u32_b32, bld.def(v1), lo, Operand(0u))); + + Definition cmp_def = Definition(); + if (op == nir_op_iand) + cmp_def = bld.vopc(aco_opcode::v_cmp_eq_u32, bld.def(s2), Operand(0u), mbcnt).def(0); + else if (op == nir_op_ior) + cmp_def = bld.vopc(aco_opcode::v_cmp_lg_u32, bld.def(s2), Operand(0u), mbcnt).def(0); + else if (op == nir_op_ixor) + cmp_def = bld.vopc(aco_opcode::v_cmp_lg_u32, bld.def(s2), Operand(0u), + bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(1u), mbcnt)).def(0); + cmp_def.setHint(vcc); + return cmp_def.getTemp(); +} + +Temp emit_boolean_inclusive_scan(isel_context *ctx, nir_op op, Temp src) +{ + Builder bld(ctx->program, ctx->block); + + //subgroupInclusiveAnd(val) -> subgroupExclusiveAnd(val) && val + //subgroupInclusiveOr(val) -> subgroupExclusiveOr(val) || val + //subgroupInclusiveXor(val) -> subgroupExclusiveXor(val) ^^ val + Temp tmp = emit_boolean_exclusive_scan(ctx, op, src); + if (op == nir_op_iand) + return bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), tmp, src); + else if (op == nir_op_ior) + return bld.sop2(aco_opcode::s_or_b64, bld.def(s2), bld.def(s1, scc), tmp, src); + else if (op == nir_op_ixor) + return bld.sop2(aco_opcode::s_xor_b64, bld.def(s2), bld.def(s1, scc), tmp, src); + + assert(false); + return Temp(); +} + +void emit_uniform_subgroup(isel_context *ctx, nir_intrinsic_instr *instr, Temp src) +{ + Builder bld(ctx->program, ctx->block); + Definition dst(get_ssa_temp(ctx, &instr->dest.ssa)); + if (src.regClass().type() == RegType::vgpr) { + bld.pseudo(aco_opcode::p_as_uniform, dst, src); + } else if (instr->dest.ssa.bit_size == 1 && src.regClass() == s2) { + bld.sopc(aco_opcode::s_cmp_lg_u64, bld.scc(dst), Operand(0u), Operand(src)); + } else if (src.regClass() == s1) { + bld.sop1(aco_opcode::s_mov_b32, dst, src); + } else if (src.regClass() == s2) { + bld.sop1(aco_opcode::s_mov_b64, dst, src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } +} + +void emit_interp_center(isel_context *ctx, Temp dst, Temp pos1, Temp pos2) +{ + Builder bld(ctx->program, ctx->block); + Temp p1 = ctx->fs_inputs[fs_input::persp_center_p1]; + Temp p2 = ctx->fs_inputs[fs_input::persp_center_p2]; + + /* Build DD X/Y */ + Temp tl_1 = bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), p1, dpp_quad_perm(0, 0, 0, 0)); + Temp ddx_1 = bld.vop2_dpp(aco_opcode::v_sub_f32, bld.def(v1), p1, tl_1, dpp_quad_perm(1, 1, 1, 1)); + Temp ddy_1 = bld.vop2_dpp(aco_opcode::v_sub_f32, bld.def(v1), p1, tl_1, dpp_quad_perm(2, 2, 2, 2)); + Temp tl_2 = bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), p2, dpp_quad_perm(0, 0, 0, 0)); + Temp ddx_2 = bld.vop2_dpp(aco_opcode::v_sub_f32, bld.def(v1), p2, tl_2, dpp_quad_perm(1, 1, 1, 1)); + Temp ddy_2 = bld.vop2_dpp(aco_opcode::v_sub_f32, bld.def(v1), p2, tl_2, dpp_quad_perm(2, 2, 2, 2)); + + /* res_k = p_k + ddx_k * pos1 + ddy_k * pos2 */ + Temp tmp1 = bld.vop3(aco_opcode::v_mad_f32, bld.def(v1), ddx_1, pos1, p1); + Temp tmp2 = bld.vop3(aco_opcode::v_mad_f32, bld.def(v1), ddx_2, pos1, p2); + tmp1 = bld.vop3(aco_opcode::v_mad_f32, bld.def(v1), ddy_1, pos2, tmp1); + tmp2 = bld.vop3(aco_opcode::v_mad_f32, bld.def(v1), ddy_2, pos2, tmp2); + Temp wqm1 = bld.tmp(v1); + emit_wqm(ctx, tmp1, wqm1, true); + Temp wqm2 = bld.tmp(v1); + emit_wqm(ctx, tmp2, wqm2, true); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), wqm1, wqm2); + return; +} + +void visit_intrinsic(isel_context *ctx, nir_intrinsic_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + switch(instr->intrinsic) { + case nir_intrinsic_load_barycentric_sample: + case nir_intrinsic_load_barycentric_pixel: + case nir_intrinsic_load_barycentric_centroid: { + glsl_interp_mode mode = (glsl_interp_mode)nir_intrinsic_interp_mode(instr); + fs_input input = get_interp_input(instr->intrinsic, mode); + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + if (input == fs_input::max_inputs) { + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + Operand(0u), Operand(0u)); + } else { + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + ctx->fs_inputs[input], + ctx->fs_inputs[input + 1]); + } + emit_split_vector(ctx, dst, 2); + break; + } + case nir_intrinsic_load_barycentric_at_sample: { + uint32_t sample_pos_offset = RING_PS_SAMPLE_POSITIONS * 16; + switch (ctx->options->key.fs.num_samples) { + case 2: sample_pos_offset += 1 << 3; break; + case 4: sample_pos_offset += 3 << 3; break; + case 8: sample_pos_offset += 7 << 3; break; + default: break; + } + Temp sample_pos; + Temp addr = get_ssa_temp(ctx, instr->src[0].ssa); + nir_const_value* const_addr = nir_src_as_const_value(instr->src[0]); + if (addr.type() == RegType::sgpr) { + Operand offset; + if (const_addr) { + sample_pos_offset += const_addr->u32 << 3; + offset = Operand(sample_pos_offset); + } else if (ctx->options->chip_class >= GFX9) { + offset = bld.sop2(aco_opcode::s_lshl3_add_u32, bld.def(s1), bld.def(s1, scc), addr, Operand(sample_pos_offset)); + } else { + offset = bld.sop2(aco_opcode::s_lshl_b32, bld.def(s1), bld.def(s1, scc), addr, Operand(3u)); + offset = bld.sop2(aco_opcode::s_add_u32, bld.def(s1), bld.def(s1, scc), addr, Operand(sample_pos_offset)); + } + addr = ctx->private_segment_buffer; + sample_pos = bld.smem(aco_opcode::s_load_dwordx2, bld.def(s2), addr, Operand(offset)); + + } else if (ctx->options->chip_class >= GFX9) { + addr = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(3u), addr); + sample_pos = bld.global(aco_opcode::global_load_dwordx2, bld.def(v2), addr, ctx->private_segment_buffer, sample_pos_offset); + } else { + /* addr += ctx->private_segment_buffer + sample_pos_offset */ + Temp tmp0 = bld.tmp(s1); + Temp tmp1 = bld.tmp(s1); + bld.pseudo(aco_opcode::p_split_vector, Definition(tmp0), Definition(tmp1), ctx->private_segment_buffer); + Definition scc_tmp = bld.def(s1, scc); + tmp0 = bld.sop2(aco_opcode::s_add_u32, bld.def(s1), scc_tmp, tmp0, Operand(sample_pos_offset)); + tmp1 = bld.sop2(aco_opcode::s_addc_u32, bld.def(s1), bld.def(s1, scc), tmp1, Operand(0u), scc_tmp.getTemp()); + addr = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(3u), addr); + Temp pck0 = bld.tmp(v1); + Temp carry = bld.vadd32(Definition(pck0), tmp0, addr, true).def(1).getTemp(); + tmp1 = as_vgpr(ctx, tmp1); + Temp pck1 = bld.vop2_e64(aco_opcode::v_addc_co_u32, bld.def(v1), bld.hint_vcc(bld.def(s2)), tmp1, Operand(0u), carry); + addr = bld.pseudo(aco_opcode::p_create_vector, bld.def(v2), pck0, pck1); + + /* sample_pos = flat_load_dwordx2 addr */ + sample_pos = bld.flat(aco_opcode::flat_load_dwordx2, bld.def(v2), addr, Operand(s1)); + } + + /* sample_pos -= 0.5 */ + Temp pos1 = bld.tmp(RegClass(sample_pos.type(), 1)); + Temp pos2 = bld.tmp(RegClass(sample_pos.type(), 1)); + bld.pseudo(aco_opcode::p_split_vector, Definition(pos1), Definition(pos2), sample_pos); + pos1 = bld.vop2_e64(aco_opcode::v_sub_f32, bld.def(v1), pos1, Operand(0x3f000000u)); + pos2 = bld.vop2_e64(aco_opcode::v_sub_f32, bld.def(v1), pos2, Operand(0x3f000000u)); + + emit_interp_center(ctx, get_ssa_temp(ctx, &instr->dest.ssa), pos1, pos2); + break; + } + case nir_intrinsic_load_barycentric_at_offset: { + Temp offset = get_ssa_temp(ctx, instr->src[0].ssa); + RegClass rc = RegClass(offset.type(), 1); + Temp pos1 = bld.tmp(rc), pos2 = bld.tmp(rc); + bld.pseudo(aco_opcode::p_split_vector, Definition(pos1), Definition(pos2), offset); + emit_interp_center(ctx, get_ssa_temp(ctx, &instr->dest.ssa), pos1, pos2); + break; + } + case nir_intrinsic_load_front_face: { + bld.vopc(aco_opcode::v_cmp_lg_u32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), + Operand(0u), ctx->fs_inputs[fs_input::front_face]).def(0).setHint(vcc); + break; + } + case nir_intrinsic_load_view_index: + case nir_intrinsic_load_layer_id: { + if (instr->intrinsic == nir_intrinsic_load_view_index && (ctx->stage & sw_vs)) { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.copy(Definition(dst), Operand(ctx->view_index)); + break; + } + + unsigned idx = nir_intrinsic_base(instr); + bld.vintrp(aco_opcode::v_interp_mov_f32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), + Operand(2u), bld.m0(ctx->prim_mask), idx, 0); + break; + } + case nir_intrinsic_load_frag_coord: { + emit_load_frag_coord(ctx, get_ssa_temp(ctx, &instr->dest.ssa), 4); + break; + } + case nir_intrinsic_load_sample_pos: { + Temp posx = ctx->fs_inputs[fs_input::frag_pos_0]; + Temp posy = ctx->fs_inputs[fs_input::frag_pos_1]; + bld.pseudo(aco_opcode::p_create_vector, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), + posx.id() ? bld.vop1(aco_opcode::v_fract_f32, bld.def(v1), posx) : Operand(0u), + posy.id() ? bld.vop1(aco_opcode::v_fract_f32, bld.def(v1), posy) : Operand(0u)); + break; + } + case nir_intrinsic_load_interpolated_input: + visit_load_interpolated_input(ctx, instr); + break; + case nir_intrinsic_store_output: + visit_store_output(ctx, instr); + break; + case nir_intrinsic_load_input: + visit_load_input(ctx, instr); + break; + case nir_intrinsic_load_ubo: + visit_load_ubo(ctx, instr); + break; + case nir_intrinsic_load_push_constant: + visit_load_push_constant(ctx, instr); + break; + case nir_intrinsic_load_constant: + visit_load_constant(ctx, instr); + break; + case nir_intrinsic_vulkan_resource_index: + visit_load_resource(ctx, instr); + break; + case nir_intrinsic_discard: + visit_discard(ctx, instr); + break; + case nir_intrinsic_discard_if: + visit_discard_if(ctx, instr); + break; + case nir_intrinsic_load_shared: + visit_load_shared(ctx, instr); + break; + case nir_intrinsic_store_shared: + visit_store_shared(ctx, instr); + break; + case nir_intrinsic_shared_atomic_add: + case nir_intrinsic_shared_atomic_imin: + case nir_intrinsic_shared_atomic_umin: + case nir_intrinsic_shared_atomic_imax: + case nir_intrinsic_shared_atomic_umax: + case nir_intrinsic_shared_atomic_and: + case nir_intrinsic_shared_atomic_or: + case nir_intrinsic_shared_atomic_xor: + case nir_intrinsic_shared_atomic_exchange: + case nir_intrinsic_shared_atomic_comp_swap: + visit_shared_atomic(ctx, instr); + break; + case nir_intrinsic_image_deref_load: + visit_image_load(ctx, instr); + break; + case nir_intrinsic_image_deref_store: + visit_image_store(ctx, instr); + break; + case nir_intrinsic_image_deref_atomic_add: + case nir_intrinsic_image_deref_atomic_umin: + case nir_intrinsic_image_deref_atomic_imin: + case nir_intrinsic_image_deref_atomic_umax: + case nir_intrinsic_image_deref_atomic_imax: + case nir_intrinsic_image_deref_atomic_and: + case nir_intrinsic_image_deref_atomic_or: + case nir_intrinsic_image_deref_atomic_xor: + case nir_intrinsic_image_deref_atomic_exchange: + case nir_intrinsic_image_deref_atomic_comp_swap: + visit_image_atomic(ctx, instr); + break; + case nir_intrinsic_image_deref_size: + visit_image_size(ctx, instr); + break; + case nir_intrinsic_load_ssbo: + visit_load_ssbo(ctx, instr); + break; + case nir_intrinsic_store_ssbo: + visit_store_ssbo(ctx, instr); + break; + case nir_intrinsic_load_global: + visit_load_global(ctx, instr); + break; + case nir_intrinsic_store_global: + visit_store_global(ctx, instr); + break; + case nir_intrinsic_ssbo_atomic_add: + case nir_intrinsic_ssbo_atomic_imin: + case nir_intrinsic_ssbo_atomic_umin: + case nir_intrinsic_ssbo_atomic_imax: + case nir_intrinsic_ssbo_atomic_umax: + case nir_intrinsic_ssbo_atomic_and: + case nir_intrinsic_ssbo_atomic_or: + case nir_intrinsic_ssbo_atomic_xor: + case nir_intrinsic_ssbo_atomic_exchange: + case nir_intrinsic_ssbo_atomic_comp_swap: + visit_atomic_ssbo(ctx, instr); + break; + case nir_intrinsic_load_scratch: + visit_load_scratch(ctx, instr); + break; + case nir_intrinsic_store_scratch: + visit_store_scratch(ctx, instr); + break; + case nir_intrinsic_get_buffer_size: + visit_get_buffer_size(ctx, instr); + break; + case nir_intrinsic_barrier: { + unsigned* bsize = ctx->program->info->cs.block_size; + unsigned workgroup_size = bsize[0] * bsize[1] * bsize[2]; + if (workgroup_size > 64) + bld.sopp(aco_opcode::s_barrier); + break; + } + case nir_intrinsic_group_memory_barrier: + case nir_intrinsic_memory_barrier: + case nir_intrinsic_memory_barrier_atomic_counter: + case nir_intrinsic_memory_barrier_buffer: + case nir_intrinsic_memory_barrier_image: + case nir_intrinsic_memory_barrier_shared: + emit_memory_barrier(ctx, instr); + break; + case nir_intrinsic_load_num_work_groups: + case nir_intrinsic_load_work_group_id: + case nir_intrinsic_load_local_invocation_id: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + Temp* ids; + if (instr->intrinsic == nir_intrinsic_load_num_work_groups) + ids = ctx->num_workgroups; + else if (instr->intrinsic == nir_intrinsic_load_work_group_id) + ids = ctx->workgroup_ids; + else + ids = ctx->local_invocation_ids; + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), + ids[0].id() ? Operand(ids[0]) : Operand(1u), + ids[1].id() ? Operand(ids[1]) : Operand(1u), + ids[2].id() ? Operand(ids[2]) : Operand(1u)); + emit_split_vector(ctx, dst, 3); + break; + } + case nir_intrinsic_load_local_invocation_index: { + Temp id = bld.vop3(aco_opcode::v_mbcnt_hi_u32_b32, bld.def(v1), Operand((uint32_t) -1), + bld.vop3(aco_opcode::v_mbcnt_lo_u32_b32, bld.def(v1), Operand((uint32_t) -1), Operand(0u))); + Temp tg_num = bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), Operand(0xfc0u), ctx->tg_size); + bld.vop2(aco_opcode::v_or_b32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), tg_num, id); + break; + } + case nir_intrinsic_load_subgroup_id: { + if (ctx->stage == compute_cs) { + Temp tg_num = bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), Operand(0xfc0u), ctx->tg_size); + bld.sop2(aco_opcode::s_lshr_b32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), bld.def(s1, scc), tg_num, Operand(0x6u)); + } else { + bld.sop1(aco_opcode::s_mov_b32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), Operand(0x0u)); + } + break; + } + case nir_intrinsic_load_subgroup_invocation: { + bld.vop3(aco_opcode::v_mbcnt_hi_u32_b32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), Operand((uint32_t) -1), + bld.vop3(aco_opcode::v_mbcnt_lo_u32_b32, bld.def(v1), Operand((uint32_t) -1), Operand(0u))); + break; + } + case nir_intrinsic_load_num_subgroups: { + if (ctx->stage == compute_cs) + bld.sop2(aco_opcode::s_and_b32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), bld.def(s1, scc), Operand(0x3fu), ctx->tg_size); + else + bld.sop1(aco_opcode::s_mov_b32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), Operand(0x1u)); + break; + } + case nir_intrinsic_ballot: { + Definition tmp = bld.def(s2); + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + if (instr->src[0].ssa->bit_size == 1 && src.regClass() == s2) { + bld.sop2(aco_opcode::s_and_b64, tmp, bld.def(s1, scc), Operand(exec, s2), src); + } else if (instr->src[0].ssa->bit_size == 1 && src.regClass() == s1) { + bld.sop2(aco_opcode::s_cselect_b64, tmp, Operand(exec, s2), Operand(0u), bld.scc(src)); + } else if (instr->src[0].ssa->bit_size == 32 && src.regClass() == v1) { + bld.vopc(aco_opcode::v_cmp_lg_u32, tmp, Operand(0u), src); + } else if (instr->src[0].ssa->bit_size == 64 && src.regClass() == v2) { + bld.vopc(aco_opcode::v_cmp_lg_u64, tmp, Operand(0u), src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + emit_wqm(ctx, tmp.getTemp(), get_ssa_temp(ctx, &instr->dest.ssa)); + break; + } + case nir_intrinsic_shuffle: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + if (!ctx->divergent_vals[instr->dest.ssa.index]) { + emit_uniform_subgroup(ctx, instr, src); + } else { + Temp tid = get_ssa_temp(ctx, instr->src[1].ssa); + assert(tid.regClass() == v1); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + if (src.regClass() == v1) { + tid = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(2u), tid); + emit_wqm(ctx, bld.ds(aco_opcode::ds_bpermute_b32, bld.def(v1), tid, src), dst); + } else if (src.regClass() == v2) { + tid = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(2u), tid); + + Temp lo = bld.tmp(v1), hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lo), Definition(hi), src); + lo = emit_wqm(ctx, bld.ds(aco_opcode::ds_bpermute_b32, bld.def(v1), tid, lo)); + hi = emit_wqm(ctx, bld.ds(aco_opcode::ds_bpermute_b32, bld.def(v1), tid, hi)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lo, hi); + emit_split_vector(ctx, dst, 2); + } else if (instr->dest.ssa.bit_size == 1 && src.regClass() == s2) { + Temp tmp = bld.vop3(aco_opcode::v_lshrrev_b64, bld.def(v2), tid, src); + tmp = emit_extract_vector(ctx, tmp, 0, v1); + tmp = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(1u), tmp); + emit_wqm(ctx, bld.vopc(aco_opcode::v_cmp_lg_u32, bld.def(s2), Operand(0u), tmp), dst); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + } + break; + } + case nir_intrinsic_load_sample_id: { + bld.vop3(aco_opcode::v_bfe_u32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), + ctx->fs_inputs[ancillary], Operand(8u), Operand(4u)); + break; + } + case nir_intrinsic_load_sample_mask_in: { + visit_load_sample_mask_in(ctx, instr); + break; + } + case nir_intrinsic_read_first_invocation: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + if (src.regClass() == v1) { + emit_wqm(ctx, + bld.vop1(aco_opcode::v_readfirstlane_b32, bld.def(s1), src), + dst); + } else if (src.regClass() == v2) { + Temp lo = bld.tmp(v1), hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lo), Definition(hi), src); + lo = emit_wqm(ctx, bld.vop1(aco_opcode::v_readfirstlane_b32, bld.def(s1), lo)); + hi = emit_wqm(ctx, bld.vop1(aco_opcode::v_readfirstlane_b32, bld.def(s1), hi)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lo, hi); + emit_split_vector(ctx, dst, 2); + } else if (instr->dest.ssa.bit_size == 1 && src.regClass() == s2) { + emit_wqm(ctx, + bld.sopc(aco_opcode::s_bitcmp1_b64, bld.def(s1, scc), src, + bld.sop1(aco_opcode::s_ff1_i32_b64, bld.def(s1), Operand(exec, s2))), + dst); + } else if (src.regClass() == s1) { + bld.sop1(aco_opcode::s_mov_b32, Definition(dst), src); + } else if (src.regClass() == s2) { + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_intrinsic_read_invocation: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + Temp lane = get_ssa_temp(ctx, instr->src[1].ssa); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + assert(lane.regClass() == s1); + if (src.regClass() == v1) { + emit_wqm(ctx, bld.vop3(aco_opcode::v_readlane_b32, bld.def(s1), src, lane), dst); + } else if (src.regClass() == v2) { + Temp lo = bld.tmp(v1), hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lo), Definition(hi), src); + lo = emit_wqm(ctx, bld.vop3(aco_opcode::v_readlane_b32, bld.def(s1), lo, lane)); + hi = emit_wqm(ctx, bld.vop3(aco_opcode::v_readlane_b32, bld.def(s1), hi, lane)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lo, hi); + emit_split_vector(ctx, dst, 2); + } else if (instr->dest.ssa.bit_size == 1 && src.regClass() == s2) { + emit_wqm(ctx, bld.sopc(aco_opcode::s_bitcmp1_b64, bld.def(s1, scc), src, lane), dst); + } else if (src.regClass() == s1) { + bld.sop1(aco_opcode::s_mov_b32, Definition(dst), src); + } else if (src.regClass() == s2) { + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), src); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_intrinsic_vote_all: { + Temp src = as_divergent_bool(ctx, get_ssa_temp(ctx, instr->src[0].ssa), false); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + assert(src.regClass() == s2); + assert(dst.regClass() == s1); + + Definition tmp = bld.def(s1); + bld.sopc(aco_opcode::s_cmp_eq_u64, bld.scc(tmp), + bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)), + Operand(exec, s2)); + emit_wqm(ctx, tmp.getTemp(), dst); + break; + } + case nir_intrinsic_vote_any: { + Temp src = as_divergent_bool(ctx, get_ssa_temp(ctx, instr->src[0].ssa), false); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + assert(src.regClass() == s2); + assert(dst.regClass() == s1); + + Definition tmp = bld.def(s1); + bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.scc(tmp), src, Operand(exec, s2)); + emit_wqm(ctx, tmp.getTemp(), dst); + break; + } + case nir_intrinsic_reduce: + case nir_intrinsic_inclusive_scan: + case nir_intrinsic_exclusive_scan: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + nir_op op = (nir_op) nir_intrinsic_reduction_op(instr); + unsigned cluster_size = instr->intrinsic == nir_intrinsic_reduce ? + nir_intrinsic_cluster_size(instr) : 0; + cluster_size = util_next_power_of_two(MIN2(cluster_size ? cluster_size : 64, 64)); + + if (!ctx->divergent_vals[instr->src[0].ssa->index] && (op == nir_op_ior || op == nir_op_iand)) { + emit_uniform_subgroup(ctx, instr, src); + } else if (instr->dest.ssa.bit_size == 1) { + if (op == nir_op_imul || op == nir_op_umin || op == nir_op_imin) + op = nir_op_iand; + else if (op == nir_op_iadd) + op = nir_op_ixor; + else if (op == nir_op_umax || op == nir_op_imax) + op = nir_op_ior; + assert(op == nir_op_iand || op == nir_op_ior || op == nir_op_ixor); + + switch (instr->intrinsic) { + case nir_intrinsic_reduce: + emit_wqm(ctx, emit_boolean_reduce(ctx, op, cluster_size, src), dst); + break; + case nir_intrinsic_exclusive_scan: + emit_wqm(ctx, emit_boolean_exclusive_scan(ctx, op, src), dst); + break; + case nir_intrinsic_inclusive_scan: + emit_wqm(ctx, emit_boolean_inclusive_scan(ctx, op, src), dst); + break; + default: + assert(false); + } + } else if (cluster_size == 1) { + bld.copy(Definition(dst), src); + } else { + src = as_vgpr(ctx, src); + + ReduceOp reduce_op; + switch (op) { + #define CASE(name) case nir_op_##name: reduce_op = (src.regClass() == v1) ? name##32 : name##64; break; + CASE(iadd) + CASE(imul) + CASE(fadd) + CASE(fmul) + CASE(imin) + CASE(umin) + CASE(fmin) + CASE(imax) + CASE(umax) + CASE(fmax) + CASE(iand) + CASE(ior) + CASE(ixor) + default: + unreachable("unknown reduction op"); + #undef CASE + } + + aco_opcode aco_op; + switch (instr->intrinsic) { + case nir_intrinsic_reduce: aco_op = aco_opcode::p_reduce; break; + case nir_intrinsic_inclusive_scan: aco_op = aco_opcode::p_inclusive_scan; break; + case nir_intrinsic_exclusive_scan: aco_op = aco_opcode::p_exclusive_scan; break; + default: + unreachable("unknown reduce intrinsic"); + } + + aco_ptr reduce{create_instruction(aco_op, Format::PSEUDO_REDUCTION, 3, 5)}; + reduce->operands[0] = Operand(src); + // filled in by aco_reduce_assign.cpp, used internally as part of the + // reduce sequence + assert(dst.size() == 1 || dst.size() == 2); + reduce->operands[1] = Operand(RegClass(RegType::vgpr, dst.size()).as_linear()); + reduce->operands[2] = Operand(v1.as_linear()); + + Temp tmp_dst = bld.tmp(dst.regClass()); + reduce->definitions[0] = Definition(tmp_dst); + reduce->definitions[1] = bld.def(s2); // used internally + reduce->definitions[2] = Definition(); + reduce->definitions[3] = Definition(scc, s1); + reduce->definitions[4] = Definition(); + reduce->reduce_op = reduce_op; + reduce->cluster_size = cluster_size; + ctx->block->instructions.emplace_back(std::move(reduce)); + + emit_wqm(ctx, tmp_dst, dst); + } + break; + } + case nir_intrinsic_quad_broadcast: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + if (!ctx->divergent_vals[instr->dest.ssa.index]) { + emit_uniform_subgroup(ctx, instr, src); + } else { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + unsigned lane = nir_src_as_const_value(instr->src[1])->u32; + if (instr->dest.ssa.bit_size == 1 && src.regClass() == s2) { + uint32_t half_mask = 0x11111111u << lane; + Temp mask_tmp = bld.pseudo(aco_opcode::p_create_vector, bld.def(s2), Operand(half_mask), Operand(half_mask)); + Temp tmp = bld.tmp(s2); + bld.sop1(aco_opcode::s_wqm_b64, Definition(tmp), + bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), mask_tmp, + bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), src, Operand(exec, s2)))); + emit_wqm(ctx, tmp, dst); + } else if (instr->dest.ssa.bit_size == 32) { + emit_wqm(ctx, + bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), src, + dpp_quad_perm(lane, lane, lane, lane)), + dst); + } else if (instr->dest.ssa.bit_size == 64) { + Temp lo = bld.tmp(v1), hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lo), Definition(hi), src); + lo = emit_wqm(ctx, bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), lo, dpp_quad_perm(lane, lane, lane, lane))); + hi = emit_wqm(ctx, bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), hi, dpp_quad_perm(lane, lane, lane, lane))); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lo, hi); + emit_split_vector(ctx, dst, 2); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + } + break; + } + case nir_intrinsic_quad_swap_horizontal: + case nir_intrinsic_quad_swap_vertical: + case nir_intrinsic_quad_swap_diagonal: + case nir_intrinsic_quad_swizzle_amd: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + if (!ctx->divergent_vals[instr->dest.ssa.index]) { + emit_uniform_subgroup(ctx, instr, src); + break; + } + uint16_t dpp_ctrl = 0; + switch (instr->intrinsic) { + case nir_intrinsic_quad_swap_horizontal: + dpp_ctrl = dpp_quad_perm(1, 0, 3, 2); + break; + case nir_intrinsic_quad_swap_vertical: + dpp_ctrl = dpp_quad_perm(2, 3, 0, 1); + break; + case nir_intrinsic_quad_swap_diagonal: + dpp_ctrl = dpp_quad_perm(3, 2, 1, 0); + break; + case nir_intrinsic_quad_swizzle_amd: { + dpp_ctrl = nir_intrinsic_swizzle_mask(instr); + break; + } + default: + break; + } + + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + if (instr->dest.ssa.bit_size == 1 && src.regClass() == s2) { + src = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), Operand(0u), Operand((uint32_t)-1), src); + src = bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), src, dpp_ctrl); + Temp tmp = bld.vopc(aco_opcode::v_cmp_lg_u32, bld.def(s2), Operand(0u), src); + emit_wqm(ctx, tmp, dst); + } else if (instr->dest.ssa.bit_size == 32) { + Temp tmp = bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), src, dpp_ctrl); + emit_wqm(ctx, tmp, dst); + } else if (instr->dest.ssa.bit_size == 64) { + Temp lo = bld.tmp(v1), hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lo), Definition(hi), src); + lo = emit_wqm(ctx, bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), lo, dpp_ctrl)); + hi = emit_wqm(ctx, bld.vop1_dpp(aco_opcode::v_mov_b32, bld.def(v1), hi, dpp_ctrl)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lo, hi); + emit_split_vector(ctx, dst, 2); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_intrinsic_masked_swizzle_amd: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + if (!ctx->divergent_vals[instr->dest.ssa.index]) { + emit_uniform_subgroup(ctx, instr, src); + break; + } + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + uint32_t mask = nir_intrinsic_swizzle_mask(instr); + if (dst.regClass() == v1) { + emit_wqm(ctx, + bld.ds(aco_opcode::ds_swizzle_b32, bld.def(v1), src, mask, 0, false), + dst); + } else if (dst.regClass() == v2) { + Temp lo = bld.tmp(v1), hi = bld.tmp(v1); + bld.pseudo(aco_opcode::p_split_vector, Definition(lo), Definition(hi), src); + lo = emit_wqm(ctx, bld.ds(aco_opcode::ds_swizzle_b32, bld.def(v1), lo, mask, 0, false)); + hi = emit_wqm(ctx, bld.ds(aco_opcode::ds_swizzle_b32, bld.def(v1), hi, mask, 0, false)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lo, hi); + emit_split_vector(ctx, dst, 2); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_intrinsic_write_invocation_amd: { + Temp src = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[0].ssa)); + Temp val = bld.as_uniform(get_ssa_temp(ctx, instr->src[1].ssa)); + Temp lane = bld.as_uniform(get_ssa_temp(ctx, instr->src[2].ssa)); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + if (dst.regClass() == v1) { + /* src2 is ignored for writelane. RA assigns the same reg for dst */ + emit_wqm(ctx, bld.vop3(aco_opcode::v_writelane_b32, bld.def(v1), val, lane, src), dst); + } else if (dst.regClass() == v2) { + Temp src_lo = bld.tmp(v1), src_hi = bld.tmp(v1); + Temp val_lo = bld.tmp(s1), val_hi = bld.tmp(s1); + bld.pseudo(aco_opcode::p_split_vector, Definition(src_lo), Definition(src_hi), src); + bld.pseudo(aco_opcode::p_split_vector, Definition(val_lo), Definition(val_hi), val); + Temp lo = emit_wqm(ctx, bld.vop3(aco_opcode::v_writelane_b32, bld.def(v1), val_lo, lane, src_hi)); + Temp hi = emit_wqm(ctx, bld.vop3(aco_opcode::v_writelane_b32, bld.def(v1), val_hi, lane, src_hi)); + bld.pseudo(aco_opcode::p_create_vector, Definition(dst), lo, hi); + emit_split_vector(ctx, dst, 2); + } else { + fprintf(stderr, "Unimplemented NIR instr bit size: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + } + break; + } + case nir_intrinsic_mbcnt_amd: { + Temp src = get_ssa_temp(ctx, instr->src[0].ssa); + RegClass rc = RegClass(src.type(), 1); + Temp mask_lo = bld.tmp(rc), mask_hi = bld.tmp(rc); + bld.pseudo(aco_opcode::p_split_vector, Definition(mask_lo), Definition(mask_hi), src); + Temp tmp = bld.vop3(aco_opcode::v_mbcnt_lo_u32_b32, bld.def(v1), mask_lo, Operand(0u)); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + Temp wqm_tmp = bld.vop3(aco_opcode::v_mbcnt_hi_u32_b32, bld.def(v1), mask_hi, tmp); + emit_wqm(ctx, wqm_tmp, dst); + break; + } + case nir_intrinsic_load_helper_invocation: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.pseudo(aco_opcode::p_load_helper, Definition(dst)); + ctx->block->kind |= block_kind_needs_lowering; + ctx->program->needs_exact = true; + break; + } + case nir_intrinsic_is_helper_invocation: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.pseudo(aco_opcode::p_is_helper, Definition(dst)); + ctx->block->kind |= block_kind_needs_lowering; + ctx->program->needs_exact = true; + break; + } + case nir_intrinsic_demote: + bld.pseudo(aco_opcode::p_demote_to_helper); + ctx->block->kind |= block_kind_needs_lowering; + ctx->program->needs_exact = true; + break; + case nir_intrinsic_demote_if: { + Temp cond = bld.sop2(aco_opcode::s_and_b64, bld.def(s2), bld.def(s1, scc), + as_divergent_bool(ctx, get_ssa_temp(ctx, instr->src[0].ssa), false), + Operand(exec, s2)); + bld.pseudo(aco_opcode::p_demote_to_helper, cond); + ctx->block->kind |= block_kind_needs_lowering; + ctx->program->needs_exact = true; + break; + } + case nir_intrinsic_first_invocation: { + emit_wqm(ctx, bld.sop1(aco_opcode::s_ff1_i32_b64, bld.def(s1), Operand(exec, s2)), + get_ssa_temp(ctx, &instr->dest.ssa)); + break; + } + case nir_intrinsic_shader_clock: + bld.smem(aco_opcode::s_memtime, Definition(get_ssa_temp(ctx, &instr->dest.ssa))); + break; + case nir_intrinsic_load_vertex_id_zero_base: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.copy(Definition(dst), ctx->vertex_id); + break; + } + case nir_intrinsic_load_first_vertex: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.copy(Definition(dst), ctx->base_vertex); + break; + } + case nir_intrinsic_load_base_instance: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.copy(Definition(dst), ctx->start_instance); + break; + } + case nir_intrinsic_load_instance_id: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.copy(Definition(dst), ctx->instance_id); + break; + } + case nir_intrinsic_load_draw_id: { + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + bld.copy(Definition(dst), ctx->draw_id); + break; + } + default: + fprintf(stderr, "Unimplemented intrinsic instr: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + abort(); + + break; + } +} + + +void tex_fetch_ptrs(isel_context *ctx, nir_tex_instr *instr, + Temp *res_ptr, Temp *samp_ptr, Temp *fmask_ptr, + enum glsl_base_type *stype) +{ + nir_deref_instr *texture_deref_instr = NULL; + nir_deref_instr *sampler_deref_instr = NULL; + int plane = -1; + + for (unsigned i = 0; i < instr->num_srcs; i++) { + switch (instr->src[i].src_type) { + case nir_tex_src_texture_deref: + texture_deref_instr = nir_src_as_deref(instr->src[i].src); + break; + case nir_tex_src_sampler_deref: + sampler_deref_instr = nir_src_as_deref(instr->src[i].src); + break; + case nir_tex_src_plane: + plane = nir_src_as_int(instr->src[i].src); + break; + default: + break; + } + } + + *stype = glsl_get_sampler_result_type(texture_deref_instr->type); + + if (!sampler_deref_instr) + sampler_deref_instr = texture_deref_instr; + + if (plane >= 0) { + assert(instr->op != nir_texop_txf_ms && + instr->op != nir_texop_samples_identical); + assert(instr->sampler_dim != GLSL_SAMPLER_DIM_BUF); + *res_ptr = get_sampler_desc(ctx, texture_deref_instr, (aco_descriptor_type)(ACO_DESC_PLANE_0 + plane), instr, false, false); + } else if (instr->sampler_dim == GLSL_SAMPLER_DIM_BUF) { + *res_ptr = get_sampler_desc(ctx, texture_deref_instr, ACO_DESC_BUFFER, instr, false, false); + } else { + *res_ptr = get_sampler_desc(ctx, texture_deref_instr, ACO_DESC_IMAGE, instr, false, false); + } + if (samp_ptr) { + *samp_ptr = get_sampler_desc(ctx, sampler_deref_instr, ACO_DESC_SAMPLER, instr, false, false); + if (instr->sampler_dim < GLSL_SAMPLER_DIM_RECT && ctx->options->chip_class < GFX8) { + fprintf(stderr, "Unimplemented sampler descriptor: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + abort(); + // TODO: build samp_ptr = and(samp_ptr, res_ptr) + } + } + if (fmask_ptr && (instr->op == nir_texop_txf_ms || + instr->op == nir_texop_samples_identical)) + *fmask_ptr = get_sampler_desc(ctx, texture_deref_instr, ACO_DESC_FMASK, instr, false, false); +} + +void build_cube_select(isel_context *ctx, Temp ma, Temp id, Temp deriv, + Temp *out_ma, Temp *out_sc, Temp *out_tc) +{ + Builder bld(ctx->program, ctx->block); + + Temp deriv_x = emit_extract_vector(ctx, deriv, 0, v1); + Temp deriv_y = emit_extract_vector(ctx, deriv, 1, v1); + Temp deriv_z = emit_extract_vector(ctx, deriv, 2, v1); + + Operand neg_one(0xbf800000u); + Operand one(0x3f800000u); + Operand two(0x40000000u); + Operand four(0x40800000u); + + Temp is_ma_positive = bld.vopc(aco_opcode::v_cmp_le_f32, bld.hint_vcc(bld.def(s2)), Operand(0u), ma); + Temp sgn_ma = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), neg_one, one, is_ma_positive); + Temp neg_sgn_ma = bld.vop2(aco_opcode::v_sub_f32, bld.def(v1), Operand(0u), sgn_ma); + + Temp is_ma_z = bld.vopc(aco_opcode::v_cmp_le_f32, bld.hint_vcc(bld.def(s2)), four, id); + Temp is_ma_y = bld.vopc(aco_opcode::v_cmp_le_f32, bld.def(s2), two, id); + is_ma_y = bld.sop2(aco_opcode::s_andn2_b64, bld.hint_vcc(bld.def(s2)), is_ma_y, is_ma_z); + Temp is_not_ma_x = bld.sop2(aco_opcode::s_or_b64, bld.hint_vcc(bld.def(s2)), bld.def(s1, scc), is_ma_z, is_ma_y); + + // select sc + Temp tmp = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), deriv_z, deriv_x, is_not_ma_x); + Temp sgn = bld.vop2_e64(aco_opcode::v_cndmask_b32, bld.def(v1), + bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), neg_sgn_ma, sgn_ma, is_ma_z), + one, is_ma_y); + *out_sc = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), tmp, sgn); + + // select tc + tmp = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), deriv_y, deriv_z, is_ma_y); + sgn = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), neg_one, sgn_ma, is_ma_y); + *out_tc = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), tmp, sgn); + + // select ma + tmp = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), + bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), deriv_x, deriv_y, is_ma_y), + deriv_z, is_ma_z); + tmp = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(0x7fffffffu), tmp); + *out_ma = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), two, tmp); +} + +void prepare_cube_coords(isel_context *ctx, Temp* coords, Temp* ddx, Temp* ddy, bool is_deriv, bool is_array) +{ + Builder bld(ctx->program, ctx->block); + Temp coord_args[4], ma, tc, sc, id; + for (unsigned i = 0; i < (is_array ? 4 : 3); i++) + coord_args[i] = emit_extract_vector(ctx, *coords, i, v1); + + if (is_array) { + coord_args[3] = bld.vop1(aco_opcode::v_rndne_f32, bld.def(v1), coord_args[3]); + + // see comment in ac_prepare_cube_coords() + if (ctx->options->chip_class <= GFX8) + coord_args[3] = bld.vop2(aco_opcode::v_max_f32, bld.def(v1), Operand(0u), coord_args[3]); + } + + ma = bld.vop3(aco_opcode::v_cubema_f32, bld.def(v1), coord_args[0], coord_args[1], coord_args[2]); + + aco_ptr vop3a{create_instruction(aco_opcode::v_rcp_f32, asVOP3(Format::VOP1), 1, 1)}; + vop3a->operands[0] = Operand(ma); + vop3a->abs[0] = true; + Temp invma = bld.tmp(v1); + vop3a->definitions[0] = Definition(invma); + ctx->block->instructions.emplace_back(std::move(vop3a)); + + sc = bld.vop3(aco_opcode::v_cubesc_f32, bld.def(v1), coord_args[0], coord_args[1], coord_args[2]); + if (!is_deriv) + sc = bld.vop2(aco_opcode::v_madak_f32, bld.def(v1), sc, invma, Operand(0x3fc00000u/*1.5*/)); + + tc = bld.vop3(aco_opcode::v_cubetc_f32, bld.def(v1), coord_args[0], coord_args[1], coord_args[2]); + if (!is_deriv) + tc = bld.vop2(aco_opcode::v_madak_f32, bld.def(v1), tc, invma, Operand(0x3fc00000u/*1.5*/)); + + id = bld.vop3(aco_opcode::v_cubeid_f32, bld.def(v1), coord_args[0], coord_args[1], coord_args[2]); + + if (is_deriv) { + sc = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), sc, invma); + tc = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), tc, invma); + + for (unsigned i = 0; i < 2; i++) { + // see comment in ac_prepare_cube_coords() + Temp deriv_ma; + Temp deriv_sc, deriv_tc; + build_cube_select(ctx, ma, id, i ? *ddy : *ddx, + &deriv_ma, &deriv_sc, &deriv_tc); + + deriv_ma = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), deriv_ma, invma); + + Temp x = bld.vop2(aco_opcode::v_sub_f32, bld.def(v1), + bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), deriv_sc, invma), + bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), deriv_ma, sc)); + Temp y = bld.vop2(aco_opcode::v_sub_f32, bld.def(v1), + bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), deriv_tc, invma), + bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), deriv_ma, tc)); + *(i ? ddy : ddx) = bld.pseudo(aco_opcode::p_create_vector, bld.def(v2), x, y); + } + + sc = bld.vop2(aco_opcode::v_add_f32, bld.def(v1), Operand(0x3fc00000u/*1.5*/), sc); + tc = bld.vop2(aco_opcode::v_add_f32, bld.def(v1), Operand(0x3fc00000u/*1.5*/), tc); + } + + if (is_array) + id = bld.vop2(aco_opcode::v_madmk_f32, bld.def(v1), coord_args[3], id, Operand(0x41000000u/*8.0*/)); + *coords = bld.pseudo(aco_opcode::p_create_vector, bld.def(v3), sc, tc, id); + +} + +Temp apply_round_slice(isel_context *ctx, Temp coords, unsigned idx) +{ + Temp coord_vec[3]; + for (unsigned i = 0; i < coords.size(); i++) + coord_vec[i] = emit_extract_vector(ctx, coords, i, v1); + + Builder bld(ctx->program, ctx->block); + coord_vec[idx] = bld.vop1(aco_opcode::v_rndne_f32, bld.def(v1), coord_vec[idx]); + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, coords.size(), 1)}; + for (unsigned i = 0; i < coords.size(); i++) + vec->operands[i] = Operand(coord_vec[i]); + Temp res = bld.tmp(RegType::vgpr, coords.size()); + vec->definitions[0] = Definition(res); + ctx->block->instructions.emplace_back(std::move(vec)); + return res; +} + +void get_const_vec(nir_ssa_def *vec, nir_const_value *cv[4]) +{ + if (vec->parent_instr->type != nir_instr_type_alu) + return; + nir_alu_instr *vec_instr = nir_instr_as_alu(vec->parent_instr); + if (vec_instr->op != nir_op_vec(vec->num_components)) + return; + + for (unsigned i = 0; i < vec->num_components; i++) { + cv[i] = vec_instr->src[i].swizzle[0] == 0 ? + nir_src_as_const_value(vec_instr->src[i].src) : NULL; + } +} + +void visit_tex(isel_context *ctx, nir_tex_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + bool has_bias = false, has_lod = false, level_zero = false, has_compare = false, + has_offset = false, has_ddx = false, has_ddy = false, has_derivs = false, has_sample_index = false; + Temp resource, sampler, fmask_ptr, bias = Temp(), coords, compare = Temp(), sample_index = Temp(), + lod = Temp(), offset = Temp(), ddx = Temp(), ddy = Temp(), derivs = Temp(); + nir_const_value *sample_index_cv = NULL; + nir_const_value *const_offset[4] = {NULL, NULL, NULL, NULL}; + enum glsl_base_type stype; + tex_fetch_ptrs(ctx, instr, &resource, &sampler, &fmask_ptr, &stype); + + bool tg4_integer_workarounds = ctx->options->chip_class <= GFX8 && instr->op == nir_texop_tg4 && + (stype == GLSL_TYPE_UINT || stype == GLSL_TYPE_INT); + bool tg4_integer_cube_workaround = tg4_integer_workarounds && + instr->sampler_dim == GLSL_SAMPLER_DIM_CUBE; + + for (unsigned i = 0; i < instr->num_srcs; i++) { + switch (instr->src[i].src_type) { + case nir_tex_src_coord: + coords = as_vgpr(ctx, get_ssa_temp(ctx, instr->src[i].src.ssa)); + break; + case nir_tex_src_bias: + if (instr->op == nir_texop_txb) { + bias = get_ssa_temp(ctx, instr->src[i].src.ssa); + has_bias = true; + } + break; + case nir_tex_src_lod: { + nir_const_value *val = nir_src_as_const_value(instr->src[i].src); + + if (val && val->f32 <= 0.0) { + level_zero = true; + } else { + lod = get_ssa_temp(ctx, instr->src[i].src.ssa); + has_lod = true; + } + break; + } + case nir_tex_src_comparator: + if (instr->is_shadow) { + compare = get_ssa_temp(ctx, instr->src[i].src.ssa); + has_compare = true; + } + break; + case nir_tex_src_offset: + offset = get_ssa_temp(ctx, instr->src[i].src.ssa); + get_const_vec(instr->src[i].src.ssa, const_offset); + has_offset = true; + break; + case nir_tex_src_ddx: + ddx = get_ssa_temp(ctx, instr->src[i].src.ssa); + has_ddx = true; + break; + case nir_tex_src_ddy: + ddy = get_ssa_temp(ctx, instr->src[i].src.ssa); + has_ddy = true; + break; + case nir_tex_src_ms_index: + sample_index = get_ssa_temp(ctx, instr->src[i].src.ssa); + sample_index_cv = nir_src_as_const_value(instr->src[i].src); + has_sample_index = true; + break; + case nir_tex_src_texture_offset: + case nir_tex_src_sampler_offset: + default: + break; + } + } +// TODO: all other cases: structure taken from ac_nir_to_llvm.c + if (instr->op == nir_texop_txs && instr->sampler_dim == GLSL_SAMPLER_DIM_BUF) + return get_buffer_size(ctx, resource, get_ssa_temp(ctx, &instr->dest.ssa), true); + + if (instr->op == nir_texop_texture_samples) { + Temp dword3 = emit_extract_vector(ctx, resource, 3, s1); + + Temp samples_log2 = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), dword3, Operand(16u | 4u<<16)); + Temp samples = bld.sop2(aco_opcode::s_lshl_b32, bld.def(s1), bld.def(s1, scc), Operand(1u), samples_log2); + Temp type = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), dword3, Operand(28u | 4u<<16 /* offset=28, width=4 */)); + Temp is_msaa = bld.sopc(aco_opcode::s_cmp_ge_u32, bld.def(s1, scc), type, Operand(14u)); + + bld.sop2(aco_opcode::s_cselect_b32, Definition(get_ssa_temp(ctx, &instr->dest.ssa)), + samples, Operand(1u), bld.scc(is_msaa)); + return; + } + + if (has_offset && instr->op != nir_texop_txf && instr->op != nir_texop_txf_ms) { + aco_ptr tmp_instr; + Temp acc, pack = Temp(); + + uint32_t pack_const = 0; + for (unsigned i = 0; i < offset.size(); i++) { + if (!const_offset[i]) + continue; + pack_const |= (const_offset[i]->u32 & 0x3Fu) << (8u * i); + } + + if (offset.type() == RegType::sgpr) { + for (unsigned i = 0; i < offset.size(); i++) { + if (const_offset[i]) + continue; + + acc = emit_extract_vector(ctx, offset, i, s1); + acc = bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), acc, Operand(0x3Fu)); + + if (i) { + acc = bld.sop2(aco_opcode::s_lshl_b32, bld.def(s1), bld.def(s1, scc), acc, Operand(8u * i)); + } + + if (pack == Temp()) { + pack = acc; + } else { + pack = bld.sop2(aco_opcode::s_or_b32, bld.def(s1), bld.def(s1, scc), pack, acc); + } + } + + if (pack_const && pack != Temp()) + pack = bld.sop2(aco_opcode::s_or_b32, bld.def(s1), bld.def(s1, scc), Operand(pack_const), pack); + } else { + for (unsigned i = 0; i < offset.size(); i++) { + if (const_offset[i]) + continue; + + acc = emit_extract_vector(ctx, offset, i, v1); + acc = bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(0x3Fu), acc); + + if (i) { + acc = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(8u * i), acc); + } + + if (pack == Temp()) { + pack = acc; + } else { + pack = bld.vop2(aco_opcode::v_or_b32, bld.def(v1), pack, acc); + } + } + + if (pack_const && pack != Temp()) + pack = bld.sop2(aco_opcode::v_or_b32, bld.def(v1), Operand(pack_const), pack); + } + if (pack_const && pack == Temp()) + offset = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand(pack_const)); + else if (pack == Temp()) + has_offset = false; + else + offset = pack; + } + + if (instr->sampler_dim == GLSL_SAMPLER_DIM_CUBE && instr->coord_components) + prepare_cube_coords(ctx, &coords, &ddx, &ddy, instr->op == nir_texop_txd, instr->is_array && instr->op != nir_texop_lod); + + /* pack derivatives */ + if (has_ddx || has_ddy) { + if (instr->sampler_dim == GLSL_SAMPLER_DIM_1D && ctx->options->chip_class >= GFX9) { + derivs = bld.pseudo(aco_opcode::p_create_vector, bld.def(v4), + ddx, Operand(0u), ddy, Operand(0u)); + } else { + derivs = bld.pseudo(aco_opcode::p_create_vector, bld.def(RegType::vgpr, ddx.size() + ddy.size()), ddx, ddy); + } + has_derivs = true; + } + + if (instr->coord_components > 1 && + instr->sampler_dim == GLSL_SAMPLER_DIM_1D && + instr->is_array && + instr->op != nir_texop_txf) + coords = apply_round_slice(ctx, coords, 1); + + if (instr->coord_components > 2 && + (instr->sampler_dim == GLSL_SAMPLER_DIM_2D || + instr->sampler_dim == GLSL_SAMPLER_DIM_MS || + instr->sampler_dim == GLSL_SAMPLER_DIM_SUBPASS || + instr->sampler_dim == GLSL_SAMPLER_DIM_SUBPASS_MS) && + instr->is_array && + instr->op != nir_texop_txf && instr->op != nir_texop_txf_ms) + coords = apply_round_slice(ctx, coords, 2); + + if (ctx->options->chip_class >= GFX9 && + instr->sampler_dim == GLSL_SAMPLER_DIM_1D && + instr->op != nir_texop_lod && instr->coord_components) { + assert(coords.size() > 0 && coords.size() < 3); + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, coords.size() + 1, 1)}; + vec->operands[0] = Operand(emit_extract_vector(ctx, coords, 0, v1)); + vec->operands[1] = instr->op == nir_texop_txf ? Operand((uint32_t) 0) : Operand((uint32_t) 0x3f000000); + if (coords.size() > 1) + vec->operands[2] = Operand(emit_extract_vector(ctx, coords, 1, v1)); + coords = bld.tmp(RegType::vgpr, coords.size() + 1); + vec->definitions[0] = Definition(coords); + ctx->block->instructions.emplace_back(std::move(vec)); + } + + bool da = should_declare_array(ctx, instr->sampler_dim, instr->is_array); + + if (instr->op == nir_texop_samples_identical) + resource = fmask_ptr; + + else if ((instr->sampler_dim == GLSL_SAMPLER_DIM_MS || + instr->sampler_dim == GLSL_SAMPLER_DIM_SUBPASS_MS) && + instr->op != nir_texop_txs) { + assert(has_sample_index); + Operand op(sample_index); + if (sample_index_cv) + op = Operand(sample_index_cv->u32); + sample_index = adjust_sample_index_using_fmask(ctx, da, coords, op, fmask_ptr); + } + + if (has_offset && (instr->op == nir_texop_txf || instr->op == nir_texop_txf_ms)) { + Temp split_coords[coords.size()]; + emit_split_vector(ctx, coords, coords.size()); + for (unsigned i = 0; i < coords.size(); i++) + split_coords[i] = emit_extract_vector(ctx, coords, i, v1); + + unsigned i = 0; + for (; i < std::min(offset.size(), instr->coord_components); i++) { + Temp off = emit_extract_vector(ctx, offset, i, v1); + split_coords[i] = bld.vadd32(bld.def(v1), split_coords[i], off); + } + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, coords.size(), 1)}; + for (unsigned i = 0; i < coords.size(); i++) + vec->operands[i] = Operand(split_coords[i]); + coords = bld.tmp(coords.regClass()); + vec->definitions[0] = Definition(coords); + ctx->block->instructions.emplace_back(std::move(vec)); + + has_offset = false; + } + + /* Build tex instruction */ + unsigned dmask = nir_ssa_def_components_read(&instr->dest.ssa); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + Temp tmp_dst = dst; + + /* gather4 selects the component by dmask and always returns vec4 */ + if (instr->op == nir_texop_tg4) { + assert(instr->dest.ssa.num_components == 4); + if (instr->is_shadow) + dmask = 1; + else + dmask = 1 << instr->component; + if (tg4_integer_cube_workaround || dst.type() == RegType::sgpr) + tmp_dst = bld.tmp(v4); + } else if (instr->op == nir_texop_samples_identical) { + tmp_dst = bld.tmp(v1); + } else if (util_bitcount(dmask) != instr->dest.ssa.num_components || dst.type() == RegType::sgpr) { + tmp_dst = bld.tmp(RegClass(RegType::vgpr, util_bitcount(dmask))); + } + + aco_ptr tex; + if (instr->op == nir_texop_txs || instr->op == nir_texop_query_levels) { + if (!has_lod) + lod = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand(0u)); + + bool div_by_6 = instr->op == nir_texop_txs && + instr->sampler_dim == GLSL_SAMPLER_DIM_CUBE && + instr->is_array && + (dmask & (1 << 2)); + if (tmp_dst.id() == dst.id() && div_by_6) + tmp_dst = bld.tmp(tmp_dst.regClass()); + + tex.reset(create_instruction(aco_opcode::image_get_resinfo, Format::MIMG, 2, 1)); + tex->operands[0] = Operand(as_vgpr(ctx,lod)); + tex->operands[1] = Operand(resource); + if (ctx->options->chip_class >= GFX9 && + instr->op == nir_texop_txs && + instr->sampler_dim == GLSL_SAMPLER_DIM_1D && + instr->is_array) { + tex->dmask = (dmask & 0x1) | ((dmask & 0x2) << 1); + } else if (instr->op == nir_texop_query_levels) { + tex->dmask = 1 << 3; + } else { + tex->dmask = dmask; + } + tex->da = da; + tex->definitions[0] = Definition(tmp_dst); + tex->can_reorder = true; + ctx->block->instructions.emplace_back(std::move(tex)); + + if (div_by_6) { + /* divide 3rd value by 6 by multiplying with magic number */ + emit_split_vector(ctx, tmp_dst, tmp_dst.size()); + Temp c = bld.copy(bld.def(s1), Operand((uint32_t) 0x2AAAAAAB)); + Temp by_6 = bld.vop3(aco_opcode::v_mul_hi_i32, bld.def(v1), emit_extract_vector(ctx, tmp_dst, 2, v1), c); + assert(instr->dest.ssa.num_components == 3); + Temp tmp = dst.type() == RegType::vgpr ? dst : bld.tmp(v3); + tmp_dst = bld.pseudo(aco_opcode::p_create_vector, Definition(tmp), + emit_extract_vector(ctx, tmp_dst, 0, v1), + emit_extract_vector(ctx, tmp_dst, 1, v1), + by_6); + + } + + expand_vector(ctx, tmp_dst, dst, instr->dest.ssa.num_components, dmask); + return; + } + + Temp tg4_compare_cube_wa64 = Temp(); + + if (tg4_integer_workarounds) { + tex.reset(create_instruction(aco_opcode::image_get_resinfo, Format::MIMG, 2, 1)); + tex->operands[0] = bld.vop1(aco_opcode::v_mov_b32, bld.def(v1), Operand(0u)); + tex->operands[1] = Operand(resource); + tex->dmask = 0x3; + tex->da = da; + Temp size = bld.tmp(v2); + tex->definitions[0] = Definition(size); + tex->can_reorder = true; + ctx->block->instructions.emplace_back(std::move(tex)); + emit_split_vector(ctx, size, size.size()); + + Temp half_texel[2]; + for (unsigned i = 0; i < 2; i++) { + half_texel[i] = emit_extract_vector(ctx, size, i, v1); + half_texel[i] = bld.vop1(aco_opcode::v_cvt_f32_i32, bld.def(v1), half_texel[i]); + half_texel[i] = bld.vop1(aco_opcode::v_rcp_iflag_f32, bld.def(v1), half_texel[i]); + half_texel[i] = bld.vop2(aco_opcode::v_mul_f32, bld.def(v1), Operand(0xbf000000/*-0.5*/), half_texel[i]); + } + + Temp orig_coords[2] = { + emit_extract_vector(ctx, coords, 0, v1), + emit_extract_vector(ctx, coords, 1, v1)}; + Temp new_coords[2] = { + bld.vop2(aco_opcode::v_add_f32, bld.def(v1), orig_coords[0], half_texel[0]), + bld.vop2(aco_opcode::v_add_f32, bld.def(v1), orig_coords[1], half_texel[1]) + }; + + if (tg4_integer_cube_workaround) { + // see comment in ac_nir_to_llvm.c's lower_gather4_integer() + Temp desc[resource.size()]; + aco_ptr split{create_instruction(aco_opcode::p_split_vector, + Format::PSEUDO, 1, resource.size())}; + split->operands[0] = Operand(resource); + for (unsigned i = 0; i < resource.size(); i++) { + desc[i] = bld.tmp(s1); + split->definitions[i] = Definition(desc[i]); + } + ctx->block->instructions.emplace_back(std::move(split)); + + Temp dfmt = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), desc[1], Operand(20u | (6u << 16))); + Temp compare_cube_wa = bld.sopc(aco_opcode::s_cmp_eq_u32, bld.def(s1, scc), dfmt, + Operand((uint32_t)V_008F14_IMG_DATA_FORMAT_8_8_8_8)); + + Temp nfmt; + if (stype == GLSL_TYPE_UINT) { + nfmt = bld.sop2(aco_opcode::s_cselect_b32, bld.def(s1), + Operand((uint32_t)V_008F14_IMG_NUM_FORMAT_USCALED), + Operand((uint32_t)V_008F14_IMG_NUM_FORMAT_UINT), + bld.scc(compare_cube_wa)); + } else { + nfmt = bld.sop2(aco_opcode::s_cselect_b32, bld.def(s1), + Operand((uint32_t)V_008F14_IMG_NUM_FORMAT_SSCALED), + Operand((uint32_t)V_008F14_IMG_NUM_FORMAT_SINT), + bld.scc(compare_cube_wa)); + } + tg4_compare_cube_wa64 = as_divergent_bool(ctx, compare_cube_wa, true); + nfmt = bld.sop2(aco_opcode::s_lshl_b32, bld.def(s1), bld.def(s1, scc), nfmt, Operand(26u)); + + desc[1] = bld.sop2(aco_opcode::s_and_b32, bld.def(s1), bld.def(s1, scc), desc[1], + Operand((uint32_t)C_008F14_NUM_FORMAT)); + desc[1] = bld.sop2(aco_opcode::s_or_b32, bld.def(s1), bld.def(s1, scc), desc[1], nfmt); + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, + Format::PSEUDO, resource.size(), 1)}; + for (unsigned i = 0; i < resource.size(); i++) + vec->operands[i] = Operand(desc[i]); + resource = bld.tmp(resource.regClass()); + vec->definitions[0] = Definition(resource); + ctx->block->instructions.emplace_back(std::move(vec)); + + new_coords[0] = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), + new_coords[0], orig_coords[0], tg4_compare_cube_wa64); + new_coords[1] = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), + new_coords[1], orig_coords[1], tg4_compare_cube_wa64); + } + + if (coords.size() == 3) { + coords = bld.pseudo(aco_opcode::p_create_vector, bld.def(v3), + new_coords[0], new_coords[1], + emit_extract_vector(ctx, coords, 2, v1)); + } else { + assert(coords.size() == 2); + coords = bld.pseudo(aco_opcode::p_create_vector, bld.def(v2), + new_coords[0], new_coords[1]); + } + } + + if (!(has_ddx && has_ddy) && !has_lod && !level_zero && + instr->sampler_dim != GLSL_SAMPLER_DIM_MS && + instr->sampler_dim != GLSL_SAMPLER_DIM_SUBPASS_MS) + coords = emit_wqm(ctx, coords, bld.tmp(coords.regClass()), true); + + std::vector args; + if (has_offset) + args.emplace_back(Operand(offset)); + if (has_bias) + args.emplace_back(Operand(bias)); + if (has_compare) + args.emplace_back(Operand(compare)); + if (has_derivs) + args.emplace_back(Operand(derivs)); + args.emplace_back(Operand(coords)); + if (has_sample_index) + args.emplace_back(Operand(sample_index)); + if (has_lod) + args.emplace_back(lod); + + Operand arg; + if (args.size() > 1) { + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, args.size(), 1)}; + unsigned size = 0; + for (unsigned i = 0; i < args.size(); i++) { + size += args[i].size(); + vec->operands[i] = args[i]; + } + RegClass rc = RegClass(RegType::vgpr, size); + Temp tmp = bld.tmp(rc); + vec->definitions[0] = Definition(tmp); + ctx->block->instructions.emplace_back(std::move(vec)); + arg = Operand(tmp); + } else { + assert(args[0].isTemp()); + arg = Operand(as_vgpr(ctx, args[0].getTemp())); + } + + if (instr->sampler_dim == GLSL_SAMPLER_DIM_BUF) { + //FIXME: if (ctx->abi->gfx9_stride_size_workaround) return ac_build_buffer_load_format_gfx9_safe() + + assert(coords.size() == 1); + unsigned last_bit = util_last_bit(nir_ssa_def_components_read(&instr->dest.ssa)); + aco_opcode op; + switch (last_bit) { + case 1: + op = aco_opcode::buffer_load_format_x; break; + case 2: + op = aco_opcode::buffer_load_format_xy; break; + case 3: + op = aco_opcode::buffer_load_format_xyz; break; + case 4: + op = aco_opcode::buffer_load_format_xyzw; break; + default: + unreachable("Tex instruction loads more than 4 components."); + } + + /* if the instruction return value matches exactly the nir dest ssa, we can use it directly */ + if (last_bit == instr->dest.ssa.num_components && dst.type() == RegType::vgpr) + tmp_dst = dst; + else + tmp_dst = bld.tmp(RegType::vgpr, last_bit); + + aco_ptr mubuf{create_instruction(op, Format::MUBUF, 3, 1)}; + mubuf->operands[0] = Operand(coords); + mubuf->operands[1] = Operand(resource); + mubuf->operands[2] = Operand((uint32_t) 0); + mubuf->definitions[0] = Definition(tmp_dst); + mubuf->idxen = true; + mubuf->can_reorder = true; + ctx->block->instructions.emplace_back(std::move(mubuf)); + + expand_vector(ctx, tmp_dst, dst, instr->dest.ssa.num_components, (1 << last_bit) - 1); + return; + } + + + if (instr->op == nir_texop_txf || + instr->op == nir_texop_txf_ms || + instr->op == nir_texop_samples_identical) { + aco_opcode op = level_zero || instr->sampler_dim == GLSL_SAMPLER_DIM_MS ? aco_opcode::image_load : aco_opcode::image_load_mip; + tex.reset(create_instruction(op, Format::MIMG, 2, 1)); + tex->operands[0] = Operand(arg); + tex->operands[1] = Operand(resource); + tex->dmask = dmask; + tex->unrm = true; + tex->da = da; + tex->definitions[0] = Definition(tmp_dst); + tex->can_reorder = true; + ctx->block->instructions.emplace_back(std::move(tex)); + + if (instr->op == nir_texop_samples_identical) { + assert(dmask == 1 && dst.regClass() == v1); + assert(dst.id() != tmp_dst.id()); + + Temp tmp = bld.tmp(s2); + bld.vopc(aco_opcode::v_cmp_eq_u32, Definition(tmp), Operand(0u), tmp_dst).def(0).setHint(vcc); + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(dst), Operand(0u), Operand((uint32_t)-1), tmp); + + } else { + expand_vector(ctx, tmp_dst, dst, instr->dest.ssa.num_components, dmask); + } + return; + } + + // TODO: would be better to do this by adding offsets, but needs the opcodes ordered. + aco_opcode opcode = aco_opcode::image_sample; + if (has_offset) { /* image_sample_*_o */ + if (has_compare) { + opcode = aco_opcode::image_sample_c_o; + if (has_derivs) + opcode = aco_opcode::image_sample_c_d_o; + if (has_bias) + opcode = aco_opcode::image_sample_c_b_o; + if (level_zero) + opcode = aco_opcode::image_sample_c_lz_o; + if (has_lod) + opcode = aco_opcode::image_sample_c_l_o; + } else { + opcode = aco_opcode::image_sample_o; + if (has_derivs) + opcode = aco_opcode::image_sample_d_o; + if (has_bias) + opcode = aco_opcode::image_sample_b_o; + if (level_zero) + opcode = aco_opcode::image_sample_lz_o; + if (has_lod) + opcode = aco_opcode::image_sample_l_o; + } + } else { /* no offset */ + if (has_compare) { + opcode = aco_opcode::image_sample_c; + if (has_derivs) + opcode = aco_opcode::image_sample_c_d; + if (has_bias) + opcode = aco_opcode::image_sample_c_b; + if (level_zero) + opcode = aco_opcode::image_sample_c_lz; + if (has_lod) + opcode = aco_opcode::image_sample_c_l; + } else { + opcode = aco_opcode::image_sample; + if (has_derivs) + opcode = aco_opcode::image_sample_d; + if (has_bias) + opcode = aco_opcode::image_sample_b; + if (level_zero) + opcode = aco_opcode::image_sample_lz; + if (has_lod) + opcode = aco_opcode::image_sample_l; + } + } + + if (instr->op == nir_texop_tg4) { + if (has_offset) { + opcode = aco_opcode::image_gather4_lz_o; + if (has_compare) + opcode = aco_opcode::image_gather4_c_lz_o; + } else { + opcode = aco_opcode::image_gather4_lz; + if (has_compare) + opcode = aco_opcode::image_gather4_c_lz; + } + } else if (instr->op == nir_texop_lod) { + opcode = aco_opcode::image_get_lod; + } + + tex.reset(create_instruction(opcode, Format::MIMG, 3, 1)); + tex->operands[0] = arg; + tex->operands[1] = Operand(resource); + tex->operands[2] = Operand(sampler); + tex->dmask = dmask; + tex->da = da; + tex->definitions[0] = Definition(tmp_dst); + tex->can_reorder = true; + ctx->block->instructions.emplace_back(std::move(tex)); + + if (tg4_integer_cube_workaround) { + assert(tmp_dst.id() != dst.id()); + assert(tmp_dst.size() == dst.size() && dst.size() == 4); + + emit_split_vector(ctx, tmp_dst, tmp_dst.size()); + Temp val[4]; + for (unsigned i = 0; i < dst.size(); i++) { + val[i] = emit_extract_vector(ctx, tmp_dst, i, v1); + Temp cvt_val; + if (stype == GLSL_TYPE_UINT) + cvt_val = bld.vop1(aco_opcode::v_cvt_u32_f32, bld.def(v1), val[i]); + else + cvt_val = bld.vop1(aco_opcode::v_cvt_i32_f32, bld.def(v1), val[i]); + val[i] = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), val[i], cvt_val, tg4_compare_cube_wa64); + } + Temp tmp = dst.regClass() == v4 ? dst : bld.tmp(v4); + tmp_dst = bld.pseudo(aco_opcode::p_create_vector, Definition(tmp), + val[0], val[1], val[2], val[3]); + } + unsigned mask = instr->op == nir_texop_tg4 ? 0xF : dmask; + expand_vector(ctx, tmp_dst, dst, instr->dest.ssa.num_components, mask); + +} + + +Operand get_phi_operand(isel_context *ctx, nir_ssa_def *ssa) +{ + Temp tmp = get_ssa_temp(ctx, ssa); + if (ssa->parent_instr->type == nir_instr_type_ssa_undef) + return Operand(tmp.regClass()); + else + return Operand(tmp); +} + +void visit_phi(isel_context *ctx, nir_phi_instr *instr) +{ + aco_ptr phi; + unsigned num_src = exec_list_length(&instr->srcs); + Temp dst = get_ssa_temp(ctx, &instr->dest.ssa); + + aco_opcode opcode = !dst.is_linear() || ctx->divergent_vals[instr->dest.ssa.index] ? aco_opcode::p_phi : aco_opcode::p_linear_phi; + + std::map phi_src; + bool all_undef = true; + nir_foreach_phi_src(src, instr) { + phi_src[src->pred->index] = src->src.ssa; + if (src->src.ssa->parent_instr->type != nir_instr_type_ssa_undef) + all_undef = false; + } + if (all_undef) { + Builder bld(ctx->program, ctx->block); + if (dst.regClass() == s1) { + bld.sop1(aco_opcode::s_mov_b32, Definition(dst), Operand(0u)); + } else if (dst.regClass() == v1) { + bld.vop1(aco_opcode::v_mov_b32, Definition(dst), Operand(0u)); + } else { + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, dst.size(), 1)}; + for (unsigned i = 0; i < dst.size(); i++) + vec->operands[i] = Operand(0u); + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + } + return; + } + + /* try to scalarize vector phis */ + if (dst.size() > 1) { + // TODO: scalarize linear phis on divergent ifs + bool can_scalarize = (opcode == aco_opcode::p_phi || !(ctx->block->kind & block_kind_merge)); + std::array new_vec; + for (std::pair& pair : phi_src) { + Operand src = get_phi_operand(ctx, pair.second); + if (src.isTemp() && ctx->allocated_vec.find(src.tempId()) == ctx->allocated_vec.end()) { + can_scalarize = false; + break; + } + } + if (can_scalarize) { + unsigned num_components = instr->dest.ssa.num_components; + assert(dst.size() % num_components == 0); + RegClass rc = RegClass(dst.type(), dst.size() / num_components); + + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, num_components, 1)}; + for (unsigned k = 0; k < num_components; k++) { + phi.reset(create_instruction(opcode, Format::PSEUDO, num_src, 1)); + std::map::iterator it = phi_src.begin(); + for (unsigned i = 0; i < num_src; i++) { + Operand src = get_phi_operand(ctx, it->second); + phi->operands[i] = src.isTemp() ? Operand(ctx->allocated_vec[src.tempId()][k]) : Operand(rc); + ++it; + } + Temp phi_dst = {ctx->program->allocateId(), rc}; + phi->definitions[0] = Definition(phi_dst); + ctx->block->instructions.emplace(ctx->block->instructions.begin(), std::move(phi)); + new_vec[k] = phi_dst; + vec->operands[k] = Operand(phi_dst); + } + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + ctx->allocated_vec.emplace(dst.id(), new_vec); + return; + } + } + + unsigned extra_src = 0; + if (opcode == aco_opcode::p_linear_phi && (ctx->block->kind & block_kind_loop_exit) && + ctx->program->blocks[ctx->block->index-2].kind & block_kind_continue_or_break) { + extra_src++; + } + + phi.reset(create_instruction(opcode, Format::PSEUDO, num_src + extra_src, 1)); + + /* if we have a linear phi on a divergent if, we know that one src is undef */ + if (opcode == aco_opcode::p_linear_phi && ctx->block->kind & block_kind_merge) { + assert(extra_src == 0); + Block* block; + /* we place the phi either in the invert-block or in the current block */ + if (phi_src.begin()->second->parent_instr->type != nir_instr_type_ssa_undef) { + assert((++phi_src.begin())->second->parent_instr->type == nir_instr_type_ssa_undef); + Block& linear_else = ctx->program->blocks[ctx->block->linear_preds[1]]; + block = &ctx->program->blocks[linear_else.linear_preds[0]]; + assert(block->kind & block_kind_invert); + phi->operands[0] = get_phi_operand(ctx, phi_src.begin()->second); + } else { + assert((++phi_src.begin())->second->parent_instr->type != nir_instr_type_ssa_undef); + block = ctx->block; + phi->operands[0] = get_phi_operand(ctx, (++phi_src.begin())->second); + } + phi->operands[1] = Operand(dst.regClass()); + phi->definitions[0] = Definition(dst); + block->instructions.emplace(block->instructions.begin(), std::move(phi)); + return; + } + + std::map::iterator it = phi_src.begin(); + for (unsigned i = 0; i < num_src; i++) { + phi->operands[i] = get_phi_operand(ctx, it->second); + ++it; + } + for (unsigned i = 0; i < extra_src; i++) + phi->operands[num_src + i] = Operand(dst.regClass()); + phi->definitions[0] = Definition(dst); + ctx->block->instructions.emplace(ctx->block->instructions.begin(), std::move(phi)); +} + + +void visit_undef(isel_context *ctx, nir_ssa_undef_instr *instr) +{ + Temp dst = get_ssa_temp(ctx, &instr->def); + + assert(dst.type() == RegType::sgpr); + + if (dst.size() == 1) { + Builder(ctx->program, ctx->block).copy(Definition(dst), Operand(0u)); + } else { + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, dst.size(), 1)}; + for (unsigned i = 0; i < dst.size(); i++) + vec->operands[i] = Operand(0u); + vec->definitions[0] = Definition(dst); + ctx->block->instructions.emplace_back(std::move(vec)); + } +} + +void visit_jump(isel_context *ctx, nir_jump_instr *instr) +{ + Builder bld(ctx->program, ctx->block); + Block *logical_target; + append_logical_end(ctx->block); + unsigned idx = ctx->block->index; + + switch (instr->type) { + case nir_jump_break: + logical_target = ctx->cf_info.parent_loop.exit; + add_logical_edge(idx, logical_target); + ctx->block->kind |= block_kind_break; + + if (!ctx->cf_info.parent_if.is_divergent && + !ctx->cf_info.parent_loop.has_divergent_continue) { + /* uniform break - directly jump out of the loop */ + ctx->block->kind |= block_kind_uniform; + ctx->cf_info.has_branch = true; + bld.branch(aco_opcode::p_branch); + add_linear_edge(idx, logical_target); + return; + } + ctx->cf_info.parent_loop.has_divergent_branch = true; + break; + case nir_jump_continue: + logical_target = &ctx->program->blocks[ctx->cf_info.parent_loop.header_idx]; + add_logical_edge(idx, logical_target); + ctx->block->kind |= block_kind_continue; + + if (ctx->cf_info.parent_if.is_divergent) { + /* for potential uniform breaks after this continue, + we must ensure that they are handled correctly */ + ctx->cf_info.parent_loop.has_divergent_continue = true; + ctx->cf_info.parent_loop.has_divergent_branch = true; + } else { + /* uniform continue - directly jump to the loop header */ + ctx->block->kind |= block_kind_uniform; + ctx->cf_info.has_branch = true; + bld.branch(aco_opcode::p_branch); + add_linear_edge(idx, logical_target); + return; + } + break; + default: + fprintf(stderr, "Unknown NIR jump instr: "); + nir_print_instr(&instr->instr, stderr); + fprintf(stderr, "\n"); + abort(); + } + + /* remove critical edges from linear CFG */ + bld.branch(aco_opcode::p_branch); + Block* break_block = ctx->program->create_and_insert_block(); + break_block->loop_nest_depth = ctx->cf_info.loop_nest_depth; + break_block->kind |= block_kind_uniform; + add_linear_edge(idx, break_block); + /* the loop_header pointer might be invalidated by this point */ + if (instr->type == nir_jump_continue) + logical_target = &ctx->program->blocks[ctx->cf_info.parent_loop.header_idx]; + add_linear_edge(break_block->index, logical_target); + bld.reset(break_block); + bld.branch(aco_opcode::p_branch); + + Block* continue_block = ctx->program->create_and_insert_block(); + continue_block->loop_nest_depth = ctx->cf_info.loop_nest_depth; + add_linear_edge(idx, continue_block); + append_logical_start(continue_block); + ctx->block = continue_block; + return; +} + +void visit_block(isel_context *ctx, nir_block *block) +{ + nir_foreach_instr(instr, block) { + switch (instr->type) { + case nir_instr_type_alu: + visit_alu_instr(ctx, nir_instr_as_alu(instr)); + break; + case nir_instr_type_load_const: + visit_load_const(ctx, nir_instr_as_load_const(instr)); + break; + case nir_instr_type_intrinsic: + visit_intrinsic(ctx, nir_instr_as_intrinsic(instr)); + break; + case nir_instr_type_tex: + visit_tex(ctx, nir_instr_as_tex(instr)); + break; + case nir_instr_type_phi: + visit_phi(ctx, nir_instr_as_phi(instr)); + break; + case nir_instr_type_ssa_undef: + visit_undef(ctx, nir_instr_as_ssa_undef(instr)); + break; + case nir_instr_type_deref: + break; + case nir_instr_type_jump: + visit_jump(ctx, nir_instr_as_jump(instr)); + break; + default: + fprintf(stderr, "Unknown NIR instr type: "); + nir_print_instr(instr, stderr); + fprintf(stderr, "\n"); + //abort(); + } + } +} + + + +static void visit_loop(isel_context *ctx, nir_loop *loop) +{ + append_logical_end(ctx->block); + ctx->block->kind |= block_kind_loop_preheader | block_kind_uniform; + Builder bld(ctx->program, ctx->block); + bld.branch(aco_opcode::p_branch); + unsigned loop_preheader_idx = ctx->block->index; + + Block loop_exit = Block(); + loop_exit.loop_nest_depth = ctx->cf_info.loop_nest_depth; + loop_exit.kind |= (block_kind_loop_exit | (ctx->block->kind & block_kind_top_level)); + + Block* loop_header = ctx->program->create_and_insert_block(); + loop_header->loop_nest_depth = ctx->cf_info.loop_nest_depth + 1; + loop_header->kind |= block_kind_loop_header; + add_edge(loop_preheader_idx, loop_header); + ctx->block = loop_header; + + /* emit loop body */ + unsigned loop_header_idx = loop_header->index; + loop_info_RAII loop_raii(ctx, loop_header_idx, &loop_exit); + append_logical_start(ctx->block); + visit_cf_list(ctx, &loop->body); + + //TODO: what if a loop ends with a unconditional or uniformly branched continue and this branch is never taken? + if (!ctx->cf_info.has_branch) { + append_logical_end(ctx->block); + if (ctx->cf_info.exec_potentially_empty) { + /* Discards can result in code running with an empty exec mask. + * This would result in divergent breaks not ever being taken. As a + * workaround, break the loop when the loop mask is empty instead of + * always continuing. */ + ctx->block->kind |= (block_kind_continue_or_break | block_kind_uniform); + + /* create "loop_almost_exit" to avoid critical edges */ + unsigned block_idx = ctx->block->index; + Block *loop_almost_exit = ctx->program->create_and_insert_block(); + loop_almost_exit->loop_nest_depth = ctx->cf_info.loop_nest_depth; + loop_almost_exit->kind = block_kind_uniform; + bld.reset(loop_almost_exit); + bld.branch(aco_opcode::p_branch); + + add_linear_edge(block_idx, loop_almost_exit); + add_linear_edge(loop_almost_exit->index, &loop_exit); + + ctx->block = &ctx->program->blocks[block_idx]; + } else { + ctx->block->kind |= (block_kind_continue | block_kind_uniform); + } + if (!ctx->cf_info.parent_loop.has_divergent_branch) + add_edge(ctx->block->index, &ctx->program->blocks[loop_header_idx]); + else + add_linear_edge(ctx->block->index, &ctx->program->blocks[loop_header_idx]); + bld.reset(ctx->block); + bld.branch(aco_opcode::p_branch); + } + + /* fixup phis in loop header from unreachable blocks */ + if (ctx->cf_info.has_branch || ctx->cf_info.parent_loop.has_divergent_branch) { + bool linear = ctx->cf_info.has_branch; + bool logical = ctx->cf_info.has_branch || ctx->cf_info.parent_loop.has_divergent_branch; + for (aco_ptr& instr : ctx->program->blocks[loop_header_idx].instructions) { + if ((logical && instr->opcode == aco_opcode::p_phi) || + (linear && instr->opcode == aco_opcode::p_linear_phi)) { + /* the last operand should be the one that needs to be removed */ + instr->operands.pop_back(); + } else if (!is_phi(instr)) { + break; + } + } + } + + ctx->cf_info.has_branch = false; + + // TODO: if the loop has not a single exit, we must add one °° + /* emit loop successor block */ + ctx->block = ctx->program->insert_block(std::move(loop_exit)); + append_logical_start(ctx->block); + + #if 0 + // TODO: check if it is beneficial to not branch on continues + /* trim linear phis in loop header */ + for (auto&& instr : loop_entry->instructions) { + if (instr->opcode == aco_opcode::p_linear_phi) { + aco_ptr new_phi{create_instruction(aco_opcode::p_linear_phi, Format::PSEUDO, loop_entry->linear_predecessors.size(), 1)}; + new_phi->definitions[0] = instr->definitions[0]; + for (unsigned i = 0; i < new_phi->operands.size(); i++) + new_phi->operands[i] = instr->operands[i]; + /* check that the remaining operands are all the same */ + for (unsigned i = new_phi->operands.size(); i < instr->operands.size(); i++) + assert(instr->operands[i].tempId() == instr->operands.back().tempId()); + instr.swap(new_phi); + } else if (instr->opcode == aco_opcode::p_phi) { + continue; + } else { + break; + } + } + #endif +} + +static void begin_divergent_if_then(isel_context *ctx, if_context *ic, Temp cond) +{ + ic->cond = cond; + + append_logical_end(ctx->block); + ctx->block->kind |= block_kind_branch; + + /* branch to linear then block */ + assert(cond.regClass() == s2); + aco_ptr branch; + branch.reset(create_instruction(aco_opcode::p_cbranch_z, Format::PSEUDO_BRANCH, 1, 0)); + branch->operands[0] = Operand(cond); + ctx->block->instructions.push_back(std::move(branch)); + + ic->BB_if_idx = ctx->block->index; + ic->BB_invert = Block(); + ic->BB_invert.loop_nest_depth = ctx->cf_info.loop_nest_depth; + /* Invert blocks are intentionally not marked as top level because they + * are not part of the logical cfg. */ + ic->BB_invert.kind |= block_kind_invert; + ic->BB_endif = Block(); + ic->BB_endif.loop_nest_depth = ctx->cf_info.loop_nest_depth; + ic->BB_endif.kind |= (block_kind_merge | (ctx->block->kind & block_kind_top_level)); + + ic->exec_potentially_empty_old = ctx->cf_info.exec_potentially_empty; + ic->divergent_old = ctx->cf_info.parent_if.is_divergent; + ctx->cf_info.parent_if.is_divergent = true; + ctx->cf_info.exec_potentially_empty = false; /* divergent branches use cbranch_execz */ + + /** emit logical then block */ + Block* BB_then_logical = ctx->program->create_and_insert_block(); + BB_then_logical->loop_nest_depth = ctx->cf_info.loop_nest_depth; + add_edge(ic->BB_if_idx, BB_then_logical); + ctx->block = BB_then_logical; + append_logical_start(BB_then_logical); +} + +static void begin_divergent_if_else(isel_context *ctx, if_context *ic) +{ + Block *BB_then_logical = ctx->block; + append_logical_end(BB_then_logical); + /* branch from logical then block to invert block */ + aco_ptr branch; + branch.reset(create_instruction(aco_opcode::p_branch, Format::PSEUDO_BRANCH, 0, 0)); + BB_then_logical->instructions.emplace_back(std::move(branch)); + add_linear_edge(BB_then_logical->index, &ic->BB_invert); + if (!ctx->cf_info.parent_loop.has_divergent_branch) + add_logical_edge(BB_then_logical->index, &ic->BB_endif); + BB_then_logical->kind |= block_kind_uniform; + assert(!ctx->cf_info.has_branch); + ic->then_branch_divergent = ctx->cf_info.parent_loop.has_divergent_branch; + ctx->cf_info.parent_loop.has_divergent_branch = false; + + /** emit linear then block */ + Block* BB_then_linear = ctx->program->create_and_insert_block(); + BB_then_linear->loop_nest_depth = ctx->cf_info.loop_nest_depth; + BB_then_linear->kind |= block_kind_uniform; + add_linear_edge(ic->BB_if_idx, BB_then_linear); + /* branch from linear then block to invert block */ + branch.reset(create_instruction(aco_opcode::p_branch, Format::PSEUDO_BRANCH, 0, 0)); + BB_then_linear->instructions.emplace_back(std::move(branch)); + add_linear_edge(BB_then_linear->index, &ic->BB_invert); + + /** emit invert merge block */ + ctx->block = ctx->program->insert_block(std::move(ic->BB_invert)); + ic->invert_idx = ctx->block->index; + + /* branch to linear else block (skip else) */ + branch.reset(create_instruction(aco_opcode::p_cbranch_nz, Format::PSEUDO_BRANCH, 1, 0)); + branch->operands[0] = Operand(ic->cond); + ctx->block->instructions.push_back(std::move(branch)); + + ic->exec_potentially_empty_old |= ctx->cf_info.exec_potentially_empty; + ctx->cf_info.exec_potentially_empty = false; /* divergent branches use cbranch_execz */ + + /** emit logical else block */ + Block* BB_else_logical = ctx->program->create_and_insert_block(); + BB_else_logical->loop_nest_depth = ctx->cf_info.loop_nest_depth; + add_logical_edge(ic->BB_if_idx, BB_else_logical); + add_linear_edge(ic->invert_idx, BB_else_logical); + ctx->block = BB_else_logical; + append_logical_start(BB_else_logical); +} + +static void end_divergent_if(isel_context *ctx, if_context *ic) +{ + Block *BB_else_logical = ctx->block; + append_logical_end(BB_else_logical); + + /* branch from logical else block to endif block */ + aco_ptr branch; + branch.reset(create_instruction(aco_opcode::p_branch, Format::PSEUDO_BRANCH, 0, 0)); + BB_else_logical->instructions.emplace_back(std::move(branch)); + add_linear_edge(BB_else_logical->index, &ic->BB_endif); + if (!ctx->cf_info.parent_loop.has_divergent_branch) + add_logical_edge(BB_else_logical->index, &ic->BB_endif); + BB_else_logical->kind |= block_kind_uniform; + + assert(!ctx->cf_info.has_branch); + ctx->cf_info.parent_loop.has_divergent_branch &= ic->then_branch_divergent; + + + /** emit linear else block */ + Block* BB_else_linear = ctx->program->create_and_insert_block(); + BB_else_linear->loop_nest_depth = ctx->cf_info.loop_nest_depth; + BB_else_linear->kind |= block_kind_uniform; + add_linear_edge(ic->invert_idx, BB_else_linear); + + /* branch from linear else block to endif block */ + branch.reset(create_instruction(aco_opcode::p_branch, Format::PSEUDO_BRANCH, 0, 0)); + BB_else_linear->instructions.emplace_back(std::move(branch)); + add_linear_edge(BB_else_linear->index, &ic->BB_endif); + + + /** emit endif merge block */ + ctx->block = ctx->program->insert_block(std::move(ic->BB_endif)); + append_logical_start(ctx->block); + + + ctx->cf_info.parent_if.is_divergent = ic->divergent_old; + ctx->cf_info.exec_potentially_empty |= ic->exec_potentially_empty_old; + /* uniform control flow never has an empty exec-mask */ + if (!ctx->cf_info.loop_nest_depth && !ctx->cf_info.parent_if.is_divergent) + ctx->cf_info.exec_potentially_empty = false; +} + +static void visit_if(isel_context *ctx, nir_if *if_stmt) +{ + Temp cond = get_ssa_temp(ctx, if_stmt->condition.ssa); + Builder bld(ctx->program, ctx->block); + aco_ptr branch; + + if (!ctx->divergent_vals[if_stmt->condition.ssa->index]) { /* uniform condition */ + /** + * Uniform conditionals are represented in the following way*) : + * + * The linear and logical CFG: + * BB_IF + * / \ + * BB_THEN (logical) BB_ELSE (logical) + * \ / + * BB_ENDIF + * + * *) Exceptions may be due to break and continue statements within loops + * If a break/continue happens within uniform control flow, it branches + * to the loop exit/entry block. Otherwise, it branches to the next + * merge block. + **/ + append_logical_end(ctx->block); + ctx->block->kind |= block_kind_uniform; + + /* emit branch */ + if (cond.regClass() == s2) { + // TODO: in a post-RA optimizer, we could check if the condition is in VCC and omit this instruction + cond = as_uniform_bool(ctx, cond); + } + branch.reset(create_instruction(aco_opcode::p_cbranch_z, Format::PSEUDO_BRANCH, 1, 0)); + branch->operands[0] = Operand(cond); + branch->operands[0].setFixed(scc); + ctx->block->instructions.emplace_back(std::move(branch)); + + unsigned BB_if_idx = ctx->block->index; + Block BB_endif = Block(); + BB_endif.loop_nest_depth = ctx->cf_info.loop_nest_depth; + BB_endif.kind |= ctx->block->kind & block_kind_top_level; + + /** emit then block */ + Block* BB_then = ctx->program->create_and_insert_block(); + BB_then->loop_nest_depth = ctx->cf_info.loop_nest_depth; + add_edge(BB_if_idx, BB_then); + append_logical_start(BB_then); + ctx->block = BB_then; + visit_cf_list(ctx, &if_stmt->then_list); + BB_then = ctx->block; + bool then_branch = ctx->cf_info.has_branch; + bool then_branch_divergent = ctx->cf_info.parent_loop.has_divergent_branch; + + if (!then_branch) { + append_logical_end(BB_then); + /* branch from then block to endif block */ + branch.reset(create_instruction(aco_opcode::p_branch, Format::PSEUDO_BRANCH, 0, 0)); + BB_then->instructions.emplace_back(std::move(branch)); + add_linear_edge(BB_then->index, &BB_endif); + if (!then_branch_divergent) + add_logical_edge(BB_then->index, &BB_endif); + BB_then->kind |= block_kind_uniform; + } + + ctx->cf_info.has_branch = false; + ctx->cf_info.parent_loop.has_divergent_branch = false; + + /** emit else block */ + Block* BB_else = ctx->program->create_and_insert_block(); + BB_else->loop_nest_depth = ctx->cf_info.loop_nest_depth; + add_edge(BB_if_idx, BB_else); + append_logical_start(BB_else); + ctx->block = BB_else; + visit_cf_list(ctx, &if_stmt->else_list); + BB_else = ctx->block; + + if (!ctx->cf_info.has_branch) { + append_logical_end(BB_else); + /* branch from then block to endif block */ + branch.reset(create_instruction(aco_opcode::p_branch, Format::PSEUDO_BRANCH, 0, 0)); + BB_else->instructions.emplace_back(std::move(branch)); + add_linear_edge(BB_else->index, &BB_endif); + if (!ctx->cf_info.parent_loop.has_divergent_branch) + add_logical_edge(BB_else->index, &BB_endif); + BB_else->kind |= block_kind_uniform; + } + + ctx->cf_info.has_branch &= then_branch; + ctx->cf_info.parent_loop.has_divergent_branch &= then_branch_divergent; + + /** emit endif merge block */ + if (!ctx->cf_info.has_branch) { + ctx->block = ctx->program->insert_block(std::move(BB_endif)); + append_logical_start(ctx->block); + } + } else { /* non-uniform condition */ + /** + * To maintain a logical and linear CFG without critical edges, + * non-uniform conditionals are represented in the following way*) : + * + * The linear CFG: + * BB_IF + * / \ + * BB_THEN (logical) BB_THEN (linear) + * \ / + * BB_INVERT (linear) + * / \ + * BB_ELSE (logical) BB_ELSE (linear) + * \ / + * BB_ENDIF + * + * The logical CFG: + * BB_IF + * / \ + * BB_THEN (logical) BB_ELSE (logical) + * \ / + * BB_ENDIF + * + * *) Exceptions may be due to break and continue statements within loops + **/ + + if_context ic; + + begin_divergent_if_then(ctx, &ic, cond); + visit_cf_list(ctx, &if_stmt->then_list); + + begin_divergent_if_else(ctx, &ic); + visit_cf_list(ctx, &if_stmt->else_list); + + end_divergent_if(ctx, &ic); + } +} + +static void visit_cf_list(isel_context *ctx, + struct exec_list *list) +{ + foreach_list_typed(nir_cf_node, node, node, list) { + switch (node->type) { + case nir_cf_node_block: + visit_block(ctx, nir_cf_node_as_block(node)); + break; + case nir_cf_node_if: + visit_if(ctx, nir_cf_node_as_if(node)); + break; + case nir_cf_node_loop: + visit_loop(ctx, nir_cf_node_as_loop(node)); + break; + default: + unreachable("unimplemented cf list type"); + } + } +} + +static void export_vs_varying(isel_context *ctx, int slot, bool is_pos, int *next_pos) +{ + int offset = ctx->program->info->vs.outinfo.vs_output_param_offset[slot]; + uint64_t mask = ctx->vs_output.mask[slot]; + if (!is_pos && !mask) + return; + if (!is_pos && offset == AC_EXP_PARAM_UNDEFINED) + return; + aco_ptr exp{create_instruction(aco_opcode::exp, Format::EXP, 4, 0)}; + exp->enabled_mask = mask; + for (unsigned i = 0; i < 4; ++i) { + if (mask & (1 << i)) + exp->operands[i] = Operand(ctx->vs_output.outputs[slot][i]); + else + exp->operands[i] = Operand(v1); + } + exp->valid_mask = false; + exp->done = false; + exp->compressed = false; + if (is_pos) + exp->dest = V_008DFC_SQ_EXP_POS + (*next_pos)++; + else + exp->dest = V_008DFC_SQ_EXP_PARAM + offset; + ctx->block->instructions.emplace_back(std::move(exp)); +} + +static void export_vs_psiz_layer_viewport(isel_context *ctx, int *next_pos) +{ + aco_ptr exp{create_instruction(aco_opcode::exp, Format::EXP, 4, 0)}; + exp->enabled_mask = 0; + for (unsigned i = 0; i < 4; ++i) + exp->operands[i] = Operand(v1); + if (ctx->vs_output.mask[VARYING_SLOT_PSIZ]) { + exp->operands[0] = Operand(ctx->vs_output.outputs[VARYING_SLOT_PSIZ][0]); + exp->enabled_mask |= 0x1; + } + if (ctx->vs_output.mask[VARYING_SLOT_LAYER]) { + exp->operands[2] = Operand(ctx->vs_output.outputs[VARYING_SLOT_LAYER][0]); + exp->enabled_mask |= 0x4; + } + if (ctx->vs_output.mask[VARYING_SLOT_VIEWPORT]) { + if (ctx->options->chip_class < GFX9) { + exp->operands[3] = Operand(ctx->vs_output.outputs[VARYING_SLOT_VIEWPORT][0]); + exp->enabled_mask |= 0x8; + } else { + Builder bld(ctx->program, ctx->block); + + Temp out = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(16u), + Operand(ctx->vs_output.outputs[VARYING_SLOT_VIEWPORT][0])); + if (exp->operands[2].isTemp()) + out = bld.vop2(aco_opcode::v_or_b32, bld.def(v1), Operand(out), exp->operands[2]); + + exp->operands[2] = Operand(out); + exp->enabled_mask |= 0x4; + } + } + exp->valid_mask = false; + exp->done = false; + exp->compressed = false; + exp->dest = V_008DFC_SQ_EXP_POS + (*next_pos)++; + ctx->block->instructions.emplace_back(std::move(exp)); +} + +static void create_vs_exports(isel_context *ctx) +{ + radv_vs_output_info *outinfo = &ctx->program->info->vs.outinfo; + + if (outinfo->export_prim_id) { + ctx->vs_output.mask[VARYING_SLOT_PRIMITIVE_ID] |= 0x1; + ctx->vs_output.outputs[VARYING_SLOT_PRIMITIVE_ID][0] = ctx->vs_prim_id; + } + + if (ctx->options->key.has_multiview_view_index) { + ctx->vs_output.mask[VARYING_SLOT_LAYER] |= 0x1; + ctx->vs_output.outputs[VARYING_SLOT_LAYER][0] = as_vgpr(ctx, ctx->view_index); + } + + /* the order these position exports are created is important */ + int next_pos = 0; + export_vs_varying(ctx, VARYING_SLOT_POS, true, &next_pos); + if (outinfo->writes_pointsize || outinfo->writes_layer || outinfo->writes_viewport_index) { + export_vs_psiz_layer_viewport(ctx, &next_pos); + } + if (ctx->num_clip_distances + ctx->num_cull_distances > 0) + export_vs_varying(ctx, VARYING_SLOT_CLIP_DIST0, true, &next_pos); + if (ctx->num_clip_distances + ctx->num_cull_distances > 4) + export_vs_varying(ctx, VARYING_SLOT_CLIP_DIST1, true, &next_pos); + + if (ctx->options->key.vs_common_out.export_clip_dists) { + if (ctx->num_clip_distances + ctx->num_cull_distances > 0) + export_vs_varying(ctx, VARYING_SLOT_CLIP_DIST0, false, &next_pos); + if (ctx->num_clip_distances + ctx->num_cull_distances > 4) + export_vs_varying(ctx, VARYING_SLOT_CLIP_DIST1, false, &next_pos); + } + + for (unsigned i = 0; i <= VARYING_SLOT_VAR31; ++i) { + if (i < VARYING_SLOT_VAR0 && i != VARYING_SLOT_LAYER && + i != VARYING_SLOT_PRIMITIVE_ID) + continue; + + export_vs_varying(ctx, i, false, NULL); + } +} + +static void emit_stream_output(isel_context *ctx, + Temp const *so_buffers, + Temp const *so_write_offset, + const struct radv_stream_output *output) +{ + unsigned num_comps = util_bitcount(output->component_mask); + unsigned loc = output->location; + unsigned buf = output->buffer; + unsigned offset = output->offset; + + assert(num_comps && num_comps <= 4); + if (!num_comps || num_comps > 4) + return; + + unsigned start = ffs(output->component_mask) - 1; + + Temp out[4]; + bool all_undef = true; + assert(ctx->stage == vertex_vs); + for (unsigned i = 0; i < num_comps; i++) { + out[i] = ctx->vs_output.outputs[loc][start + i]; + all_undef = all_undef && !out[i].id(); + } + if (all_undef) + return; + + Temp write_data = {ctx->program->allocateId(), RegClass(RegType::vgpr, num_comps)}; + aco_ptr vec{create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, num_comps, 1)}; + for (unsigned i = 0; i < num_comps; ++i) + vec->operands[i] = (ctx->vs_output.mask[loc] & 1 << i) ? Operand(out[i]) : Operand(0u); + vec->definitions[0] = Definition(write_data); + ctx->block->instructions.emplace_back(std::move(vec)); + + aco_opcode opcode; + switch (num_comps) { + case 1: + opcode = aco_opcode::buffer_store_dword; + break; + case 2: + opcode = aco_opcode::buffer_store_dwordx2; + break; + case 3: + opcode = aco_opcode::buffer_store_dwordx3; + break; + case 4: + opcode = aco_opcode::buffer_store_dwordx4; + break; + } + + aco_ptr store{create_instruction(opcode, Format::MUBUF, 4, 0)}; + store->operands[0] = Operand(so_write_offset[buf]); + store->operands[1] = Operand(so_buffers[buf]); + store->operands[2] = Operand((uint32_t) 0); + store->operands[3] = Operand(write_data); + if (offset > 4095) { + /* Don't think this can happen in RADV, but maybe GL? It's easy to do this anyway. */ + Builder bld(ctx->program, ctx->block); + store->operands[0] = bld.vadd32(bld.def(v1), Operand(offset), Operand(so_write_offset[buf])); + } else { + store->offset = offset; + } + store->offen = true; + store->glc = true; + store->slc = true; + store->can_reorder = true; + ctx->block->instructions.emplace_back(std::move(store)); +} + +static void emit_streamout(isel_context *ctx, unsigned stream) +{ + Builder bld(ctx->program, ctx->block); + + Temp so_buffers[4]; + Temp buf_ptr = convert_pointer_to_64_bit(ctx, ctx->streamout_buffers); + for (unsigned i = 0; i < 4; i++) { + unsigned stride = ctx->program->info->so.strides[i]; + if (!stride) + continue; + + so_buffers[i] = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), buf_ptr, Operand(i * 16u)); + } + + Temp so_vtx_count = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), + ctx->streamout_config, Operand(0x70010u)); + + Temp tid = bld.vop3(aco_opcode::v_mbcnt_hi_u32_b32, bld.def(v1), Operand((uint32_t) -1), + bld.vop3(aco_opcode::v_mbcnt_lo_u32_b32, bld.def(v1), Operand((uint32_t) -1), Operand(0u))); + + Temp can_emit = bld.vopc(aco_opcode::v_cmp_gt_i32, bld.def(s2), so_vtx_count, tid); + + if_context ic; + begin_divergent_if_then(ctx, &ic, can_emit); + + bld.reset(ctx->block); + + Temp so_write_index = bld.vadd32(bld.def(v1), ctx->streamout_write_idx, tid); + + Temp so_write_offset[4]; + + for (unsigned i = 0; i < 4; i++) { + unsigned stride = ctx->program->info->so.strides[i]; + if (!stride) + continue; + + if (stride == 1) { + Temp offset = bld.sop2(aco_opcode::s_add_i32, bld.def(s1), bld.def(s1, scc), + ctx->streamout_write_idx, ctx->streamout_offset[i]); + Temp new_offset = bld.vadd32(bld.def(v1), offset, tid); + + so_write_offset[i] = bld.vop2(aco_opcode::v_lshlrev_b32, bld.def(v1), Operand(2u), new_offset); + } else { + Temp offset = bld.v_mul_imm(bld.def(v1), so_write_index, stride * 4u); + Temp offset2 = bld.sop2(aco_opcode::s_mul_i32, bld.def(s1), Operand(4u), ctx->streamout_offset[i]); + so_write_offset[i] = bld.vadd32(bld.def(v1), offset, offset2); + } + } + + for (unsigned i = 0; i < ctx->program->info->so.num_outputs; i++) { + struct radv_stream_output *output = + &ctx->program->info->so.outputs[i]; + if (stream != output->stream) + continue; + + emit_stream_output(ctx, so_buffers, so_write_offset, output); + } + + begin_divergent_if_else(ctx, &ic); + end_divergent_if(ctx, &ic); +} + +} /* end namespace */ + +void handle_bc_optimize(isel_context *ctx) +{ + /* needed when SPI_PS_IN_CONTROL.BC_OPTIMIZE_DISABLE is set to 0 */ + Builder bld(ctx->program, ctx->block); + uint32_t spi_ps_input_ena = ctx->program->config->spi_ps_input_ena; + bool uses_center = G_0286CC_PERSP_CENTER_ENA(spi_ps_input_ena) || G_0286CC_LINEAR_CENTER_ENA(spi_ps_input_ena); + bool uses_centroid = G_0286CC_PERSP_CENTROID_ENA(spi_ps_input_ena) || G_0286CC_LINEAR_CENTROID_ENA(spi_ps_input_ena); + if (uses_center && uses_centroid) { + Temp sel = bld.vopc_e64(aco_opcode::v_cmp_lt_i32, bld.hint_vcc(bld.def(s2)), ctx->prim_mask, Operand(0u)); + + if (G_0286CC_PERSP_CENTROID_ENA(spi_ps_input_ena)) { + for (unsigned i = 0; i < 2; i++) { + Temp new_coord = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), + ctx->fs_inputs[fs_input::persp_centroid_p1 + i], + ctx->fs_inputs[fs_input::persp_center_p1 + i], + sel); + ctx->fs_inputs[fs_input::persp_centroid_p1 + i] = new_coord; + } + } + + if (G_0286CC_LINEAR_CENTROID_ENA(spi_ps_input_ena)) { + for (unsigned i = 0; i < 2; i++) { + Temp new_coord = bld.vop2(aco_opcode::v_cndmask_b32, bld.def(v1), + ctx->fs_inputs[fs_input::linear_centroid_p1 + i], + ctx->fs_inputs[fs_input::linear_center_p1 + i], + sel); + ctx->fs_inputs[fs_input::linear_centroid_p1 + i] = new_coord; + } + } + } +} + +void select_program(Program *program, + unsigned shader_count, + struct nir_shader *const *shaders, + ac_shader_config* config, + struct radv_shader_info *info, + struct radv_nir_compiler_options *options) +{ + isel_context ctx = setup_isel_context(program, shader_count, shaders, config, info, options); + + for (unsigned i = 0; i < shader_count; i++) { + nir_shader *nir = shaders[i]; + init_context(&ctx, nir); + + if (!i) { + add_startpgm(&ctx); /* needs to be after init_context() for FS */ + append_logical_start(ctx.block); + } + + if_context ic; + if (shader_count >= 2) { + Builder bld(ctx.program, ctx.block); + Temp count = bld.sop2(aco_opcode::s_bfe_u32, bld.def(s1), bld.def(s1, scc), ctx.merged_wave_info, Operand((8u << 16) | (i * 8u))); + Temp thread_id = bld.vop3(aco_opcode::v_mbcnt_hi_u32_b32, bld.def(v1), Operand((uint32_t) -1), + bld.vop3(aco_opcode::v_mbcnt_lo_u32_b32, bld.def(v1), Operand((uint32_t) -1), Operand(0u))); + Temp cond = bld.vopc(aco_opcode::v_cmp_gt_u32, bld.hint_vcc(bld.def(s2)), count, thread_id); + + begin_divergent_if_then(&ctx, &ic, cond); + } + + if (i) { + Builder bld(ctx.program, ctx.block); + bld.barrier(aco_opcode::p_memory_barrier_shared); //TODO: different barriers are needed for different stages + bld.sopp(aco_opcode::s_barrier); + } + + if (ctx.stage == fragment_fs) + handle_bc_optimize(&ctx); + + nir_function_impl *func = nir_shader_get_entrypoint(nir); + visit_cf_list(&ctx, &func->body); + + if (ctx.program->info->so.num_outputs/*&& !ctx->is_gs_copy_shader */) + emit_streamout(&ctx, 0); + + if (ctx.stage == vertex_vs) + create_vs_exports(&ctx); + + if (shader_count >= 2) { + begin_divergent_if_else(&ctx, &ic); + end_divergent_if(&ctx, &ic); + } + + ralloc_free(ctx.divergent_vals); + } + + append_logical_end(ctx.block); + ctx.block->kind |= block_kind_uniform; + Builder bld(ctx.program, ctx.block); + if (ctx.program->wb_smem_l1_on_end) + bld.smem(aco_opcode::s_dcache_wb, false); + bld.sopp(aco_opcode::s_endpgm); + + /* cleanup CFG */ + for (Block& BB : program->blocks) { + for (unsigned idx : BB.linear_preds) + program->blocks[idx].linear_succs.emplace_back(BB.index); + for (unsigned idx : BB.logical_preds) + program->blocks[idx].logical_succs.emplace_back(BB.index); + } +} +} diff --git a/src/amd/compiler/aco_instruction_selection_setup.cpp b/src/amd/compiler/aco_instruction_selection_setup.cpp new file mode 100644 index 00000000000..6c4c408e659 --- /dev/null +++ b/src/amd/compiler/aco_instruction_selection_setup.cpp @@ -0,0 +1,1366 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include +#include "aco_ir.h" +#include "nir.h" +#include "vulkan/radv_shader.h" +#include "vulkan/radv_descriptor_set.h" +#include "sid.h" +#include "ac_exp_param.h" + +#include "util/u_math.h" + +#define MAX_INLINE_PUSH_CONSTS 8 + +namespace aco { + +enum fs_input { + persp_sample_p1, + persp_sample_p2, + persp_center_p1, + persp_center_p2, + persp_centroid_p1, + persp_centroid_p2, + persp_pull_model, + linear_sample_p1, + linear_sample_p2, + linear_center_p1, + linear_center_p2, + linear_centroid_p1, + linear_centroid_p2, + line_stipple, + frag_pos_0, + frag_pos_1, + frag_pos_2, + frag_pos_3, + front_face, + ancillary, + sample_coverage, + fixed_pt, + max_inputs, +}; + +struct vs_output_state { + uint8_t mask[VARYING_SLOT_VAR31 + 1]; + Temp outputs[VARYING_SLOT_VAR31 + 1][4]; +}; + +struct isel_context { + struct radv_nir_compiler_options *options; + Program *program; + nir_shader *shader; + uint32_t constant_data_offset; + Block *block; + bool *divergent_vals; + std::unique_ptr allocated; + std::unordered_map> allocated_vec; + Stage stage; /* Stage */ + struct { + bool has_branch; + uint16_t loop_nest_depth = 0; + struct { + unsigned header_idx; + Block* exit; + bool has_divergent_continue = false; + bool has_divergent_branch = false; + } parent_loop; + struct { + bool is_divergent = false; + } parent_if; + bool exec_potentially_empty = false; + } cf_info; + + /* scratch */ + bool scratch_enabled = false; + Temp private_segment_buffer = Temp(0, s2); /* also the part of the scratch descriptor on compute */ + Temp scratch_offset = Temp(0, s1); + + /* inputs common for merged stages */ + Temp merged_wave_info = Temp(0, s1); + + /* FS inputs */ + bool fs_vgpr_args[fs_input::max_inputs]; + Temp fs_inputs[fs_input::max_inputs]; + Temp prim_mask = Temp(0, s1); + Temp descriptor_sets[MAX_SETS]; + Temp push_constants = Temp(0, s1); + Temp inline_push_consts[MAX_INLINE_PUSH_CONSTS]; + unsigned num_inline_push_consts = 0; + unsigned base_inline_push_consts = 0; + + /* VS inputs */ + Temp vertex_buffers = Temp(0, s1); + Temp base_vertex = Temp(0, s1); + Temp start_instance = Temp(0, s1); + Temp draw_id = Temp(0, s1); + Temp view_index = Temp(0, s1); + Temp es2gs_offset = Temp(0, s1); + Temp vertex_id = Temp(0, v1); + Temp rel_auto_id = Temp(0, v1); + Temp instance_id = Temp(0, v1); + Temp vs_prim_id = Temp(0, v1); + bool needs_instance_id; + + /* CS inputs */ + Temp num_workgroups[3] = {Temp(0, s1), Temp(0, s1), Temp(0, s1)}; + Temp workgroup_ids[3] = {Temp(0, s1), Temp(0, s1), Temp(0, s1)}; + Temp tg_size = Temp(0, s1); + Temp local_invocation_ids[3] = {Temp(0, v1), Temp(0, v1), Temp(0, v1)}; + + /* VS output information */ + unsigned num_clip_distances; + unsigned num_cull_distances; + vs_output_state vs_output; + + /* Streamout */ + Temp streamout_buffers = Temp(0, s1); + Temp streamout_write_idx = Temp(0, s1); + Temp streamout_config = Temp(0, s1); + Temp streamout_offset[4] = {Temp(0, s1), Temp(0, s1), Temp(0, s1), Temp(0, s1)}; +}; + +fs_input get_interp_input(nir_intrinsic_op intrin, enum glsl_interp_mode interp) +{ + switch (interp) { + case INTERP_MODE_SMOOTH: + case INTERP_MODE_NONE: + if (intrin == nir_intrinsic_load_barycentric_pixel || + intrin == nir_intrinsic_load_barycentric_at_sample || + intrin == nir_intrinsic_load_barycentric_at_offset) + return fs_input::persp_center_p1; + else if (intrin == nir_intrinsic_load_barycentric_centroid) + return fs_input::persp_centroid_p1; + else if (intrin == nir_intrinsic_load_barycentric_sample) + return fs_input::persp_sample_p1; + break; + case INTERP_MODE_NOPERSPECTIVE: + if (intrin == nir_intrinsic_load_barycentric_pixel) + return fs_input::linear_center_p1; + else if (intrin == nir_intrinsic_load_barycentric_centroid) + return fs_input::linear_centroid_p1; + else if (intrin == nir_intrinsic_load_barycentric_sample) + return fs_input::linear_sample_p1; + break; + default: + break; + } + return fs_input::max_inputs; +} + +void init_context(isel_context *ctx, nir_shader *shader) +{ + nir_function_impl *impl = nir_shader_get_entrypoint(shader); + + ctx->shader = shader; + ctx->divergent_vals = nir_divergence_analysis(shader, nir_divergence_view_index_uniform); + + std::unique_ptr allocated{new Temp[impl->ssa_alloc]()}; + memset(&ctx->fs_vgpr_args, false, sizeof(ctx->fs_vgpr_args)); + + bool done = false; + while (!done) { + done = true; + nir_foreach_block(block, impl) { + nir_foreach_instr(instr, block) { + switch(instr->type) { + case nir_instr_type_alu: { + nir_alu_instr *alu_instr = nir_instr_as_alu(instr); + unsigned size = alu_instr->dest.dest.ssa.num_components; + if (alu_instr->dest.dest.ssa.bit_size == 64) + size *= 2; + RegType type = RegType::sgpr; + switch(alu_instr->op) { + case nir_op_fmul: + case nir_op_fadd: + case nir_op_fsub: + case nir_op_fmax: + case nir_op_fmin: + case nir_op_fmax3: + case nir_op_fmin3: + case nir_op_fmed3: + case nir_op_fmod: + case nir_op_frem: + case nir_op_fneg: + case nir_op_fabs: + case nir_op_fsat: + case nir_op_fsign: + case nir_op_frcp: + case nir_op_frsq: + case nir_op_fsqrt: + case nir_op_fexp2: + case nir_op_flog2: + case nir_op_ffract: + case nir_op_ffloor: + case nir_op_fceil: + case nir_op_ftrunc: + case nir_op_fround_even: + case nir_op_fsin: + case nir_op_fcos: + case nir_op_f2f32: + case nir_op_f2f64: + case nir_op_u2f32: + case nir_op_u2f64: + case nir_op_i2f32: + case nir_op_i2f64: + case nir_op_pack_half_2x16: + case nir_op_unpack_half_2x16_split_x: + case nir_op_unpack_half_2x16_split_y: + case nir_op_fddx: + case nir_op_fddy: + case nir_op_fddx_fine: + case nir_op_fddy_fine: + case nir_op_fddx_coarse: + case nir_op_fddy_coarse: + case nir_op_fquantize2f16: + case nir_op_ldexp: + case nir_op_frexp_sig: + case nir_op_frexp_exp: + case nir_op_cube_face_index: + case nir_op_cube_face_coord: + type = RegType::vgpr; + break; + case nir_op_flt: + case nir_op_fge: + case nir_op_feq: + case nir_op_fne: + size = 2; + break; + case nir_op_ilt: + case nir_op_ige: + case nir_op_ult: + case nir_op_uge: + size = alu_instr->src[0].src.ssa->bit_size == 64 ? 2 : 1; + /* fallthrough */ + case nir_op_ieq: + case nir_op_ine: + case nir_op_i2b1: + if (ctx->divergent_vals[alu_instr->dest.dest.ssa.index]) { + size = 2; + } else { + for (unsigned i = 0; i < nir_op_infos[alu_instr->op].num_inputs; i++) { + if (allocated[alu_instr->src[i].src.ssa->index].type() == RegType::vgpr) + size = 2; + } + } + break; + case nir_op_f2i64: + case nir_op_f2u64: + case nir_op_b2i32: + case nir_op_b2f32: + case nir_op_f2i32: + case nir_op_f2u32: + type = ctx->divergent_vals[alu_instr->dest.dest.ssa.index] ? RegType::vgpr : RegType::sgpr; + break; + case nir_op_bcsel: + if (alu_instr->dest.dest.ssa.bit_size == 1) { + if (ctx->divergent_vals[alu_instr->dest.dest.ssa.index]) + size = 2; + else if (allocated[alu_instr->src[1].src.ssa->index].regClass() == s2 && + allocated[alu_instr->src[2].src.ssa->index].regClass() == s2) + size = 2; + else + size = 1; + } else { + if (ctx->divergent_vals[alu_instr->dest.dest.ssa.index]) { + type = RegType::vgpr; + } else { + if (allocated[alu_instr->src[1].src.ssa->index].type() == RegType::vgpr || + allocated[alu_instr->src[2].src.ssa->index].type() == RegType::vgpr) { + type = RegType::vgpr; + } + } + if (alu_instr->src[1].src.ssa->num_components == 1 && alu_instr->src[2].src.ssa->num_components == 1) { + assert(allocated[alu_instr->src[1].src.ssa->index].size() == allocated[alu_instr->src[2].src.ssa->index].size()); + size = allocated[alu_instr->src[1].src.ssa->index].size(); + } + } + break; + case nir_op_mov: + if (alu_instr->dest.dest.ssa.bit_size == 1) { + size = allocated[alu_instr->src[0].src.ssa->index].size(); + } else { + type = ctx->divergent_vals[alu_instr->dest.dest.ssa.index] ? RegType::vgpr : RegType::sgpr; + } + break; + case nir_op_inot: + case nir_op_ixor: + if (alu_instr->dest.dest.ssa.bit_size == 1) { + size = ctx->divergent_vals[alu_instr->dest.dest.ssa.index] ? 2 : 1; + break; + } else { + /* fallthrough */ + } + default: + if (alu_instr->dest.dest.ssa.bit_size == 1) { + if (ctx->divergent_vals[alu_instr->dest.dest.ssa.index]) { + size = 2; + } else { + size = 2; + for (unsigned i = 0; i < nir_op_infos[alu_instr->op].num_inputs; i++) { + if (allocated[alu_instr->src[i].src.ssa->index].regClass() == s1) { + size = 1; + break; + } + } + } + } else { + for (unsigned i = 0; i < nir_op_infos[alu_instr->op].num_inputs; i++) { + if (allocated[alu_instr->src[i].src.ssa->index].type() == RegType::vgpr) + type = RegType::vgpr; + } + } + break; + } + allocated[alu_instr->dest.dest.ssa.index] = Temp(0, RegClass(type, size)); + break; + } + case nir_instr_type_load_const: { + unsigned size = nir_instr_as_load_const(instr)->def.num_components; + if (nir_instr_as_load_const(instr)->def.bit_size == 64) + size *= 2; + allocated[nir_instr_as_load_const(instr)->def.index] = Temp(0, RegClass(RegType::sgpr, size)); + break; + } + case nir_instr_type_intrinsic: { + nir_intrinsic_instr *intrinsic = nir_instr_as_intrinsic(instr); + if (!nir_intrinsic_infos[intrinsic->intrinsic].has_dest) + break; + unsigned size = intrinsic->dest.ssa.num_components; + if (intrinsic->dest.ssa.bit_size == 64) + size *= 2; + RegType type = RegType::sgpr; + switch(intrinsic->intrinsic) { + case nir_intrinsic_load_push_constant: + case nir_intrinsic_load_work_group_id: + case nir_intrinsic_load_num_work_groups: + case nir_intrinsic_load_subgroup_id: + case nir_intrinsic_load_num_subgroups: + case nir_intrinsic_load_first_vertex: + case nir_intrinsic_load_base_instance: + case nir_intrinsic_get_buffer_size: + case nir_intrinsic_vote_all: + case nir_intrinsic_vote_any: + case nir_intrinsic_read_first_invocation: + case nir_intrinsic_read_invocation: + case nir_intrinsic_first_invocation: + case nir_intrinsic_vulkan_resource_index: + type = RegType::sgpr; + break; + case nir_intrinsic_ballot: + type = RegType::sgpr; + size = 2; + break; + case nir_intrinsic_load_sample_id: + case nir_intrinsic_load_sample_mask_in: + case nir_intrinsic_load_input: + case nir_intrinsic_load_vertex_id: + case nir_intrinsic_load_vertex_id_zero_base: + case nir_intrinsic_load_barycentric_sample: + case nir_intrinsic_load_barycentric_pixel: + case nir_intrinsic_load_barycentric_centroid: + case nir_intrinsic_load_barycentric_at_sample: + case nir_intrinsic_load_barycentric_at_offset: + case nir_intrinsic_load_interpolated_input: + case nir_intrinsic_load_frag_coord: + case nir_intrinsic_load_sample_pos: + case nir_intrinsic_load_layer_id: + case nir_intrinsic_load_local_invocation_id: + case nir_intrinsic_load_local_invocation_index: + case nir_intrinsic_load_subgroup_invocation: + case nir_intrinsic_write_invocation_amd: + case nir_intrinsic_mbcnt_amd: + case nir_intrinsic_load_instance_id: + case nir_intrinsic_ssbo_atomic_add: + case nir_intrinsic_ssbo_atomic_imin: + case nir_intrinsic_ssbo_atomic_umin: + case nir_intrinsic_ssbo_atomic_imax: + case nir_intrinsic_ssbo_atomic_umax: + case nir_intrinsic_ssbo_atomic_and: + case nir_intrinsic_ssbo_atomic_or: + case nir_intrinsic_ssbo_atomic_xor: + case nir_intrinsic_ssbo_atomic_exchange: + case nir_intrinsic_ssbo_atomic_comp_swap: + case nir_intrinsic_image_deref_atomic_add: + case nir_intrinsic_image_deref_atomic_umin: + case nir_intrinsic_image_deref_atomic_imin: + case nir_intrinsic_image_deref_atomic_umax: + case nir_intrinsic_image_deref_atomic_imax: + case nir_intrinsic_image_deref_atomic_and: + case nir_intrinsic_image_deref_atomic_or: + case nir_intrinsic_image_deref_atomic_xor: + case nir_intrinsic_image_deref_atomic_exchange: + case nir_intrinsic_image_deref_atomic_comp_swap: + case nir_intrinsic_image_deref_size: + case nir_intrinsic_shared_atomic_add: + case nir_intrinsic_shared_atomic_imin: + case nir_intrinsic_shared_atomic_umin: + case nir_intrinsic_shared_atomic_imax: + case nir_intrinsic_shared_atomic_umax: + case nir_intrinsic_shared_atomic_and: + case nir_intrinsic_shared_atomic_or: + case nir_intrinsic_shared_atomic_xor: + case nir_intrinsic_shared_atomic_exchange: + case nir_intrinsic_shared_atomic_comp_swap: + case nir_intrinsic_load_scratch: + type = RegType::vgpr; + break; + case nir_intrinsic_shuffle: + case nir_intrinsic_quad_broadcast: + case nir_intrinsic_quad_swap_horizontal: + case nir_intrinsic_quad_swap_vertical: + case nir_intrinsic_quad_swap_diagonal: + case nir_intrinsic_quad_swizzle_amd: + case nir_intrinsic_masked_swizzle_amd: + case nir_intrinsic_inclusive_scan: + case nir_intrinsic_exclusive_scan: + if (!ctx->divergent_vals[intrinsic->dest.ssa.index]) { + type = RegType::sgpr; + } else if (intrinsic->src[0].ssa->bit_size == 1) { + type = RegType::sgpr; + size = 2; + } else { + type = RegType::vgpr; + } + break; + case nir_intrinsic_load_view_index: + type = ctx->stage == fragment_fs ? RegType::vgpr : RegType::sgpr; + break; + case nir_intrinsic_load_front_face: + case nir_intrinsic_load_helper_invocation: + case nir_intrinsic_is_helper_invocation: + type = RegType::sgpr; + size = 2; + break; + case nir_intrinsic_reduce: + if (nir_intrinsic_cluster_size(intrinsic) == 0 || + !ctx->divergent_vals[intrinsic->dest.ssa.index]) { + type = RegType::sgpr; + } else if (intrinsic->src[0].ssa->bit_size == 1) { + type = RegType::sgpr; + size = 2; + } else { + type = RegType::vgpr; + } + break; + case nir_intrinsic_load_ubo: + case nir_intrinsic_load_ssbo: + case nir_intrinsic_load_global: + type = ctx->divergent_vals[intrinsic->dest.ssa.index] ? RegType::vgpr : RegType::sgpr; + break; + /* due to copy propagation, the swizzled imov is removed if num dest components == 1 */ + case nir_intrinsic_load_shared: + if (ctx->divergent_vals[intrinsic->dest.ssa.index]) + type = RegType::vgpr; + else + type = RegType::sgpr; + break; + default: + for (unsigned i = 0; i < nir_intrinsic_infos[intrinsic->intrinsic].num_srcs; i++) { + if (allocated[intrinsic->src[i].ssa->index].type() == RegType::vgpr) + type = RegType::vgpr; + } + break; + } + allocated[intrinsic->dest.ssa.index] = Temp(0, RegClass(type, size)); + + switch(intrinsic->intrinsic) { + case nir_intrinsic_load_barycentric_sample: + case nir_intrinsic_load_barycentric_pixel: + case nir_intrinsic_load_barycentric_centroid: + case nir_intrinsic_load_barycentric_at_sample: + case nir_intrinsic_load_barycentric_at_offset: { + glsl_interp_mode mode = (glsl_interp_mode)nir_intrinsic_interp_mode(intrinsic); + ctx->fs_vgpr_args[get_interp_input(intrinsic->intrinsic, mode)] = true; + break; + } + case nir_intrinsic_load_front_face: + ctx->fs_vgpr_args[fs_input::front_face] = true; + break; + case nir_intrinsic_load_frag_coord: + case nir_intrinsic_load_sample_pos: { + uint8_t mask = nir_ssa_def_components_read(&intrinsic->dest.ssa); + for (unsigned i = 0; i < 4; i++) { + if (mask & (1 << i)) + ctx->fs_vgpr_args[fs_input::frag_pos_0 + i] = true; + + } + break; + } + case nir_intrinsic_load_sample_id: + ctx->fs_vgpr_args[fs_input::ancillary] = true; + break; + case nir_intrinsic_load_sample_mask_in: + ctx->fs_vgpr_args[fs_input::ancillary] = true; + ctx->fs_vgpr_args[fs_input::sample_coverage] = true; + break; + default: + break; + } + break; + } + case nir_instr_type_tex: { + nir_tex_instr* tex = nir_instr_as_tex(instr); + unsigned size = tex->dest.ssa.num_components; + + if (tex->dest.ssa.bit_size == 64) + size *= 2; + if (tex->op == nir_texop_texture_samples) + assert(!ctx->divergent_vals[tex->dest.ssa.index]); + if (ctx->divergent_vals[tex->dest.ssa.index]) + allocated[tex->dest.ssa.index] = Temp(0, RegClass(RegType::vgpr, size)); + else + allocated[tex->dest.ssa.index] = Temp(0, RegClass(RegType::sgpr, size)); + break; + } + case nir_instr_type_parallel_copy: { + nir_foreach_parallel_copy_entry(entry, nir_instr_as_parallel_copy(instr)) { + allocated[entry->dest.ssa.index] = allocated[entry->src.ssa->index]; + } + break; + } + case nir_instr_type_ssa_undef: { + unsigned size = nir_instr_as_ssa_undef(instr)->def.num_components; + if (nir_instr_as_ssa_undef(instr)->def.bit_size == 64) + size *= 2; + allocated[nir_instr_as_ssa_undef(instr)->def.index] = Temp(0, RegClass(RegType::sgpr, size)); + break; + } + case nir_instr_type_phi: { + nir_phi_instr* phi = nir_instr_as_phi(instr); + RegType type; + unsigned size = phi->dest.ssa.num_components; + + if (phi->dest.ssa.bit_size == 1) { + assert(size == 1 && "multiple components not yet supported on boolean phis."); + type = RegType::sgpr; + size *= ctx->divergent_vals[phi->dest.ssa.index] ? 2 : 1; + allocated[phi->dest.ssa.index] = Temp(0, RegClass(type, size)); + break; + } + + if (ctx->divergent_vals[phi->dest.ssa.index]) { + type = RegType::vgpr; + } else { + type = RegType::sgpr; + nir_foreach_phi_src (src, phi) { + if (allocated[src->src.ssa->index].type() == RegType::vgpr) + type = RegType::vgpr; + if (allocated[src->src.ssa->index].type() == RegType::none) + done = false; + } + } + + size *= phi->dest.ssa.bit_size == 64 ? 2 : 1; + RegClass rc = RegClass(type, size); + if (rc != allocated[phi->dest.ssa.index].regClass()) { + done = false; + } else { + nir_foreach_phi_src(src, phi) + assert(allocated[src->src.ssa->index].size() == rc.size()); + } + allocated[phi->dest.ssa.index] = Temp(0, rc); + break; + } + default: + break; + } + } + } + } + + for (unsigned i = 0; i < impl->ssa_alloc; i++) + allocated[i] = Temp(ctx->program->allocateId(), allocated[i].regClass()); + + ctx->allocated.reset(allocated.release()); +} + +struct user_sgpr_info { + uint8_t num_sgpr; + uint8_t remaining_sgprs; + uint8_t user_sgpr_idx; + bool need_ring_offsets; + bool indirect_all_descriptor_sets; +}; + +static void allocate_inline_push_consts(isel_context *ctx, + user_sgpr_info& user_sgpr_info) +{ + uint8_t remaining_sgprs = user_sgpr_info.remaining_sgprs; + + /* Only supported if shaders use push constants. */ + if (ctx->program->info->min_push_constant_used == UINT8_MAX) + return; + + /* Only supported if shaders don't have indirect push constants. */ + if (ctx->program->info->has_indirect_push_constants) + return; + + /* Only supported for 32-bit push constants. */ + //TODO: it's possible that some day, the load/store vectorization could make this inaccurate + if (!ctx->program->info->has_only_32bit_push_constants) + return; + + uint8_t num_push_consts = + (ctx->program->info->max_push_constant_used - + ctx->program->info->min_push_constant_used) / 4; + + /* Check if the number of user SGPRs is large enough. */ + if (num_push_consts < remaining_sgprs) { + ctx->program->info->num_inline_push_consts = num_push_consts; + } else { + ctx->program->info->num_inline_push_consts = remaining_sgprs; + } + + /* Clamp to the maximum number of allowed inlined push constants. */ + if (ctx->program->info->num_inline_push_consts > MAX_INLINE_PUSH_CONSTS) + ctx->program->info->num_inline_push_consts = MAX_INLINE_PUSH_CONSTS; + + if (ctx->program->info->num_inline_push_consts == num_push_consts && + !ctx->program->info->loads_dynamic_offsets) { + /* Disable the default push constants path if all constants are + * inlined and if shaders don't use dynamic descriptors. + */ + ctx->program->info->loads_push_constants = false; + user_sgpr_info.num_sgpr--; + user_sgpr_info.remaining_sgprs++; + } + + ctx->program->info->base_inline_push_consts = + ctx->program->info->min_push_constant_used / 4; + + user_sgpr_info.num_sgpr += ctx->program->info->num_inline_push_consts; + user_sgpr_info.remaining_sgprs -= ctx->program->info->num_inline_push_consts; +} + +static void allocate_user_sgprs(isel_context *ctx, + bool needs_view_index, user_sgpr_info& user_sgpr_info) +{ + memset(&user_sgpr_info, 0, sizeof(struct user_sgpr_info)); + uint32_t user_sgpr_count = 0; + + /* until we sort out scratch/global buffers always assign ring offsets for gs/vs/es */ + if (ctx->stage != fragment_fs && + ctx->stage != compute_cs + /*|| ctx->is_gs_copy_shader */) + user_sgpr_info.need_ring_offsets = true; + + if (ctx->stage == fragment_fs && + ctx->program->info->ps.needs_sample_positions) + user_sgpr_info.need_ring_offsets = true; + + /* 2 user sgprs will nearly always be allocated for scratch/rings */ + if (ctx->options->supports_spill || user_sgpr_info.need_ring_offsets || ctx->scratch_enabled) + user_sgpr_count += 2; + + switch (ctx->stage) { + case vertex_vs: + /* if (!ctx->is_gs_copy_shader) */ { + if (ctx->program->info->vs.has_vertex_buffers) + user_sgpr_count++; + user_sgpr_count += ctx->program->info->vs.needs_draw_id ? 3 : 2; + } + break; + case fragment_fs: + //user_sgpr_count += ctx->program->info->ps.needs_sample_positions; + break; + case compute_cs: + if (ctx->program->info->cs.uses_grid_size) + user_sgpr_count += 3; + break; + default: + unreachable("Shader stage not implemented"); + } + + if (needs_view_index) + user_sgpr_count++; + + if (ctx->program->info->loads_push_constants) + user_sgpr_count += 1; /* we use 32bit pointers */ + + if (ctx->program->info->so.num_outputs) + user_sgpr_count += 1; /* we use 32bit pointers */ + + uint32_t available_sgprs = ctx->options->chip_class >= GFX9 && !(ctx->stage & hw_cs) ? 32 : 16; + uint32_t remaining_sgprs = available_sgprs - user_sgpr_count; + uint32_t num_desc_set = util_bitcount(ctx->program->info->desc_set_used_mask); + + if (available_sgprs < user_sgpr_count + num_desc_set) { + user_sgpr_info.indirect_all_descriptor_sets = true; + user_sgpr_info.num_sgpr = user_sgpr_count + 1; + user_sgpr_info.remaining_sgprs = remaining_sgprs - 1; + } else { + user_sgpr_info.num_sgpr = user_sgpr_count + num_desc_set; + user_sgpr_info.remaining_sgprs = remaining_sgprs - num_desc_set; + } + + allocate_inline_push_consts(ctx, user_sgpr_info); +} + +#define MAX_ARGS 64 +struct arg_info { + RegClass types[MAX_ARGS]; + Temp *assign[MAX_ARGS]; + PhysReg reg[MAX_ARGS]; + unsigned array_params_mask; + uint8_t count; + uint8_t sgpr_count; + uint8_t num_sgprs_used; + uint8_t num_vgprs_used; +}; + +static void +add_arg(arg_info *info, RegClass rc, Temp *param_ptr, unsigned reg) +{ + assert(info->count < MAX_ARGS); + + info->assign[info->count] = param_ptr; + info->types[info->count] = rc; + + if (rc.type() == RegType::sgpr) { + info->num_sgprs_used += rc.size(); + info->sgpr_count++; + info->reg[info->count] = PhysReg{reg}; + } else { + assert(rc.type() == RegType::vgpr); + info->num_vgprs_used += rc.size(); + info->reg[info->count] = PhysReg{reg + 256}; + } + info->count++; +} + +static void +set_loc(struct radv_userdata_info *ud_info, uint8_t *sgpr_idx, uint8_t num_sgprs) +{ + ud_info->sgpr_idx = *sgpr_idx; + ud_info->num_sgprs = num_sgprs; + *sgpr_idx += num_sgprs; +} + +static void +set_loc_shader(isel_context *ctx, int idx, uint8_t *sgpr_idx, + uint8_t num_sgprs) +{ + struct radv_userdata_info *ud_info = &ctx->program->info->user_sgprs_locs.shader_data[idx]; + assert(ud_info); + + set_loc(ud_info, sgpr_idx, num_sgprs); +} + +static void +set_loc_shader_ptr(isel_context *ctx, int idx, uint8_t *sgpr_idx) +{ + bool use_32bit_pointers = idx != AC_UD_SCRATCH_RING_OFFSETS; + + set_loc_shader(ctx, idx, sgpr_idx, use_32bit_pointers ? 1 : 2); +} + +static void +set_loc_desc(isel_context *ctx, int idx, uint8_t *sgpr_idx) +{ + struct radv_userdata_locations *locs = &ctx->program->info->user_sgprs_locs; + struct radv_userdata_info *ud_info = &locs->descriptor_sets[idx]; + assert(ud_info); + + set_loc(ud_info, sgpr_idx, 1); + locs->descriptor_sets_enabled |= 1 << idx; +} + +static void +declare_global_input_sgprs(isel_context *ctx, + /* bool has_previous_stage, gl_shader_stage previous_stage, */ + user_sgpr_info *user_sgpr_info, + struct arg_info *args, + Temp *desc_sets) +{ + /* 1 for each descriptor set */ + if (!user_sgpr_info->indirect_all_descriptor_sets) { + uint32_t mask = ctx->program->info->desc_set_used_mask; + while (mask) { + int i = u_bit_scan(&mask); + add_arg(args, s1, &desc_sets[i], user_sgpr_info->user_sgpr_idx); + set_loc_desc(ctx, i, &user_sgpr_info->user_sgpr_idx); + } + /* NIR->LLVM might have set this to true if RADV_DEBUG=compiletime */ + ctx->program->info->need_indirect_descriptor_sets = false; + } else { + add_arg(args, s1, desc_sets, user_sgpr_info->user_sgpr_idx); + set_loc_shader_ptr(ctx, AC_UD_INDIRECT_DESCRIPTOR_SETS, &user_sgpr_info->user_sgpr_idx); + ctx->program->info->need_indirect_descriptor_sets = true; + } + + if (ctx->program->info->loads_push_constants) { + /* 1 for push constants and dynamic descriptors */ + add_arg(args, s1, &ctx->push_constants, user_sgpr_info->user_sgpr_idx); + set_loc_shader_ptr(ctx, AC_UD_PUSH_CONSTANTS, &user_sgpr_info->user_sgpr_idx); + } + + if (ctx->program->info->num_inline_push_consts) { + unsigned count = ctx->program->info->num_inline_push_consts; + for (unsigned i = 0; i < count; i++) + add_arg(args, s1, &ctx->inline_push_consts[i], user_sgpr_info->user_sgpr_idx + i); + set_loc_shader(ctx, AC_UD_INLINE_PUSH_CONSTANTS, &user_sgpr_info->user_sgpr_idx, count); + + ctx->num_inline_push_consts = ctx->program->info->num_inline_push_consts; + ctx->base_inline_push_consts = ctx->program->info->base_inline_push_consts; + } + + if (ctx->program->info->so.num_outputs) { + add_arg(args, s1, &ctx->streamout_buffers, user_sgpr_info->user_sgpr_idx); + set_loc_shader_ptr(ctx, AC_UD_STREAMOUT_BUFFERS, &user_sgpr_info->user_sgpr_idx); + } +} + +static void +declare_vs_input_vgprs(isel_context *ctx, struct arg_info *args) +{ + unsigned vgpr_idx = 0; + add_arg(args, v1, &ctx->vertex_id, vgpr_idx++); +/* if (!ctx->is_gs_copy_shader) */ { + if (ctx->options->key.vs.out.as_ls) { + add_arg(args, v1, &ctx->rel_auto_id, vgpr_idx++); + add_arg(args, v1, &ctx->instance_id, vgpr_idx++); + } else { + add_arg(args, v1, &ctx->instance_id, vgpr_idx++); + add_arg(args, v1, &ctx->vs_prim_id, vgpr_idx++); + } + add_arg(args, v1, NULL, vgpr_idx); /* unused */ + } +} + +static void +declare_streamout_sgprs(isel_context *ctx, struct arg_info *args, unsigned *idx) +{ + /* Streamout SGPRs. */ + if (ctx->program->info->so.num_outputs) { + assert(ctx->stage & hw_vs); + + if (ctx->stage != tess_eval_vs) { + add_arg(args, s1, &ctx->streamout_config, (*idx)++); + } else { + args->assign[args->count - 1] = &ctx->streamout_config; + args->types[args->count - 1] = s1; + } + + add_arg(args, s1, &ctx->streamout_write_idx, (*idx)++); + } + + /* A streamout buffer offset is loaded if the stride is non-zero. */ + for (unsigned i = 0; i < 4; i++) { + if (!ctx->program->info->so.strides[i]) + continue; + + add_arg(args, s1, &ctx->streamout_offset[i], (*idx)++); + } +} + +static bool needs_view_index_sgpr(isel_context *ctx) +{ + switch (ctx->stage) { + case vertex_vs: + return ctx->program->info->needs_multiview_view_index || ctx->options->key.has_multiview_view_index; + case tess_eval_vs: + return ctx->program->info->needs_multiview_view_index && ctx->options->key.has_multiview_view_index; + case vertex_ls: + case vertex_tess_control_ls: + case vertex_geometry_es: + case tess_control_hs: + case tess_eval_es: + case tess_eval_geometry_es: + case geometry_gs: + return ctx->program->info->needs_multiview_view_index; + default: + return false; + } +} + +static inline bool +add_fs_arg(isel_context *ctx, arg_info *args, unsigned &vgpr_idx, fs_input input, unsigned value, bool enable_next = false, RegClass rc = v1) +{ + if (!ctx->fs_vgpr_args[input]) + return false; + + add_arg(args, rc, &ctx->fs_inputs[input], vgpr_idx); + vgpr_idx += rc.size(); + + if (enable_next) { + add_arg(args, rc, &ctx->fs_inputs[input + 1], vgpr_idx); + vgpr_idx += rc.size(); + } + + ctx->program->config->spi_ps_input_addr |= value; + ctx->program->config->spi_ps_input_ena |= value; + return true; +} + +void add_startpgm(struct isel_context *ctx) +{ + user_sgpr_info user_sgpr_info; + bool needs_view_index = needs_view_index_sgpr(ctx); + allocate_user_sgprs(ctx, needs_view_index, user_sgpr_info); + arg_info args = {}; + + /* this needs to be in sgprs 0 and 1 */ + if (ctx->options->supports_spill || user_sgpr_info.need_ring_offsets || ctx->scratch_enabled) { + add_arg(&args, s2, &ctx->private_segment_buffer, 0); + set_loc_shader_ptr(ctx, AC_UD_SCRATCH_RING_OFFSETS, &user_sgpr_info.user_sgpr_idx); + } + + unsigned vgpr_idx = 0; + switch (ctx->stage) { + case vertex_vs: { + declare_global_input_sgprs(ctx, &user_sgpr_info, &args, ctx->descriptor_sets); + if (ctx->program->info->vs.has_vertex_buffers) { + add_arg(&args, s1, &ctx->vertex_buffers, user_sgpr_info.user_sgpr_idx); + set_loc_shader_ptr(ctx, AC_UD_VS_VERTEX_BUFFERS, &user_sgpr_info.user_sgpr_idx); + } + add_arg(&args, s1, &ctx->base_vertex, user_sgpr_info.user_sgpr_idx); + add_arg(&args, s1, &ctx->start_instance, user_sgpr_info.user_sgpr_idx + 1); + if (ctx->program->info->vs.needs_draw_id) { + add_arg(&args, s1, &ctx->draw_id, user_sgpr_info.user_sgpr_idx + 2); + set_loc_shader(ctx, AC_UD_VS_BASE_VERTEX_START_INSTANCE, &user_sgpr_info.user_sgpr_idx, 3); + } else + set_loc_shader(ctx, AC_UD_VS_BASE_VERTEX_START_INSTANCE, &user_sgpr_info.user_sgpr_idx, 2); + + if (needs_view_index) { + add_arg(&args, s1, &ctx->view_index, user_sgpr_info.user_sgpr_idx); + set_loc_shader(ctx, AC_UD_VIEW_INDEX, &user_sgpr_info.user_sgpr_idx, 1); + } + + assert(user_sgpr_info.user_sgpr_idx == user_sgpr_info.num_sgpr); + unsigned idx = user_sgpr_info.user_sgpr_idx; + if (ctx->options->key.vs.out.as_es) + add_arg(&args, s1, &ctx->es2gs_offset, idx++); + else + declare_streamout_sgprs(ctx, &args, &idx); + + if (ctx->scratch_enabled) + add_arg(&args, s1, &ctx->scratch_offset, idx++); + + declare_vs_input_vgprs(ctx, &args); + break; + } + case fragment_fs: { + declare_global_input_sgprs(ctx, &user_sgpr_info, &args, ctx->descriptor_sets); + + assert(user_sgpr_info.user_sgpr_idx == user_sgpr_info.num_sgpr); + add_arg(&args, s1, &ctx->prim_mask, user_sgpr_info.user_sgpr_idx); + + if (ctx->scratch_enabled) + add_arg(&args, s1, &ctx->scratch_offset, user_sgpr_info.user_sgpr_idx + 1); + + ctx->program->config->spi_ps_input_addr = 0; + ctx->program->config->spi_ps_input_ena = 0; + + bool has_interp_mode = false; + + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::persp_sample_p1, S_0286CC_PERSP_SAMPLE_ENA(1), true); + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::persp_center_p1, S_0286CC_PERSP_CENTER_ENA(1), true); + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::persp_centroid_p1, S_0286CC_PERSP_CENTROID_ENA(1), true); + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::persp_pull_model, S_0286CC_PERSP_PULL_MODEL_ENA(1), false, v3); + + if (!has_interp_mode && ctx->fs_vgpr_args[fs_input::frag_pos_3]) { + /* If POS_W_FLOAT (11) is enabled, at least one of PERSP_* must be enabled too */ + ctx->fs_vgpr_args[fs_input::persp_center_p1] = true; + has_interp_mode = add_fs_arg(ctx, &args, vgpr_idx, fs_input::persp_center_p1, S_0286CC_PERSP_CENTER_ENA(1), true); + } + + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::linear_sample_p1, S_0286CC_LINEAR_SAMPLE_ENA(1), true); + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::linear_center_p1, S_0286CC_LINEAR_CENTER_ENA(1), true); + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::linear_centroid_p1, S_0286CC_LINEAR_CENTROID_ENA(1), true); + has_interp_mode |= add_fs_arg(ctx, &args, vgpr_idx, fs_input::line_stipple, S_0286CC_LINE_STIPPLE_TEX_ENA(1)); + + if (!has_interp_mode) { + /* At least one of PERSP_* (0xF) or LINEAR_* (0x70) must be enabled */ + ctx->fs_vgpr_args[fs_input::persp_center_p1] = true; + has_interp_mode = add_fs_arg(ctx, &args, vgpr_idx, fs_input::persp_center_p1, S_0286CC_PERSP_CENTER_ENA(1), true); + } + + add_fs_arg(ctx, &args, vgpr_idx, fs_input::frag_pos_0, S_0286CC_POS_X_FLOAT_ENA(1)); + add_fs_arg(ctx, &args, vgpr_idx, fs_input::frag_pos_1, S_0286CC_POS_Y_FLOAT_ENA(1)); + add_fs_arg(ctx, &args, vgpr_idx, fs_input::frag_pos_2, S_0286CC_POS_Z_FLOAT_ENA(1)); + add_fs_arg(ctx, &args, vgpr_idx, fs_input::frag_pos_3, S_0286CC_POS_W_FLOAT_ENA(1)); + + add_fs_arg(ctx, &args, vgpr_idx, fs_input::front_face, S_0286CC_FRONT_FACE_ENA(1)); + add_fs_arg(ctx, &args, vgpr_idx, fs_input::ancillary, S_0286CC_ANCILLARY_ENA(1)); + add_fs_arg(ctx, &args, vgpr_idx, fs_input::sample_coverage, S_0286CC_SAMPLE_COVERAGE_ENA(1)); + add_fs_arg(ctx, &args, vgpr_idx, fs_input::fixed_pt, S_0286CC_POS_FIXED_PT_ENA(1)); + + ASSERTED bool unset_interp_mode = !(ctx->program->config->spi_ps_input_addr & 0x7F) || + (G_0286CC_POS_W_FLOAT_ENA(ctx->program->config->spi_ps_input_addr) + && !(ctx->program->config->spi_ps_input_addr & 0xF)); + + assert(has_interp_mode); + assert(!unset_interp_mode); + break; + } + case compute_cs: { + declare_global_input_sgprs(ctx, &user_sgpr_info, &args, ctx->descriptor_sets); + + if (ctx->program->info->cs.uses_grid_size) { + add_arg(&args, s1, &ctx->num_workgroups[0], user_sgpr_info.user_sgpr_idx); + add_arg(&args, s1, &ctx->num_workgroups[1], user_sgpr_info.user_sgpr_idx + 1); + add_arg(&args, s1, &ctx->num_workgroups[2], user_sgpr_info.user_sgpr_idx + 2); + set_loc_shader(ctx, AC_UD_CS_GRID_SIZE, &user_sgpr_info.user_sgpr_idx, 3); + } + assert(user_sgpr_info.user_sgpr_idx == user_sgpr_info.num_sgpr); + unsigned idx = user_sgpr_info.user_sgpr_idx; + for (unsigned i = 0; i < 3; i++) { + if (ctx->program->info->cs.uses_block_id[i]) + add_arg(&args, s1, &ctx->workgroup_ids[i], idx++); + } + + if (ctx->program->info->cs.uses_local_invocation_idx) + add_arg(&args, s1, &ctx->tg_size, idx++); + if (ctx->scratch_enabled) + add_arg(&args, s1, &ctx->scratch_offset, idx++); + + add_arg(&args, v1, &ctx->local_invocation_ids[0], vgpr_idx++); + add_arg(&args, v1, &ctx->local_invocation_ids[1], vgpr_idx++); + add_arg(&args, v1, &ctx->local_invocation_ids[2], vgpr_idx++); + break; + } + default: + unreachable("Shader stage not implemented"); + } + + ctx->program->info->num_input_vgprs = 0; + ctx->program->info->num_input_sgprs = args.num_sgprs_used; + ctx->program->info->num_user_sgprs = user_sgpr_info.num_sgpr; + ctx->program->info->num_input_vgprs = args.num_vgprs_used; + + aco_ptr startpgm{create_instruction(aco_opcode::p_startpgm, Format::PSEUDO, 0, args.count + 1)}; + for (unsigned i = 0; i < args.count; i++) { + if (args.assign[i]) { + *args.assign[i] = Temp{ctx->program->allocateId(), args.types[i]}; + startpgm->definitions[i] = Definition(*args.assign[i]); + startpgm->definitions[i].setFixed(args.reg[i]); + } + } + startpgm->definitions[args.count] = Definition{ctx->program->allocateId(), exec, s2}; + ctx->block->instructions.push_back(std::move(startpgm)); +} + +int +type_size(const struct glsl_type *type, bool bindless) +{ + // TODO: don't we need type->std430_base_alignment() here? + return glsl_count_attribute_slots(type, false); +} + +void +shared_var_info(const struct glsl_type *type, unsigned *size, unsigned *align) +{ + assert(glsl_type_is_vector_or_scalar(type)); + + uint32_t comp_size = glsl_type_is_boolean(type) + ? 4 : glsl_get_bit_size(type) / 8; + unsigned length = glsl_get_vector_elements(type); + *size = comp_size * length, + *align = comp_size; +} + +int +get_align(nir_variable_mode mode, bool is_store, unsigned bit_size, unsigned num_components) +{ + /* TODO: ACO doesn't have good support for non-32-bit reads/writes yet */ + if (bit_size != 32) + return -1; + + switch (mode) { + case nir_var_mem_ubo: + case nir_var_mem_ssbo: + //case nir_var_mem_push_const: enable with 1240! + case nir_var_mem_shared: + /* TODO: what are the alignment requirements for LDS? */ + return num_components <= 4 ? 4 : -1; + default: + return -1; + } +} + +void +setup_vs_variables(isel_context *ctx, nir_shader *nir) +{ + nir_foreach_variable(variable, &nir->inputs) + { + variable->data.driver_location = variable->data.location * 4; + } + nir_foreach_variable(variable, &nir->outputs) + { + variable->data.driver_location = variable->data.location * 4; + } + + radv_vs_output_info *outinfo = &ctx->program->info->vs.outinfo; + + memset(outinfo->vs_output_param_offset, AC_EXP_PARAM_UNDEFINED, + sizeof(outinfo->vs_output_param_offset)); + + ctx->needs_instance_id = ctx->program->info->vs.needs_instance_id; + + bool export_clip_dists = ctx->options->key.vs_common_out.export_clip_dists; + + outinfo->param_exports = 0; + int pos_written = 0x1; + if (outinfo->writes_pointsize || outinfo->writes_viewport_index || outinfo->writes_layer) + pos_written |= 1 << 1; + + nir_foreach_variable(variable, &nir->outputs) + { + int idx = variable->data.location; + unsigned slots = variable->type->count_attribute_slots(false); + if (variable->data.compact) { + unsigned component_count = variable->data.location_frac + variable->type->length; + slots = (component_count + 3) / 4; + } + + if (idx >= VARYING_SLOT_VAR0 || idx == VARYING_SLOT_LAYER || idx == VARYING_SLOT_PRIMITIVE_ID || + ((idx == VARYING_SLOT_CLIP_DIST0 || idx == VARYING_SLOT_CLIP_DIST1) && export_clip_dists)) { + for (unsigned i = 0; i < slots; i++) { + if (outinfo->vs_output_param_offset[idx + i] == AC_EXP_PARAM_UNDEFINED) + outinfo->vs_output_param_offset[idx + i] = outinfo->param_exports++; + } + } + } + if (outinfo->writes_layer && + outinfo->vs_output_param_offset[VARYING_SLOT_LAYER] == AC_EXP_PARAM_UNDEFINED) { + /* when ctx->options->key.has_multiview_view_index = true, the layer + * variable isn't declared in NIR and it's isel's job to get the layer */ + outinfo->vs_output_param_offset[VARYING_SLOT_LAYER] = outinfo->param_exports++; + } + + if (outinfo->export_prim_id) { + assert(outinfo->vs_output_param_offset[VARYING_SLOT_PRIMITIVE_ID] == AC_EXP_PARAM_UNDEFINED); + outinfo->vs_output_param_offset[VARYING_SLOT_PRIMITIVE_ID] = outinfo->param_exports++; + } + + ctx->num_clip_distances = util_bitcount(outinfo->clip_dist_mask); + ctx->num_cull_distances = util_bitcount(outinfo->cull_dist_mask); + + assert(ctx->num_clip_distances + ctx->num_cull_distances <= 8); + + if (ctx->num_clip_distances + ctx->num_cull_distances > 0) + pos_written |= 1 << 2; + if (ctx->num_clip_distances + ctx->num_cull_distances > 4) + pos_written |= 1 << 3; + + outinfo->pos_exports = util_bitcount(pos_written); +} + +void +setup_variables(isel_context *ctx, nir_shader *nir) +{ + switch (nir->info.stage) { + case MESA_SHADER_FRAGMENT: { + nir_foreach_variable(variable, &nir->outputs) + { + int idx = variable->data.location + variable->data.index; + variable->data.driver_location = idx * 4; + } + break; + } + case MESA_SHADER_COMPUTE: { + unsigned lds_allocation_size_unit = 4 * 64; + if (ctx->program->chip_class >= GFX7) + lds_allocation_size_unit = 4 * 128; + ctx->program->config->lds_size = (nir->info.cs.shared_size + lds_allocation_size_unit - 1) / lds_allocation_size_unit; + break; + } + case MESA_SHADER_VERTEX: { + setup_vs_variables(ctx, nir); + break; + } + default: + unreachable("Unhandled shader stage."); + } +} + +isel_context +setup_isel_context(Program* program, + unsigned shader_count, + struct nir_shader *const *shaders, + ac_shader_config* config, + radv_shader_info *info, + radv_nir_compiler_options *options) +{ + program->stage = 0; + for (unsigned i = 0; i < shader_count; i++) { + switch (shaders[i]->info.stage) { + case MESA_SHADER_VERTEX: + program->stage |= sw_vs; + break; + case MESA_SHADER_TESS_CTRL: + program->stage |= sw_tcs; + break; + case MESA_SHADER_TESS_EVAL: + program->stage |= sw_tes; + break; + case MESA_SHADER_GEOMETRY: + program->stage |= sw_gs; + break; + case MESA_SHADER_FRAGMENT: + program->stage |= sw_fs; + break; + case MESA_SHADER_COMPUTE: + program->stage |= sw_cs; + break; + default: + unreachable("Shader stage not implemented"); + } + } + if (program->stage == sw_vs) + program->stage |= hw_vs; + else if (program->stage == sw_fs) + program->stage |= hw_fs; + else if (program->stage == sw_cs) + program->stage |= hw_cs; + else + unreachable("Shader stage not implemented"); + + program->config = config; + program->info = info; + program->chip_class = options->chip_class; + program->family = options->family; + program->sgpr_limit = options->chip_class >= GFX8 ? 102 : 104; + if (options->family == CHIP_TONGA || options->family == CHIP_ICELAND) + program->sgpr_limit = 94; /* workaround hardware bug */ + + for (unsigned i = 0; i < MAX_SETS; ++i) + program->info->user_sgprs_locs.descriptor_sets[i].sgpr_idx = -1; + for (unsigned i = 0; i < AC_UD_MAX_UD; ++i) + program->info->user_sgprs_locs.shader_data[i].sgpr_idx = -1; + + isel_context ctx = {}; + ctx.program = program; + ctx.options = options; + ctx.stage = program->stage; + + for (unsigned i = 0; i < fs_input::max_inputs; ++i) + ctx.fs_inputs[i] = Temp(0, v1); + ctx.fs_inputs[fs_input::persp_pull_model] = Temp(0, v3); + for (unsigned i = 0; i < MAX_SETS; ++i) + ctx.descriptor_sets[i] = Temp(0, s1); + for (unsigned i = 0; i < MAX_INLINE_PUSH_CONSTS; ++i) + ctx.inline_push_consts[i] = Temp(0, s1); + for (unsigned i = 0; i <= VARYING_SLOT_VAR31; ++i) { + for (unsigned j = 0; j < 4; ++j) + ctx.vs_output.outputs[i][j] = Temp(0, v1); + } + + for (unsigned i = 0; i < shader_count; i++) { + nir_shader *nir = shaders[i]; + + /* align and copy constant data */ + while (program->constant_data.size() % 4u) + program->constant_data.push_back(0); + ctx.constant_data_offset = program->constant_data.size(); + program->constant_data.insert(program->constant_data.end(), + (uint8_t*)nir->constant_data, + (uint8_t*)nir->constant_data + nir->constant_data_size); + + /* the variable setup has to be done before lower_io / CSE */ + if (nir->info.stage == MESA_SHADER_COMPUTE) + nir_lower_vars_to_explicit_types(nir, nir_var_mem_shared, shared_var_info); + setup_variables(&ctx, nir); + + /* optimize and lower memory operations */ + bool lower_to_scalar = false; + bool lower_pack = false; + // TODO: uncomment this once !1240 is merged + /*if (nir_opt_load_store_vectorize(nir, + (nir_variable_mode)(nir_var_mem_ssbo | nir_var_mem_ubo | + nir_var_mem_push_const | nir_var_mem_shared), + get_align)) { + lower_to_scalar = true; + lower_pack = true; + }*/ + if (nir->info.stage == MESA_SHADER_COMPUTE) + lower_to_scalar |= nir_lower_explicit_io(nir, nir_var_mem_shared, nir_address_format_32bit_offset); + else + nir_lower_io(nir, (nir_variable_mode)(nir_var_shader_in | nir_var_shader_out), type_size, (nir_lower_io_options)0); + nir_lower_explicit_io(nir, nir_var_mem_global, nir_address_format_64bit_global); + + if (lower_to_scalar) + nir_lower_alu_to_scalar(nir, NULL, NULL); + if (lower_pack) + nir_lower_pack(nir); + + /* lower ALU operations */ + nir_opt_idiv_const(nir, 32); + nir_lower_idiv(nir); // TODO: use the LLVM path once !1239 is merged + + // TODO: implement logic64 in aco, it's more effective for sgprs + nir_lower_int64(nir, (nir_lower_int64_options) (nir_lower_imul64 | + nir_lower_imul_high64 | + nir_lower_imul_2x32_64 | + nir_lower_divmod64 | + nir_lower_logic64 | + nir_lower_minmax64 | + nir_lower_iabs64 | + nir_lower_ineg64)); + + /* optimize the lowered ALU operations */ + nir_copy_prop(nir); + nir_opt_constant_folding(nir); + nir_opt_algebraic(nir); + nir_opt_algebraic_late(nir); + nir_opt_constant_folding(nir); + + /* cleanup passes */ + nir_lower_load_const_to_scalar(nir); + nir_opt_cse(nir); + nir_opt_dce(nir); + nir_opt_shrink_load(nir); + nir_move_options move_opts = (nir_move_options)( + nir_move_const_undef | nir_move_load_ubo | nir_move_load_input | nir_move_comparisons); + //nir_opt_sink(nir, move_opts); // TODO: enable this once !1664 is merged + nir_opt_move(nir, move_opts); + nir_convert_to_lcssa(nir, true, false); + nir_lower_phis_to_scalar(nir); + + nir_function_impl *func = nir_shader_get_entrypoint(nir); + nir_index_ssa_defs(func); + + if (options->dump_preoptir) { + fprintf(stderr, "NIR shader before instruction selection:\n"); + nir_print_shader(nir, stderr); + } + } + + unsigned scratch_size = 0; + for (unsigned i = 0; i < shader_count; i++) + scratch_size = std::max(scratch_size, shaders[i]->scratch_size); + ctx.scratch_enabled = scratch_size > 0; + ctx.program->config->scratch_bytes_per_wave = align(scratch_size * ctx.options->wave_size, 1024); + ctx.program->config->float_mode = V_00B028_FP_64_DENORMS; + ctx.program->info->wave_size = ctx.options->wave_size; + + ctx.block = ctx.program->create_and_insert_block(); + ctx.block->loop_nest_depth = 0; + ctx.block->kind = block_kind_top_level; + + return ctx; +} + +} diff --git a/src/amd/compiler/aco_interface.cpp b/src/amd/compiler/aco_interface.cpp new file mode 100644 index 00000000000..6adb911e4b3 --- /dev/null +++ b/src/amd/compiler/aco_interface.cpp @@ -0,0 +1,166 @@ +/* + * Copyright © 2018 Google + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include "aco_interface.h" +#include "aco_ir.h" +#include "vulkan/radv_shader.h" +#include "c11/threads.h" +#include "util/debug.h" + +#include +#include + +namespace aco { +uint64_t debug_flags = 0; + +static const struct debug_control aco_debug_options[] = { + {"validateir", DEBUG_VALIDATE}, + {"validatera", DEBUG_VALIDATE_RA}, + {"perfwarn", DEBUG_PERFWARN}, + {NULL, 0} +}; + +static once_flag init_once_flag = ONCE_FLAG_INIT; + +static void init() +{ + debug_flags = parse_debug_string(getenv("ACO_DEBUG"), aco_debug_options); + + #ifndef NDEBUG + /* enable some flags by default on debug builds */ + debug_flags |= aco::DEBUG_VALIDATE; + #endif +} +} + +void aco_compile_shader(unsigned shader_count, + struct nir_shader *const *shaders, + struct radv_shader_binary **binary, + struct radv_shader_info *info, + struct radv_nir_compiler_options *options) +{ + call_once(&aco::init_once_flag, aco::init); + + ac_shader_config config = {0}; + std::unique_ptr program{new aco::Program}; + + /* Instruction Selection */ + aco::select_program(program.get(), shader_count, shaders, &config, info, options); + if (options->dump_preoptir) { + std::cerr << "After Instruction Selection:\n"; + aco_print_program(program.get(), stderr); + } + aco::validate(program.get(), stderr); + + /* Boolean phi lowering */ + aco::lower_bool_phis(program.get()); + //std::cerr << "After Boolean Phi Lowering:\n"; + //aco_print_program(program.get(), stderr); + + aco::dominator_tree(program.get()); + + /* Optimization */ + aco::value_numbering(program.get()); + aco::optimize(program.get()); + aco::validate(program.get(), stderr); + + aco::setup_reduce_temp(program.get()); + aco::insert_exec_mask(program.get()); + aco::validate(program.get(), stderr); + + aco::live live_vars = aco::live_var_analysis(program.get(), options); + aco::spill(program.get(), live_vars, options); + + //std::cerr << "Before Schedule:\n"; + //aco_print_program(program.get(), stderr); + aco::schedule_program(program.get(), live_vars); + + /* Register Allocation */ + aco::register_allocation(program.get(), live_vars.live_out); + if (options->dump_shader) { + std::cerr << "After RA:\n"; + aco_print_program(program.get(), stderr); + } + + if (aco::validate_ra(program.get(), options, stderr)) { + std::cerr << "Program after RA validation failure:\n"; + aco_print_program(program.get(), stderr); + abort(); + } + + aco::ssa_elimination(program.get()); + /* Lower to HW Instructions */ + aco::lower_to_hw_instr(program.get()); + //std::cerr << "After Eliminate Pseudo Instr:\n"; + //aco_print_program(program.get(), stderr); + + /* Insert Waitcnt */ + aco::insert_wait_states(program.get()); + aco::insert_NOPs(program.get()); + + //std::cerr << "After Insert-Waitcnt:\n"; + //aco_print_program(program.get(), stderr); + + /* Assembly */ + std::vector code; + unsigned exec_size = aco::emit_program(program.get(), code); + + bool get_disasm = options->dump_shader; +#ifndef NDEBUG + get_disasm |= options->record_llvm_ir; +#endif + + size_t size = 0; + + std::string disasm; + if (get_disasm) { + std::ostringstream stream; + aco::print_asm(program.get(), code, exec_size / 4u, options->family, stream); + stream << '\0'; + disasm = stream.str(); + size += disasm.size(); + } + + size += code.size() * sizeof(uint32_t) + sizeof(radv_shader_binary_legacy); + radv_shader_binary_legacy* legacy_binary = (radv_shader_binary_legacy*) malloc(size); + + legacy_binary->base.type = RADV_BINARY_TYPE_LEGACY; + legacy_binary->base.stage = shaders[shader_count-1]->info.stage; + legacy_binary->base.is_gs_copy_shader = false; + legacy_binary->base.total_size = size; + + memcpy(legacy_binary->data, code.data(), code.size() * sizeof(uint32_t)); + legacy_binary->exec_size = exec_size; + legacy_binary->code_size = code.size() * sizeof(uint32_t); + + legacy_binary->config = config; + legacy_binary->disasm_size = 0; + legacy_binary->llvm_ir_size = 0; + + if (get_disasm) { + disasm.copy((char*) legacy_binary->data + legacy_binary->code_size, disasm.size()); + legacy_binary->disasm_size = disasm.size() - 1; + } + + *binary = (radv_shader_binary*) legacy_binary; +} diff --git a/src/amd/compiler/aco_interface.h b/src/amd/compiler/aco_interface.h new file mode 100644 index 00000000000..1425a0997a0 --- /dev/null +++ b/src/amd/compiler/aco_interface.h @@ -0,0 +1,45 @@ +/* + * Copyright © 2018 Google + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef ACO_INTERFACE_H +#define ACO_INTERFACE_H + +#include "nir.h" + +#ifdef __cplusplus +extern "C" { +#endif + +struct ac_shader_config; + +void aco_compile_shader(unsigned shader_count, + struct nir_shader *const *shaders, + struct radv_shader_binary** binary, + struct radv_shader_info *info, + struct radv_nir_compiler_options *options); + +#ifdef __cplusplus +} +#endif + +#endif diff --git a/src/amd/compiler/aco_ir.h b/src/amd/compiler/aco_ir.h new file mode 100644 index 00000000000..663635e5b93 --- /dev/null +++ b/src/amd/compiler/aco_ir.h @@ -0,0 +1,1169 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#ifndef ACO_IR_H +#define ACO_IR_H + +#include +#include +#include +#include + +#include "nir.h" +#include "ac_binary.h" +#include "amd_family.h" +#include "aco_opcodes.h" +#include "aco_util.h" + +struct radv_nir_compiler_options; +struct radv_shader_info; + +namespace aco { + +extern uint64_t debug_flags; + +enum { + DEBUG_VALIDATE = 0x1, + DEBUG_VALIDATE_RA = 0x2, + DEBUG_PERFWARN = 0x4, +}; + +/** + * Representation of the instruction's microcode encoding format + * Note: Some Vector ALU Formats can be combined, such that: + * - VOP2* | VOP3A represents a VOP2 instruction in VOP3A encoding + * - VOP2* | DPP represents a VOP2 instruction with data parallel primitive. + * - VOP2* | SDWA represents a VOP2 instruction with sub-dword addressing. + * + * (*) The same is applicable for VOP1 and VOPC instructions. + */ +enum class Format : std::uint16_t { + /* Pseudo Instruction Format */ + PSEUDO = 0, + /* Scalar ALU & Control Formats */ + SOP1 = 1, + SOP2 = 2, + SOPK = 3, + SOPP = 4, + SOPC = 5, + /* Scalar Memory Format */ + SMEM = 6, + /* LDS/GDS Format */ + DS = 8, + /* Vector Memory Buffer Formats */ + MTBUF = 9, + MUBUF = 10, + /* Vector Memory Image Format */ + MIMG = 11, + /* Export Format */ + EXP = 12, + /* Flat Formats */ + FLAT = 13, + GLOBAL = 14, + SCRATCH = 15, + + PSEUDO_BRANCH = 16, + PSEUDO_BARRIER = 17, + PSEUDO_REDUCTION = 18, + + /* Vector ALU Formats */ + VOP1 = 1 << 8, + VOP2 = 1 << 9, + VOPC = 1 << 10, + VOP3 = 1 << 11, + VOP3A = 1 << 11, + VOP3B = 1 << 11, + VOP3P = 1 << 12, + /* Vector Parameter Interpolation Format */ + VINTRP = 1 << 13, + DPP = 1 << 14, + SDWA = 1 << 15, +}; + +enum barrier_interaction { + barrier_none = 0, + barrier_buffer = 0x1, + barrier_image = 0x2, + barrier_atomic = 0x4, + barrier_shared = 0x8, + barrier_count = 4, +}; + +constexpr Format asVOP3(Format format) { + return (Format) ((uint32_t) Format::VOP3 | (uint32_t) format); +}; + +enum class RegType { + none = 0, + sgpr, + vgpr, + linear_vgpr, +}; + +struct RegClass { + + enum RC : uint8_t { + s1 = 1, + s2 = 2, + s3 = 3, + s4 = 4, + s6 = 6, + s8 = 8, + s16 = 16, + v1 = s1 | (1 << 5), + v2 = s2 | (1 << 5), + v3 = s3 | (1 << 5), + v4 = s4 | (1 << 5), + v5 = 5 | (1 << 5), + v6 = 6 | (1 << 5), + v7 = 7 | (1 << 5), + v8 = 8 | (1 << 5), + /* these are used for WWM and spills to vgpr */ + v1_linear = v1 | (1 << 6), + v2_linear = v2 | (1 << 6), + }; + + RegClass() = default; + constexpr RegClass(RC rc) + : rc(rc) {} + constexpr RegClass(RegType type, unsigned size) + : rc((RC) ((type == RegType::vgpr ? 1 << 5 : 0) | size)) {} + + constexpr operator RC() const { return rc; } + explicit operator bool() = delete; + + constexpr RegType type() const { return rc <= RC::s16 ? RegType::sgpr : RegType::vgpr; } + constexpr unsigned size() const { return (unsigned) rc & 0x1F; } + constexpr bool is_linear() const { return rc <= RC::s16 || rc & (1 << 6); } + constexpr RegClass as_linear() const { return RegClass((RC) (rc | (1 << 6))); } + +private: + RC rc; +}; + +/* transitional helper expressions */ +static constexpr RegClass s1{RegClass::s1}; +static constexpr RegClass s2{RegClass::s2}; +static constexpr RegClass s3{RegClass::s3}; +static constexpr RegClass s4{RegClass::s4}; +static constexpr RegClass s8{RegClass::s8}; +static constexpr RegClass s16{RegClass::s16}; +static constexpr RegClass v1{RegClass::v1}; +static constexpr RegClass v2{RegClass::v2}; +static constexpr RegClass v3{RegClass::v3}; +static constexpr RegClass v4{RegClass::v4}; +static constexpr RegClass v5{RegClass::v5}; +static constexpr RegClass v6{RegClass::v6}; +static constexpr RegClass v7{RegClass::v7}; +static constexpr RegClass v8{RegClass::v8}; + +/** + * Temp Class + * Each temporary virtual register has a + * register class (i.e. size and type) + * and SSA id. + */ +struct Temp { + Temp() = default; + constexpr Temp(uint32_t id, RegClass cls) noexcept + : id_(id), reg_class(cls) {} + + constexpr uint32_t id() const noexcept { return id_; } + constexpr RegClass regClass() const noexcept { return reg_class; } + + constexpr unsigned size() const noexcept { return reg_class.size(); } + constexpr RegType type() const noexcept { return reg_class.type(); } + constexpr bool is_linear() const noexcept { return reg_class.is_linear(); } + + constexpr bool operator <(Temp other) const noexcept { return id() < other.id(); } + constexpr bool operator==(Temp other) const noexcept { return id() == other.id(); } + constexpr bool operator!=(Temp other) const noexcept { return id() != other.id(); } + +private: + uint32_t id_:24; + RegClass reg_class; +}; + +/** + * PhysReg + * Represents the physical register for each + * Operand and Definition. + */ +struct PhysReg { + constexpr PhysReg() = default; + explicit constexpr PhysReg(unsigned r) : reg(r) {} + constexpr operator unsigned() const { return reg; } + + uint16_t reg = 0; +}; + +/* helper expressions for special registers */ +static constexpr PhysReg m0{124}; +static constexpr PhysReg vcc{106}; +static constexpr PhysReg exec{126}; +static constexpr PhysReg exec_lo{126}; +static constexpr PhysReg exec_hi{127}; +static constexpr PhysReg scc{253}; + +/** + * Operand Class + * Initially, each Operand refers to either + * a temporary virtual register + * or to a constant value + * Temporary registers get mapped to physical register during RA + * Constant values are inlined into the instruction sequence. + */ +class Operand final +{ +public: + constexpr Operand() + : reg_(PhysReg{128}), isTemp_(false), isFixed_(true), isConstant_(false), + isKill_(false), isUndef_(true), isFirstKill_(false), is64BitConst_(false) {} + + explicit Operand(Temp r) noexcept + { + data_.temp = r; + if (r.id()) { + isTemp_ = true; + } else { + isUndef_ = true; + setFixed(PhysReg{128}); + } + }; + explicit Operand(uint32_t v) noexcept + { + data_.i = v; + isConstant_ = true; + if (v <= 64) + setFixed(PhysReg{128 + v}); + else if (v >= 0xFFFFFFF0) /* [-16 .. -1] */ + setFixed(PhysReg{192 - v}); + else if (v == 0x3f000000) /* 0.5 */ + setFixed(PhysReg{240}); + else if (v == 0xbf000000) /* -0.5 */ + setFixed(PhysReg{241}); + else if (v == 0x3f800000) /* 1.0 */ + setFixed(PhysReg{242}); + else if (v == 0xbf800000) /* -1.0 */ + setFixed(PhysReg{243}); + else if (v == 0x40000000) /* 2.0 */ + setFixed(PhysReg{244}); + else if (v == 0xc0000000) /* -2.0 */ + setFixed(PhysReg{245}); + else if (v == 0x40800000) /* 4.0 */ + setFixed(PhysReg{246}); + else if (v == 0xc0800000) /* -4.0 */ + setFixed(PhysReg{247}); + else if (v == 0x3e22f983) /* 1/(2*PI) */ + setFixed(PhysReg{248}); + else /* Literal Constant */ + setFixed(PhysReg{255}); + }; + explicit Operand(uint64_t v) noexcept + { + isConstant_ = true; + is64BitConst_ = true; + if (v <= 64) + setFixed(PhysReg{128 + (uint32_t) v}); + else if (v >= 0xFFFFFFFFFFFFFFF0) /* [-16 .. -1] */ + setFixed(PhysReg{192 - (uint32_t) v}); + else if (v == 0x3FE0000000000000) /* 0.5 */ + setFixed(PhysReg{240}); + else if (v == 0xBFE0000000000000) /* -0.5 */ + setFixed(PhysReg{241}); + else if (v == 0x3FF0000000000000) /* 1.0 */ + setFixed(PhysReg{242}); + else if (v == 0xBFF0000000000000) /* -1.0 */ + setFixed(PhysReg{243}); + else if (v == 0x4000000000000000) /* 2.0 */ + setFixed(PhysReg{244}); + else if (v == 0xC000000000000000) /* -2.0 */ + setFixed(PhysReg{245}); + else if (v == 0x4010000000000000) /* 4.0 */ + setFixed(PhysReg{246}); + else if (v == 0xC010000000000000) /* -4.0 */ + setFixed(PhysReg{247}); + else if (v == 0x3fc45f306dc9c882) /* 1/(2*PI) */ + setFixed(PhysReg{248}); + else { /* Literal Constant: we don't know if it is a long or double.*/ + isConstant_ = 0; + assert(false && "attempt to create a 64-bit literal constant"); + } + }; + explicit Operand(RegClass type) noexcept + { + isUndef_ = true; + data_.temp = Temp(0, type); + setFixed(PhysReg{128}); + }; + explicit Operand(PhysReg reg, RegClass type) noexcept + { + data_.temp = Temp(0, type); + setFixed(reg); + } + + constexpr bool isTemp() const noexcept + { + return isTemp_; + } + + constexpr void setTemp(Temp t) noexcept { + assert(!isConstant_); + isTemp_ = true; + data_.temp = t; + } + + constexpr Temp getTemp() const noexcept + { + return data_.temp; + } + + constexpr uint32_t tempId() const noexcept + { + return data_.temp.id(); + } + + constexpr bool hasRegClass() const noexcept + { + return isTemp() || isUndefined(); + } + + constexpr RegClass regClass() const noexcept + { + return data_.temp.regClass(); + } + + constexpr unsigned size() const noexcept + { + if (isConstant()) + return is64BitConst_ ? 2 : 1; + else + return data_.temp.size(); + } + + constexpr bool isFixed() const noexcept + { + return isFixed_; + } + + constexpr PhysReg physReg() const noexcept + { + return reg_; + } + + constexpr void setFixed(PhysReg reg) noexcept + { + isFixed_ = reg != unsigned(-1); + reg_ = reg; + } + + constexpr bool isConstant() const noexcept + { + return isConstant_; + } + + constexpr bool isLiteral() const noexcept + { + return isConstant() && reg_ == 255; + } + + constexpr bool isUndefined() const noexcept + { + return isUndef_; + } + + constexpr uint32_t constantValue() const noexcept + { + return data_.i; + } + + constexpr bool constantEquals(uint32_t cmp) const noexcept + { + return isConstant() && constantValue() == cmp; + } + + constexpr void setKill(bool flag) noexcept + { + isKill_ = flag; + if (!flag) + setFirstKill(false); + } + + constexpr bool isKill() const noexcept + { + return isKill_ || isFirstKill(); + } + + constexpr void setFirstKill(bool flag) noexcept + { + isFirstKill_ = flag; + if (flag) + setKill(flag); + } + + /* When there are multiple operands killing the same temporary, + * isFirstKill() is only returns true for the first one. */ + constexpr bool isFirstKill() const noexcept + { + return isFirstKill_; + } + +private: + union { + uint32_t i; + float f; + Temp temp = Temp(0, s1); + } data_; + PhysReg reg_; + union { + struct { + uint8_t isTemp_:1; + uint8_t isFixed_:1; + uint8_t isConstant_:1; + uint8_t isKill_:1; + uint8_t isUndef_:1; + uint8_t isFirstKill_:1; + uint8_t is64BitConst_:1; + }; + /* can't initialize bit-fields in c++11, so work around using a union */ + uint8_t control_ = 0; + }; +}; + +/** + * Definition Class + * Definitions are the results of Instructions + * and refer to temporary virtual registers + * which are later mapped to physical registers + */ +class Definition final +{ +public: + constexpr Definition() : temp(Temp(0, s1)), reg_(0), isFixed_(0), hasHint_(0), isKill_(0) {} + Definition(uint32_t index, RegClass type) noexcept + : temp(index, type) {} + explicit Definition(Temp tmp) noexcept + : temp(tmp) {} + Definition(PhysReg reg, RegClass type) noexcept + : temp(Temp(0, type)) + { + setFixed(reg); + } + Definition(uint32_t tmpId, PhysReg reg, RegClass type) noexcept + : temp(Temp(tmpId, type)) + { + setFixed(reg); + } + + constexpr bool isTemp() const noexcept + { + return tempId() > 0; + } + + constexpr Temp getTemp() const noexcept + { + return temp; + } + + constexpr uint32_t tempId() const noexcept + { + return temp.id(); + } + + constexpr void setTemp(Temp t) noexcept { + temp = t; + } + + constexpr RegClass regClass() const noexcept + { + return temp.regClass(); + } + + constexpr unsigned size() const noexcept + { + return temp.size(); + } + + constexpr bool isFixed() const noexcept + { + return isFixed_; + } + + constexpr PhysReg physReg() const noexcept + { + return reg_; + } + + constexpr void setFixed(PhysReg reg) noexcept + { + isFixed_ = 1; + reg_ = reg; + } + + constexpr void setHint(PhysReg reg) noexcept + { + hasHint_ = 1; + reg_ = reg; + } + + constexpr bool hasHint() const noexcept + { + return hasHint_; + } + + constexpr void setKill(bool flag) noexcept + { + isKill_ = flag; + } + + constexpr bool isKill() const noexcept + { + return isKill_; + } + +private: + Temp temp = Temp(0, s1); + PhysReg reg_; + union { + struct { + uint8_t isFixed_:1; + uint8_t hasHint_:1; + uint8_t isKill_:1; + }; + /* can't initialize bit-fields in c++11, so work around using a union */ + uint8_t control_ = 0; + }; +}; + +class Block; + +struct Instruction { + aco_opcode opcode; + Format format; + + aco::span operands; + aco::span definitions; + + constexpr bool isVALU() const noexcept + { + return ((uint16_t) format & (uint16_t) Format::VOP1) == (uint16_t) Format::VOP1 + || ((uint16_t) format & (uint16_t) Format::VOP2) == (uint16_t) Format::VOP2 + || ((uint16_t) format & (uint16_t) Format::VOPC) == (uint16_t) Format::VOPC + || ((uint16_t) format & (uint16_t) Format::VOP3A) == (uint16_t) Format::VOP3A + || ((uint16_t) format & (uint16_t) Format::VOP3B) == (uint16_t) Format::VOP3B + || ((uint16_t) format & (uint16_t) Format::VOP3P) == (uint16_t) Format::VOP3P; + } + + constexpr bool isSALU() const noexcept + { + return format == Format::SOP1 || + format == Format::SOP2 || + format == Format::SOPC || + format == Format::SOPK || + format == Format::SOPP; + } + + constexpr bool isVMEM() const noexcept + { + return format == Format::MTBUF || + format == Format::MUBUF || + format == Format::MIMG; + } + + constexpr bool isDPP() const noexcept + { + return (uint16_t) format & (uint16_t) Format::DPP; + } + + constexpr bool isVOP3() const noexcept + { + return ((uint16_t) format & (uint16_t) Format::VOP3A) || + ((uint16_t) format & (uint16_t) Format::VOP3B) || + format == Format::VOP3P; + } + + constexpr bool isSDWA() const noexcept + { + return (uint16_t) format & (uint16_t) Format::SDWA; + } + + constexpr bool isFlatOrGlobal() const noexcept + { + return format == Format::FLAT || format == Format::GLOBAL; + } +}; + +struct SOPK_instruction : public Instruction { + uint16_t imm; +}; + +struct SOPP_instruction : public Instruction { + uint32_t imm; + int block; +}; + +struct SOPC_instruction : public Instruction { +}; + +struct SOP1_instruction : public Instruction { +}; + +struct SOP2_instruction : public Instruction { +}; + +/** + * Scalar Memory Format: + * For s_(buffer_)load_dword*: + * Operand(0): SBASE - SGPR-pair which provides base address + * Operand(1): Offset - immediate (un)signed offset or SGPR + * Operand(2) / Definition(0): SDATA - SGPR for read / write result + * Operand(n-1): SOffset - SGPR offset (Vega only) + * + * Having no operands is also valid for instructions such as s_dcache_inv. + * + */ +struct SMEM_instruction : public Instruction { + bool glc; /* VI+: globally coherent */ + bool dlc; /* NAVI: device level coherent */ + bool nv; /* VEGA only: Non-volatile */ + bool can_reorder; + bool disable_wqm; + barrier_interaction barrier; +}; + +struct VOP1_instruction : public Instruction { +}; + +struct VOP2_instruction : public Instruction { +}; + +struct VOPC_instruction : public Instruction { +}; + +struct VOP3A_instruction : public Instruction { + bool abs[3]; + bool opsel[3]; + bool clamp; + unsigned omod; + bool neg[3]; +}; + +/** + * Data Parallel Primitives Format: + * This format can be used for VOP1, VOP2 or VOPC instructions. + * The swizzle applies to the src0 operand. + * + */ +struct DPP_instruction : public Instruction { + uint16_t dpp_ctrl; + uint8_t row_mask; + uint8_t bank_mask; + bool abs[2]; + bool neg[2]; + bool bound_ctrl; +}; + +struct Interp_instruction : public Instruction { + unsigned attribute; + unsigned component; +}; + +/** + * Local and Global Data Sharing instructions + * Operand(0): ADDR - VGPR which supplies the address. + * Operand(1): DATA0 - First data VGPR. + * Operand(2): DATA1 - Second data VGPR. + * Operand(n-1): M0 - LDS size. + * Definition(0): VDST - Destination VGPR when results returned to VGPRs. + * + */ +struct DS_instruction : public Instruction { + int16_t offset0; + int8_t offset1; + bool gds; +}; + +/** + * Vector Memory Untyped-buffer Instructions + * Operand(0): VADDR - Address source. Can carry an index and/or offset + * Operand(1): SRSRC - Specifies which SGPR supplies T# (resource constant) + * Operand(2): SOFFSET - SGPR to supply unsigned byte offset. (SGPR, M0, or inline constant) + * Operand(3) / Definition(0): VDATA - Vector GPR for write result / read data + * + */ +struct MUBUF_instruction : public Instruction { + unsigned offset; /* Unsigned byte offset - 12 bit */ + bool offen; /* Supply an offset from VGPR (VADDR) */ + bool idxen; /* Supply an index from VGPR (VADDR) */ + bool glc; /* globally coherent */ + bool dlc; /* NAVI: device level coherent */ + bool slc; /* system level coherent */ + bool tfe; /* texture fail enable */ + bool lds; /* Return read-data to LDS instead of VGPRs */ + bool disable_wqm; /* Require an exec mask without helper invocations */ + bool can_reorder; + barrier_interaction barrier; +}; + +/** + * Vector Memory Typed-buffer Instructions + * Operand(0): VADDR - Address source. Can carry an index and/or offset + * Operand(1): SRSRC - Specifies which SGPR supplies T# (resource constant) + * Operand(2): SOFFSET - SGPR to supply unsigned byte offset. (SGPR, M0, or inline constant) + * Operand(3) / Definition(0): VDATA - Vector GPR for write result / read data + * + */ +struct MTBUF_instruction : public Instruction { + union { + struct { + uint8_t dfmt : 4; /* Data Format of data in memory buffer */ + uint8_t nfmt : 3; /* Numeric format of data in memory */ + }; + uint8_t img_format; /* Buffer or image format as used by GFX10 */ + }; + unsigned offset; /* Unsigned byte offset - 12 bit */ + bool offen; /* Supply an offset from VGPR (VADDR) */ + bool idxen; /* Supply an index from VGPR (VADDR) */ + bool glc; /* globally coherent */ + bool dlc; /* NAVI: device level coherent */ + bool slc; /* system level coherent */ + bool tfe; /* texture fail enable */ + bool disable_wqm; /* Require an exec mask without helper invocations */ + bool can_reorder; + barrier_interaction barrier; +}; + +/** + * Vector Memory Image Instructions + * Operand(0): VADDR - Address source. Can carry an offset or an index. + * Operand(1): SRSRC - Scalar GPR that specifies the resource constant. + * Operand(2): SSAMP - Scalar GPR that specifies sampler constant. + * Operand(3) / Definition(0): VDATA - Vector GPR for read / write result. + * + */ +struct MIMG_instruction : public Instruction { + unsigned dmask; /* Data VGPR enable mask */ + bool unrm; /* Force address to be un-normalized */ + bool dlc; /* NAVI: device level coherent */ + bool glc; /* globally coherent */ + bool slc; /* system level coherent */ + bool tfe; /* texture fail enable */ + bool da; /* declare an array */ + bool lwe; /* Force data to be un-normalized */ + bool r128; /* NAVI: Texture resource size */ + bool a16; /* VEGA, NAVI: Address components are 16-bits */ + bool d16; /* Convert 32-bit data to 16-bit data */ + bool disable_wqm; /* Require an exec mask without helper invocations */ + bool can_reorder; + barrier_interaction barrier; +}; + +/** + * Flat/Scratch/Global Instructions + * Operand(0): ADDR + * Operand(1): SADDR + * Operand(2) / Definition(0): DATA/VDST + * + */ +struct FLAT_instruction : public Instruction { + uint16_t offset; /* Vega only */ + bool slc; + bool glc; + bool lds; + bool nv; +}; + +struct Export_instruction : public Instruction { + unsigned enabled_mask; + unsigned dest; + bool compressed; + bool done; + bool valid_mask; +}; + +struct Pseudo_instruction : public Instruction { + bool tmp_in_scc; + PhysReg scratch_sgpr; /* might not be valid if it's not needed */ +}; + +struct Pseudo_branch_instruction : public Instruction { + /* target[0] is the block index of the branch target. + * For conditional branches, target[1] contains the fall-through alternative. + * A value of 0 means the target has not been initialized (BB0 cannot be a branch target). + */ + uint32_t target[2]; +}; + +struct Pseudo_barrier_instruction : public Instruction { +}; + +enum ReduceOp { + iadd32, iadd64, + imul32, imul64, + fadd32, fadd64, + fmul32, fmul64, + imin32, imin64, + imax32, imax64, + umin32, umin64, + umax32, umax64, + fmin32, fmin64, + fmax32, fmax64, + iand32, iand64, + ior32, ior64, + ixor32, ixor64, +}; + +/** + * Subgroup Reduction Instructions, everything except for the data to be + * reduced and the result as inserted by setup_reduce_temp(). + * Operand(0): data to be reduced + * Operand(1): reduce temporary + * Operand(2): vector temporary + * Definition(0): result + * Definition(1): scalar temporary + * Definition(2): scalar identity temporary + * Definition(3): scc clobber + * Definition(4): vcc clobber + * + */ +struct Pseudo_reduction_instruction : public Instruction { + ReduceOp reduce_op; + unsigned cluster_size; // must be 0 for scans +}; + +struct instr_deleter_functor { + void operator()(void* p) { + free(p); + } +}; + +template +using aco_ptr = std::unique_ptr; + +template +T* create_instruction(aco_opcode opcode, Format format, uint32_t num_operands, uint32_t num_definitions) +{ + std::size_t size = sizeof(T) + num_operands * sizeof(Operand) + num_definitions * sizeof(Definition); + char *data = (char*) calloc(1, size); + T* inst = (T*) data; + + inst->opcode = opcode; + inst->format = format; + + inst->operands = aco::span((Operand*)(data + sizeof(T)), num_operands); + inst->definitions = aco::span((Definition*)inst->operands.end(), num_definitions); + + return inst; +} + +constexpr bool is_phi(Instruction* instr) +{ + return instr->opcode == aco_opcode::p_phi || instr->opcode == aco_opcode::p_linear_phi; +} + +static inline bool is_phi(aco_ptr& instr) +{ + return is_phi(instr.get()); +} + +constexpr barrier_interaction get_barrier_interaction(Instruction* instr) +{ + switch (instr->format) { + case Format::SMEM: + return static_cast(instr)->barrier; + case Format::MUBUF: + return static_cast(instr)->barrier; + case Format::MIMG: + return static_cast(instr)->barrier; + case Format::FLAT: + case Format::GLOBAL: + return barrier_buffer; + case Format::DS: + return barrier_shared; + default: + return barrier_none; + } +} + +enum block_kind { + /* uniform indicates that leaving this block, + * all actives lanes stay active */ + block_kind_uniform = 1 << 0, + block_kind_top_level = 1 << 1, + block_kind_loop_preheader = 1 << 2, + block_kind_loop_header = 1 << 3, + block_kind_loop_exit = 1 << 4, + block_kind_continue = 1 << 5, + block_kind_break = 1 << 6, + block_kind_continue_or_break = 1 << 7, + block_kind_discard = 1 << 8, + block_kind_branch = 1 << 9, + block_kind_merge = 1 << 10, + block_kind_invert = 1 << 11, + block_kind_uses_discard_if = 1 << 12, + block_kind_needs_lowering = 1 << 13, +}; + + +struct RegisterDemand { + constexpr RegisterDemand() = default; + constexpr RegisterDemand(const int16_t v, const int16_t s) noexcept + : vgpr{v}, sgpr{s} {} + int16_t vgpr = 0; + int16_t sgpr = 0; + + constexpr friend bool operator==(const RegisterDemand a, const RegisterDemand b) noexcept { + return a.vgpr == b.vgpr && a.sgpr == b.sgpr; + } + + constexpr bool exceeds(const RegisterDemand other) const noexcept { + return vgpr > other.vgpr || sgpr > other.sgpr; + } + + constexpr RegisterDemand operator+(const Temp t) const noexcept { + if (t.type() == RegType::sgpr) + return RegisterDemand( vgpr, sgpr + t.size() ); + else + return RegisterDemand( vgpr + t.size(), sgpr ); + } + + constexpr RegisterDemand operator+(const RegisterDemand other) const noexcept { + return RegisterDemand(vgpr + other.vgpr, sgpr + other.sgpr); + } + + constexpr RegisterDemand operator-(const RegisterDemand other) const noexcept { + return RegisterDemand(vgpr - other.vgpr, sgpr - other.sgpr); + } + + constexpr RegisterDemand& operator+=(const RegisterDemand other) noexcept { + vgpr += other.vgpr; + sgpr += other.sgpr; + return *this; + } + + constexpr RegisterDemand& operator-=(const RegisterDemand other) noexcept { + vgpr -= other.vgpr; + sgpr -= other.sgpr; + return *this; + } + + constexpr RegisterDemand& operator+=(const Temp t) noexcept { + if (t.type() == RegType::sgpr) + sgpr += t.size(); + else + vgpr += t.size(); + return *this; + } + + constexpr RegisterDemand& operator-=(const Temp t) noexcept { + if (t.type() == RegType::sgpr) + sgpr -= t.size(); + else + vgpr -= t.size(); + return *this; + } + + constexpr void update(const RegisterDemand other) noexcept { + vgpr = std::max(vgpr, other.vgpr); + sgpr = std::max(sgpr, other.sgpr); + } + +}; + +/* CFG */ +struct Block { + unsigned index; + unsigned offset = 0; + std::vector> instructions; + std::vector logical_preds; + std::vector linear_preds; + std::vector logical_succs; + std::vector linear_succs; + RegisterDemand register_demand = RegisterDemand(); + uint16_t loop_nest_depth = 0; + uint16_t kind = 0; + int logical_idom = -1; + int linear_idom = -1; + Temp live_out_exec = Temp(); + + /* this information is needed for predecessors to blocks with phis when + * moving out of ssa */ + bool scc_live_out = false; + PhysReg scratch_sgpr = PhysReg(); /* only needs to be valid if scc_live_out != false */ + + Block(unsigned idx) : index(idx) {} + Block() : index(0) {} +}; + +using Stage = uint16_t; + +/* software stages */ +static constexpr Stage sw_vs = 1 << 0; +static constexpr Stage sw_gs = 1 << 1; +static constexpr Stage sw_tcs = 1 << 2; +static constexpr Stage sw_tes = 1 << 3; +static constexpr Stage sw_fs = 1 << 4; +static constexpr Stage sw_cs = 1 << 5; +static constexpr Stage sw_mask = 0x3f; + +/* hardware stages (can't be OR'd, just a mask for convenience when testing multiple) */ +static constexpr Stage hw_vs = 1 << 6; +static constexpr Stage hw_es = 1 << 7; +static constexpr Stage hw_gs = 1 << 8; /* not on GFX9. combined into ES on GFX9 (and GFX10/legacy). */ +static constexpr Stage hw_ls = 1 << 9; +static constexpr Stage hw_hs = 1 << 10; /* not on GFX9. combined into LS on GFX9 (and GFX10/legacy). */ +static constexpr Stage hw_fs = 1 << 11; +static constexpr Stage hw_cs = 1 << 12; +static constexpr Stage hw_mask = 0x7f << 6; + +/* possible settings of Program::stage */ +static constexpr Stage vertex_vs = sw_vs | hw_vs; +static constexpr Stage fragment_fs = sw_fs | hw_fs; +static constexpr Stage compute_cs = sw_cs | hw_cs; +static constexpr Stage tess_eval_vs = sw_tes | hw_vs; +/* GFX10/NGG */ +static constexpr Stage ngg_vertex_gs = sw_vs | hw_gs; +static constexpr Stage ngg_vertex_geometry_gs = sw_vs | sw_gs | hw_gs; +static constexpr Stage ngg_tess_eval_geometry_gs = sw_tes | sw_gs | hw_gs; +static constexpr Stage ngg_vertex_tess_control_hs = sw_vs | sw_tcs | hw_hs; +/* GFX9 (and GFX10 if NGG isn't used) */ +static constexpr Stage vertex_geometry_es = sw_vs | sw_gs | hw_es; +static constexpr Stage vertex_tess_control_ls = sw_vs | sw_tcs | hw_ls; +static constexpr Stage tess_eval_geometry_es = sw_tes | sw_gs | hw_es; +/* pre-GFX9 */ +static constexpr Stage vertex_ls = sw_vs | hw_ls; /* vertex before tesselation control */ +static constexpr Stage tess_control_hs = sw_tcs | hw_hs; +static constexpr Stage tess_eval_es = sw_tes | hw_gs; /* tesselation evaluation before GS */ +static constexpr Stage geometry_gs = sw_gs | hw_gs; + +class Program final { +public: + std::vector blocks; + RegisterDemand max_reg_demand = RegisterDemand(); + uint16_t sgpr_limit = 0; + uint16_t num_waves = 0; + ac_shader_config* config; + struct radv_shader_info *info; + enum chip_class chip_class; + enum radeon_family family; + Stage stage; /* Stage */ + bool needs_exact = false; /* there exists an instruction with disable_wqm = true */ + bool needs_wqm = false; /* there exists a p_wqm instruction */ + bool wb_smem_l1_on_end = false; + + std::vector constant_data; + + uint32_t allocateId() + { + assert(allocationID <= 16777215); + return allocationID++; + } + + uint32_t peekAllocationId() + { + return allocationID; + } + + void setAllocationId(uint32_t id) + { + allocationID = id; + } + + Block* create_and_insert_block() { + blocks.emplace_back(blocks.size()); + return &blocks.back(); + } + + Block* insert_block(Block&& block) { + block.index = blocks.size(); + blocks.emplace_back(std::move(block)); + return &blocks.back(); + } + +private: + uint32_t allocationID = 1; +}; + +struct live { + /* live temps out per block */ + std::vector> live_out; + /* register demand (sgpr/vgpr) per instruction per block */ + std::vector> register_demand; +}; + +void select_program(Program *program, + unsigned shader_count, + struct nir_shader *const *shaders, + ac_shader_config* config, + struct radv_shader_info *info, + struct radv_nir_compiler_options *options); + +void lower_wqm(Program* program, live& live_vars, + const struct radv_nir_compiler_options *options); +void lower_bool_phis(Program* program); +void update_vgpr_sgpr_demand(Program* program, const RegisterDemand new_demand); +live live_var_analysis(Program* program, const struct radv_nir_compiler_options *options); +std::vector dead_code_analysis(Program *program); +void dominator_tree(Program* program); +void insert_exec_mask(Program *program); +void value_numbering(Program* program); +void optimize(Program* program); +void setup_reduce_temp(Program* program); +void lower_to_cssa(Program* program, live& live_vars, const struct radv_nir_compiler_options *options); +void register_allocation(Program *program, std::vector> live_out_per_block); +void ssa_elimination(Program* program); +void lower_to_hw_instr(Program* program); +void schedule_program(Program* program, live& live_vars); +void spill(Program* program, live& live_vars, const struct radv_nir_compiler_options *options); +void insert_wait_states(Program* program); +void insert_NOPs(Program* program); +unsigned emit_program(Program* program, std::vector& code); +void print_asm(Program *program, std::vector& binary, unsigned exec_size, + enum radeon_family family, std::ostream& out); +void validate(Program* program, FILE *output); +bool validate_ra(Program* program, const struct radv_nir_compiler_options *options, FILE *output); +#ifndef NDEBUG +void perfwarn(bool cond, const char *msg, Instruction *instr=NULL); +#else +#define perfwarn(program, cond, msg, ...) +#endif + +void aco_print_instr(Instruction *instr, FILE *output); +void aco_print_program(Program *program, FILE *output); + +typedef struct { + const int16_t opcode_gfx9[static_cast(aco_opcode::num_opcodes)]; + const int16_t opcode_gfx10[static_cast(aco_opcode::num_opcodes)]; + const std::bitset(aco_opcode::num_opcodes)> can_use_input_modifiers; + const std::bitset(aco_opcode::num_opcodes)> can_use_output_modifiers; + const char *name[static_cast(aco_opcode::num_opcodes)]; + const aco::Format format[static_cast(aco_opcode::num_opcodes)]; +} Info; + +extern const Info instr_info; + +} + +#endif /* ACO_IR_H */ + diff --git a/src/amd/compiler/aco_live_var_analysis.cpp b/src/amd/compiler/aco_live_var_analysis.cpp new file mode 100644 index 00000000000..f99e57c8b3a --- /dev/null +++ b/src/amd/compiler/aco_live_var_analysis.cpp @@ -0,0 +1,243 @@ +/* + * Copyright © 2018 Valve Corporation + * Copyright © 2018 Google + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Daniel Schürmann (daniel.schuermann@campus.tu-berlin.de) + * Bas Nieuwenhuizen (bas@basnieuwenhuizen.nl) + * + */ + +#include "aco_ir.h" + +#include +#include + +#include "vulkan/radv_shader.h" + +namespace aco { +namespace { + +void process_live_temps_per_block(Program *program, live& lives, Block* block, + std::set& worklist, std::vector& phi_sgpr_ops) +{ + std::vector& register_demand = lives.register_demand[block->index]; + RegisterDemand new_demand; + + register_demand.resize(block->instructions.size()); + block->register_demand = RegisterDemand(); + + std::set live_sgprs; + std::set live_vgprs; + + /* add the live_out_exec to live */ + bool exec_live = false; + if (block->live_out_exec != Temp()) { + live_sgprs.insert(block->live_out_exec); + new_demand.sgpr += 2; + exec_live = true; + } + + /* split the live-outs from this block into the temporary sets */ + std::vector>& live_temps = lives.live_out; + for (const Temp temp : live_temps[block->index]) { + const bool inserted = temp.is_linear() + ? live_sgprs.insert(temp).second + : live_vgprs.insert(temp).second; + if (inserted) { + new_demand += temp; + } + } + new_demand.sgpr -= phi_sgpr_ops[block->index]; + + /* traverse the instructions backwards */ + for (int idx = block->instructions.size() -1; idx >= 0; idx--) + { + /* substract the 2 sgprs from exec */ + if (exec_live) + assert(new_demand.sgpr >= 2); + register_demand[idx] = RegisterDemand(new_demand.vgpr, new_demand.sgpr - (exec_live ? 2 : 0)); + + Instruction *insn = block->instructions[idx].get(); + /* KILL */ + for (Definition& definition : insn->definitions) { + if (!definition.isTemp()) { + continue; + } + + const Temp temp = definition.getTemp(); + size_t n = 0; + if (temp.is_linear()) + n = live_sgprs.erase(temp); + else + n = live_vgprs.erase(temp); + + if (n) { + new_demand -= temp; + definition.setKill(false); + } else { + register_demand[idx] += temp; + definition.setKill(true); + } + + if (definition.isFixed() && definition.physReg() == exec) + exec_live = false; + } + + /* GEN */ + if (insn->opcode == aco_opcode::p_phi || + insn->opcode == aco_opcode::p_linear_phi) { + /* directly insert into the predecessors live-out set */ + std::vector& preds = insn->opcode == aco_opcode::p_phi + ? block->logical_preds + : block->linear_preds; + for (unsigned i = 0; i < preds.size(); ++i) + { + Operand &operand = insn->operands[i]; + if (!operand.isTemp()) { + continue; + } + /* check if we changed an already processed block */ + const bool inserted = live_temps[preds[i]].insert(operand.getTemp()).second; + if (inserted) { + operand.setFirstKill(true); + worklist.insert(preds[i]); + if (insn->opcode == aco_opcode::p_phi && operand.getTemp().type() == RegType::sgpr) + phi_sgpr_ops[preds[i]] += operand.size(); + } + } + } else if (insn->opcode == aco_opcode::p_logical_end) { + new_demand.sgpr += phi_sgpr_ops[block->index]; + } else { + for (unsigned i = 0; i < insn->operands.size(); ++i) + { + Operand& operand = insn->operands[i]; + if (!operand.isTemp()) { + continue; + } + const Temp temp = operand.getTemp(); + const bool inserted = temp.is_linear() + ? live_sgprs.insert(temp).second + : live_vgprs.insert(temp).second; + if (inserted) { + operand.setFirstKill(true); + for (unsigned j = i + 1; j < insn->operands.size(); ++j) { + if (insn->operands[j].isTemp() && insn->operands[j].tempId() == operand.tempId()) { + insn->operands[j].setFirstKill(false); + insn->operands[j].setKill(true); + } + } + new_demand += temp; + } else { + operand.setKill(false); + } + + if (operand.isFixed() && operand.physReg() == exec) + exec_live = true; + } + } + + block->register_demand.update(register_demand[idx]); + } + + /* now, we have the live-in sets and need to merge them into the live-out sets */ + for (unsigned pred_idx : block->logical_preds) { + for (Temp vgpr : live_vgprs) { + auto it = live_temps[pred_idx].insert(vgpr); + if (it.second) + worklist.insert(pred_idx); + } + } + + for (unsigned pred_idx : block->linear_preds) { + for (Temp sgpr : live_sgprs) { + auto it = live_temps[pred_idx].insert(sgpr); + if (it.second) + worklist.insert(pred_idx); + } + } + + if (!(block->index != 0 || (live_vgprs.empty() && live_sgprs.empty()))) { + aco_print_program(program, stderr); + fprintf(stderr, "These temporaries are never defined or are defined after use:\n"); + for (Temp vgpr : live_vgprs) + fprintf(stderr, "%%%d\n", vgpr.id()); + for (Temp sgpr : live_sgprs) + fprintf(stderr, "%%%d\n", sgpr.id()); + abort(); + } + + assert(block->index != 0 || new_demand == RegisterDemand()); +} +} /* end namespace */ + +void update_vgpr_sgpr_demand(Program* program, const RegisterDemand new_demand) +{ + // TODO: also take shared mem into account + const int16_t total_sgpr_regs = program->chip_class >= GFX8 ? 800 : 512; + const int16_t max_addressible_sgpr = program->sgpr_limit; + /* VGPRs are allocated in chunks of 4 */ + const int16_t rounded_vgpr_demand = std::max(4, (new_demand.vgpr + 3) & ~3); + /* SGPRs are allocated in chunks of 16 between 8 and 104. VCC occupies the last 2 registers */ + const int16_t rounded_sgpr_demand = std::min(std::max(8, (new_demand.sgpr + 2 + 7) & ~7), max_addressible_sgpr); + /* this won't compile, register pressure reduction necessary */ + if (new_demand.vgpr > 256 || new_demand.sgpr > max_addressible_sgpr) { + program->num_waves = 0; + program->max_reg_demand = new_demand; + } else { + program->num_waves = std::min(10, + std::min(256 / rounded_vgpr_demand, + total_sgpr_regs / rounded_sgpr_demand)); + + program->max_reg_demand = { int16_t((256 / program->num_waves) & ~3), std::min(((total_sgpr_regs / program->num_waves) & ~7) - 2, max_addressible_sgpr)}; + } +} + +live live_var_analysis(Program* program, + const struct radv_nir_compiler_options *options) +{ + live result; + result.live_out.resize(program->blocks.size()); + result.register_demand.resize(program->blocks.size()); + std::set worklist; + std::vector phi_sgpr_ops(program->blocks.size()); + RegisterDemand new_demand; + + /* this implementation assumes that the block idx corresponds to the block's position in program->blocks vector */ + for (Block& block : program->blocks) + worklist.insert(block.index); + while (!worklist.empty()) { + std::set::reverse_iterator b_it = worklist.rbegin(); + unsigned block_idx = *b_it; + worklist.erase(block_idx); + process_live_temps_per_block(program, result, &program->blocks[block_idx], worklist, phi_sgpr_ops); + new_demand.update(program->blocks[block_idx].register_demand); + } + + /* calculate the program's register demand and number of waves */ + update_vgpr_sgpr_demand(program, new_demand); + + return result; +} + +} + diff --git a/src/amd/compiler/aco_lower_bool_phis.cpp b/src/amd/compiler/aco_lower_bool_phis.cpp new file mode 100644 index 00000000000..0c56ca07214 --- /dev/null +++ b/src/amd/compiler/aco_lower_bool_phis.cpp @@ -0,0 +1,241 @@ +/* + * Copyright © 2019 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Rhys Perry (pendingchaos02@gmail.com) + * + */ + +#include + +#include "aco_ir.h" +#include "aco_builder.h" + + +namespace aco { + +struct phi_use { + Block *block; + unsigned phi_def; + + bool operator<(const phi_use& other) const { + return std::make_tuple(block, phi_def) < + std::make_tuple(other.block, other.phi_def); + } +}; + +struct ssa_state { + std::map latest; + std::map> phis; +}; + +Operand get_ssa(Program *program, unsigned block_idx, ssa_state *state) +{ + while (true) { + auto pos = state->latest.find(block_idx); + if (pos != state->latest.end()) + return Operand({pos->second, s2}); + + Block& block = program->blocks[block_idx]; + size_t pred = block.linear_preds.size(); + if (pred == 0) { + return Operand(s2); + } else if (pred == 1) { + block_idx = block.linear_preds[0]; + continue; + } else { + unsigned res = program->allocateId(); + state->latest[block_idx] = res; + + aco_ptr phi{create_instruction(aco_opcode::p_linear_phi, Format::PSEUDO, pred, 1)}; + for (unsigned i = 0; i < pred; i++) { + phi->operands[i] = get_ssa(program, block.linear_preds[i], state); + if (phi->operands[i].isTemp()) { + assert(i < 64); + state->phis[phi->operands[i].tempId()][(phi_use){&block, res}] |= (uint64_t)1 << i; + } + } + phi->definitions[0] = Definition(Temp{res, s2}); + block.instructions.emplace(block.instructions.begin(), std::move(phi)); + + return Operand({res, s2}); + } + } +} + +void update_phi(Program *program, ssa_state *state, Block *block, unsigned phi_def, uint64_t operand_mask) { + for (auto& phi : block->instructions) { + if (phi->opcode != aco_opcode::p_phi && phi->opcode != aco_opcode::p_linear_phi) + break; + if (phi->opcode != aco_opcode::p_linear_phi) + continue; + if (phi->definitions[0].tempId() != phi_def) + continue; + assert(ffsll(operand_mask) <= phi->operands.size()); + + uint64_t operands = operand_mask; + while (operands) { + unsigned operand = u_bit_scan64(&operands); + Operand new_operand = get_ssa(program, block->linear_preds[operand], state); + phi->operands[operand] = new_operand; + if (!new_operand.isUndefined()) + state->phis[new_operand.tempId()][(phi_use){block, phi_def}] |= (uint64_t)1 << operand; + } + return; + } + assert(false); +} + +Temp write_ssa(Program *program, Block *block, ssa_state *state, unsigned previous) { + unsigned id = program->allocateId(); + state->latest[block->index] = id; + + /* update phis */ + if (previous) { + std::map phis; + phis.swap(state->phis[previous]); + for (auto& phi : phis) + update_phi(program, state, phi.first.block, phi.first.phi_def, phi.second); + } + + return {id, s2}; +} + +void insert_before_branch(Block *block, aco_ptr instr) +{ + int end = block->instructions.size() - 1; + if (block->instructions[end]->format == Format::PSEUDO_BRANCH) + block->instructions.emplace(std::prev(block->instructions.end()), std::move(instr)); + else + block->instructions.emplace_back(std::move(instr)); +} + +void insert_before_logical_end(Block *block, aco_ptr instr) +{ + for (int i = block->instructions.size() - 1; i >= 0; --i) { + if (block->instructions[i]->opcode == aco_opcode::p_logical_end) { + block->instructions.emplace(std::next(block->instructions.begin(), i), std::move(instr)); + return; + } + } + insert_before_branch(block, std::move(instr)); +} + +aco_ptr lower_divergent_bool_phi(Program *program, Block *block, aco_ptr& phi) +{ + Builder bld(program); + + ssa_state state; + for (unsigned i = 0; i < phi->operands.size(); i++) { + Block *pred = &program->blocks[block->logical_preds[i]]; + + if (phi->operands[i].isUndefined()) + continue; + + assert(phi->operands[i].isTemp()); + Temp phi_src = phi->operands[i].getTemp(); + if (phi_src.regClass() == s1) { + Temp new_phi_src = bld.tmp(s2); + insert_before_logical_end(pred, + bld.sop2(aco_opcode::s_cselect_b64, Definition(new_phi_src), + Operand((uint32_t)-1), Operand(0u), bld.scc(phi_src)).get_ptr()); + phi_src = new_phi_src; + } + assert(phi_src.regClass() == s2); + + Operand cur = get_ssa(program, pred->index, &state); + Temp new_cur = write_ssa(program, pred, &state, cur.isTemp() ? cur.tempId() : 0); + + if (cur.isUndefined()) { + insert_before_logical_end(pred, bld.sop1(aco_opcode::s_mov_b64, Definition(new_cur), phi_src).get_ptr()); + } else { + Temp tmp1 = bld.tmp(s2), tmp2 = bld.tmp(s2); + insert_before_logical_end(pred, + bld.sop2(aco_opcode::s_andn2_b64, Definition(tmp1), bld.def(s1, scc), + cur, Operand(exec, s2)).get_ptr()); + insert_before_logical_end(pred, + bld.sop2(aco_opcode::s_and_b64, Definition(tmp2), bld.def(s1, scc), + phi_src, Operand(exec, s2)).get_ptr()); + insert_before_logical_end(pred, + bld.sop2(aco_opcode::s_or_b64, Definition(new_cur), bld.def(s1, scc), + tmp1, tmp2).get_ptr()); + } + } + + return bld.sop1(aco_opcode::s_mov_b64, phi->definitions[0], get_ssa(program, block->index, &state)).get_ptr(); +} + +void lower_linear_bool_phi(Program *program, Block *block, aco_ptr& phi) +{ + Builder bld(program); + + for (unsigned i = 0; i < phi->operands.size(); i++) { + if (!phi->operands[i].isTemp()) + continue; + + Temp phi_src = phi->operands[i].getTemp(); + if (phi_src.regClass() == s2) { + Temp new_phi_src = bld.tmp(s1); + insert_before_logical_end(&program->blocks[block->linear_preds[i]], + bld.sopc(aco_opcode::s_cmp_lg_u64, bld.scc(Definition(new_phi_src)), + Operand(0u), phi_src).get_ptr()); + phi->operands[i].setTemp(new_phi_src); + } + } +} + +void lower_bool_phis(Program* program) +{ + for (Block& block : program->blocks) { + std::vector> instructions; + std::vector> non_phi; + instructions.swap(block.instructions); + block.instructions.reserve(instructions.size()); + unsigned i = 0; + for (; i < instructions.size(); i++) + { + aco_ptr& phi = instructions[i]; + if (phi->opcode != aco_opcode::p_phi && phi->opcode != aco_opcode::p_linear_phi) + break; + if (phi->opcode == aco_opcode::p_phi && phi->definitions[0].regClass() == s2) { + non_phi.emplace_back(std::move(lower_divergent_bool_phi(program, &block, phi))); + } else if (phi->opcode == aco_opcode::p_linear_phi && phi->definitions[0].regClass() == s1) { + /* if it's a valid non-boolean phi, this should be a no-op */ + lower_linear_bool_phi(program, &block, phi); + block.instructions.emplace_back(std::move(phi)); + } else { + block.instructions.emplace_back(std::move(phi)); + } + } + for (auto&& instr : non_phi) { + assert(instr->opcode != aco_opcode::p_phi && instr->opcode != aco_opcode::p_linear_phi); + block.instructions.emplace_back(std::move(instr)); + } + for (; i < instructions.size(); i++) { + aco_ptr instr = std::move(instructions[i]); + assert(instr->opcode != aco_opcode::p_phi && instr->opcode != aco_opcode::p_linear_phi); + block.instructions.emplace_back(std::move(instr)); + } + } +} + +} diff --git a/src/amd/compiler/aco_lower_to_hw_instr.cpp b/src/amd/compiler/aco_lower_to_hw_instr.cpp new file mode 100644 index 00000000000..8fd33e47d92 --- /dev/null +++ b/src/amd/compiler/aco_lower_to_hw_instr.cpp @@ -0,0 +1,765 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Daniel Schürmann (daniel.schuermann@campus.tu-berlin.de) + * + */ + +#include + +#include "aco_ir.h" +#include "aco_builder.h" +#include "util/u_math.h" +#include "sid.h" + + +namespace aco { + +struct lower_context { + Program *program; + std::vector> instructions; +}; + +void emit_dpp_op(lower_context *ctx, PhysReg dst, PhysReg src0, PhysReg src1, PhysReg vtmp, PhysReg wrtmp, + aco_opcode op, Format format, bool clobber_vcc, unsigned dpp_ctrl, + unsigned row_mask, unsigned bank_mask, bool bound_ctrl_zero, unsigned size, + Operand *identity=NULL) /* for VOP3 with sparse writes */ +{ + RegClass rc = RegClass(RegType::vgpr, size); + if (format == Format::VOP3) { + Builder bld(ctx->program, &ctx->instructions); + + if (identity) + bld.vop1(aco_opcode::v_mov_b32, Definition(vtmp, v1), identity[0]); + if (identity && size >= 2) + bld.vop1(aco_opcode::v_mov_b32, Definition(PhysReg{vtmp+1}, v1), identity[1]); + + for (unsigned i = 0; i < size; i++) + bld.vop1_dpp(aco_opcode::v_mov_b32, Definition(PhysReg{vtmp+i}, v1), Operand(PhysReg{src0+i}, v1), + dpp_ctrl, row_mask, bank_mask, bound_ctrl_zero); + + if (clobber_vcc) + bld.vop3(op, Definition(dst, rc), Definition(vcc, s2), Operand(vtmp, rc), Operand(src1, rc)); + else + bld.vop3(op, Definition(dst, rc), Operand(vtmp, rc), Operand(src1, rc)); + } else { + assert(format == Format::VOP2 || format == Format::VOP1); + assert(size == 1 || (op == aco_opcode::v_mov_b32)); + + for (unsigned i = 0; i < size; i++) { + aco_ptr dpp{create_instruction( + op, (Format) ((uint32_t) format | (uint32_t) Format::DPP), + format == Format::VOP2 ? 2 : 1, clobber_vcc ? 2 : 1)}; + dpp->operands[0] = Operand(PhysReg{src0+i}, rc); + if (format == Format::VOP2) + dpp->operands[1] = Operand(PhysReg{src1+i}, rc); + dpp->definitions[0] = Definition(PhysReg{dst+i}, rc); + if (clobber_vcc) + dpp->definitions[1] = Definition(vcc, s2); + dpp->dpp_ctrl = dpp_ctrl; + dpp->row_mask = row_mask; + dpp->bank_mask = bank_mask; + dpp->bound_ctrl = bound_ctrl_zero; + ctx->instructions.emplace_back(std::move(dpp)); + } + } +} + +uint32_t get_reduction_identity(ReduceOp op, unsigned idx) +{ + switch (op) { + case iadd32: + case iadd64: + case fadd32: + case fadd64: + case ior32: + case ior64: + case ixor32: + case ixor64: + case umax32: + case umax64: + return 0; + case imul32: + case imul64: + return idx ? 0 : 1; + case fmul32: + return 0x3f800000u; /* 1.0 */ + case fmul64: + return idx ? 0x3ff00000u : 0u; /* 1.0 */ + case imin32: + return INT32_MAX; + case imin64: + return idx ? 0x7fffffffu : 0xffffffffu; + case imax32: + return INT32_MIN; + case imax64: + return idx ? 0x80000000u : 0; + case umin32: + case umin64: + case iand32: + case iand64: + return 0xffffffffu; + case fmin32: + return 0x7f800000u; /* infinity */ + case fmin64: + return idx ? 0x7ff00000u : 0u; /* infinity */ + case fmax32: + return 0xff800000u; /* negative infinity */ + case fmax64: + return idx ? 0xfff00000u : 0u; /* negative infinity */ + } + unreachable("Invalid reduction operation"); +} + +aco_opcode get_reduction_opcode(lower_context *ctx, ReduceOp op, bool *clobber_vcc, Format *format) +{ + *clobber_vcc = false; + *format = Format::VOP2; + switch (op) { + case iadd32: + *clobber_vcc = ctx->program->chip_class < GFX9; + return ctx->program->chip_class < GFX9 ? aco_opcode::v_add_co_u32 : aco_opcode::v_add_u32; + case imul32: + *format = Format::VOP3; + return aco_opcode::v_mul_lo_u32; + case fadd32: + return aco_opcode::v_add_f32; + case fmul32: + return aco_opcode::v_mul_f32; + case imax32: + return aco_opcode::v_max_i32; + case imin32: + return aco_opcode::v_min_i32; + case umin32: + return aco_opcode::v_min_u32; + case umax32: + return aco_opcode::v_max_u32; + case fmin32: + return aco_opcode::v_min_f32; + case fmax32: + return aco_opcode::v_max_f32; + case iand32: + return aco_opcode::v_and_b32; + case ixor32: + return aco_opcode::v_xor_b32; + case ior32: + return aco_opcode::v_or_b32; + case iadd64: + case imul64: + assert(false); + break; + case fadd64: + *format = Format::VOP3; + return aco_opcode::v_add_f64; + case fmul64: + *format = Format::VOP3; + return aco_opcode::v_mul_f64; + case imin64: + case imax64: + case umin64: + case umax64: + assert(false); + break; + case fmin64: + *format = Format::VOP3; + return aco_opcode::v_min_f64; + case fmax64: + *format = Format::VOP3; + return aco_opcode::v_max_f64; + case iand64: + case ior64: + case ixor64: + assert(false); + break; + } + unreachable("Invalid reduction operation"); + return aco_opcode::v_min_u32; +} + +void emit_vopn(lower_context *ctx, PhysReg dst, PhysReg src0, PhysReg src1, + RegClass rc, aco_opcode op, Format format, bool clobber_vcc) +{ + aco_ptr instr; + switch (format) { + case Format::VOP2: + instr.reset(create_instruction(op, format, 2, clobber_vcc ? 2 : 1)); + break; + case Format::VOP3: + instr.reset(create_instruction(op, format, 2, clobber_vcc ? 2 : 1)); + break; + default: + assert(false); + } + instr->operands[0] = Operand(src0, rc); + instr->operands[1] = Operand(src1, rc); + instr->definitions[0] = Definition(dst, rc); + if (clobber_vcc) + instr->definitions[1] = Definition(vcc, s2); + ctx->instructions.emplace_back(std::move(instr)); +} + +void emit_reduction(lower_context *ctx, aco_opcode op, ReduceOp reduce_op, unsigned cluster_size, PhysReg tmp, + PhysReg stmp, PhysReg vtmp, PhysReg sitmp, Operand src, Definition dst) +{ + assert(cluster_size == 64 || op == aco_opcode::p_reduce); + + Builder bld(ctx->program, &ctx->instructions); + + PhysReg wrtmp{0}; /* should never be needed */ + + Format format; + bool should_clobber_vcc; + aco_opcode reduce_opcode = get_reduction_opcode(ctx, reduce_op, &should_clobber_vcc, &format); + Operand identity[2]; + identity[0] = Operand(get_reduction_identity(reduce_op, 0)); + identity[1] = Operand(get_reduction_identity(reduce_op, 1)); + Operand vcndmask_identity[2] = {identity[0], identity[1]}; + + /* First, copy the source to tmp and set inactive lanes to the identity */ + // note: this clobbers SCC! + bld.sop1(aco_opcode::s_or_saveexec_b64, Definition(stmp, s2), Definition(scc, s1), Definition(exec, s2), Operand(UINT64_MAX), Operand(exec, s2)); + + for (unsigned i = 0; i < src.size(); i++) { + /* p_exclusive_scan needs it to be a sgpr or inline constant for the v_writelane_b32 */ + if (identity[i].isLiteral() && op == aco_opcode::p_exclusive_scan) { + bld.sop1(aco_opcode::s_mov_b32, Definition(PhysReg{sitmp+i}, s1), identity[i]); + identity[i] = Operand(PhysReg{sitmp+i}, s1); + + bld.vop1(aco_opcode::v_mov_b32, Definition(PhysReg{tmp+i}, v1), identity[i]); + vcndmask_identity[i] = Operand(PhysReg{tmp+i}, v1); + } else if (identity[i].isLiteral()) { + bld.vop1(aco_opcode::v_mov_b32, Definition(PhysReg{tmp+i}, v1), identity[i]); + vcndmask_identity[i] = Operand(PhysReg{tmp+i}, v1); + } + } + + for (unsigned i = 0; i < src.size(); i++) { + bld.vop2_e64(aco_opcode::v_cndmask_b32, Definition(PhysReg{tmp + i}, v1), + vcndmask_identity[i], Operand(PhysReg{src.physReg() + i}, v1), + Operand(stmp, s2)); + } + + bool exec_restored = false; + bool dst_written = false; + switch (op) { + case aco_opcode::p_reduce: + if (cluster_size == 1) break; + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_quad_perm(1, 0, 3, 2), 0xf, 0xf, false, src.size()); + if (cluster_size == 2) break; + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_quad_perm(2, 3, 0, 1), 0xf, 0xf, false, src.size()); + if (cluster_size == 4) break; + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_half_mirror, 0xf, 0xf, false, src.size()); + if (cluster_size == 8) break; + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_mirror, 0xf, 0xf, false, src.size()); + if (cluster_size == 16) break; + if (cluster_size == 32) { + for (unsigned i = 0; i < src.size(); i++) + bld.ds(aco_opcode::ds_swizzle_b32, Definition(PhysReg{vtmp+i}, v1), Operand(PhysReg{tmp+i}, s1), ds_pattern_bitmode(0x1f, 0, 0x10)); + bld.sop1(aco_opcode::s_mov_b64, Definition(exec, s2), Operand(stmp, s2)); + exec_restored = true; + emit_vopn(ctx, dst.physReg(), vtmp, tmp, src.regClass(), reduce_opcode, format, should_clobber_vcc); + dst_written = true; + } else { + assert(cluster_size == 64); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_bcast15, 0xa, 0xf, false, src.size()); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_bcast31, 0xc, 0xf, false, src.size()); + } + break; + case aco_opcode::p_exclusive_scan: + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, aco_opcode::v_mov_b32, Format::VOP1, false, + dpp_wf_sr1, 0xf, 0xf, true, src.size()); + for (unsigned i = 0; i < src.size(); i++) { + if (!identity[i].isConstant() || identity[i].constantValue()) { /* bound_ctrl should take case of this overwise */ + assert((identity[i].isConstant() && !identity[i].isLiteral()) || identity[i].physReg() == PhysReg{sitmp+i}); + bld.vop3(aco_opcode::v_writelane_b32, Definition(PhysReg{tmp+i}, v1), + identity[i], Operand(0u)); + } + } + /* fall through */ + case aco_opcode::p_inclusive_scan: + assert(cluster_size == 64); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_sr(1), 0xf, 0xf, false, src.size(), identity); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_sr(2), 0xf, 0xf, false, src.size(), identity); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_sr(4), 0xf, 0xf, false, src.size(), identity); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_sr(8), 0xf, 0xf, false, src.size(), identity); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_bcast15, 0xa, 0xf, false, src.size(), identity); + emit_dpp_op(ctx, tmp, tmp, tmp, vtmp, wrtmp, reduce_opcode, format, should_clobber_vcc, + dpp_row_bcast31, 0xc, 0xf, false, src.size(), identity); + break; + default: + unreachable("Invalid reduction mode"); + } + + if (!exec_restored) + bld.sop1(aco_opcode::s_mov_b64, Definition(exec, s2), Operand(stmp, s2)); + + if (op == aco_opcode::p_reduce && cluster_size == 64) { + for (unsigned k = 0; k < src.size(); k++) { + bld.vop3(aco_opcode::v_readlane_b32, Definition(PhysReg{dst.physReg() + k}, s1), + Operand(PhysReg{tmp + k}, v1), Operand(63u)); + } + } else if (!(dst.physReg() == tmp) && !dst_written) { + for (unsigned k = 0; k < src.size(); k++) { + bld.vop1(aco_opcode::v_mov_b32, Definition(PhysReg{dst.physReg() + k}, s1), + Operand(PhysReg{tmp + k}, v1)); + } + } +} + +struct copy_operation { + Operand op; + Definition def; + unsigned uses; + unsigned size; +}; + +void handle_operands(std::map& copy_map, lower_context* ctx, chip_class chip_class, Pseudo_instruction *pi) +{ + Builder bld(ctx->program, &ctx->instructions); + aco_ptr mov; + std::map::iterator it = copy_map.begin(); + std::map::iterator target; + bool writes_scc = false; + + /* count the number of uses for each dst reg */ + while (it != copy_map.end()) { + if (it->second.op.isConstant()) { + ++it; + continue; + } + + if (it->second.def.physReg() == scc) + writes_scc = true; + + assert(!pi->tmp_in_scc || !(it->second.def.physReg() == pi->scratch_sgpr)); + + /* if src and dst reg are the same, remove operation */ + if (it->first == it->second.op.physReg()) { + it = copy_map.erase(it); + continue; + } + /* check if the operand reg may be overwritten by another copy operation */ + target = copy_map.find(it->second.op.physReg()); + if (target != copy_map.end()) { + target->second.uses++; + } + + ++it; + } + + /* first, handle paths in the location transfer graph */ + bool preserve_scc = pi->tmp_in_scc && !writes_scc; + it = copy_map.begin(); + while (it != copy_map.end()) { + + /* the target reg is not used as operand for any other copy */ + if (it->second.uses == 0) { + + /* try to coalesce 32-bit sgpr copies to 64-bit copies */ + if (it->second.def.getTemp().type() == RegType::sgpr && it->second.size == 1 && + !it->second.op.isConstant() && it->first % 2 == it->second.op.physReg() % 2) { + + PhysReg other_def_reg = PhysReg{it->first % 2 ? it->first - 1 : it->first + 1}; + PhysReg other_op_reg = PhysReg{it->first % 2 ? it->second.op.physReg() - 1 : it->second.op.physReg() + 1}; + std::map::iterator other = copy_map.find(other_def_reg); + + if (other != copy_map.end() && !other->second.uses && other->second.size == 1 && + other->second.op.physReg() == other_op_reg && !other->second.op.isConstant()) { + std::map::iterator to_erase = it->first % 2 ? it : other; + it = it->first % 2 ? other : it; + copy_map.erase(to_erase); + it->second.size = 2; + } + } + + if (it->second.def.physReg() == scc) { + bld.sopc(aco_opcode::s_cmp_lg_i32, it->second.def, it->second.op, Operand(0u)); + preserve_scc = true; + } else if (it->second.size == 2 && it->second.def.getTemp().type() == RegType::sgpr) { + bld.sop1(aco_opcode::s_mov_b64, it->second.def, Operand(it->second.op.physReg(), s2)); + } else { + bld.copy(it->second.def, it->second.op); + } + + /* reduce the number of uses of the operand reg by one */ + if (!it->second.op.isConstant()) { + for (unsigned i = 0; i < it->second.size; i++) { + target = copy_map.find(PhysReg{it->second.op.physReg() + i}); + if (target != copy_map.end()) + target->second.uses--; + } + } + + copy_map.erase(it); + it = copy_map.begin(); + continue; + } else { + /* the target reg is used as operand, check the next entry */ + ++it; + } + } + + if (copy_map.empty()) + return; + + /* all target regs are needed as operand somewhere which means, all entries are part of a cycle */ + bool constants = false; + for (it = copy_map.begin(); it != copy_map.end(); ++it) { + assert(it->second.op.isFixed()); + if (it->first == it->second.op.physReg()) + continue; + /* do constants later */ + if (it->second.op.isConstant()) { + constants = true; + continue; + } + + if (preserve_scc && it->second.def.getTemp().type() == RegType::sgpr) + assert(!(it->second.def.physReg() == pi->scratch_sgpr)); + + /* to resolve the cycle, we have to swap the src reg with the dst reg */ + copy_operation swap = it->second; + assert(swap.op.regClass() == swap.def.regClass()); + Operand def_as_op = Operand(swap.def.physReg(), swap.def.regClass()); + Definition op_as_def = Definition(swap.op.physReg(), swap.op.regClass()); + if (chip_class >= GFX9 && swap.def.getTemp().type() == RegType::vgpr) { + bld.vop1(aco_opcode::v_swap_b32, swap.def, op_as_def, swap.op, def_as_op); + } else if (swap.op.physReg() == scc || swap.def.physReg() == scc) { + /* we need to swap scc and another sgpr */ + assert(!preserve_scc); + + PhysReg other = swap.op.physReg() == scc ? swap.def.physReg() : swap.op.physReg(); + + bld.sop1(aco_opcode::s_mov_b32, Definition(pi->scratch_sgpr, s1), Operand(scc, s1)); + bld.sopc(aco_opcode::s_cmp_lg_i32, Definition(scc, s1), Operand(other, s1), Operand(0u)); + bld.sop1(aco_opcode::s_mov_b32, Definition(other, s1), Operand(pi->scratch_sgpr, s1)); + } else if (swap.def.getTemp().type() == RegType::sgpr) { + if (preserve_scc) { + bld.sop1(aco_opcode::s_mov_b32, Definition(pi->scratch_sgpr, s1), swap.op); + bld.sop1(aco_opcode::s_mov_b32, op_as_def, def_as_op); + bld.sop1(aco_opcode::s_mov_b32, swap.def, Operand(pi->scratch_sgpr, s1)); + } else { + bld.sop2(aco_opcode::s_xor_b32, op_as_def, Definition(scc, s1), swap.op, def_as_op); + bld.sop2(aco_opcode::s_xor_b32, swap.def, Definition(scc, s1), swap.op, def_as_op); + bld.sop2(aco_opcode::s_xor_b32, op_as_def, Definition(scc, s1), swap.op, def_as_op); + } + } else { + bld.vop2(aco_opcode::v_xor_b32, op_as_def, swap.op, def_as_op); + bld.vop2(aco_opcode::v_xor_b32, swap.def, swap.op, def_as_op); + bld.vop2(aco_opcode::v_xor_b32, op_as_def, swap.op, def_as_op); + } + + /* change the operand reg of the target's use */ + assert(swap.uses == 1); + target = it; + for (++target; target != copy_map.end(); ++target) { + if (target->second.op.physReg() == it->first) { + target->second.op.setFixed(swap.op.physReg()); + break; + } + } + } + + /* copy constants into a registers which were operands */ + if (constants) { + for (it = copy_map.begin(); it != copy_map.end(); ++it) { + if (!it->second.op.isConstant()) + continue; + if (it->second.def.physReg() == scc) { + bld.sopc(aco_opcode::s_cmp_lg_i32, Definition(scc, s1), Operand(0u), Operand(it->second.op.constantValue() ? 1u : 0u)); + } else { + bld.copy(it->second.def, it->second.op); + } + } + } +} + +void lower_to_hw_instr(Program* program) +{ + Block *discard_block = NULL; + + for (size_t i = 0; i < program->blocks.size(); i++) + { + Block *block = &program->blocks[i]; + lower_context ctx; + ctx.program = program; + Builder bld(program, &ctx.instructions); + + for (size_t j = 0; j < block->instructions.size(); j++) { + aco_ptr& instr = block->instructions[j]; + aco_ptr mov; + if (instr->format == Format::PSEUDO) { + Pseudo_instruction *pi = (Pseudo_instruction*)instr.get(); + + switch (instr->opcode) + { + case aco_opcode::p_extract_vector: + { + unsigned reg = instr->operands[0].physReg() + instr->operands[1].constantValue() * instr->definitions[0].size(); + RegClass rc = RegClass(instr->operands[0].getTemp().type(), 1); + RegClass rc_def = RegClass(instr->definitions[0].getTemp().type(), 1); + if (reg == instr->definitions[0].physReg()) + break; + + std::map copy_operations; + for (unsigned i = 0; i < instr->definitions[0].size(); i++) { + Definition def = Definition(PhysReg{instr->definitions[0].physReg() + i}, rc_def); + copy_operations[def.physReg()] = {Operand(PhysReg{reg + i}, rc), def, 0, 1}; + } + handle_operands(copy_operations, &ctx, program->chip_class, pi); + break; + } + case aco_opcode::p_create_vector: + { + std::map copy_operations; + RegClass rc_def = RegClass(instr->definitions[0].getTemp().type(), 1); + unsigned reg_idx = 0; + for (const Operand& op : instr->operands) { + if (op.isConstant()) { + const PhysReg reg = PhysReg{instr->definitions[0].physReg() + reg_idx}; + const Definition def = Definition(reg, rc_def); + copy_operations[reg] = {op, def, 0, 1}; + reg_idx++; + continue; + } + + RegClass rc_op = RegClass(op.getTemp().type(), 1); + for (unsigned j = 0; j < op.size(); j++) + { + const Operand copy_op = Operand(PhysReg{op.physReg() + j}, rc_op); + const Definition def = Definition(PhysReg{instr->definitions[0].physReg() + reg_idx}, rc_def); + copy_operations[def.physReg()] = {copy_op, def, 0, 1}; + reg_idx++; + } + } + handle_operands(copy_operations, &ctx, program->chip_class, pi); + break; + } + case aco_opcode::p_split_vector: + { + std::map copy_operations; + RegClass rc_op = instr->operands[0].isConstant() ? s1 : RegClass(instr->operands[0].regClass().type(), 1); + for (unsigned i = 0; i < instr->definitions.size(); i++) { + unsigned k = instr->definitions[i].size(); + RegClass rc_def = RegClass(instr->definitions[i].getTemp().type(), 1); + for (unsigned j = 0; j < k; j++) { + Operand op = Operand(PhysReg{instr->operands[0].physReg() + (i*k+j)}, rc_op); + Definition def = Definition(PhysReg{instr->definitions[i].physReg() + j}, rc_def); + copy_operations[def.physReg()] = {op, def, 0, 1}; + } + } + handle_operands(copy_operations, &ctx, program->chip_class, pi); + break; + } + case aco_opcode::p_parallelcopy: + case aco_opcode::p_wqm: + { + std::map copy_operations; + for (unsigned i = 0; i < instr->operands.size(); i++) + { + Operand operand = instr->operands[i]; + if (operand.isConstant() || operand.size() == 1) { + assert(instr->definitions[i].size() == 1); + copy_operations[instr->definitions[i].physReg()] = {operand, instr->definitions[i], 0, 1}; + } else { + RegClass def_rc = RegClass(instr->definitions[i].regClass().type(), 1); + RegClass op_rc = RegClass(operand.getTemp().type(), 1); + for (unsigned j = 0; j < operand.size(); j++) + { + Operand op = Operand(PhysReg{instr->operands[i].physReg() + j}, op_rc); + Definition def = Definition(PhysReg{instr->definitions[i].physReg() + j}, def_rc); + copy_operations[def.physReg()] = {op, def, 0, 1}; + } + } + } + handle_operands(copy_operations, &ctx, program->chip_class, pi); + break; + } + case aco_opcode::p_discard_if: + { + bool early_exit = false; + if (block->instructions[j + 1]->opcode != aco_opcode::p_logical_end || + block->instructions[j + 2]->opcode != aco_opcode::s_endpgm) { + early_exit = true; + } + + if (early_exit && !discard_block) { + discard_block = program->create_and_insert_block(); + block = &program->blocks[i]; + + bld.reset(discard_block); + bld.exp(aco_opcode::exp, Operand(v1), Operand(v1), Operand(v1), Operand(v1), + 0, V_008DFC_SQ_EXP_NULL, false, true, true); + if (program->wb_smem_l1_on_end) + bld.smem(aco_opcode::s_dcache_wb); + bld.sopp(aco_opcode::s_endpgm); + + bld.reset(&ctx.instructions); + } + + // TODO: optimize uniform conditions + Definition branch_cond = instr->definitions.back(); + Operand discard_cond = instr->operands.back(); + aco_ptr sop2; + /* backwards, to finally branch on the global exec mask */ + for (int i = instr->operands.size() - 2; i >= 0; i--) { + bld.sop2(aco_opcode::s_andn2_b64, + instr->definitions[i], /* new mask */ + branch_cond, /* scc */ + instr->operands[i], /* old mask */ + discard_cond); + } + + if (early_exit) { + bld.sopp(aco_opcode::s_cbranch_scc0, bld.scc(branch_cond.getTemp()), discard_block->index); + + discard_block->linear_preds.push_back(block->index); + block->linear_succs.push_back(discard_block->index); + } + + break; + } + case aco_opcode::p_spill: + { + assert(instr->operands[0].regClass() == v1.as_linear()); + for (unsigned i = 0; i < instr->operands[2].size(); i++) { + bld.vop3(aco_opcode::v_writelane_b32, bld.def(v1, instr->operands[0].physReg()), + Operand(PhysReg{instr->operands[2].physReg() + i}, s1), + Operand(instr->operands[1].constantValue() + i)); + } + break; + } + case aco_opcode::p_reload: + { + assert(instr->operands[0].regClass() == v1.as_linear()); + for (unsigned i = 0; i < instr->definitions[0].size(); i++) { + bld.vop3(aco_opcode::v_readlane_b32, + bld.def(s1, PhysReg{instr->definitions[0].physReg() + i}), + instr->operands[0], Operand(instr->operands[1].constantValue() + i)); + } + break; + } + case aco_opcode::p_as_uniform: + { + if (instr->operands[0].isConstant() || instr->operands[0].regClass().type() == RegType::sgpr) { + std::map copy_operations; + Operand operand = instr->operands[0]; + if (operand.isConstant() || operand.size() == 1) { + assert(instr->definitions[0].size() == 1); + copy_operations[instr->definitions[0].physReg()] = {operand, instr->definitions[0], 0, 1}; + } else { + for (unsigned i = 0; i < operand.size(); i++) + { + Operand op = Operand(PhysReg{operand.physReg() + i}, s1); + Definition def = Definition(PhysReg{instr->definitions[0].physReg() + i}, s1); + copy_operations[def.physReg()] = {op, def, 0, 1}; + } + } + + handle_operands(copy_operations, &ctx, program->chip_class, pi); + } else { + assert(instr->operands[0].regClass().type() == RegType::vgpr); + assert(instr->definitions[0].regClass().type() == RegType::sgpr); + assert(instr->operands[0].size() == instr->definitions[0].size()); + for (unsigned i = 0; i < instr->definitions[0].size(); i++) { + bld.vop1(aco_opcode::v_readfirstlane_b32, + bld.def(s1, PhysReg{instr->definitions[0].physReg() + i}), + Operand(PhysReg{instr->operands[0].physReg() + i}, v1)); + } + } + break; + } + default: + break; + } + } else if (instr->format == Format::PSEUDO_BRANCH) { + Pseudo_branch_instruction* branch = static_cast(instr.get()); + /* check if all blocks from current to target are empty */ + bool can_remove = block->index < branch->target[0]; + for (unsigned i = block->index + 1; can_remove && i < branch->target[0]; i++) { + if (program->blocks[i].instructions.size()) + can_remove = false; + } + if (can_remove) + continue; + + switch (instr->opcode) { + case aco_opcode::p_branch: + assert(block->linear_succs[0] == branch->target[0]); + bld.sopp(aco_opcode::s_branch, branch->target[0]); + break; + case aco_opcode::p_cbranch_nz: + assert(block->linear_succs[1] == branch->target[0]); + if (branch->operands[0].physReg() == exec) + bld.sopp(aco_opcode::s_cbranch_execnz, branch->target[0]); + else if (branch->operands[0].physReg() == vcc) + bld.sopp(aco_opcode::s_cbranch_vccnz, branch->target[0]); + else { + assert(branch->operands[0].physReg() == scc); + bld.sopp(aco_opcode::s_cbranch_scc1, branch->target[0]); + } + break; + case aco_opcode::p_cbranch_z: + assert(block->linear_succs[1] == branch->target[0]); + if (branch->operands[0].physReg() == exec) + bld.sopp(aco_opcode::s_cbranch_execz, branch->target[0]); + else if (branch->operands[0].physReg() == vcc) + bld.sopp(aco_opcode::s_cbranch_vccz, branch->target[0]); + else { + assert(branch->operands[0].physReg() == scc); + bld.sopp(aco_opcode::s_cbranch_scc0, branch->target[0]); + } + break; + default: + unreachable("Unknown Pseudo branch instruction!"); + } + + } else if (instr->format == Format::PSEUDO_REDUCTION) { + Pseudo_reduction_instruction* reduce = static_cast(instr.get()); + emit_reduction(&ctx, reduce->opcode, reduce->reduce_op, reduce->cluster_size, + reduce->operands[1].physReg(), // tmp + reduce->definitions[1].physReg(), // stmp + reduce->operands[2].physReg(), // vtmp + reduce->definitions[2].physReg(), // sitmp + reduce->operands[0], reduce->definitions[0]); + } else { + ctx.instructions.emplace_back(std::move(instr)); + } + + } + block->instructions.swap(ctx.instructions); + } +} + +} diff --git a/src/amd/compiler/aco_opcodes.py b/src/amd/compiler/aco_opcodes.py new file mode 100644 index 00000000000..2221e2817af --- /dev/null +++ b/src/amd/compiler/aco_opcodes.py @@ -0,0 +1,1552 @@ +# +# Copyright (c) 2018 Valve Corporation +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice (including the next +# paragraph) shall be included in all copies or substantial portions of the +# Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING +# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS +# IN THE SOFTWARE. +# +# Authors: +# Daniel Schuermann (daniel.schuermann@campus.tu-berlin.de) + + +# Class that represents all the information we have about the opcode +# NOTE: this must be kept in sync with aco_op_info + +from enum import Enum + +class Format(Enum): + PSEUDO = 0 + SOP1 = 1 + SOP2 = 2 + SOPK = 3 + SOPP = 4 + SOPC = 5 + SMEM = 6 + DS = 8 + MTBUF = 9 + MUBUF = 10 + MIMG = 11 + EXP = 12 + FLAT = 13 + GLOBAL = 14 + SCRATCH = 15 + PSEUDO_BRANCH = 16 + PSEUDO_BARRIER = 17 + PSEUDO_REDUCTION = 18 + VOP1 = 1 << 8 + VOP2 = 1 << 9 + VOPC = 1 << 10 + VOP3A = 1 << 11 + VOP3B = 1 << 11 + VOP3P = 1 << 12 + VINTRP = 1 << 13 + DPP = 1 << 14 + SDWA = 1 << 15 + + def get_builder_fields(self): + if self == Format.SOPK: + return [('uint16_t', 'imm', None)] + elif self == Format.SOPP: + return [('uint32_t', 'block', '-1'), + ('uint32_t', 'imm', '0')] + elif self == Format.SMEM: + return [('bool', 'can_reorder', 'true'), + ('bool', 'glc', 'false'), + ('bool', 'dlc', 'false'), + ('bool', 'nv', 'false')] + elif self == Format.DS: + return [('int16_t', 'offset0', '0'), + ('int8_t', 'offset1', '0'), + ('bool', 'gds', 'false')] + elif self == Format.MTBUF: + return [('unsigned', 'dfmt', None), + ('unsigned', 'nfmt', None), + ('unsigned', 'img_format', None), + ('unsigned', 'offset', None), + ('bool', 'offen', None), + ('bool', 'idxen', 'false'), + ('bool', 'disable_wqm', 'false'), + ('bool', 'glc', 'false'), + ('bool', 'dlc', 'false'), + ('bool', 'slc', 'false'), + ('bool', 'tfe', 'false'), + ('bool', 'lds', 'false')] + elif self == Format.MUBUF: + return [('unsigned', 'offset', None), + ('bool', 'offen', None), + ('bool', 'idxen', 'false'), + ('bool', 'disable_wqm', 'false'), + ('bool', 'glc', 'false'), + ('bool', 'dlc', 'false'), + ('bool', 'slc', 'false'), + ('bool', 'tfe', 'false'), + ('bool', 'lds', 'false')] + elif self == Format.MIMG: + return [('unsigned', 'dmask', '0xF'), + ('bool', 'da', 'false'), + ('bool', 'unrm', 'true'), + ('bool', 'disable_wqm', 'false'), + ('bool', 'glc', 'false'), + ('bool', 'dlc', 'false'), + ('bool', 'slc', 'false'), + ('bool', 'tfe', 'false'), + ('bool', 'lwe', 'false'), + ('bool', 'r128_a16', 'false', 'r128'), + ('bool', 'd16', 'false')] + return [('unsigned', 'attribute', None), + ('unsigned', 'component', None)] + elif self == Format.EXP: + return [('unsigned', 'enabled_mask', None), + ('unsigned', 'dest', None), + ('bool', 'compr', 'false', 'compressed'), + ('bool', 'done', 'false'), + ('bool', 'vm', 'false', 'valid_mask')] + elif self == Format.PSEUDO_BRANCH: + return [('uint32_t', 'target0', '0', 'target[0]'), + ('uint32_t', 'target1', '0', 'target[1]')] + elif self == Format.PSEUDO_REDUCTION: + return [('ReduceOp', 'op', None, 'reduce_op'), + ('unsigned', 'cluster_size', '0')] + elif self == Format.VINTRP: + return [('unsigned', 'attribute', None), + ('unsigned', 'component', None)] + elif self == Format.DPP: + return [('uint16_t', 'dpp_ctrl', None), + ('uint8_t', 'row_mask', '0xF'), + ('uint8_t', 'bank_mask', '0xF'), + ('bool', 'bound_ctrl', 'false')] + elif self in [Format.FLAT, Format.GLOBAL, Format.SCRATCH]: + return [('uint16_t', 'offset', 0), + ('bool', 'glc', 'false'), + ('bool', 'slc', 'false'), + ('bool', 'lds', 'false'), + ('bool', 'nv', 'false')] + else: + return [] + + def get_builder_field_names(self): + return [f[1] for f in self.get_builder_fields()] + + def get_builder_field_dests(self): + return [(f[3] if len(f) >= 4 else f[1]) for f in self.get_builder_fields()] + + def get_builder_field_decls(self): + return [('%s %s=%s' % (f[0], f[1], f[2]) if f[2] != None else '%s %s' % (f[0], f[1])) for f in self.get_builder_fields()] + + +class Opcode(object): + """Class that represents all the information we have about the opcode + NOTE: this must be kept in sync with aco_op_info + """ + def __init__(self, name, opcode_gfx9, opcode_gfx10, format, input_mod, output_mod): + """Parameters: + + - name is the name of the opcode (prepend nir_op_ for the enum name) + - all types are strings that get nir_type_ prepended to them + - input_types is a list of types + - algebraic_properties is a space-seperated string, where nir_op_is_ is + prepended before each entry + - const_expr is an expression or series of statements that computes the + constant value of the opcode given the constant values of its inputs. + """ + assert isinstance(name, str) + assert isinstance(opcode_gfx9, int) + assert isinstance(opcode_gfx10, int) + assert isinstance(format, Format) + assert isinstance(input_mod, bool) + assert isinstance(output_mod, bool) + + self.name = name + self.opcode_gfx9 = opcode_gfx9 + self.opcode_gfx10 = opcode_gfx10 + self.input_mod = "1" if input_mod else "0" + self.output_mod = "1" if output_mod else "0" + self.format = format + + +# global dictionary of opcodes +opcodes = {} + +# VOPC to GFX6 opcode translation map +VOPC_GFX6 = [0] * 256 + +def opcode(name, opcode_gfx9 = -1, opcode_gfx10 = -1, format = Format.PSEUDO, input_mod = False, output_mod = False): + assert name not in opcodes + opcodes[name] = Opcode(name, opcode_gfx9, opcode_gfx10, format, input_mod, output_mod) + +opcode("exp", 0, 0, format = Format.EXP) +opcode("p_parallelcopy") +opcode("p_startpgm") +opcode("p_phi") +opcode("p_linear_phi") +opcode("p_as_uniform") + +opcode("p_create_vector") +opcode("p_extract_vector") +opcode("p_split_vector") + +# start/end the parts where we can use exec based instructions +# implicitly +opcode("p_logical_start") +opcode("p_logical_end") + +# e.g. subgroupMin() in SPIR-V +opcode("p_reduce", format=Format.PSEUDO_REDUCTION) +# e.g. subgroupInclusiveMin() +opcode("p_inclusive_scan", format=Format.PSEUDO_REDUCTION) +# e.g. subgroupExclusiveMin() +opcode("p_exclusive_scan", format=Format.PSEUDO_REDUCTION) + +opcode("p_branch", format=Format.PSEUDO_BRANCH) +opcode("p_cbranch", format=Format.PSEUDO_BRANCH) +opcode("p_cbranch_z", format=Format.PSEUDO_BRANCH) +opcode("p_cbranch_nz", format=Format.PSEUDO_BRANCH) + +opcode("p_memory_barrier_all", format=Format.PSEUDO_BARRIER) +opcode("p_memory_barrier_atomic", format=Format.PSEUDO_BARRIER) +opcode("p_memory_barrier_buffer", format=Format.PSEUDO_BARRIER) +opcode("p_memory_barrier_image", format=Format.PSEUDO_BARRIER) +opcode("p_memory_barrier_shared", format=Format.PSEUDO_BARRIER) + +opcode("p_spill") +opcode("p_reload") + +# start/end linear vgprs +opcode("p_start_linear_vgpr") +opcode("p_end_linear_vgpr") + +opcode("p_wqm") +opcode("p_discard_if") +opcode("p_load_helper") +opcode("p_demote_to_helper") +opcode("p_is_helper") + +opcode("p_fs_buffer_store_smem", format=Format.SMEM) + + +# SOP2 instructions: 2 scalar inputs, 1 scalar output (+optional scc) +SOP2 = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name + (0x00, 0x00, 0x00, 0x00, 0x00, "s_add_u32"), + (0x01, 0x01, 0x01, 0x01, 0x01, "s_sub_u32"), + (0x02, 0x02, 0x02, 0x02, 0x02, "s_add_i32"), + (0x03, 0x03, 0x03, 0x03, 0x03, "s_sub_i32"), + (0x04, 0x04, 0x04, 0x04, 0x04, "s_addc_u32"), + (0x05, 0x05, 0x05, 0x05, 0x05, "s_subb_u32"), + (0x06, 0x06, 0x06, 0x06, 0x06, "s_min_i32"), + (0x07, 0x07, 0x07, 0x07, 0x07, "s_min_u32"), + (0x08, 0x08, 0x08, 0x08, 0x08, "s_max_i32"), + (0x09, 0x09, 0x09, 0x09, 0x09, "s_max_u32"), + (0x0a, 0x0a, 0x0a, 0x0a, 0x0a, "s_cselect_b32"), + (0x0b, 0x0b, 0x0b, 0x0b, 0x0b, "s_cselect_b64"), + (0x0e, 0x0e, 0x0c, 0x0c, 0x0e, "s_and_b32"), + (0x0f, 0x0f, 0x0d, 0x0d, 0x0f, "s_and_b64"), + (0x10, 0x10, 0x0e, 0x0e, 0x10, "s_or_b32"), + (0x11, 0x11, 0x0f, 0x0f, 0x11, "s_or_b64"), + (0x12, 0x12, 0x10, 0x10, 0x12, "s_xor_b32"), + (0x13, 0x13, 0x11, 0x11, 0x13, "s_xor_b64"), + (0x14, 0x14, 0x12, 0x12, 0x14, "s_andn2_b32"), + (0x15, 0x15, 0x13, 0x13, 0x15, "s_andn2_b64"), + (0x16, 0x16, 0x14, 0x14, 0x16, "s_orn2_b32"), + (0x17, 0x17, 0x15, 0x15, 0x17, "s_orn2_b64"), + (0x18, 0x18, 0x16, 0x16, 0x18, "s_nand_b32"), + (0x19, 0x19, 0x17, 0x17, 0x19, "s_nand_b64"), + (0x1a, 0x1a, 0x18, 0x18, 0x1a, "s_nor_b32"), + (0x1b, 0x1b, 0x19, 0x19, 0x1b, "s_nor_b64"), + (0x1c, 0x1c, 0x1a, 0x1a, 0x1c, "s_xnor_b32"), + (0x1d, 0x1d, 0x1b, 0x1b, 0x1d, "s_xnor_b64"), + (0x1e, 0x1e, 0x1c, 0x1c, 0x1e, "s_lshl_b32"), + (0x1f, 0x1f, 0x1d, 0x1d, 0x1f, "s_lshl_b64"), + (0x20, 0x20, 0x1e, 0x1e, 0x20, "s_lshr_b32"), + (0x21, 0x21, 0x1f, 0x1f, 0x21, "s_lshr_b64"), + (0x22, 0x22, 0x20, 0x20, 0x22, "s_ashr_i32"), + (0x23, 0x23, 0x21, 0x21, 0x23, "s_ashr_i64"), + (0x24, 0x24, 0x22, 0x22, 0x24, "s_bfm_b32"), + (0x25, 0x25, 0x23, 0x23, 0x25, "s_bfm_b64"), + (0x26, 0x26, 0x24, 0x24, 0x26, "s_mul_i32"), + (0x27, 0x27, 0x25, 0x25, 0x27, "s_bfe_u32"), + (0x28, 0x28, 0x26, 0x26, 0x28, "s_bfe_i32"), + (0x29, 0x29, 0x27, 0x27, 0x29, "s_bfe_u64"), + (0x2a, 0x2a, 0x28, 0x28, 0x2a, "s_bfe_i64"), + (0x2b, 0x2b, 0x29, 0x29, -1, "s_cbranch_g_fork"), + (0x2c, 0x2c, 0x2a, 0x2a, 0x2c, "s_absdiff_i32"), + ( -1, -1, 0x2b, 0x2b, -1, "s_rfe_restore_b64"), + ( -1, -1, -1, 0x2e, 0x2e, "s_lshl1_add_u32"), + ( -1, -1, -1, 0x2f, 0x2f, "s_lshl2_add_u32"), + ( -1, -1, -1, 0x30, 0x30, "s_lshl3_add_u32"), + ( -1, -1, -1, 0x31, 0x31, "s_lshl4_add_u32"), + ( -1, -1, -1, 0x32, 0x32, "s_pack_ll_b32_b16"), + ( -1, -1, -1, 0x33, 0x33, "s_pack_lh_b32_b16"), + ( -1, -1, -1, 0x34, 0x34, "s_pack_hh_b32_b16"), + ( -1, -1, -1, 0x2c, 0x35, "s_mul_hi_u32"), + ( -1, -1, -1, 0x2c, 0x35, "s_mul_hi_i32"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in SOP2: + opcode(name, gfx9, gfx10, Format.SOP2) + + +# SOPK instructions: 0 input (+ imm), 1 output + optional scc +SOPK = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name + (0x00, 0x00, 0x00, 0x00, 0x00, "s_movk_i32"), + ( -1, -1, -1, -1, 0x01, "s_version"), # GFX10+ + (0x02, 0x02, 0x01, 0x01, 0x02, "s_cmovk_i32"), # GFX8_GFX9 + (0x03, 0x03, 0x02, 0x02, 0x03, "s_cmpk_eq_i32"), + (0x04, 0x04, 0x03, 0x03, 0x04, "s_cmpk_lg_i32"), + (0x05, 0x05, 0x04, 0x04, 0x05, "s_cmpk_gt_i32"), + (0x06, 0x06, 0x05, 0x05, 0x06, "s_cmpk_ge_i32"), + (0x07, 0x07, 0x06, 0x06, 0x07, "s_cmpk_lt_i32"), + (0x08, 0x08, 0x07, 0x07, 0x08, "s_cmpk_le_i32"), + (0x09, 0x09, 0x08, 0x08, 0x09, "s_cmpk_eq_u32"), + (0x0a, 0x0a, 0x09, 0x09, 0x0a, "s_cmpk_lg_u32"), + (0x0b, 0x0b, 0x0a, 0x0a, 0x0b, "s_cmpk_gt_u32"), + (0x0c, 0x0c, 0x0b, 0x0b, 0x0c, "s_cmpk_ge_u32"), + (0x0d, 0x0d, 0x0c, 0x0c, 0x0d, "s_cmpk_lt_u32"), + (0x0e, 0x0e, 0x0d, 0x0d, 0x0e, "s_cmpk_le_u32"), + (0x0f, 0x0f, 0x0e, 0x0e, 0x0f, "s_addk_i32"), + (0x10, 0x10, 0x0f, 0x0f, 0x10, "s_mulk_i32"), + (0x11, 0x11, 0x10, 0x10, -1, "s_cbranch_i_fork"), + (0x12, 0x12, 0x11, 0x11, 0x12, "s_getreg_b32"), + (0x13, 0x13, 0x12, 0x12, 0x13, "s_setreg_b32"), + (0x15, 0x15, 0x14, 0x14, 0x15, "s_setreg_imm32_b32"), # requires 32bit literal + ( -1, -1, 0x15, 0x15, 0x16, "s_call_b64"), + ( -1, -1, -1, -1, 0x17, "s_waitcnt_vscnt"), + ( -1, -1, -1, -1, 0x18, "s_waitcnt_vmcnt"), + ( -1, -1, -1, -1, 0x19, "s_waitcnt_expcnt"), + ( -1, -1, -1, -1, 0x1a, "s_waitcnt_lgkmcnt"), + ( -1, -1, -1, -1, 0x1b, "s_subvector_loop_begin"), + ( -1, -1, -1, -1, 0x1c, "s_subvector_loop_end"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in SOPK: + opcode(name, gfx9, gfx10, Format.SOPK) + + +# SOP1 instructions: 1 input, 1 output (+optional SCC) +SOP1 = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name + (0x03, 0x03, 0x00, 0x00, 0x03, "s_mov_b32"), + (0x04, 0x04, 0x01, 0x01, 0x04, "s_mov_b64"), + (0x05, 0x05, 0x02, 0x02, 0x05, "s_cmov_b32"), + (0x06, 0x06, 0x03, 0x03, 0x06, "s_cmov_b64"), + (0x07, 0x07, 0x04, 0x04, 0x07, "s_not_b32"), + (0x08, 0x08, 0x05, 0x05, 0x08, "s_not_b64"), + (0x09, 0x09, 0x06, 0x06, 0x09, "s_wqm_b32"), + (0x0a, 0x0a, 0x07, 0x07, 0x0a, "s_wqm_b64"), + (0x0b, 0x0b, 0x08, 0x08, 0x0b, "s_brev_b32"), + (0x0c, 0x0c, 0x09, 0x09, 0x0c, "s_brev_b64"), + (0x0d, 0x0d, 0x0a, 0x0a, 0x0d, "s_bcnt0_i32_b32"), + (0x0e, 0x0e, 0x0b, 0x0b, 0x0e, "s_bcnt0_i32_b64"), + (0x0f, 0x0f, 0x0c, 0x0c, 0x0f, "s_bcnt1_i32_b32"), + (0x10, 0x10, 0x0d, 0x0d, 0x10, "s_bcnt1_i32_b64"), + (0x11, 0x11, 0x0e, 0x0e, 0x11, "s_ff0_i32_b32"), + (0x12, 0x12, 0x0f, 0x0f, 0x12, "s_ff0_i32_b64"), + (0x13, 0x13, 0x10, 0x10, 0x13, "s_ff1_i32_b32"), + (0x14, 0x14, 0x11, 0x11, 0x14, "s_ff1_i32_b64"), + (0x15, 0x15, 0x12, 0x12, 0x15, "s_flbit_i32_b32"), + (0x16, 0x16, 0x13, 0x13, 0x16, "s_flbit_i32_b64"), + (0x17, 0x17, 0x14, 0x14, 0x17, "s_flbit_i32"), + (0x18, 0x18, 0x15, 0x15, 0x18, "s_flbit_i32_i64"), + (0x19, 0x19, 0x16, 0x16, 0x19, "s_sext_i32_i8"), + (0x1a, 0x1a, 0x17, 0x17, 0x1a, "s_sext_i32_i16"), + (0x1b, 0x1b, 0x18, 0x18, 0x1b, "s_bitset0_b32"), + (0x1c, 0x1c, 0x19, 0x19, 0x1c, "s_bitset0_b64"), + (0x1d, 0x1d, 0x1a, 0x1a, 0x1d, "s_bitset1_b32"), + (0x1e, 0x1e, 0x1b, 0x1b, 0x1e, "s_bitset1_b64"), + (0x1f, 0x1f, 0x1c, 0x1c, 0x1f, "s_getpc_b64"), + (0x20, 0x20, 0x1d, 0x1d, 0x20, "s_setpc_b64"), + (0x21, 0x21, 0x1e, 0x1e, 0x21, "s_swappc_b64"), + (0x22, 0x22, 0x1f, 0x1f, 0x22, "s_rfe_b64"), + (0x24, 0x24, 0x20, 0x20, 0x24, "s_and_saveexec_b64"), + (0x25, 0x25, 0x21, 0x21, 0x25, "s_or_saveexec_b64"), + (0x26, 0x26, 0x22, 0x22, 0x26, "s_xor_saveexec_b64"), + (0x27, 0x27, 0x23, 0x23, 0x27, "s_andn2_saveexec_b64"), + (0x28, 0x28, 0x24, 0x24, 0x28, "s_orn2_saveexec_b64"), + (0x29, 0x29, 0x25, 0x25, 0x29, "s_nand_saveexec_b64"), + (0x2a, 0x2a, 0x26, 0x26, 0x2a, "s_nor_saveexec_b64"), + (0x2b, 0x2b, 0x27, 0x27, 0x2b, "s_xnor_saveexec_b64"), + (0x2c, 0x2c, 0x28, 0x28, 0x2c, "s_quadmask_b32"), + (0x2d, 0x2d, 0x29, 0x29, 0x2d, "s_quadmask_b64"), + (0x2e, 0x2e, 0x2a, 0x2a, 0x2e, "s_movrels_b32"), + (0x2f, 0x2f, 0x2b, 0x2b, 0x2f, "s_movrels_b64"), + (0x30, 0x30, 0x2c, 0x2c, 0x30, "s_movreld_b32"), + (0x31, 0x31, 0x2d, 0x2d, 0x31, "s_movreld_b64"), + (0x32, 0x32, 0x2e, 0x2e, -1, "s_cbranch_join"), + (0x34, 0x34, 0x30, 0x30, 0x34, "s_abs_i32"), + (0x35, 0x35, -1, -1, 0x35, "s_mov_fed_b32"), + ( -1, -1, 0x32, 0x32, -1, "s_set_gpr_idx_idx"), + ( -1, -1, -1, 0x33, 0x37, "s_andn1_saveexec_b64"), + ( -1, -1, -1, 0x34, 0x38, "s_orn1_saveexec_b64"), + ( -1, -1, -1, 0x35, 0x39, "s_andn1_wrexec_b64"), + ( -1, -1, -1, 0x36, 0x3a, "s_andn2_wrexec_b64"), + ( -1, -1, -1, 0x37, 0x3b, "s_bitreplicate_b64_b32"), + ( -1, -1, -1, -1, 0x3c, "s_and_saveexec_b32"), + ( -1, -1, -1, -1, 0x3d, "s_or_saveexec_b32"), + ( -1, -1, -1, -1, 0x3e, "s_xor_saveexec_b32"), + ( -1, -1, -1, -1, 0x3f, "s_andn2_saveexec_b32"), + ( -1, -1, -1, -1, 0x40, "s_orn2_saveexec_b32"), + ( -1, -1, -1, -1, 0x41, "s_nand_saveexec_b32"), + ( -1, -1, -1, -1, 0x42, "s_nor_saveexec_b32"), + ( -1, -1, -1, -1, 0x43, "s_xnor_saveexec_b32"), + ( -1, -1, -1, -1, 0x44, "s_andn1_saveexec_b32"), + ( -1, -1, -1, -1, 0x45, "s_orn1_saveexec_b32"), + ( -1, -1, -1, -1, 0x46, "s_andn1_wrexec_b32"), + ( -1, -1, -1, -1, 0x47, "s_andn2_wrexec_b32"), + ( -1, -1, -1, -1, 0x49, "s_movrelsd_2_b32"), + # actually a pseudo-instruction. it's lowered to SALU during assembly though, so it's useful to identify it as a SOP1. + ( -1, -1, -1, -1, -1, "p_constaddr"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in SOP1: + opcode(name, gfx9, gfx10, Format.SOP1) + + +# SOPC instructions: 2 inputs and 0 outputs (+SCC) +SOPC = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name + (0x00, 0x00, 0x00, 0x00, 0x00, "s_cmp_eq_i32"), + (0x01, 0x01, 0x01, 0x01, 0x01, "s_cmp_lg_i32"), + (0x02, 0x02, 0x02, 0x02, 0x02, "s_cmp_gt_i32"), + (0x03, 0x03, 0x03, 0x03, 0x03, "s_cmp_ge_i32"), + (0x04, 0x04, 0x04, 0x04, 0x04, "s_cmp_lt_i32"), + (0x05, 0x05, 0x05, 0x05, 0x05, "s_cmp_le_i32"), + (0x06, 0x06, 0x06, 0x06, 0x06, "s_cmp_eq_u32"), + (0x07, 0x07, 0x07, 0x07, 0x07, "s_cmp_lg_u32"), + (0x08, 0x08, 0x08, 0x08, 0x08, "s_cmp_gt_u32"), + (0x09, 0x09, 0x09, 0x09, 0x09, "s_cmp_ge_u32"), + (0x0a, 0x0a, 0x0a, 0x0a, 0x0a, "s_cmp_lt_u32"), + (0x0b, 0x0b, 0x0b, 0x0b, 0x0b, "s_cmp_le_u32"), + (0x0c, 0x0c, 0x0c, 0x0c, 0x0c, "s_bitcmp0_b32"), + (0x0d, 0x0d, 0x0d, 0x0d, 0x0d, "s_bitcmp1_b32"), + (0x0e, 0x0e, 0x0e, 0x0e, 0x0e, "s_bitcmp0_b64"), + (0x0f, 0x0f, 0x0f, 0x0f, 0x0f, "s_bitcmp1_b64"), + (0x10, 0x10, 0x10, 0x10, -1, "s_setvskip"), + ( -1, -1, 0x11, 0x11, -1, "s_set_gpr_idx_on"), + ( -1, -1, 0x12, 0x12, 0x12, "s_cmp_eq_u64"), + ( -1, -1, 0x13, 0x13, 0x13, "s_cmp_lg_u64"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in SOPC: + opcode(name, gfx9, gfx10, Format.SOPC) + + +# SOPP instructions: 0 inputs (+optional scc/vcc), 0 outputs +SOPP = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name + (0x00, 0x00, 0x00, 0x00, 0x00, "s_nop"), + (0x01, 0x01, 0x01, 0x01, 0x01, "s_endpgm"), + (0x02, 0x02, 0x02, 0x02, 0x02, "s_branch"), + ( -1, -1, 0x03, 0x03, 0x03, "s_wakeup"), + (0x04, 0x04, 0x04, 0x04, 0x04, "s_cbranch_scc0"), + (0x05, 0x05, 0x05, 0x05, 0x05, "s_cbranch_scc1"), + (0x06, 0x06, 0x06, 0x06, 0x06, "s_cbranch_vccz"), + (0x07, 0x07, 0x07, 0x07, 0x07, "s_cbranch_vccnz"), + (0x08, 0x08, 0x08, 0x08, 0x08, "s_cbranch_execz"), + (0x09, 0x09, 0x09, 0x09, 0x09, "s_cbranch_execnz"), + (0x0a, 0x0a, 0x0a, 0x0a, 0x0a, "s_barrier"), + ( -1, 0x0b, 0x0b, 0x0b, 0x0b, "s_setkill"), + (0x0c, 0x0c, 0x0c, 0x0c, 0x0c, "s_waitcnt"), + (0x0d, 0x0d, 0x0d, 0x0d, 0x0d, "s_sethalt"), + (0x0e, 0x0e, 0x0e, 0x0e, 0x0e, "s_sleep"), + (0x0f, 0x0f, 0x0f, 0x0f, 0x0f, "s_setprio"), + (0x10, 0x10, 0x10, 0x10, 0x10, "s_sendmsg"), + (0x11, 0x11, 0x11, 0x11, 0x11, "s_sendmsghalt"), + (0x12, 0x12, 0x12, 0x12, 0x12, "s_trap"), + (0x13, 0x13, 0x13, 0x13, 0x13, "s_icache_inv"), + (0x14, 0x14, 0x14, 0x14, 0x14, "s_incperflevel"), + (0x15, 0x15, 0x15, 0x15, 0x15, "s_decperflevel"), + (0x16, 0x16, 0x16, 0x16, 0x16, "s_ttracedata"), + ( -1, 0x17, 0x17, 0x17, 0x17, "s_cbranch_cdbgsys"), + ( -1, 0x18, 0x18, 0x18, 0x18, "s_cbranch_cdbguser"), + ( -1, 0x19, 0x19, 0x19, 0x19, "s_cbranch_cdbgsys_or_user"), + ( -1, 0x1a, 0x1a, 0x1a, 0x1a, "s_cbranch_cdbgsys_and_user"), + ( -1, -1, 0x1b, 0x1b, 0x1b, "s_endpgm_saved"), + ( -1, -1, 0x1c, 0x1c, -1, "s_set_gpr_idx_off"), + ( -1, -1, 0x1d, 0x1d, -1, "s_set_gpr_idx_mode"), + ( -1, -1, -1, 0x1e, 0x1e, "s_endpgm_ordered_ps_done"), + ( -1, -1, -1, -1, 0x1f, "s_code_end"), + ( -1, -1, -1, -1, 0x20, "s_inst_prefetch"), + ( -1, -1, -1, -1, 0x21, "s_clause"), + ( -1, -1, -1, -1, 0x22, "s_wait_idle"), + ( -1, -1, -1, -1, 0x23, "s_waitcnt_depctr"), + ( -1, -1, -1, -1, 0x24, "s_round_mode"), + ( -1, -1, -1, -1, 0x25, "s_denorm_mode"), + ( -1, -1, -1, -1, 0x26, "s_ttracedata_imm"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in SOPP: + opcode(name, gfx9, gfx10, Format.SOPP) + + +# SMEM instructions: sbase input (2 sgpr), potentially 2 offset inputs, 1 sdata input/output +SMEM = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name + (0x00, 0x00, 0x00, 0x00, 0x00, "s_load_dword"), + (0x01, 0x01, 0x01, 0x01, 0x01, "s_load_dwordx2"), + (0x02, 0x02, 0x02, 0x02, 0x02, "s_load_dwordx4"), + (0x03, 0x03, 0x03, 0x03, 0x03, "s_load_dwordx8"), + (0x04, 0x04, 0x04, 0x04, 0x04, "s_load_dwordx16"), + ( -1, -1, -1, 0x05, 0x05, "s_scratch_load_dword"), + ( -1, -1, -1, 0x06, 0x06, "s_scratch_load_dwordx2"), + ( -1, -1, -1, 0x07, 0x07, "s_scratch_load_dwordx4"), + (0x08, 0x08, 0x08, 0x08, 0x08, "s_buffer_load_dword"), + (0x09, 0x09, 0x09, 0x09, 0x09, "s_buffer_load_dwordx2"), + (0x0a, 0x0a, 0x0a, 0x0a, 0x0a, "s_buffer_load_dwordx4"), + (0x0b, 0x0b, 0x0b, 0x0b, 0x0b, "s_buffer_load_dwordx8"), + (0x0c, 0x0c, 0x0c, 0x0c, 0x0c, "s_buffer_load_dwordx16"), + ( -1, -1, 0x10, 0x10, 0x10, "s_store_dword"), + ( -1, -1, 0x11, 0x11, 0x11, "s_store_dwordx2"), + ( -1, -1, 0x12, 0x12, 0x12, "s_store_dwordx4"), + ( -1, -1, -1, 0x15, 0x15, "s_scratch_store_dword"), + ( -1, -1, -1, 0x16, 0x16, "s_scratch_store_dwordx2"), + ( -1, -1, -1, 0x17, 0x17, "s_scratch_store_dwordx4"), + ( -1, -1, 0x18, 0x18, 0x18, "s_buffer_store_dword"), + ( -1, -1, 0x19, 0x19, 0x19, "s_buffer_store_dwordx2"), + ( -1, -1, 0x1a, 0x1a, 0x1a, "s_buffer_store_dwordx4"), + ( -1, -1, 0x1f, 0x1f, 0x1f, "s_gl1_inv"), + (0x1f, 0x1f, 0x20, 0x20, 0x20, "s_dcache_inv"), + ( -1, -1, 0x21, 0x21, 0x21, "s_dcache_wb"), + ( -1, 0x1d, 0x22, 0x22, -1, "s_dcache_inv_vol"), + ( -1, -1, 0x23, 0x23, -1, "s_dcache_wb_vol"), + (0x1e, 0x1e, 0x24, 0x24, 0x24, "s_memtime"), + ( -1, -1, 0x25, 0x25, 0x25, "s_memrealtime"), + ( -1, -1, 0x26, 0x26, 0x26, "s_atc_probe"), + ( -1, -1, 0x27, 0x27, 0x27, "s_atc_probe_buffer"), + ( -1, -1, -1, 0x28, 0x28, "s_dcache_discard"), + ( -1, -1, -1, 0x29, 0x29, "s_dcache_discard_x2"), + ( -1, -1, -1, -1, 0x2a, "s_get_waveid_in_workgroup"), + ( -1, -1, -1, 0x40, 0x40, "s_buffer_atomic_swap"), + ( -1, -1, -1, 0x41, 0x41, "s_buffer_atomic_cmpswap"), + ( -1, -1, -1, 0x42, 0x42, "s_buffer_atomic_add"), + ( -1, -1, -1, 0x43, 0x43, "s_buffer_atomic_sub"), + ( -1, -1, -1, 0x44, 0x44, "s_buffer_atomic_smin"), + ( -1, -1, -1, 0x45, 0x45, "s_buffer_atomic_umin"), + ( -1, -1, -1, 0x46, 0x46, "s_buffer_atomic_smax"), + ( -1, -1, -1, 0x47, 0x47, "s_buffer_atomic_umax"), + ( -1, -1, -1, 0x48, 0x48, "s_buffer_atomic_and"), + ( -1, -1, -1, 0x49, 0x49, "s_buffer_atomic_or"), + ( -1, -1, -1, 0x4a, 0x4a, "s_buffer_atomic_xor"), + ( -1, -1, -1, 0x4b, 0x4b, "s_buffer_atomic_inc"), + ( -1, -1, -1, 0x4c, 0x4c, "s_buffer_atomic_dec"), + ( -1, -1, -1, 0x60, 0x60, "s_buffer_atomic_swap_x2"), + ( -1, -1, -1, 0x61, 0x61, "s_buffer_atomic_cmpswap_x2"), + ( -1, -1, -1, 0x62, 0x62, "s_buffer_atomic_add_x2"), + ( -1, -1, -1, 0x63, 0x63, "s_buffer_atomic_sub_x2"), + ( -1, -1, -1, 0x64, 0x64, "s_buffer_atomic_smin_x2"), + ( -1, -1, -1, 0x65, 0x65, "s_buffer_atomic_umin_x2"), + ( -1, -1, -1, 0x66, 0x66, "s_buffer_atomic_smax_x2"), + ( -1, -1, -1, 0x67, 0x67, "s_buffer_atomic_umax_x2"), + ( -1, -1, -1, 0x68, 0x68, "s_buffer_atomic_and_x2"), + ( -1, -1, -1, 0x69, 0x69, "s_buffer_atomic_or_x2"), + ( -1, -1, -1, 0x6a, 0x6a, "s_buffer_atomic_xor_x2"), + ( -1, -1, -1, 0x6b, 0x6b, "s_buffer_atomic_inc_x2"), + ( -1, -1, -1, 0x6c, 0x6c, "s_buffer_atomic_dec_x2"), + ( -1, -1, -1, 0x80, 0x80, "s_atomic_swap"), + ( -1, -1, -1, 0x81, 0x81, "s_atomic_cmpswap"), + ( -1, -1, -1, 0x82, 0x82, "s_atomic_add"), + ( -1, -1, -1, 0x83, 0x83, "s_atomic_sub"), + ( -1, -1, -1, 0x84, 0x84, "s_atomic_smin"), + ( -1, -1, -1, 0x85, 0x85, "s_atomic_umin"), + ( -1, -1, -1, 0x86, 0x86, "s_atomic_smax"), + ( -1, -1, -1, 0x87, 0x87, "s_atomic_umax"), + ( -1, -1, -1, 0x88, 0x88, "s_atomic_and"), + ( -1, -1, -1, 0x89, 0x89, "s_atomic_or"), + ( -1, -1, -1, 0x8a, 0x8a, "s_atomic_xor"), + ( -1, -1, -1, 0x8b, 0x8b, "s_atomic_inc"), + ( -1, -1, -1, 0x8c, 0x8c, "s_atomic_dec"), + ( -1, -1, -1, 0xa0, 0xa0, "s_atomic_swap_x2"), + ( -1, -1, -1, 0xa1, 0xa1, "s_atomic_cmpswap_x2"), + ( -1, -1, -1, 0xa2, 0xa2, "s_atomic_add_x2"), + ( -1, -1, -1, 0xa3, 0xa3, "s_atomic_sub_x2"), + ( -1, -1, -1, 0xa4, 0xa4, "s_atomic_smin_x2"), + ( -1, -1, -1, 0xa5, 0xa5, "s_atomic_umin_x2"), + ( -1, -1, -1, 0xa6, 0xa6, "s_atomic_smax_x2"), + ( -1, -1, -1, 0xa7, 0xa7, "s_atomic_umax_x2"), + ( -1, -1, -1, 0xa8, 0xa8, "s_atomic_and_x2"), + ( -1, -1, -1, 0xa9, 0xa9, "s_atomic_or_x2"), + ( -1, -1, -1, 0xaa, 0xaa, "s_atomic_xor_x2"), + ( -1, -1, -1, 0xab, 0xab, "s_atomic_inc_x2"), + ( -1, -1, -1, 0xac, 0xac, "s_atomic_dec_x2"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in SMEM: + opcode(name, gfx9, gfx10, Format.SMEM) + + +# VOP2 instructions: 2 inputs, 1 output (+ optional vcc) +# TODO: misses some GFX6_7 opcodes which were shifted to VOP3 in GFX8 +VOP2 = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name, input/output modifiers + (0x00, 0x00, 0x00, 0x00, 0x01, "v_cndmask_b32", False), + (0x03, 0x03, 0x01, 0x01, 0x03, "v_add_f32", True), + (0x04, 0x04, 0x02, 0x02, 0x04, "v_sub_f32", True), + (0x05, 0x05, 0x03, 0x03, 0x05, "v_subrev_f32", True), + (0x06, 0x06, -1, -1, 0x06, "v_mac_legacy_f32", True), + (0x07, 0x07, 0x04, 0x04, 0x07, "v_mul_legacy_f32", True), + (0x08, 0x08, 0x05, 0x05, 0x08, "v_mul_f32", True), + (0x09, 0x09, 0x06, 0x06, 0x09, "v_mul_i32_i24", False), + (0x0a, 0x0a, 0x07, 0x07, 0x0a, "v_mul_hi_i32_i24", False), + (0x0b, 0x0b, 0x08, 0x08, 0x0b, "v_mul_u32_u24", False), + (0x0c, 0x0c, 0x09, 0x09, 0x0c, "v_mul_hi_u32_u24", False), + (0x0d, 0x0d, -1, -1, -1, "v_min_legacy_f32", True), + (0x0e, 0x0e, -1, -1, -1, "v_max_legacy_f32", True), + (0x0f, 0x0f, 0x0a, 0x0a, 0x0f, "v_min_f32", True), + (0x10, 0x10, 0x0b, 0x0b, 0x10, "v_max_f32", True), + (0x11, 0x11, 0x0c, 0x0c, 0x11, "v_min_i32", False), + (0x12, 0x12, 0x0d, 0x0d, 0x12, "v_max_i32", False), + (0x13, 0x13, 0x0e, 0x0e, 0x13, "v_min_u32", False), + (0x14, 0x14, 0x0f, 0x0f, 0x14, "v_max_u32", False), + (0x15, 0x15, -1, -1, -1, "v_lshr_b32", False), + (0x16, 0x16, 0x10, 0x10, 0x16, "v_lshrrev_b32", False), + (0x17, 0x17, -1, -1, -1, "v_ashr_i32", False), + (0x18, 0x18, 0x11, 0x11, 0x18, "v_ashrrev_i32", False), + (0x19, 0x19, -1, -1, -1, "v_lshl_b32", False), + (0x1a, 0x1a, 0x12, 0x12, 0x1a, "v_lshlrev_b32", False), + (0x1b, 0x1b, 0x13, 0x13, 0x1b, "v_and_b32", False), + (0x1c, 0x1c, 0x14, 0x14, 0x1c, "v_or_b32", False), + (0x1d, 0x1d, 0x15, 0x15, 0x1d, "v_xor_b32", False), + ( -1, -1, -1, -1, 0x1e, "v_xnor_b32", False), + (0x1f, 0x1f, 0x16, 0x16, 0x1f, "v_mac_f32", True), + (0x20, 0x20, 0x17, 0x17, 0x20, "v_madmk_f32", False), + (0x21, 0x21, 0x18, 0x18, 0x21, "v_madak_f32", False), + (0x25, 0x25, 0x19, 0x19, -1, "v_add_co_u32", False), # VOP3B only in RDNA + (0x26, 0x26, 0x1a, 0x1a, -1, "v_sub_co_u32", False), # VOP3B only in RDNA + (0x27, 0x27, 0x1b, 0x1b, -1, "v_subrev_co_u32", False), # VOP3B only in RDNA + (0x28, 0x28, 0x1c, 0x1c, 0x28, "v_addc_co_u32", False), # v_add_co_ci_u32 in RDNA + (0x29, 0x29, 0x1d, 0x1d, 0x29, "v_subb_co_u32", False), # v_sub_co_ci_u32 in RDNA + (0x2a, 0x2a, 0x1e, 0x1e, 0x2a, "v_subbrev_co_u32", False), # v_subrev_co_ci_u32 in RDNA + ( -1, -1, -1, -1, 0x2b, "v_fmac_f32", True), + ( -1, -1, -1, -1, 0x2c, "v_fmamk_f32", True), + ( -1, -1, -1, -1, 0x2d, "v_fmaak_f32", True), + ( -1, -1, 0x1f, 0x1f, 0x32, "v_add_f16", True), + ( -1, -1, 0x20, 0x20, 0x33, "v_sub_f16", True), + ( -1, -1, 0x21, 0x21, 0x34, "v_subrev_f16", True), + ( -1, -1, 0x22, 0x22, 0x35, "v_mul_f16", True), + ( -1, -1, 0x23, 0x23, -1, "v_mac_f16", True), + ( -1, -1, 0x24, 0x24, -1, "v_madmk_f16", False), + ( -1, -1, 0x25, 0x25, -1, "v_madak_f16", False), + ( -1, -1, 0x26, 0x26, -1, "v_add_u16", False), + ( -1, -1, 0x27, 0x27, -1, "v_sub_u16", False), + ( -1, -1, 0x28, 0x28, -1, "v_subrev_u16", False), + ( -1, -1, 0x29, 0x29, -1, "v_mul_lo_u16", False), + ( -1, -1, 0x2a, 0x2a, -1, "v_lshlrev_b16", False), + ( -1, -1, 0x2b, 0x2b, -1, "v_lshrrev_b16", False), + ( -1, -1, 0x2c, 0x2c, -1, "v_ashrrev_b16", False), + ( -1, -1, 0x2d, 0x2d, 0x39, "v_max_f16", True), + ( -1, -1, 0x2e, 0x2e, 0x3a, "v_min_f16", True), + ( -1, -1, 0x2f, 0x2f, -1, "v_max_u16", False), + ( -1, -1, 0x30, 0x30, -1, "v_max_i16", False), + ( -1, -1, 0x31, 0x31, -1, "v_min_u16", False), + ( -1, -1, 0x32, 0x32, -1, "v_min_i16", False), + ( -1, -1, 0x33, 0x33, 0x3b, "v_ldexp_f16", False), + ( -1, -1, 0x34, 0x34, 0x25, "v_add_u32", False), # v_add_nc_u32 in RDNA + ( -1, -1, 0x35, 0x35, 0x26, "v_sub_u32", False), # v_sub_nc_u32 in RDNA + ( -1, -1, 0x36, 0x36, 0x27, "v_subrev_u32", False), # v_subrev_nc_u32 in RDNA + ( -1, -1, -1, -1, 0x36, "v_fmac_f16", False), + ( -1, -1, -1, -1, 0x37, "v_fmamk_f16", False), + ( -1, -1, -1, -1, 0x38, "v_fmaak_f16", False), + ( -1, -1, -1, -1, 0x3c, "v_pk_fmac_f16", False), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name, modifiers) in VOP2: + opcode(name, gfx9, gfx10, Format.VOP2, modifiers, modifiers) + + +# VOP1 instructions: instructions with 1 input and 1 output +VOP1 = { + # GFX6, GFX7, GFX8, GFX9, GFX10, name, input_modifiers, output_modifiers + (0x00, 0x00, 0x00, 0x00, 0x00, "v_nop", False, False), + (0x01, 0x01, 0x01, 0x01, 0x01, "v_mov_b32", False, False), + (0x02, 0x02, 0x02, 0x02, 0x02, "v_readfirstlane_b32", False, False), + (0x03, 0x03, 0x03, 0x03, 0x03, "v_cvt_i32_f64", True, False), + (0x04, 0x04, 0x04, 0x04, 0x04, "v_cvt_f64_i32", False, True), + (0x05, 0x05, 0x05, 0x05, 0x05, "v_cvt_f32_i32", False, True), + (0x06, 0x06, 0x06, 0x06, 0x06, "v_cvt_f32_u32", False, True), + (0x07, 0x07, 0x07, 0x07, 0x07, "v_cvt_u32_f32", True, False), + (0x08, 0x08, 0x08, 0x08, 0x08, "v_cvt_i32_f32", True, False), + (0x09, 0x09, -1, -1, 0x09, "v_mov_fed_b32", True, False), # LLVM mentions it for GFX8_9 + (0x0a, 0x0a, 0x0a, 0x0a, 0x0a, "v_cvt_f16_f32", True, True), + (0x0b, 0x0b, 0x0b, 0x0b, 0x0b, "v_cvt_f32_f16", True, True), + (0x0c, 0x0c, 0x0c, 0x0c, 0x0c, "v_cvt_rpi_i32_f32", True, False), + (0x0d, 0x0d, 0x0d, 0x0d, 0x0d, "v_cvt_flr_i32_f32", True, False), + (0x0e, 0x0e, 0x0e, 0x0e, 0x0e, "v_cvt_off_f32_i4", False, True), + (0x0f, 0x0f, 0x0f, 0x0f, 0x0f, "v_cvt_f32_f64", True, True), + (0x10, 0x10, 0x10, 0x10, 0x10, "v_cvt_f64_f32", True, True), + (0x11, 0x11, 0x11, 0x11, 0x11, "v_cvt_f32_ubyte0", False, True), + (0x12, 0x12, 0x12, 0x12, 0x12, "v_cvt_f32_ubyte1", False, True), + (0x13, 0x13, 0x13, 0x13, 0x13, "v_cvt_f32_ubyte2", False, True), + (0x14, 0x14, 0x14, 0x14, 0x14, "v_cvt_f32_ubyte3", False, True), + (0x15, 0x15, 0x15, 0x15, 0x15, "v_cvt_u32_f64", True, False), + (0x16, 0x16, 0x16, 0x16, 0x16, "v_cvt_f64_u32", False, True), + ( -1, 0x17, 0x17, 0x17, 0x17, "v_trunc_f64", True, True), + ( -1, 0x18, 0x18, 0x18, 0x18, "v_ceil_f64", True, True), + ( -1, 0x19, 0x19, 0x19, 0x19, "v_rndne_f64", True, True), + ( -1, 0x1a, 0x1a, 0x1a, 0x1a, "v_floor_f64", True, True), + ( -1, -1, -1, -1, 0x1b, "v_pipeflush", False, False), + (0x20, 0x20, 0x1b, 0x1b, 0x20, "v_fract_f32", True, True), + (0x21, 0x21, 0x1c, 0x1c, 0x21, "v_trunc_f32", True, True), + (0x22, 0x22, 0x1d, 0x1d, 0x22, "v_ceil_f32", True, True), + (0x23, 0x23, 0x1e, 0x1e, 0x23, "v_rndne_f32", True, True), + (0x24, 0x24, 0x1f, 0x1f, 0x24, "v_floor_f32", True, True), + (0x25, 0x25, 0x20, 0x20, 0x25, "v_exp_f32", True, True), + (0x26, 0x26, -1, -1, -1, "v_log_clamp_f32", True, True), + (0x27, 0x27, 0x21, 0x21, 0x27, "v_log_f32", True, True), + (0x28, 0x28, -1, -1, -1, "v_rcp_clamp_f32", True, True), + (0x29, 0x29, -1, -1, -1, "v_rcp_legacy_f32", True, True), + (0x2a, 0x2a, 0x22, 0x22, 0x2a, "v_rcp_f32", True, True), + (0x2b, 0x2b, 0x23, 0x23, 0x2b, "v_rcp_iflag_f32", True, True), + (0x2c, 0x2c, -1, -1, -1, "v_rsq_clamp_f32", True, True), + (0x2d, 0x2d, -1, -1, -1, "v_rsq_legacy_f32", True, True), + (0x2e, 0x2e, 0x24, 0x24, 0x2e, "v_rsq_f32", True, True), + (0x2f, 0x2f, 0x25, 0x25, 0x2f, "v_rcp_f64", True, True), + (0x30, 0x30, -1, -1, -1, "v_rcp_clamp_f64", True, True), + (0x31, 0x31, 0x26, 0x26, 0x31, "v_rsq_f64", True, True), + (0x32, 0x32, -1, -1, -1, "v_rsq_clamp_f64", True, True), + (0x33, 0x33, 0x27, 0x27, 0x33, "v_sqrt_f32", True, True), + (0x34, 0x34, 0x28, 0x28, 0x34, "v_sqrt_f64", True, True), + (0x35, 0x35, 0x29, 0x29, 0x35, "v_sin_f32", True, True), + (0x36, 0x36, 0x2a, 0x2a, 0x36, "v_cos_f32", True, True), + (0x37, 0x37, 0x2b, 0x2b, 0x37, "v_not_b32", False, False), + (0x38, 0x38, 0x2c, 0x2c, 0x38, "v_bfrev_b32", False, False), + (0x39, 0x39, 0x2d, 0x2d, 0x39, "v_ffbh_u32", False, False), + (0x3a, 0x3a, 0x2e, 0x2e, 0x3a, "v_ffbl_b32", False, False), + (0x3b, 0x3b, 0x2f, 0x2f, 0x3b, "v_ffbh_i32", False, False), + (0x3c, 0x3c, 0x30, 0x30, 0x3c, "v_frexp_exp_i32_f64", True, False), + (0x3d, 0x3d, 0x31, 0x31, 0x3d, "v_frexp_mant_f64", True, False), + (0x3e, 0x3e, 0x32, 0x32, 0x3e, "v_fract_f64", True, True), + (0x3f, 0x3f, 0x33, 0x33, 0x3f, "v_frexp_exp_i32_f32", True, False), + (0x40, 0x40, 0x34, 0x34, 0x40, "v_frexp_mant_f32", True, False), + (0x41, 0x41, 0x35, 0x35, 0x41, "v_clrexcp", False, False), + (0x42, 0x42, 0x36, -1, 0x42, "v_movreld_b32", False, False), + (0x43, 0x43, 0x37, -1, 0x43, "v_movrels_b32", False, False), + (0x44, 0x44, 0x38, -1, 0x44, "v_movrelsd_b32", False, False), + ( -1, -1, -1, -1, 0x48, "v_movrelsd_2_b32", False, False), + ( -1, -1, -1, 0x37, -1, "v_screen_partition_4se_b32", False, False), + ( -1, -1, 0x39, 0x39, 0x50, "v_cvt_f16_u16", False, True), + ( -1, -1, 0x3a, 0x3a, 0x51, "v_cvt_f16_i16", False, True), + ( -1, -1, 0x3b, 0x3b, 0x52, "v_cvt_u16_f16", True, False), + ( -1, -1, 0x3c, 0x3c, 0x53, "v_cvt_i16_f16", True, False), + ( -1, -1, 0x3d, 0x3d, 0x54, "v_rcp_f16", True, True), + ( -1, -1, 0x3e, 0x3e, 0x55, "v_sqrt_f16", True, True), + ( -1, -1, 0x3f, 0x3f, 0x56, "v_rsq_f16", True, True), + ( -1, -1, 0x40, 0x40, 0x57, "v_log_f16", True, True), + ( -1, -1, 0x41, 0x41, 0x58, "v_exp_f16", True, True), + ( -1, -1, 0x42, 0x42, 0x59, "v_frexp_mant_f16", True, False), + ( -1, -1, 0x43, 0x43, 0x5a, "v_frexp_exp_i16_f16", True, False), + ( -1, -1, 0x44, 0x44, 0x5b, "v_floor_f16", True, True), + ( -1, -1, 0x45, 0x45, 0x5c, "v_ceil_f16", True, True), + ( -1, -1, 0x46, 0x46, 0x5d, "v_trunc_f16", True, True), + ( -1, -1, 0x47, 0x47, 0x5e, "v_rndne_f16", True, True), + ( -1, -1, 0x48, 0x48, 0x5f, "v_fract_f16", True, True), + ( -1, -1, 0x49, 0x49, 0x60, "v_sin_f16", True, True), + ( -1, -1, 0x4a, 0x4a, 0x61, "v_cos_f16", True, True), + ( -1, 0x46, 0x4b, 0x4b, -1, "v_exp_legacy_f32", True, True), + ( -1, 0x45, 0x4c, 0x4c, -1, "v_log_legacy_f32", True, True), + ( -1, -1, -1, 0x4f, 0x62, "v_sat_pk_u8_i16", False, False), + ( -1, -1, -1, 0x4d, 0x63, "v_cvt_norm_i16_f16", True, False), + ( -1, -1, -1, 0x4e, 0x64, "v_cvt_norm_u16_f16", True, False), + ( -1, -1, -1, 0x51, 0x65, "v_swap_b32", False, False), + ( -1, -1, -1, -1, 0x68, "v_swaprel_b32", False, False), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name, in_mod, out_mod) in VOP1: + opcode(name, gfx9, gfx10, Format.VOP1, in_mod, out_mod) + + +# VOPC instructions: + +VOPC_CLASS = { + (0x88, 0x88, 0x10, 0x10, 0x88, "v_cmp_class_f32"), + ( -1, -1, 0x14, 0x14, 0x8f, "v_cmp_class_f16"), + (0x98, 0x98, 0x11, 0x11, 0x98, "v_cmpx_class_f32"), + ( -1, -1, 0x15, 0x15, 0x9f, "v_cmpx_class_f16"), + (0xa8, 0xa8, 0x12, 0x12, 0xa8, "v_cmp_class_f64"), + (0xb8, 0xb8, 0x13, 0x13, 0xb8, "v_cmpx_class_f64"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in VOPC_CLASS: + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + +COMPF = ["f", "lt", "eq", "le", "gt", "lg", "ge", "o", "u", "nge", "nlg", "ngt", "nle", "neq", "nlt", "tru"] + +for i in range(8): + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0x20+i, 0x20+i, 0xc8+i, "v_cmp_"+COMPF[i]+"_f16") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0x30+i, 0x30+i, 0xd8+i, "v_cmpx_"+COMPF[i]+"_f16") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0x28+i, 0x28+i, 0xe8+i, "v_cmp_"+COMPF[i+8]+"_f16") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0x38+i, 0x38+i, 0xf8+i, "v_cmpx_"+COMPF[i+8]+"_f16") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + +for i in range(16): + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x00+i, 0x00+i, 0x40+i, 0x40+i, 0x00+i, "v_cmp_"+COMPF[i]+"_f32") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x10+i, 0x10+i, 0x50+i, 0x50+i, 0x10+i, "v_cmpx_"+COMPF[i]+"_f32") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x20+i, 0x20+i, 0x60+i, 0x60+i, 0x20+i, "v_cmp_"+COMPF[i]+"_f64") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x30+i, 0x30+i, 0x70+i, 0x70+i, 0x30+i, "v_cmpx_"+COMPF[i]+"_f64") + opcode(name, gfx9, gfx10, Format.VOPC, True, False) + # GFX_6_7 + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x40+i, 0x40+i, -1, -1, -1, "v_cmps_"+COMPF[i]+"_f32") + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x50+i, 0x50+i, -1, -1, -1, "v_cmpsx_"+COMPF[i]+"_f32") + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x60+i, 0x60+i, -1, -1, -1, "v_cmps_"+COMPF[i]+"_f64") + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x70+i, 0x70+i, -1, -1, -1, "v_cmpsx_"+COMPF[i]+"_f64") + +COMPI = ["f", "lt", "eq", "le", "gt", "lg", "ge", "tru"] + +# GFX_8_9 +for i in [0,7]: # only 0 and 7 + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xa0+i, 0xa0+i, -1, "v_cmp_"+COMPI[i]+"_i16") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xb0+i, 0xb0+i, -1, "v_cmpx_"+COMPI[i]+"_i16") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xa8+i, 0xa8+i, -1, "v_cmp_"+COMPI[i]+"_u16") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xb8+i, 0xb8+i, -1, "v_cmpx_"+COMPI[i]+"_u16") + opcode(name, gfx9, gfx10, Format.VOPC) + +for i in range(1, 7): # [1..6] + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xa0+i, 0xa0+i, 0x88+i, "v_cmp_"+COMPI[i]+"_i16") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xb0+i, 0xb0+i, 0x98+i, "v_cmpx_"+COMPI[i]+"_i16") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xa8+i, 0xa8+i, 0xa8+i, "v_cmp_"+COMPI[i]+"_u16") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, 0xb8+i, 0xb8+i, 0xb8+i, "v_cmpx_"+COMPI[i]+"_u16") + opcode(name, gfx9, gfx10, Format.VOPC) + +for i in range(8): + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x80+i, 0x80+i, 0xc0+i, 0xc0+i, 0x80+i, "v_cmp_"+COMPI[i]+"_i32") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0x90+i, 0x90+i, 0xd0+i, 0xd0+i, 0x90+i, "v_cmpx_"+COMPI[i]+"_i32") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0xa0+i, 0xa0+i, 0xe0+i, 0xe0+i, 0xa0+i, "v_cmp_"+COMPI[i]+"_i64") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0xb0+i, 0xb0+i, 0xf0+i, 0xf0+i, 0xb0+i, "v_cmpx_"+COMPI[i]+"_i64") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0xc0+i, 0xc0+i, 0xc8+i, 0xc8+i, 0xc0+i, "v_cmp_"+COMPI[i]+"_u32") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0xd0+i, 0xd0+i, 0xd8+i, 0xd8+i, 0xd0+i, "v_cmpx_"+COMPI[i]+"_u32") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0xe0+i, 0xe0+i, 0xe8+i, 0xe8+i, 0xe0+i, "v_cmp_"+COMPI[i]+"_u64") + opcode(name, gfx9, gfx10, Format.VOPC) + (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (0xf0+i, 0xf0+i, 0xf8+i, 0xf8+i, 0xf0+i, "v_cmpx_"+COMPI[i]+"_u64") + opcode(name, gfx9, gfx10, Format.VOPC) + + +# VOPP instructions: packed 16bit instructions - 1 or 2 inputs and 1 output +VOPP = { + (0x00, "v_pk_mad_i16"), + (0x01, "v_pk_mul_lo_u16"), + (0x02, "v_pk_add_i16"), + (0x03, "v_pk_sub_i16"), + (0x04, "v_pk_lshlrev_b16"), + (0x05, "v_pk_lshrrev_b16"), + (0x06, "v_pk_ashrrev_i16"), + (0x07, "v_pk_max_i16"), + (0x08, "v_pk_min_i16"), + (0x09, "v_pk_mad_u16"), + (0x0a, "v_pk_add_u16"), + (0x0b, "v_pk_sub_u16"), + (0x0c, "v_pk_max_u16"), + (0x0d, "v_pk_min_u16"), + (0x0e, "v_pk_fma_f16"), + (0x0f, "v_pk_add_f16"), + (0x10, "v_pk_mul_f16"), + (0x11, "v_pk_min_f16"), + (0x12, "v_pk_max_f16"), + (0x20, "v_pk_fma_mix_f32"), # v_mad_mix_f32 in VEGA ISA, v_fma_mix_f32 in RDNA ISA + (0x21, "v_pk_fma_mixlo_f16"), # v_mad_mixlo_f16 in VEGA ISA, v_fma_mixlo_f16 in RDNA ISA + (0x22, "v_pk_fma_mixhi_f16"), # v_mad_mixhi_f16 in VEGA ISA, v_fma_mixhi_f16 in RDNA ISA +} +# note that these are only supported on gfx9+ so we'll need to distinguish between gfx8 and gfx9 here +# (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (-1, -1, -1, code, code, name) +for (code, name) in VOPP: + opcode(name, code, code, Format.VOP3P) + + +# VINTERP instructions: +VINTRP = { + (0x00, "v_interp_p1_f32"), + (0x01, "v_interp_p2_f32"), + (0x02, "v_interp_mov_f32"), +} +# (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (code, code, code, code, code, name) +for (code, name) in VINTRP: + opcode(name, code, code, Format.VINTRP) + +# VOP3 instructions: 3 inputs, 1 output +# VOP3b instructions: have a unique scalar output, e.g. VOP2 with vcc out +VOP3 = { + (0x140, 0x140, 0x1c0, 0x1c0, 0x140, "v_mad_legacy_f32", True, True), + (0x141, 0x141, 0x1c1, 0x1c1, 0x141, "v_mad_f32", True, True), + (0x142, 0x142, 0x1c2, 0x1c2, 0x142, "v_mad_i32_i24", False, False), + (0x143, 0x143, 0x1c3, 0x1c3, 0x143, "v_mad_u32_u24", False, False), + (0x144, 0x144, 0x1c4, 0x1c4, 0x144, "v_cubeid_f32", True, True), + (0x145, 0x145, 0x1c5, 0x1c5, 0x145, "v_cubesc_f32", True, True), + (0x146, 0x146, 0x1c6, 0x1c6, 0x146, "v_cubetc_f32", True, True), + (0x147, 0x147, 0x1c7, 0x1c7, 0x147, "v_cubema_f32", True, True), + (0x148, 0x148, 0x1c8, 0x1c8, 0x148, "v_bfe_u32", False, False), + (0x149, 0x149, 0x1c9, 0x1c9, 0x149, "v_bfe_i32", False, False), + (0x14a, 0x14a, 0x1ca, 0x1ca, 0x14a, "v_bfi_b32", False, False), + (0x14b, 0x14b, 0x1cb, 0x1cb, 0x14b, "v_fma_f32", True, True), + (0x14c, 0x14c, 0x1cc, 0x1cc, 0x14c, "v_fma_f64", True, True), + (0x14d, 0x14d, 0x1cd, 0x1cd, 0x14d, "v_lerp_u8", False, False), + (0x14e, 0x14e, 0x1ce, 0x1ce, 0x14e, "v_alignbit_b32", False, False), + (0x14f, 0x14f, 0x1cf, 0x1cf, 0x14f, "v_alignbyte_b32", False, False), + (0x150, 0x150, -1, -1, 0x150, "v_mullit_f32", True, True), + (0x151, 0x151, 0x1d0, 0x1d0, 0x151, "v_min3_f32", True, True), + (0x152, 0x152, 0x1d1, 0x1d1, 0x152, "v_min3_i32", False, False), + (0x153, 0x153, 0x1d2, 0x1d2, 0x153, "v_min3_u32", False, False), + (0x154, 0x154, 0x1d3, 0x1d3, 0x154, "v_max3_f32", True, True), + (0x155, 0x155, 0x1d4, 0x1d4, 0x155, "v_max3_i32", False, False), + (0x156, 0x156, 0x1d5, 0x1d5, 0x156, "v_max3_u32", False, False), + (0x157, 0x157, 0x1d6, 0x1d6, 0x157, "v_med3_f32", True, True), + (0x158, 0x158, 0x1d7, 0x1d7, 0x158, "v_med3_i32", False, False), + (0x159, 0x159, 0x1d8, 0x1d8, 0x159, "v_med3_u32", False, False), + (0x15a, 0x15a, 0x1d9, 0x1d9, 0x15a, "v_sad_u8", False, False), + (0x15b, 0x15b, 0x1da, 0x1da, 0x15b, "v_sad_hi_u8", False, False), + (0x15c, 0x15c, 0x1db, 0x1db, 0x15c, "v_sad_u16", False, False), + (0x15d, 0x15d, 0x1dc, 0x1dc, 0x15d, "v_sad_u32", False, False), + (0x15e, 0x15e, 0x1dd, 0x1dd, 0x15e, "v_cvt_pk_u8_f32", True, False), + (0x15f, 0x15f, 0x1de, 0x1de, 0x15f, "v_div_fixup_f32", True, True), + (0x160, 0x160, 0x1df, 0x1df, 0x160, "v_div_fixup_f64", True, True), + (0x161, 0x161, -1, -1, -1, "v_lshl_b64", False, False), + (0x162, 0x162, -1, -1, -1, "v_lshr_b64", False, False), + (0x163, 0x163, -1, -1, -1, "v_ashr_i64", False, False), + (0x164, 0x164, 0x280, 0x280, 0x164, "v_add_f64", True, True), + (0x165, 0x165, 0x281, 0x281, 0x165, "v_mul_f64", True, True), + (0x166, 0x166, 0x282, 0x282, 0x166, "v_min_f64", True, True), + (0x167, 0x167, 0x283, 0x283, 0x167, "v_max_f64", True, True), + (0x168, 0x168, 0x284, 0x284, 0x168, "v_ldexp_f64", False, True), # src1 can take input modifiers + (0x169, 0x169, 0x285, 0x285, 0x169, "v_mul_lo_u32", False, False), + (0x16a, 0x16a, 0x286, 0x286, 0x16a, "v_mul_hi_u32", False, False), + (0x16b, 0x16b, 0x285, 0x285, 0x16b, "v_mul_lo_i32", False, False), # identical to v_mul_lo_u32 + (0x16c, 0x16c, 0x287, 0x287, 0x16c, "v_mul_hi_i32", False, False), + (0x16d, 0x16d, 0x1e0, 0x1e0, 0x16d, "v_div_scale_f32", True, True), # writes to VCC + (0x16e, 0x16e, 0x1e1, 0x1e1, 0x16e, "v_div_scale_f64", True, True), # writes to VCC + (0x16f, 0x16f, 0x1e2, 0x1e2, 0x16f, "v_div_fmas_f32", True, True), # takes VCC input + (0x170, 0x170, 0x1e3, 0x1e3, 0x170, "v_div_fmas_f64", True, True), # takes VCC input + (0x171, 0x171, 0x1e4, 0x1e4, 0x171, "v_msad_u8", False, False), + (0x172, 0x172, 0x1e5, 0x1e5, 0x172, "v_qsad_pk_u16_u8", False, False), + (0x172, -1, -1, -1, -1, "v_qsad_u8", False, False), # what's the difference? + (0x173, 0x173, 0x1e6, 0x1e6, 0x173, "v_mqsad_pk_u16_u8", False, False), + (0x173, -1, -1, -1, -1, "v_mqsad_u8", False, False), # what's the difference? + (0x174, 0x174, 0x292, 0x292, 0x174, "v_trig_preop_f64", False, False), + ( -1, 0x175, 0x1e7, 0x1e7, 0x175, "v_mqsad_u32_u8", False, False), + ( -1, 0x176, 0x1e8, 0x1e8, 0x176, "v_mad_u64_u32", False, False), + ( -1, 0x177, 0x1e9, 0x1e9, 0x177, "v_mad_i64_i32", False, False), + ( -1, -1, 0x1ea, 0x1ea, -1, "v_mad_legacy_f16", True, True), + ( -1, -1, 0x1eb, 0x1eb, -1, "v_mad_legacy_u16", False, False), + ( -1, -1, 0x1ec, 0x1ec, -1, "v_mad_legacy_i16", False, False), + ( -1, -1, 0x1ed, 0x1ed, 0x344, "v_perm_b32", False, False), + ( -1, -1, 0x1ee, 0x1ee, -1, "v_fma_legacy_f16", True, True), + ( -1, -1, 0x1ef, 0x1ef, -1, "v_div_fixup_legacy_f16", True, True), + (0x12c, 0x12c, 0x1f0, 0x1f0, -1, "v_cvt_pkaccum_u8_f32", True, False), + ( -1, -1, -1, 0x1f1, 0x373, "v_mad_u32_u16", False, False), + ( -1, -1, -1, 0x1f2, 0x375, "v_mad_i32_i16", False, False), + ( -1, -1, -1, 0x1f2, 0x345, "v_xad_u32", False, False), + ( -1, -1, -1, 0x1f4, 0x351, "v_min3_f16", True, True), + ( -1, -1, -1, 0x1f5, 0x352, "v_min3_i16", False, False), + ( -1, -1, -1, 0x1f6, 0x353, "v_min3_u16", False, False), + ( -1, -1, -1, 0x1f7, 0x354, "v_max3_f16", True, True), + ( -1, -1, -1, 0x1f8, 0x355, "v_max3_i16", False, False), + ( -1, -1, -1, 0x1f9, 0x356, "v_max3_u16", False, False), + ( -1, -1, -1, 0x1fa, 0x357, "v_med3_f16", True, True), + ( -1, -1, -1, 0x1fb, 0x358, "v_med3_i16", False, False), + ( -1, -1, -1, 0x1fc, 0x359, "v_med3_u16", False, False), + ( -1, -1, -1, 0x1fd, 0x346, "v_lshl_add_u32", False, False), + ( -1, -1, -1, 0x1fe, 0x347, "v_add_lshl_u32", False, False), + ( -1, -1, -1, 0x1ff, 0x36d, "v_add3_u32", False, False), + ( -1, -1, -1, 0x200, 0x36f, "v_lshl_or_b32", False, False), + ( -1, -1, -1, 0x201, 0x371, "v_and_or_b32", False, False), + ( -1, -1, -1, 0x202, 0x372, "v_or3_b32", False, False), + ( -1, -1, -1, 0x203, -1, "v_mad_f16", True, True), + ( -1, -1, -1, 0x204, 0x340, "v_mad_u16", False, False), + ( -1, -1, -1, 0x205, 0x35e, "v_mad_i16", False, False), + ( -1, -1, -1, 0x206, 0x34b, "v_fma_f16", True, True), + ( -1, -1, -1, 0x207, 0x35f, "v_div_fixup_f16", True, True), + ( -1, -1, 0x274, 0x274, 0x342, "v_interp_p1ll_f16", True, True), + ( -1, -1, 0x275, 0x275, 0x343, "v_interp_p1lv_f16", True, True), + ( -1, -1, 0x276, 0x276, -1, "v_interp_p2_legacy_f16", True, True), + ( -1, -1, -1, 0x277, 0x35a, "v_interp_p2_f16", True, True), + (0x12b, 0x12b, 0x288, 0x288, 0x362, "v_ldexp_f32", False, True), + (0x101, 0x101, 0x289, 0x289, 0x360, "v_readlane_b32", False, False), + (0x102, 0x102, 0x28a, 0x28a, 0x361, "v_writelane_b32", False, False), + (0x122, 0x122, 0x28b, 0x28b, 0x364, "v_bcnt_u32_b32", False, False), + (0x123, 0x123, 0x28c, 0x28c, 0x365, "v_mbcnt_lo_u32_b32", False, False), + (0x124, 0x124, 0x28d, 0x28d, 0x366, "v_mbcnt_hi_u32_b32", False, False), + ( -1, -1, 0x28f, 0x28f, 0x2ff, "v_lshlrev_b64", False, False), + ( -1, -1, 0x290, 0x290, 0x300, "v_lshrrev_b64", False, False), + ( -1, -1, 0x291, 0x291, 0x301, "v_ashrrev_i64", False, False), + (0x11e, 0x11e, 0x293, 0x293, 0x363, "v_bfm_b32", False, False), + (0x12d, 0x12d, 0x294, 0x294, 0x368, "v_cvt_pknorm_i16_f32", True, False), + (0x12e, 0x12e, 0x295, 0x295, 0x369, "v_cvt_pknorm_u16_f32", True, False), + (0x12f, 0x12f, 0x296, 0x296, 0x12f, "v_cvt_pkrtz_f16_f32", True, False), # GFX6_7_10 is VOP2 with opcode 0x02f + (0x130, 0x130, 0x297, 0x297, 0x36a, "v_cvt_pk_u16_u32", False, False), + (0x131, 0x131, 0x298, 0x298, 0x36b, "v_cvt_pk_i16_i32", False, False), + ( -1, -1, -1, 0x299, 0x312, "v_cvt_pknorm_i16_f16", True, False), + ( -1, -1, -1, 0x29a, 0x313, "v_cvt_pknorm_u16_f16", True, False), + ( -1, -1, -1, 0x29c, 0x37f, "v_add_i32", False, False), + ( -1, -1, -1, 0x29d, 0x376, "v_sub_i32", False, False), + ( -1, -1, -1, 0x29e, 0x30d, "v_add_i16", False, False), + ( -1, -1, -1, 0x29f, 0x30e, "v_sub_i16", False, False), + ( -1, -1, -1, 0x2a0, 0x311, "v_pack_b32_f16", True, False), + ( -1, -1, -1, -1, 0x178, "v_xor3_b32", False, False), + ( -1, -1, -1, -1, 0x377, "v_permlane16_b32", False, False), + ( -1, -1, -1, -1, 0x378, "v_permlanex16_b32", False, False), + ( -1, -1, -1, -1, 0x30f, "v_add_co_u32_e64", False, False), + ( -1, -1, -1, -1, 0x310, "v_sub_co_u32_e64", False, False), + ( -1, -1, -1, -1, 0x311, "v_subrev_co_u32_e64", False, False), +# TODO: many 16bit instructions moved from VOP2 to VOP3 on GFX10 +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name, in_mod, out_mod) in VOP3: + opcode(name, gfx9, gfx10, Format.VOP3A, in_mod, out_mod) + + +# DS instructions: 3 inputs (1 addr, 2 data), 1 output +DS = { + (0x00, 0x00, 0x00, 0x00, 0x00, "ds_add_u32"), + (0x01, 0x01, 0x01, 0x01, 0x01, "ds_sub_u32"), + (0x02, 0x02, 0x02, 0x02, 0x02, "ds_rsub_u32"), + (0x03, 0x03, 0x03, 0x03, 0x03, "ds_inc_u32"), + (0x04, 0x04, 0x04, 0x04, 0x04, "ds_dec_u32"), + (0x05, 0x05, 0x05, 0x05, 0x05, "ds_min_i32"), + (0x06, 0x06, 0x06, 0x06, 0x06, "ds_max_i32"), + (0x07, 0x07, 0x07, 0x07, 0x07, "ds_min_u32"), + (0x08, 0x08, 0x08, 0x08, 0x08, "ds_max_u32"), + (0x09, 0x09, 0x09, 0x09, 0x09, "ds_and_b32"), + (0x0a, 0x0a, 0x0a, 0x0a, 0x0a, "ds_or_b32"), + (0x0b, 0x0b, 0x0b, 0x0b, 0x0b, "ds_xor_b32"), + (0x0c, 0x0c, 0x0c, 0x0c, 0x0c, "ds_mskor_b32"), + (0x0d, 0x0d, 0x0d, 0x0d, 0x0d, "ds_write_b32"), + (0x0e, 0x0e, 0x0e, 0x0e, 0x0e, "ds_write2_b32"), + (0x0f, 0x0f, 0x0f, 0x0f, 0x0f, "ds_write2st64_b32"), + (0x10, 0x10, 0x10, 0x10, 0x10, "ds_cmpst_b32"), + (0x11, 0x11, 0x11, 0x11, 0x11, "ds_cmpst_f32"), + (0x12, 0x12, 0x12, 0x12, 0x12, "ds_min_f32"), + (0x13, 0x13, 0x13, 0x13, 0x13, "ds_max_f32"), + ( -1, 0x14, 0x14, 0x14, 0x14, "ds_nop"), + ( -1, -1, 0x15, 0x15, 0x15, "ds_add_f32"), + ( -1, -1, 0x1d, 0x1d, 0xb0, "ds_write_addtid_b32"), + (0x1e, 0x1e, 0x1e, 0x1e, 0x1e, "ds_write_b8"), + (0x1f, 0x1f, 0x1f, 0x1f, 0x1f, "ds_write_b16"), + (0x20, 0x20, 0x20, 0x20, 0x20, "ds_add_rtn_u32"), + (0x21, 0x21, 0x21, 0x21, 0x21, "ds_sub_rtn_u32"), + (0x22, 0x22, 0x22, 0x22, 0x22, "ds_rsub_rtn_u32"), + (0x23, 0x23, 0x23, 0x23, 0x23, "ds_inc_rtn_u32"), + (0x24, 0x24, 0x24, 0x24, 0x24, "ds_dec_rtn_u32"), + (0x25, 0x25, 0x25, 0x25, 0x25, "ds_min_rtn_i32"), + (0x26, 0x26, 0x26, 0x26, 0x26, "ds_max_rtn_i32"), + (0x27, 0x27, 0x27, 0x27, 0x27, "ds_min_rtn_u32"), + (0x28, 0x28, 0x28, 0x28, 0x28, "ds_max_rtn_u32"), + (0x29, 0x29, 0x29, 0x29, 0x29, "ds_and_rtn_b32"), + (0x2a, 0x2a, 0x2a, 0x2a, 0x2a, "ds_or_rtn_b32"), + (0x2b, 0x2b, 0x2b, 0x2b, 0x2b, "ds_xor_rtn_b32"), + (0x2c, 0x2c, 0x2c, 0x2c, 0x2c, "ds_mskor_rtn_b32"), + (0x2d, 0x2d, 0x2d, 0x2d, 0x2d, "ds_wrxchg_rtn_b32"), + (0x2e, 0x2e, 0x2e, 0x2e, 0x2e, "ds_wrxchg2_rtn_b32"), + (0x2f, 0x2f, 0x2f, 0x2f, 0x2f, "ds_wrxchg2st64_rtn_b32"), + (0x30, 0x30, 0x30, 0x30, 0x30, "ds_cmpst_rtn_b32"), + (0x31, 0x31, 0x31, 0x31, 0x31, "ds_cmpst_rtn_f32"), + (0x32, 0x32, 0x32, 0x32, 0x32, "ds_min_rtn_f32"), + (0x33, 0x33, 0x33, 0x33, 0x33, "ds_max_rtn_f32"), + ( -1, 0x34, 0x34, 0x34, 0x34, "ds_wrap_rtn_b32"), + ( -1, -1, 0x35, 0x35, 0x55, "ds_add_rtn_f32"), + (0x36, 0x36, 0x36, 0x36, 0x36, "ds_read_b32"), + (0x37, 0x37, 0x37, 0x37, 0x37, "ds_read2_b32"), + (0x38, 0x38, 0x38, 0x38, 0x38, "ds_read2st64_b32"), + (0x39, 0x39, 0x39, 0x39, 0x39, "ds_read_i8"), + (0x3a, 0x3a, 0x3a, 0x3a, 0x3a, "ds_read_u8"), + (0x3b, 0x3b, 0x3b, 0x3b, 0x3b, "ds_read_i16"), + (0x3c, 0x3c, 0x3c, 0x3c, 0x3c, "ds_read_u16"), + (0x35, 0x35, 0x3d, 0x3d, 0x35, "ds_swizzle_b32"), #data1 & offset, no addr/data2 + ( -1, -1, 0x3e, 0x3e, 0xb2, "ds_permute_b32"), + ( -1, -1, 0x3f, 0x3f, 0xb3, "ds_bpermute_b32"), + (0x40, 0x40, 0x40, 0x40, 0x40, "ds_add_u64"), + (0x41, 0x41, 0x41, 0x41, 0x41, "ds_sub_u64"), + (0x42, 0x42, 0x42, 0x42, 0x42, "ds_rsub_u64"), + (0x43, 0x43, 0x43, 0x43, 0x43, "ds_inc_u64"), + (0x44, 0x44, 0x44, 0x44, 0x44, "ds_dec_u64"), + (0x45, 0x45, 0x45, 0x45, 0x45, "ds_min_i64"), + (0x46, 0x46, 0x46, 0x46, 0x46, "ds_max_i64"), + (0x47, 0x47, 0x47, 0x47, 0x47, "ds_min_u64"), + (0x48, 0x48, 0x48, 0x48, 0x48, "ds_max_u64"), + (0x49, 0x49, 0x49, 0x49, 0x49, "ds_and_b64"), + (0x4a, 0x4a, 0x4a, 0x4a, 0x4a, "ds_or_b64"), + (0x4b, 0x4b, 0x4b, 0x4b, 0x4b, "ds_xor_b64"), + (0x4c, 0x4c, 0x4c, 0x4c, 0x4c, "ds_mskor_b64"), + (0x4d, 0x4d, 0x4d, 0x4d, 0x4d, "ds_write_b64"), + (0x4e, 0x4e, 0x4e, 0x4e, 0x4e, "ds_write2_b64"), + (0x4f, 0x4f, 0x4f, 0x4f, 0x4f, "ds_write2st64_b64"), + (0x50, 0x50, 0x50, 0x50, 0x50, "ds_cmpst_b64"), + (0x51, 0x51, 0x51, 0x51, 0x51, "ds_cmpst_f64"), + (0x52, 0x52, 0x52, 0x52, 0x52, "ds_min_f64"), + (0x53, 0x53, 0x53, 0x53, 0x53, "ds_max_f64"), + ( -1, -1, 0x54, 0x54, 0xa0, "ds_write_b8_d16_hi"), + ( -1, -1, 0x55, 0x55, 0xa1, "ds_write_b16_d16_hi"), + ( -1, -1, 0x56, 0x56, 0xa2, "ds_read_u8_d16"), + ( -1, -1, 0x57, 0x57, 0xa3, "ds_read_u8_d16_hi"), + ( -1, -1, 0x58, 0x58, 0xa4, "ds_read_i8_d16"), + ( -1, -1, 0x59, 0x59, 0xa5, "ds_read_i8_d16_hi"), + ( -1, -1, 0x5a, 0x5a, 0xa6, "ds_read_u16_d16"), + ( -1, -1, 0x5b, 0x5b, 0xa7, "ds_read_u16_d16_hi"), + (0x60, 0x60, 0x60, 0x60, 0x60, "ds_add_rtn_u64"), + (0x61, 0x61, 0x61, 0x61, 0x61, "ds_sub_rtn_u64"), + (0x62, 0x62, 0x62, 0x62, 0x62, "ds_rsub_rtn_u64"), + (0x63, 0x63, 0x63, 0x63, 0x63, "ds_inc_rtn_u64"), + (0x64, 0x64, 0x64, 0x64, 0x64, "ds_dec_rtn_u64"), + (0x65, 0x65, 0x65, 0x65, 0x65, "ds_min_rtn_i64"), + (0x66, 0x66, 0x66, 0x66, 0x66, "ds_max_rtn_i64"), + (0x67, 0x67, 0x67, 0x67, 0x67, "ds_min_rtn_u64"), + (0x68, 0x68, 0x68, 0x68, 0x68, "ds_max_rtn_u64"), + (0x69, 0x69, 0x69, 0x69, 0x69, "ds_and_rtn_b64"), + (0x6a, 0x6a, 0x6a, 0x6a, 0x6a, "ds_or_rtn_b64"), + (0x6b, 0x6b, 0x6b, 0x6b, 0x6b, "ds_xor_rtn_b64"), + (0x6c, 0x6c, 0x6c, 0x6c, 0x6c, "ds_mskor_rtn_b64"), + (0x6d, 0x6d, 0x6d, 0x6d, 0x6d, "ds_wrxchg_rtn_b64"), + (0x6e, 0x6e, 0x6e, 0x6e, 0x6e, "ds_wrxchg2_rtn_b64"), + (0x6f, 0x6f, 0x6f, 0x6f, 0x6f, "ds_wrxchg2st64_rtn_b64"), + (0x70, 0x70, 0x70, 0x70, 0x70, "ds_cmpst_rtn_b64"), + (0x71, 0x71, 0x71, 0x71, 0x71, "ds_cmpst_rtn_f64"), + (0x72, 0x72, 0x72, 0x72, 0x72, "ds_min_rtn_f64"), + (0x73, 0x73, 0x73, 0x73, 0x73, "ds_max_rtn_f64"), + (0x76, 0x76, 0x76, 0x76, 0x76, "ds_read_b64"), + (0x77, 0x77, 0x77, 0x77, 0x77, "ds_read2_b64"), + (0x78, 0x78, 0x78, 0x78, 0x78, "ds_read2st64_b64"), + ( -1, 0x7e, 0x7e, 0x7e, 0x7e, "ds_condxchg32_rtn_b64"), + (0x80, 0x80, 0x80, 0x80, 0x80, "ds_add_src2_u32"), + (0x81, 0x81, 0x81, 0x81, 0x81, "ds_sub_src2_u32"), + (0x82, 0x82, 0x82, 0x82, 0x82, "ds_rsub_src2_u32"), + (0x83, 0x83, 0x83, 0x83, 0x83, "ds_inc_src2_u32"), + (0x84, 0x84, 0x84, 0x84, 0x84, "ds_dec_src2_u32"), + (0x85, 0x85, 0x85, 0x85, 0x85, "ds_min_src2_i32"), + (0x86, 0x86, 0x86, 0x86, 0x86, "ds_max_src2_i32"), + (0x87, 0x87, 0x87, 0x87, 0x87, "ds_min_src2_u32"), + (0x88, 0x88, 0x88, 0x88, 0x88, "ds_max_src2_u32"), + (0x89, 0x89, 0x89, 0x89, 0x89, "ds_and_src2_b32"), + (0x8a, 0x8a, 0x8a, 0x8a, 0x8a, "ds_or_src2_b32"), + (0x8b, 0x8b, 0x8b, 0x8b, 0x8b, "ds_xor_src2_b32"), + (0x8d, 0x8d, 0x8d, 0x8d, 0x8d, "ds_write_src2_b32"), + (0x92, 0x92, 0x92, 0x92, 0x92, "ds_min_src2_f32"), + (0x93, 0x93, 0x93, 0x93, 0x93, "ds_max_src2_f32"), + ( -1, -1, 0x95, 0x95, 0x95, "ds_add_src2_f32"), + ( -1, 0x18, 0x98, 0x98, 0x18, "ds_gws_sema_release_all"), + (0x19, 0x19, 0x99, 0x99, 0x19, "ds_gws_init"), + (0x1a, 0x1a, 0x9a, 0x9a, 0x1a, "ds_gws_sema_v"), + (0x1b, 0x1b, 0x9b, 0x9b, 0x1b, "ds_gws_sema_br"), + (0x1c, 0x1c, 0x9c, 0x9c, 0x1c, "ds_gws_sema_p"), + (0x1d, 0x1d, 0x9d, 0x9d, 0x1d, "ds_gws_barrier"), + ( -1, -1, 0xb6, 0xb6, 0xb1, "ds_read_addtid_b32"), + (0x3d, 0x3d, 0xbd, 0xbd, 0x3d, "ds_consume"), + (0x3e, 0x3e, 0xbe, 0xbe, 0x3e, "ds_append"), + (0x3f, 0x3f, 0xbf, 0xbf, 0x3f, "ds_ordered_count"), + (0xc0, 0xc0, 0xc0, 0xc0, 0xc0, "ds_add_src2_u64"), + (0xc1, 0xc1, 0xc1, 0xc1, 0xc1, "ds_sub_src2_u64"), + (0xc2, 0xc2, 0xc2, 0xc2, 0xc2, "ds_rsub_src2_u64"), + (0xc3, 0xc3, 0xc3, 0xc3, 0xc3, "ds_inc_src2_u64"), + (0xc4, 0xc4, 0xc4, 0xc4, 0xc4, "ds_dec_src2_u64"), + (0xc5, 0xc5, 0xc5, 0xc5, 0xc5, "ds_min_src2_i64"), + (0xc6, 0xc6, 0xc6, 0xc6, 0xc6, "ds_max_src2_i64"), + (0xc7, 0xc7, 0xc7, 0xc7, 0xc7, "ds_min_src2_u64"), + (0xc8, 0xc8, 0xc8, 0xc8, 0xc8, "ds_max_src2_u64"), + (0xc9, 0xc9, 0xc9, 0xc9, 0xc9, "ds_and_src2_b64"), + (0xca, 0xca, 0xca, 0xca, 0xca, "ds_or_src2_b64"), + (0xcb, 0xcb, 0xcb, 0xcb, 0xcb, "ds_xor_src2_b64"), + (0xcd, 0xcd, 0xcd, 0xcd, 0xcd, "ds_write_src2_b64"), + (0xd2, 0xd2, 0xd2, 0xd2, 0xd2, "ds_min_src2_f64"), + (0xd3, 0xd3, 0xd3, 0xd3, 0xd3, "ds_max_src2_f64"), + ( -1, 0xde, 0xde, 0xde, 0xde, "ds_write_b96"), + ( -1, 0xdf, 0xdf, 0xdf, 0xdf, "ds_write_b128"), + ( -1, 0xfd, 0xfd, -1, -1, "ds_condxchg32_rtn_b128"), + ( -1, 0xfe, 0xfe, 0xfe, 0xfe, "ds_read_b96"), + ( -1, 0xff, 0xff, 0xff, 0xff, "ds_read_b128"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in DS: + opcode(name, gfx9, gfx10, Format.DS) + +# MUBUF instructions: +MUBUF = { + (0x00, 0x00, 0x00, 0x00, 0x00, "buffer_load_format_x"), + (0x01, 0x01, 0x01, 0x01, 0x01, "buffer_load_format_xy"), + (0x02, 0x02, 0x02, 0x02, 0x02, "buffer_load_format_xyz"), + (0x03, 0x03, 0x03, 0x03, 0x03, "buffer_load_format_xyzw"), + (0x04, 0x04, 0x04, 0x04, 0x04, "buffer_store_format_x"), + (0x05, 0x05, 0x05, 0x05, 0x05, "buffer_store_format_xy"), + (0x06, 0x06, 0x06, 0x06, 0x06, "buffer_store_format_xyz"), + (0x07, 0x07, 0x07, 0x07, 0x07, "buffer_store_format_xyzw"), + ( -1, -1, 0x08, 0x08, 0x80, "buffer_load_format_d16_x"), + ( -1, -1, 0x09, 0x09, 0x81, "buffer_load_format_d16_xy"), + ( -1, -1, 0x0a, 0x0a, 0x82, "buffer_load_format_d16_xyz"), + ( -1, -1, 0x0b, 0x0b, 0x83, "buffer_load_format_d16_xyzw"), + ( -1, -1, 0x0c, 0x0c, 0x84, "buffer_store_format_d16_x"), + ( -1, -1, 0x0d, 0x0d, 0x85, "buffer_store_format_d16_xy"), + ( -1, -1, 0x0e, 0x0e, 0x86, "buffer_store_format_d16_xyz"), + ( -1, -1, 0x0f, 0x0f, 0x87, "buffer_store_format_d16_xyzw"), + (0x08, 0x08, 0x10, 0x10, 0x08, "buffer_load_ubyte"), + (0x09, 0x09, 0x11, 0x11, 0x09, "buffer_load_sbyte"), + (0x0a, 0x0a, 0x12, 0x12, 0x0a, "buffer_load_ushort"), + (0x0b, 0x0b, 0x13, 0x13, 0x0b, "buffer_load_sshort"), + (0x0c, 0x0c, 0x14, 0x14, 0x0c, "buffer_load_dword"), + (0x0d, 0x0d, 0x15, 0x15, 0x0d, "buffer_load_dwordx2"), + ( -1, 0x0f, 0x16, 0x16, 0x0f, "buffer_load_dwordx3"), + (0x0f, 0x0e, 0x17, 0x17, 0x0e, "buffer_load_dwordx4"), + (0x18, 0x18, 0x18, 0x18, 0x18, "buffer_store_byte"), + ( -1, -1, -1, 0x19, 0x19, "buffer_store_byte_d16_hi"), + (0x1a, 0x1a, 0x1a, 0x1a, 0x1a, "buffer_store_short"), + ( -1, -1, -1, 0x1b, 0x1b, "buffer_store_short_d16_hi"), + (0x1c, 0x1c, 0x1c, 0x1c, 0x1c, "buffer_store_dword"), + (0x1d, 0x1d, 0x1d, 0x1d, 0x1d, "buffer_store_dwordx2"), + ( -1, 0x1f, 0x1e, 0x1e, 0x1f, "buffer_store_dwordx3"), + (0x1e, 0x1e, 0x1f, 0x1f, 0x1e, "buffer_store_dwordx4"), + ( -1, -1, -1, 0x20, 0x20, "buffer_load_ubyte_d16"), + ( -1, -1, -1, 0x21, 0x21, "buffer_load_ubyte_d16_hi"), + ( -1, -1, -1, 0x22, 0x22, "buffer_load_sbyte_d16"), + ( -1, -1, -1, 0x23, 0x23, "buffer_load_sbyte_d16_hi"), + ( -1, -1, -1, 0x24, 0x24, "buffer_load_short_d16"), + ( -1, -1, -1, 0x25, 0x25, "buffer_load_short_d16_hi"), + ( -1, -1, -1, 0x26, 0x26, "buffer_load_format_d16_hi_x"), + ( -1, -1, -1, 0x27, 0x27, "buffer_store_format_d16_hi_x"), + ( -1, -1, 0x3d, 0x3d, -1, "buffer_store_lds_dword"), + (0x71, 0x71, 0x3e, 0x3e, -1, "buffer_wbinvl1"), + (0x70, 0x70, 0x3f, 0x3f, -1, "buffer_wbinvl1_vol"), + (0x30, 0x30, 0x40, 0x40, 0x30, "buffer_atomic_swap"), + (0x31, 0x31, 0x41, 0x41, 0x31, "buffer_atomic_cmpswap"), + (0x32, 0x32, 0x42, 0x42, 0x32, "buffer_atomic_add"), + (0x33, 0x33, 0x43, 0x43, 0x33, "buffer_atomic_sub"), + (0x34, -1, -1, -1, -1, "buffer_atomic_rsub"), + (0x35, 0x35, 0x44, 0x44, 0x35, "buffer_atomic_smin"), + (0x36, 0x36, 0x45, 0x45, 0x36, "buffer_atomic_umin"), + (0x37, 0x37, 0x46, 0x46, 0x37, "buffer_atomic_smax"), + (0x38, 0x38, 0x47, 0x47, 0x38, "buffer_atomic_umax"), + (0x39, 0x39, 0x48, 0x48, 0x39, "buffer_atomic_and"), + (0x3a, 0x3a, 0x49, 0x49, 0x3a, "buffer_atomic_or"), + (0x3b, 0x3b, 0x4a, 0x4a, 0x3b, "buffer_atomic_xor"), + (0x3c, 0x3c, 0x4b, 0x4b, 0x3c, "buffer_atomic_inc"), + (0x3d, 0x3d, 0x4c, 0x4c, 0x3d, "buffer_atomic_dec"), + (0x3e, 0x3e, -1, -1, 0x3e, "buffer_atomic_fcmpswap"), + (0x3f, 0x3f, -1, -1, 0x3f, "buffer_atomic_fmin"), + (0x40, 0x40, -1, -1, 0x40, "buffer_atomic_fmax"), + (0x50, 0x50, 0x60, 0x60, 0x50, "buffer_atomic_swap_x2"), + (0x51, 0x51, 0x61, 0x61, 0x51, "buffer_atomic_cmpswap_x2"), + (0x52, 0x52, 0x62, 0x62, 0x52, "buffer_atomic_add_x2"), + (0x53, 0x53, 0x63, 0x63, 0x53, "buffer_atomic_sub_x2"), + (0x54, -1, -1, -1, -1, "buffer_atomic_rsub_x2"), + (0x55, 0x55, 0x64, 0x64, 0x55, "buffer_atomic_smin_x2"), + (0x56, 0x56, 0x65, 0x65, 0x56, "buffer_atomic_umin_x2"), + (0x57, 0x57, 0x66, 0x66, 0x57, "buffer_atomic_smax_x2"), + (0x58, 0x58, 0x67, 0x67, 0x58, "buffer_atomic_umax_x2"), + (0x59, 0x59, 0x68, 0x68, 0x59, "buffer_atomic_and_x2"), + (0x5a, 0x5a, 0x69, 0x69, 0x5a, "buffer_atomic_or_x2"), + (0x5b, 0x5b, 0x6a, 0x6a, 0x5b, "buffer_atomic_xor_x2"), + (0x5c, 0x5c, 0x6b, 0x6b, 0x5c, "buffer_atomic_inc_x2"), + (0x5d, 0x5d, 0x6c, 0x6c, 0x5d, "buffer_atomic_dec_x2"), + (0x5e, 0x5e, -1, -1, 0x5e, "buffer_atomic_fcmpswap_x2"), + (0x5f, 0x5f, -1, -1, 0x5f, "buffer_atomic_fmin_x2"), + (0x60, 0x60, -1, -1, 0x60, "buffer_atomic_fmax_x2"), + ( -1, -1, -1, -1, 0x71, "buffer_gl0_inv"), + ( -1, -1, -1, -1, 0x72, "buffer_gl1_inv"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in MUBUF: + opcode(name, gfx9, gfx10, Format.MUBUF) + +MTBUF = { + (0x00, 0x00, 0x00, 0x00, 0x00, "tbuffer_load_format_x"), + (0x01, 0x01, 0x01, 0x01, 0x01, "tbuffer_load_format_xy"), + (0x02, 0x02, 0x02, 0x02, 0x02, "tbuffer_load_format_xyz"), + (0x03, 0x03, 0x03, 0x03, 0x03, "tbuffer_load_format_xyzw"), + (0x04, 0x04, 0x04, 0x04, 0x04, "tbuffer_store_format_x"), + (0x05, 0x05, 0x05, 0x05, 0x05, "tbuffer_store_format_xy"), + (0x06, 0x06, 0x06, 0x06, 0x06, "tbuffer_store_format_xyz"), + (0x07, 0x07, 0x07, 0x07, 0x07, "tbuffer_store_format_xyzw"), + ( -1, -1, 0x08, 0x08, 0x08, "tbuffer_load_format_d16_x"), + ( -1, -1, 0x09, 0x09, 0x09, "tbuffer_load_format_d16_xy"), + ( -1, -1, 0x0a, 0x0a, 0x0a, "tbuffer_load_format_d16_xyz"), + ( -1, -1, 0x0b, 0x0b, 0x0b, "tbuffer_load_format_d16_xyzw"), + ( -1, -1, 0x0c, 0x0c, 0x0c, "tbuffer_store_format_d16_x"), + ( -1, -1, 0x0d, 0x0d, 0x0d, "tbuffer_store_format_d16_xy"), + ( -1, -1, 0x0e, 0x0e, 0x0e, "tbuffer_store_format_d16_xyz"), + ( -1, -1, 0x0f, 0x0f, 0x0f, "tbuffer_store_format_d16_xyzw"), +} +for (gfx6, gfx7, gfx8, gfx9, gfx10, name) in MTBUF: + opcode(name, gfx9, gfx10, Format.MTBUF) + + +IMAGE = { + (0x00, "image_load"), + (0x01, "image_load_mip"), + (0x02, "image_load_pck"), + (0x03, "image_load_pck_sgn"), + (0x04, "image_load_mip_pck"), + (0x05, "image_load_mip_pck_sgn"), + (0x08, "image_store"), + (0x09, "image_store_mip"), + (0x0a, "image_store_pck"), + (0x0b, "image_store_mip_pck"), + (0x0e, "image_get_resinfo"), + (0x60, "image_get_lod"), +} +# (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (code, code, code, code, code, name) +for (code, name) in IMAGE: + opcode(name, code, code, Format.MIMG) + +IMAGE_ATOMIC = { + (0x0f, 0x0f, 0x10, "image_atomic_swap"), + (0x10, 0x10, 0x11, "image_atomic_cmpswap"), + (0x11, 0x11, 0x12, "image_atomic_add"), + (0x12, 0x12, 0x13, "image_atomic_sub"), + (0x13, -1, -1, "image_atomic_rsub"), + (0x14, 0x14, 0x14, "image_atomic_smin"), + (0x15, 0x15, 0x15, "image_atomic_umin"), + (0x16, 0x16, 0x16, "image_atomic_smax"), + (0x17, 0x17, 0x17, "image_atomic_umax"), + (0x18, 0x18, 0x18, "image_atomic_and"), + (0x19, 0x19, 0x19, "image_atomic_or"), + (0x1a, 0x1a, 0x1a, "image_atomic_xor"), + (0x1b, 0x1b, 0x1b, "image_atomic_inc"), + (0x1c, 0x1c, 0x1c, "image_atomic_dec"), + (0x1d, 0x1d, -1, "image_atomic_fcmpswap"), + (0x1e, 0x1e, -1, "image_atomic_fmin"), + (0x1f, 0x1f, -1, "image_atomic_fmax"), +} +# (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (gfx6, gfx7, gfx89, gfx89, ???, name) +# gfx7 and gfx10 opcodes are the same here +for (gfx6, gfx7, gfx89, name) in IMAGE_ATOMIC: + opcode(name, gfx89, gfx7, Format.MIMG) + +IMAGE_SAMPLE = { + (0x20, "image_sample"), + (0x21, "image_sample_cl"), + (0x22, "image_sample_d"), + (0x23, "image_sample_d_cl"), + (0x24, "image_sample_l"), + (0x25, "image_sample_b"), + (0x26, "image_sample_b_cl"), + (0x27, "image_sample_lz"), + (0x28, "image_sample_c"), + (0x29, "image_sample_c_cl"), + (0x2a, "image_sample_c_d"), + (0x2b, "image_sample_c_d_cl"), + (0x2c, "image_sample_c_l"), + (0x2d, "image_sample_c_b"), + (0x2e, "image_sample_c_b_cl"), + (0x2f, "image_sample_c_lz"), + (0x30, "image_sample_o"), + (0x31, "image_sample_cl_o"), + (0x32, "image_sample_d_o"), + (0x33, "image_sample_d_cl_o"), + (0x34, "image_sample_l_o"), + (0x35, "image_sample_b_o"), + (0x36, "image_sample_b_cl_o"), + (0x37, "image_sample_lz_o"), + (0x38, "image_sample_c_o"), + (0x39, "image_sample_c_cl_o"), + (0x3a, "image_sample_c_d_o"), + (0x3b, "image_sample_c_d_cl_o"), + (0x3c, "image_sample_c_l_o"), + (0x3d, "image_sample_c_b_o"), + (0x3e, "image_sample_c_b_cl_o"), + (0x3f, "image_sample_c_lz_o"), + (0x68, "image_sample_cd"), + (0x69, "image_sample_cd_cl"), + (0x6a, "image_sample_c_cd"), + (0x6b, "image_sample_c_cd_cl"), + (0x6c, "image_sample_cd_o"), + (0x6d, "image_sample_cd_cl_o"), + (0x6e, "image_sample_c_cd_o"), + (0x6f, "image_sample_c_cd_cl_o"), +} +# (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (code, code, code, code, code, name) +for (code, name) in IMAGE_SAMPLE: + opcode(name, code, code, Format.MIMG) + +IMAGE_GATHER4 = { + (0x40, "image_gather4"), + (0x41, "image_gather4_cl"), + #(0x42, "image_gather4h"), VEGA only? + (0x44, "image_gather4_l"), # following instructions have different opcodes according to ISA sheet. + (0x45, "image_gather4_b"), + (0x46, "image_gather4_b_cl"), + (0x47, "image_gather4_lz"), + (0x48, "image_gather4_c"), + (0x49, "image_gather4_c_cl"), # previous instructions have different opcodes according to ISA sheet. + #(0x4a, "image_gather4h_pck"), VEGA only? + #(0x4b, "image_gather8h_pck"), VGEA only? + (0x4c, "image_gather4_c_l"), + (0x4d, "image_gather4_c_b"), + (0x4e, "image_gather4_c_b_cl"), + (0x4f, "image_gather4_c_lz"), + (0x50, "image_gather4_o"), + (0x51, "image_gather4_cl_o"), + (0x54, "image_gather4_l_o"), + (0x55, "image_gather4_b_o"), + (0x56, "image_gather4_b_cl_o"), + (0x57, "image_gather4_lz_o"), + (0x58, "image_gather4_c_o"), + (0x59, "image_gather4_c_cl_o"), + (0x5c, "image_gather4_c_l_o"), + (0x5d, "image_gather4_c_b_o"), + (0x5e, "image_gather4_c_b_cl_o"), + (0x5f, "image_gather4_c_lz_o"), +} +# (gfx6, gfx7, gfx8, gfx9, gfx10, name) = (code, code, code, code, code, name) +for (code, name) in IMAGE_GATHER4: + opcode(name, code, code, Format.MIMG) + + +FLAT = { + #GFX7, GFX8_9, GFX10 + (0x08, 0x10, 0x08, "flat_load_ubyte"), + (0x09, 0x11, 0x09, "flat_load_sbyte"), + (0x0a, 0x12, 0x0a, "flat_load_ushort"), + (0x0b, 0x13, 0x0b, "flat_load_sshort"), + (0x0c, 0x14, 0x0c, "flat_load_dword"), + (0x0d, 0x15, 0x0d, "flat_load_dwordx2"), + (0x0f, 0x16, 0x0f, "flat_load_dwordx3"), + (0x0e, 0x17, 0x0e, "flat_load_dwordx4"), + (0x18, 0x18, 0x18, "flat_store_byte"), + ( -1, 0x19, 0x19, "flat_store_byte_d16_hi"), + (0x1a, 0x1a, 0x1a, "flat_store_short"), + ( -1, 0x1b, 0x1b, "flat_store_short_d16_hi"), + (0x1c, 0x1c, 0x1c, "flat_store_dword"), + (0x1d, 0x1d, 0x1d, "flat_store_dwordx2"), + (0x1f, 0x1e, 0x1f, "flat_store_dwordx3"), + (0x1e, 0x1f, 0x1e, "flat_store_dwordx4"), + ( -1, 0x20, 0x20, "flat_load_ubyte_d16"), + ( -1, 0x21, 0x21, "flat_load_ubyte_d16_hi"), + ( -1, 0x22, 0x22, "flat_load_sbyte_d16"), + ( -1, 0x23, 0x23, "flat_load_sbyte_d16_hi"), + ( -1, 0x24, 0x24, "flat_load_short_d16"), + ( -1, 0x25, 0x25, "flat_load_short_d16_hi"), + (0x30, 0x40, 0x30, "flat_atomic_swap"), + (0x31, 0x41, 0x31, "flat_atomic_cmpswap"), + (0x32, 0x42, 0x32, "flat_atomic_add"), + (0x33, 0x43, 0x33, "flat_atomic_sub"), + (0x35, 0x44, 0x35, "flat_atomic_smin"), + (0x36, 0x45, 0x36, "flat_atomic_umin"), + (0x37, 0x46, 0x37, "flat_atomic_smax"), + (0x38, 0x47, 0x38, "flat_atomic_umax"), + (0x39, 0x48, 0x39, "flat_atomic_and"), + (0x3a, 0x49, 0x3a, "flat_atomic_or"), + (0x3b, 0x4a, 0x3b, "flat_atomic_xor"), + (0x3c, 0x4b, 0x3c, "flat_atomic_inc"), + (0x3d, 0x4c, 0x3d, "flat_atomic_dec"), + (0x3e, -1, 0x3e, "flat_atomic_fcmpswap"), + (0x3f, -1, 0x3f, "flat_atomic_fmin"), + (0x40, -1, 0x40, "flat_atomic_fmax"), + (0x50, 0x60, 0x50, "flat_atomic_swap_x2"), + (0x51, 0x61, 0x51, "flat_atomic_cmpswap_x2"), + (0x52, 0x62, 0x52, "flat_atomic_add_x2"), + (0x53, 0x63, 0x53, "flat_atomic_sub_x2"), + (0x55, 0x64, 0x54, "flat_atomic_smin_x2"), + (0x56, 0x65, 0x55, "flat_atomic_umin_x2"), + (0x57, 0x66, 0x56, "flat_atomic_smax_x2"), + (0x58, 0x67, 0x58, "flat_atomic_umax_x2"), + (0x59, 0x68, 0x59, "flat_atomic_and_x2"), + (0x5a, 0x69, 0x5a, "flat_atomic_or_x2"), + (0x5b, 0x6a, 0x5b, "flat_atomic_xor_x2"), + (0x5c, 0x6b, 0x5c, "flat_atomic_inc_x2"), + (0x5d, 0x6c, 0x5d, "flat_atomic_dec_x2"), + (0x5e, -1, 0x5e, "flat_atomic_fcmpswap_x2"), + (0x5f, -1, 0x5f, "flat_atomic_fmin_x2"), + (0x60, -1, 0x60, "flat_atomic_fmax_x2"), +} +for (gfx7, gfx8, gfx10, name) in FLAT: + opcode(name, gfx8, gfx10, Format.FLAT) + +GLOBAL = { + #GFX8_9, GFX10 + (0x10, 0x08, "global_load_ubyte"), + (0x11, 0x09, "global_load_sbyte"), + (0x12, 0x0a, "global_load_ushort"), + (0x13, 0x0b, "global_load_sshort"), + (0x14, 0x0c, "global_load_dword"), + (0x15, 0x0d, "global_load_dwordx2"), + (0x16, 0x0f, "global_load_dwordx3"), + (0x17, 0x0e, "global_load_dwordx4"), + (0x18, 0x18, "global_store_byte"), + (0x19, 0x19, "global_store_byte_d16_hi"), + (0x1a, 0x1a, "global_store_short"), + (0x1b, 0x1b, "global_store_short_d16_hi"), + (0x1c, 0x1c, "global_store_dword"), + (0x1d, 0x1d, "global_store_dwordx2"), + (0x1e, 0x1f, "global_store_dwordx3"), + (0x1f, 0x1e, "global_store_dwordx4"), + (0x20, 0x20, "global_load_ubyte_d16"), + (0x21, 0x21, "global_load_ubyte_d16_hi"), + (0x22, 0x22, "global_load_sbyte_d16"), + (0x23, 0x23, "global_load_sbyte_d16_hi"), + (0x24, 0x24, "global_load_short_d16"), + (0x25, 0x25, "global_load_short_d16_hi"), + (0x40, 0x30, "global_atomic_swap"), + (0x41, 0x31, "global_atomic_cmpswap"), + (0x42, 0x32, "global_atomic_add"), + (0x43, 0x33, "global_atomic_sub"), + (0x44, 0x35, "global_atomic_smin"), + (0x45, 0x36, "global_atomic_umin"), + (0x46, 0x37, "global_atomic_smax"), + (0x47, 0x38, "global_atomic_umax"), + (0x48, 0x39, "global_atomic_and"), + (0x49, 0x3a, "global_atomic_or"), + (0x4a, 0x3b, "global_atomic_xor"), + (0x4b, 0x3c, "global_atomic_inc"), + (0x4c, 0x3d, "global_atomic_dec"), + ( -1, 0x3e, "global_atomic_fcmpswap"), + ( -1, 0x3f, "global_atomic_fmin"), + ( -1, 0x40, "global_atomic_fmax"), + (0x60, 0x50, "global_atomic_swap_x2"), + (0x61, 0x51, "global_atomic_cmpswap_x2"), + (0x62, 0x52, "global_atomic_add_x2"), + (0x63, 0x53, "global_atomic_sub_x2"), + (0x64, 0x54, "global_atomic_smin_x2"), + (0x65, 0x55, "global_atomic_umin_x2"), + (0x66, 0x56, "global_atomic_smax_x2"), + (0x67, 0x58, "global_atomic_umax_x2"), + (0x68, 0x59, "global_atomic_and_x2"), + (0x69, 0x5a, "global_atomic_or_x2"), + (0x6a, 0x5b, "global_atomic_xor_x2"), + (0x6b, 0x5c, "global_atomic_inc_x2"), + (0x6c, 0x5d, "global_atomic_dec_x2"), + ( -1, 0x5e, "global_atomic_fcmpswap_x2"), + ( -1, 0x5f, "global_atomic_fmin_x2"), + ( -1, 0x60, "global_atomic_fmax_x2"), +} +for (gfx8, gfx10, name) in GLOBAL: + opcode(name, gfx8, gfx10, Format.GLOBAL) + +SCRATCH = { + #GFX8_9, GFX10 + (0x10, 0x08, "scratch_load_ubyte"), + (0x11, 0x09, "scratch_load_sbyte"), + (0x12, 0x0a, "scratch_load_ushort"), + (0x13, 0x0b, "scratch_load_sshort"), + (0x14, 0x0c, "scratch_load_dword"), + (0x15, 0x0d, "scratch_load_dwordx2"), + (0x16, 0x0f, "scratch_load_dwordx3"), + (0x17, 0x0e, "scratch_load_dwordx4"), + (0x18, 0x18, "scratch_store_byte"), + (0x19, 0x19, "scratch_store_byte_d16_hi"), + (0x1a, 0x1a, "scratch_store_short"), + (0x1b, 0x1b, "scratch_store_short_d16_hi"), + (0x1c, 0x1c, "scratch_store_dword"), + (0x1d, 0x1d, "scratch_store_dwordx2"), + (0x1e, 0x1f, "scratch_store_dwordx3"), + (0x1f, 0x1e, "scratch_store_dwordx4"), + (0x20, 0x20, "scratch_load_ubyte_d16"), + (0x21, 0x21, "scratch_load_ubyte_d16_hi"), + (0x22, 0x22, "scratch_load_sbyte_d16"), + (0x23, 0x23, "scratch_load_sbyte_d16_hi"), + (0x24, 0x24, "scratch_load_short_d16"), + (0x25, 0x25, "scratch_load_short_d16_hi"), +} +for (gfx8, gfx10, name) in SCRATCH: + opcode(name, gfx8, gfx10, Format.SCRATCH) diff --git a/src/amd/compiler/aco_opcodes_cpp.py b/src/amd/compiler/aco_opcodes_cpp.py new file mode 100644 index 00000000000..83c24e0eb44 --- /dev/null +++ b/src/amd/compiler/aco_opcodes_cpp.py @@ -0,0 +1,74 @@ + +template = """\ +/* + * Copyright (c) 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" + +namespace aco { + +const unsigned VOPC_to_GFX6[256] = { +% for code in VOPC_GFX6: + ${code}, +% endfor +}; + +<% +opcode_names = sorted(opcodes.keys()) +can_use_input_modifiers = "".join([opcodes[name].input_mod for name in reversed(opcode_names)]) +can_use_output_modifiers = "".join([opcodes[name].output_mod for name in reversed(opcode_names)]) +%> + +extern const aco::Info instr_info = { + .opcode_gfx9 = { + % for name in opcode_names: + ${opcodes[name].opcode_gfx9}, + % endfor + }, + .opcode_gfx10 = { + % for name in opcode_names: + ${opcodes[name].opcode_gfx10}, + % endfor + }, + .can_use_input_modifiers = std::bitset<${len(opcode_names)}>("${can_use_input_modifiers}"), + .can_use_output_modifiers = std::bitset<${len(opcode_names)}>("${can_use_output_modifiers}"), + .name = { + % for name in opcode_names: + "${name}", + % endfor + }, + .format = { + % for name in opcode_names: + aco::Format::${str(opcodes[name].format.name)}, + % endfor + }, +}; + +} +""" + +from aco_opcodes import opcodes, VOPC_GFX6 +from mako.template import Template + +print(Template(template).render(opcodes=opcodes, VOPC_GFX6=VOPC_GFX6)) diff --git a/src/amd/compiler/aco_opcodes_h.py b/src/amd/compiler/aco_opcodes_h.py new file mode 100644 index 00000000000..6b8bfc1ee07 --- /dev/null +++ b/src/amd/compiler/aco_opcodes_h.py @@ -0,0 +1,47 @@ + +template = """\ +/* + * Copyright (c) 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Daniel Schuermann (daniel.schuermann@campus.tu-berlin.de) + */ + +#ifndef _ACO_OPCODES_ +#define _ACO_OPCODES_ + +<% opcode_names = sorted(opcodes.keys()) %> + +enum class aco_opcode : std::uint16_t { +% for name in opcode_names: + ${name}, +% endfor + last_opcode = ${opcode_names[-1]}, + num_opcodes = last_opcode + 1 +}; + +#endif /* _ACO_OPCODES_ */""" + +from aco_opcodes import opcodes +from mako.template import Template + +print(Template(template).render(opcodes=opcodes)) diff --git a/src/amd/compiler/aco_opt_value_numbering.cpp b/src/amd/compiler/aco_opt_value_numbering.cpp new file mode 100644 index 00000000000..8071ace1f97 --- /dev/null +++ b/src/amd/compiler/aco_opt_value_numbering.cpp @@ -0,0 +1,327 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include +#include + +#include "aco_ir.h" + +/* + * Implements the algorithm for dominator-tree value numbering + * from "Value Numbering" by Briggs, Cooper, and Simpson. + */ + +namespace aco { +namespace { + +struct InstrHash { + std::size_t operator()(Instruction* instr) const + { + uint64_t hash = (uint64_t) instr->opcode + (uint64_t) instr->format; + for (unsigned i = 0; i < instr->operands.size(); i++) { + Operand op = instr->operands[i]; + uint64_t val = op.isTemp() ? op.tempId() : op.isFixed() ? op.physReg() : op.constantValue(); + hash |= val << (i+1) * 8; + } + if (instr->isVOP3()) { + VOP3A_instruction* vop3 = static_cast(instr); + for (unsigned i = 0; i < 3; i++) { + hash ^= vop3->abs[i] << (i*3 + 0); + hash ^= vop3->opsel[i] << (i*3 + 1); + hash ^= vop3->neg[i] << (i*3 + 2); + } + hash ^= (vop3->clamp << 28) * 13; + hash += vop3->omod << 19; + } + switch (instr->format) { + case Format::SMEM: + break; + case Format::VINTRP: { + Interp_instruction* interp = static_cast(instr); + hash ^= interp->attribute << 13; + hash ^= interp->component << 27; + break; + } + case Format::DS: + break; + default: + break; + } + + return hash; + } +}; + +struct InstrPred { + bool operator()(Instruction* a, Instruction* b) const + { + if (a->format != b->format) + return false; + if (a->opcode != b->opcode) + return false; + if (a->operands.size() != b->operands.size() || a->definitions.size() != b->definitions.size()) + return false; /* possible with pseudo-instructions */ + for (unsigned i = 0; i < a->operands.size(); i++) { + if (a->operands[i].isConstant()) { + if (!b->operands[i].isConstant()) + return false; + if (a->operands[i].constantValue() != b->operands[i].constantValue()) + return false; + } + else if (a->operands[i].isTemp()) { + if (!b->operands[i].isTemp()) + return false; + if (a->operands[i].tempId() != b->operands[i].tempId()) + return false; + } + else if (a->operands[i].isUndefined() ^ b->operands[i].isUndefined()) + return false; + if (a->operands[i].isFixed()) { + if (a->operands[i].physReg() == exec) + return false; + if (!b->operands[i].isFixed()) + return false; + if (!(a->operands[i].physReg() == b->operands[i].physReg())) + return false; + } + } + for (unsigned i = 0; i < a->definitions.size(); i++) { + if (a->definitions[i].isTemp()) { + if (!b->definitions[i].isTemp()) + return false; + if (a->definitions[i].regClass() != b->definitions[i].regClass()) + return false; + } + if (a->definitions[i].isFixed()) { + if (!b->definitions[i].isFixed()) + return false; + if (!(a->definitions[i].physReg() == b->definitions[i].physReg())) + return false; + } + } + if (a->format == Format::PSEUDO_BRANCH) + return false; + if (a->isVOP3()) { + VOP3A_instruction* a3 = static_cast(a); + VOP3A_instruction* b3 = static_cast(b); + for (unsigned i = 0; i < 3; i++) { + if (a3->abs[i] != b3->abs[i] || + a3->opsel[i] != b3->opsel[i] || + a3->neg[i] != b3->neg[i]) + return false; + } + return a3->clamp == b3->clamp && + a3->omod == b3->omod; + } + if (a->isDPP()) { + DPP_instruction* aDPP = static_cast(a); + DPP_instruction* bDPP = static_cast(b); + return aDPP->dpp_ctrl == bDPP->dpp_ctrl && + aDPP->bank_mask == bDPP->bank_mask && + aDPP->row_mask == bDPP->row_mask && + aDPP->bound_ctrl == bDPP->bound_ctrl && + aDPP->abs[0] == bDPP->abs[0] && + aDPP->abs[1] == bDPP->abs[1] && + aDPP->neg[0] == bDPP->neg[0] && + aDPP->neg[1] == bDPP->neg[1]; + } + switch (a->format) { + case Format::VOPC: { + /* Since the results depend on the exec mask, these shouldn't + * be value numbered (this is especially useful for subgroupBallot()). */ + return false; + } + case Format::SOPK: { + SOPK_instruction* aK = static_cast(a); + SOPK_instruction* bK = static_cast(b); + return aK->imm == bK->imm; + } + case Format::SMEM: { + SMEM_instruction* aS = static_cast(a); + SMEM_instruction* bS = static_cast(b); + return aS->can_reorder && bS->can_reorder && + aS->glc == bS->glc && aS->nv == bS->nv; + } + case Format::VINTRP: { + Interp_instruction* aI = static_cast(a); + Interp_instruction* bI = static_cast(b); + if (aI->attribute != bI->attribute) + return false; + if (aI->component != bI->component) + return false; + return true; + } + case Format::PSEUDO_REDUCTION: + return false; + case Format::MTBUF: { + /* this is fine since they are only used for vertex input fetches */ + MTBUF_instruction* aM = static_cast(a); + MTBUF_instruction* bM = static_cast(b); + return aM->dfmt == bM->dfmt && + aM->nfmt == bM->nfmt && + aM->offset == bM->offset && + aM->offen == bM->offen && + aM->idxen == bM->idxen && + aM->glc == bM->glc && + aM->slc == bM->slc && + aM->tfe == bM->tfe && + aM->disable_wqm == bM->disable_wqm; + } + /* we want to optimize these in NIR and don't hassle with load-store dependencies */ + case Format::MUBUF: + case Format::FLAT: + case Format::GLOBAL: + case Format::SCRATCH: + case Format::DS: + return false; + case Format::MIMG: { + MIMG_instruction* aM = static_cast(a); + MIMG_instruction* bM = static_cast(b); + return aM->can_reorder && bM->can_reorder && + aM->dmask == bM->dmask && + aM->unrm == bM->unrm && + aM->glc == bM->glc && + aM->slc == bM->slc && + aM->tfe == bM->tfe && + aM->da == bM->da && + aM->lwe == bM->lwe && + aM->r128 == bM->r128 && + aM->a16 == bM->a16 && + aM->d16 == bM->d16 && + aM->disable_wqm == bM->disable_wqm; + } + default: + return true; + } + } +}; + + +typedef std::unordered_set expr_set; + +void process_block(Block& block, + expr_set& expr_values, + std::map& renames) +{ + bool run = false; + std::vector>::iterator it = block.instructions.begin(); + std::vector> new_instructions; + new_instructions.reserve(block.instructions.size()); + expr_set phi_values; + + while (it != block.instructions.end()) { + aco_ptr& instr = *it; + /* first, rename operands */ + for (Operand& op : instr->operands) { + if (!op.isTemp()) + continue; + auto it = renames.find(op.tempId()); + if (it != renames.end()) + op.setTemp(it->second); + } + + if (instr->definitions.empty() || !run) { + if (instr->opcode == aco_opcode::p_logical_start) + run = true; + else if (instr->opcode == aco_opcode::p_logical_end) + run = false; + else if (instr->opcode == aco_opcode::p_phi || instr->opcode == aco_opcode::p_linear_phi) { + std::pair res = phi_values.emplace(instr.get()); + if (!res.second) { + Instruction* orig_phi = *(res.first); + renames.emplace(instr->definitions[0].tempId(), orig_phi->definitions[0].getTemp()).second; + ++it; + continue; + } + } + new_instructions.emplace_back(std::move(instr)); + ++it; + continue; + } + + /* simple copy-propagation through renaming */ + if ((instr->opcode == aco_opcode::s_mov_b32 || instr->opcode == aco_opcode::s_mov_b64 || instr->opcode == aco_opcode::v_mov_b32) && + !instr->definitions[0].isFixed() && instr->operands[0].isTemp() && instr->operands[0].regClass() == instr->definitions[0].regClass() && + !instr->isDPP() && !((int)instr->format & (int)Format::SDWA)) { + renames[instr->definitions[0].tempId()] = instr->operands[0].getTemp(); + } + + std::pair res = expr_values.emplace(instr.get()); + + /* if there was already an expression with the same value number */ + if (!res.second) { + Instruction* orig_instr = *(res.first); + assert(instr->definitions.size() == orig_instr->definitions.size()); + for (unsigned i = 0; i < instr->definitions.size(); i++) { + assert(instr->definitions[i].regClass() == orig_instr->definitions[i].regClass()); + renames.emplace(instr->definitions[i].tempId(), orig_instr->definitions[i].getTemp()).second; + } + } else { + new_instructions.emplace_back(std::move(instr)); + } + ++it; + } + + block.instructions.swap(new_instructions); +} + +void rename_phi_operands(Block& block, std::map& renames) +{ + for (aco_ptr& phi : block.instructions) { + if (phi->opcode != aco_opcode::p_phi && phi->opcode != aco_opcode::p_linear_phi) + break; + + for (Operand& op : phi->operands) { + if (!op.isTemp()) + continue; + auto it = renames.find(op.tempId()); + if (it != renames.end()) + op.setTemp(it->second); + } + } +} +} /* end namespace */ + + +void value_numbering(Program* program) +{ + std::vector expr_values(program->blocks.size()); + std::map renames; + + for (Block& block : program->blocks) { + if (block.logical_idom != -1) { + /* initialize expr_values from idom */ + expr_values[block.index] = expr_values[block.logical_idom]; + process_block(block, expr_values[block.index], renames); + } else { + expr_set empty; + process_block(block, empty, renames); + } + } + + for (Block& block : program->blocks) + rename_phi_operands(block, renames); +} + +} diff --git a/src/amd/compiler/aco_optimizer.cpp b/src/amd/compiler/aco_optimizer.cpp new file mode 100644 index 00000000000..fe05634c280 --- /dev/null +++ b/src/amd/compiler/aco_optimizer.cpp @@ -0,0 +1,2401 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Daniel Schürmann (daniel.schuermann@campus.tu-berlin.de) + * + */ + +#include +#include + +#include "aco_ir.h" +#include "util/half_float.h" +#include "util/u_math.h" + +namespace aco { + +/** + * The optimizer works in 4 phases: + * (1) The first pass collects information for each ssa-def, + * propagates reg->reg operands of the same type, inline constants + * and neg/abs input modifiers. + * (2) The second pass combines instructions like mad, omod, clamp and + * propagates sgpr's on VALU instructions. + * This pass depends on information collected in the first pass. + * (3) The third pass goes backwards, and selects instructions, + * i.e. decides if a mad instruction is profitable and eliminates dead code. + * (4) The fourth pass cleans up the sequence: literals get applied and dead + * instructions are removed from the sequence. + */ + + +struct mad_info { + aco_ptr add_instr; + uint32_t mul_temp_id; + uint32_t literal_idx; + bool needs_vop3; + bool check_literal; + + mad_info(aco_ptr instr, uint32_t id, bool vop3) + : add_instr(std::move(instr)), mul_temp_id(id), needs_vop3(vop3), check_literal(false) {} +}; + +enum Label { + label_vec = 1 << 0, + label_constant = 1 << 1, + label_abs = 1 << 2, + label_neg = 1 << 3, + label_mul = 1 << 4, + label_temp = 1 << 5, + label_literal = 1 << 6, + label_mad = 1 << 7, + label_omod2 = 1 << 8, + label_omod4 = 1 << 9, + label_omod5 = 1 << 10, + label_omod_success = 1 << 11, + label_clamp = 1 << 12, + label_clamp_success = 1 << 13, + label_undefined = 1 << 14, + label_vcc = 1 << 15, + label_b2f = 1 << 16, + label_add_sub = 1 << 17, + label_bitwise = 1 << 18, + label_minmax = 1 << 19, + label_fcmp = 1 << 20, +}; + +static constexpr uint32_t instr_labels = label_vec | label_mul | label_mad | label_omod_success | label_clamp_success | label_add_sub | label_bitwise | label_minmax | label_fcmp; +static constexpr uint32_t temp_labels = label_abs | label_neg | label_temp | label_vcc | label_b2f; +static constexpr uint32_t val_labels = label_constant | label_literal | label_mad; + +struct ssa_info { + uint32_t val; + union { + Temp temp; + Instruction* instr; + }; + uint32_t label; + + void add_label(Label new_label) + { + /* Since all labels which use "instr" use it for the same thing + * (indicating the defining instruction), there is no need to clear + * any other instr labels. */ + if (new_label & instr_labels) + label &= ~temp_labels; /* instr and temp alias */ + + if (new_label & temp_labels) { + label &= ~temp_labels; + label &= ~instr_labels; /* instr and temp alias */ + } + + if (new_label & val_labels) + label &= ~val_labels; + + label |= new_label; + } + + void set_vec(Instruction* vec) + { + add_label(label_vec); + instr = vec; + } + + bool is_vec() + { + return label & label_vec; + } + + void set_constant(uint32_t constant) + { + add_label(label_constant); + val = constant; + } + + bool is_constant() + { + return label & label_constant; + } + + void set_abs(Temp abs_temp) + { + add_label(label_abs); + temp = abs_temp; + } + + bool is_abs() + { + return label & label_abs; + } + + void set_neg(Temp neg_temp) + { + add_label(label_neg); + temp = neg_temp; + } + + bool is_neg() + { + return label & label_neg; + } + + void set_neg_abs(Temp neg_abs_temp) + { + add_label((Label)((uint32_t)label_abs | (uint32_t)label_neg)); + temp = neg_abs_temp; + } + + void set_mul(Instruction* mul) + { + add_label(label_mul); + instr = mul; + } + + bool is_mul() + { + return label & label_mul; + } + + void set_temp(Temp tmp) + { + add_label(label_temp); + temp = tmp; + } + + bool is_temp() + { + return label & label_temp; + } + + void set_literal(uint32_t lit) + { + add_label(label_literal); + val = lit; + } + + bool is_literal() + { + return label & label_literal; + } + + void set_mad(Instruction* mad, uint32_t mad_info_idx) + { + add_label(label_mad); + val = mad_info_idx; + instr = mad; + } + + bool is_mad() + { + return label & label_mad; + } + + void set_omod2() + { + add_label(label_omod2); + } + + bool is_omod2() + { + return label & label_omod2; + } + + void set_omod4() + { + add_label(label_omod4); + } + + bool is_omod4() + { + return label & label_omod4; + } + + void set_omod5() + { + add_label(label_omod5); + } + + bool is_omod5() + { + return label & label_omod5; + } + + void set_omod_success(Instruction* omod_instr) + { + add_label(label_omod_success); + instr = omod_instr; + } + + bool is_omod_success() + { + return label & label_omod_success; + } + + void set_clamp() + { + add_label(label_clamp); + } + + bool is_clamp() + { + return label & label_clamp; + } + + void set_clamp_success(Instruction* clamp_instr) + { + add_label(label_clamp_success); + instr = clamp_instr; + } + + bool is_clamp_success() + { + return label & label_clamp_success; + } + + void set_undefined() + { + add_label(label_undefined); + } + + bool is_undefined() + { + return label & label_undefined; + } + + void set_vcc(Temp vcc) + { + add_label(label_vcc); + temp = vcc; + } + + bool is_vcc() + { + return label & label_vcc; + } + + bool is_constant_or_literal() + { + return is_constant() || is_literal(); + } + + void set_b2f(Temp val) + { + add_label(label_b2f); + temp = val; + } + + bool is_b2f() + { + return label & label_b2f; + } + + void set_add_sub(Instruction *add_sub_instr) + { + add_label(label_add_sub); + instr = add_sub_instr; + } + + bool is_add_sub() + { + return label & label_add_sub; + } + + void set_bitwise(Instruction *bitwise_instr) + { + add_label(label_bitwise); + instr = bitwise_instr; + } + + bool is_bitwise() + { + return label & label_bitwise; + } + + void set_minmax(Instruction *minmax_instr) + { + add_label(label_minmax); + instr = minmax_instr; + } + + bool is_minmax() + { + return label & label_minmax; + } + + void set_fcmp(Instruction *fcmp_instr) + { + add_label(label_fcmp); + instr = fcmp_instr; + } + + bool is_fcmp() + { + return label & label_fcmp; + } + +}; + +struct opt_ctx { + Program* program; + std::vector> instructions; + ssa_info* info; + std::pair last_literal; + std::vector mad_infos; + std::vector uses; +}; + +bool can_swap_operands(aco_ptr& instr) +{ + if (instr->operands[0].isConstant() || + (instr->operands[0].isTemp() && instr->operands[0].getTemp().type() == RegType::sgpr)) + return false; + + switch (instr->opcode) { + case aco_opcode::v_add_f32: + case aco_opcode::v_mul_f32: + case aco_opcode::v_or_b32: + case aco_opcode::v_and_b32: + case aco_opcode::v_xor_b32: + case aco_opcode::v_max_f32: + case aco_opcode::v_min_f32: + case aco_opcode::v_cmp_eq_f32: + case aco_opcode::v_cmp_lg_f32: + return true; + case aco_opcode::v_sub_f32: + instr->opcode = aco_opcode::v_subrev_f32; + return true; + case aco_opcode::v_cmp_lt_f32: + instr->opcode = aco_opcode::v_cmp_gt_f32; + return true; + case aco_opcode::v_cmp_ge_f32: + instr->opcode = aco_opcode::v_cmp_le_f32; + return true; + case aco_opcode::v_cmp_lt_i32: + instr->opcode = aco_opcode::v_cmp_gt_i32; + return true; + default: + return false; + } +} + +bool can_use_VOP3(aco_ptr& instr) +{ + if (instr->operands.size() && instr->operands[0].isLiteral()) + return false; + + if (instr->isDPP() || instr->isSDWA()) + return false; + + return instr->opcode != aco_opcode::v_madmk_f32 && + instr->opcode != aco_opcode::v_madak_f32 && + instr->opcode != aco_opcode::v_madmk_f16 && + instr->opcode != aco_opcode::v_madak_f16; +} + +void to_VOP3(opt_ctx& ctx, aco_ptr& instr) +{ + if (instr->isVOP3()) + return; + + assert(!instr->operands[0].isLiteral()); + aco_ptr tmp = std::move(instr); + Format format = asVOP3(tmp->format); + instr.reset(create_instruction(tmp->opcode, format, tmp->operands.size(), tmp->definitions.size())); + std::copy(tmp->operands.cbegin(), tmp->operands.cend(), instr->operands.begin()); + for (unsigned i = 0; i < instr->definitions.size(); i++) { + instr->definitions[i] = tmp->definitions[i]; + if (instr->definitions[i].isTemp()) { + ssa_info& info = ctx.info[instr->definitions[i].tempId()]; + if (info.label & instr_labels && info.instr == tmp.get()) + info.instr = instr.get(); + } + } +} + +bool valu_can_accept_literal(opt_ctx& ctx, aco_ptr& instr) +{ + // TODO: VOP3 can take a literal on GFX10 + return !instr->isSDWA() && !instr->isDPP() && !instr->isVOP3(); +} + +/* only covers special cases */ +bool can_accept_constant(aco_ptr& instr, unsigned operand) +{ + switch (instr->opcode) { + case aco_opcode::v_interp_p2_f32: + case aco_opcode::v_mac_f32: + case aco_opcode::v_writelane_b32: + case aco_opcode::v_cndmask_b32: + return operand != 2; + case aco_opcode::s_addk_i32: + case aco_opcode::s_mulk_i32: + case aco_opcode::p_wqm: + case aco_opcode::p_extract_vector: + case aco_opcode::p_split_vector: + return operand != 0; + default: + if ((instr->format == Format::MUBUF || + instr->format == Format::MIMG) && + instr->definitions.size() == 1 && + instr->operands.size() == 4) { + return operand != 3; + } + return true; + } +} + +bool parse_base_offset(opt_ctx &ctx, Instruction* instr, unsigned op_index, Temp *base, uint32_t *offset) +{ + Operand op = instr->operands[op_index]; + + if (!op.isTemp()) + return false; + Temp tmp = op.getTemp(); + if (!ctx.info[tmp.id()].is_add_sub()) + return false; + + Instruction *add_instr = ctx.info[tmp.id()].instr; + + switch (add_instr->opcode) { + case aco_opcode::v_add_u32: + case aco_opcode::v_add_co_u32: + case aco_opcode::s_add_i32: + case aco_opcode::s_add_u32: + break; + default: + return false; + } + + for (unsigned i = 0; i < 2; i++) { + if (add_instr->operands[i].isConstant()) { + *offset = add_instr->operands[i].constantValue(); + } else if (add_instr->operands[i].isTemp() && + ctx.info[add_instr->operands[i].tempId()].is_constant_or_literal()) { + *offset = ctx.info[add_instr->operands[i].tempId()].val; + } else { + continue; + } + if (!add_instr->operands[!i].isTemp()) + continue; + + uint32_t offset2 = 0; + if (parse_base_offset(ctx, add_instr, !i, base, &offset2)) { + *offset += offset2; + } else { + *base = add_instr->operands[!i].getTemp(); + } + return true; + } + + return false; +} + +void label_instruction(opt_ctx &ctx, aco_ptr& instr) +{ + if (instr->isSALU() || instr->isVALU() || instr->format == Format::PSEUDO) { + ASSERTED bool all_const = false; + for (Operand& op : instr->operands) + all_const = all_const && (!op.isTemp() || ctx.info[op.tempId()].is_constant_or_literal()); + perfwarn(all_const, "All instruction operands are constant", instr.get()); + } + + for (unsigned i = 0; i < instr->operands.size(); i++) + { + if (!instr->operands[i].isTemp()) + continue; + + ssa_info info = ctx.info[instr->operands[i].tempId()]; + /* propagate undef */ + if (info.is_undefined() && is_phi(instr)) + instr->operands[i] = Operand(instr->operands[i].regClass()); + /* propagate reg->reg of same type */ + if (info.is_temp() && info.temp.regClass() == instr->operands[i].getTemp().regClass()) { + instr->operands[i].setTemp(ctx.info[instr->operands[i].tempId()].temp); + info = ctx.info[info.temp.id()]; + } + + /* SALU / PSEUDO: propagate inline constants */ + if (instr->isSALU() || instr->format == Format::PSEUDO) { + if (info.is_temp() && info.temp.type() == RegType::sgpr) { + instr->operands[i].setTemp(info.temp); + info = ctx.info[info.temp.id()]; + } else if (info.is_temp() && info.temp.type() == RegType::vgpr) { + /* propagate vgpr if it can take it */ + switch (instr->opcode) { + case aco_opcode::p_create_vector: + case aco_opcode::p_split_vector: + case aco_opcode::p_extract_vector: + case aco_opcode::p_phi: { + const bool all_vgpr = std::none_of(instr->definitions.begin(), instr->definitions.end(), + [] (const Definition& def) { return def.getTemp().type() != RegType::vgpr;}); + if (all_vgpr) { + instr->operands[i] = Operand(info.temp); + info = ctx.info[info.temp.id()]; + } + break; + } + default: + break; + } + } + if ((info.is_constant() || (info.is_literal() && instr->format == Format::PSEUDO)) && !instr->operands[i].isFixed() && can_accept_constant(instr, i)) { + instr->operands[i] = Operand(info.val); + continue; + } + } + + /* VALU: propagate neg, abs & inline constants */ + else if (instr->isVALU()) { + if (info.is_temp() && info.temp.type() == RegType::vgpr) { + instr->operands[i].setTemp(info.temp); + info = ctx.info[info.temp.id()]; + } + if (info.is_abs() && (can_use_VOP3(instr) || instr->isDPP()) && instr_info.can_use_input_modifiers[(int)instr->opcode]) { + if (!instr->isDPP()) + to_VOP3(ctx, instr); + instr->operands[i] = Operand(info.temp); + if (instr->isDPP()) + static_cast(instr.get())->abs[i] = true; + else + static_cast(instr.get())->abs[i] = true; + } + if (info.is_neg() && instr->opcode == aco_opcode::v_add_f32) { + instr->opcode = i ? aco_opcode::v_sub_f32 : aco_opcode::v_subrev_f32; + instr->operands[i].setTemp(info.temp); + continue; + } else if (info.is_neg() && (can_use_VOP3(instr) || instr->isDPP()) && instr_info.can_use_input_modifiers[(int)instr->opcode]) { + if (!instr->isDPP()) + to_VOP3(ctx, instr); + instr->operands[i].setTemp(info.temp); + if (instr->isDPP()) + static_cast(instr.get())->neg[i] = true; + else + static_cast(instr.get())->neg[i] = true; + continue; + } + if (info.is_constant() && can_accept_constant(instr, i)) { + perfwarn(instr->opcode == aco_opcode::v_cndmask_b32 && i == 2, "v_cndmask_b32 with a constant selector", instr.get()); + if (i == 0) { + instr->operands[i] = Operand(info.val); + continue; + } else if (!instr->isVOP3() && can_swap_operands(instr)) { + instr->operands[i] = instr->operands[0]; + instr->operands[0] = Operand(info.val); + continue; + } else if (can_use_VOP3(instr)) { + to_VOP3(ctx, instr); + instr->operands[i] = Operand(info.val); + continue; + } + } + } + + /* MUBUF: propagate constants and combine additions */ + else if (instr->format == Format::MUBUF) { + MUBUF_instruction *mubuf = static_cast(instr.get()); + Temp base; + uint32_t offset; + while (info.is_temp()) + info = ctx.info[info.temp.id()]; + + if (mubuf->offen && i == 0 && info.is_constant_or_literal() && mubuf->offset + info.val < 4096) { + assert(!mubuf->idxen); + instr->operands[i] = Operand(v1); + mubuf->offset += info.val; + mubuf->offen = false; + continue; + } else if (i == 2 && info.is_constant_or_literal() && mubuf->offset + info.val < 4096) { + instr->operands[2] = Operand((uint32_t) 0); + mubuf->offset += info.val; + continue; + } else if (mubuf->offen && i == 0 && parse_base_offset(ctx, instr.get(), i, &base, &offset) && base.regClass() == v1 && mubuf->offset + offset < 4096) { + assert(!mubuf->idxen); + instr->operands[i].setTemp(base); + mubuf->offset += offset; + continue; + } else if (i == 2 && parse_base_offset(ctx, instr.get(), i, &base, &offset) && base.regClass() == s1 && mubuf->offset + offset < 4096) { + instr->operands[i].setTemp(base); + mubuf->offset += offset; + continue; + } + } + + /* DS: combine additions */ + else if (instr->format == Format::DS) { + + DS_instruction *ds = static_cast(instr.get()); + Temp base; + uint32_t offset; + if (i == 0 && parse_base_offset(ctx, instr.get(), i, &base, &offset) && base.regClass() == instr->operands[i].regClass()) { + if (instr->opcode == aco_opcode::ds_write2_b32 || instr->opcode == aco_opcode::ds_read2_b32) { + if (offset % 4 == 0 && + ds->offset0 + (offset >> 2) <= 255 && + ds->offset1 + (offset >> 2) <= 255) { + instr->operands[i].setTemp(base); + ds->offset0 += offset >> 2; + ds->offset1 += offset >> 2; + } + } else { + if (ds->offset0 + offset <= 65535) { + instr->operands[i].setTemp(base); + ds->offset0 += offset; + } + } + } + } + + /* SMEM: propagate constants and combine additions */ + else if (instr->format == Format::SMEM) { + + SMEM_instruction *smem = static_cast(instr.get()); + Temp base; + uint32_t offset; + if (i == 1 && info.is_constant_or_literal() && info.val <= 0xFFFFF) { + instr->operands[i] = Operand(info.val); + continue; + } else if (i == 1 && parse_base_offset(ctx, instr.get(), i, &base, &offset) && base.regClass() == s1 && offset <= 0xFFFFF && ctx.program->chip_class >= GFX9) { + bool soe = smem->operands.size() >= (!smem->definitions.empty() ? 3 : 4); + if (soe && + (!ctx.info[smem->operands.back().tempId()].is_constant_or_literal() || + ctx.info[smem->operands.back().tempId()].val != 0)) { + continue; + } + if (soe) { + smem->operands[1] = Operand(offset); + smem->operands.back() = Operand(base); + } else { + SMEM_instruction *new_instr = create_instruction(smem->opcode, Format::SMEM, smem->operands.size() + 1, smem->definitions.size()); + new_instr->operands[0] = smem->operands[0]; + new_instr->operands[1] = Operand(offset); + if (smem->definitions.empty()) + new_instr->operands[2] = smem->operands[2]; + new_instr->operands.back() = Operand(base); + if (!smem->definitions.empty()) + new_instr->definitions[0] = smem->definitions[0]; + instr.reset(new_instr); + smem = static_cast(instr.get()); + } + continue; + } + } + } + + /* if this instruction doesn't define anything, return */ + if (instr->definitions.empty()) + return; + + switch (instr->opcode) { + case aco_opcode::p_create_vector: { + unsigned num_ops = instr->operands.size(); + for (const Operand& op : instr->operands) { + if (op.isTemp() && ctx.info[op.tempId()].is_vec()) + num_ops += ctx.info[op.tempId()].instr->operands.size() - 1; + } + if (num_ops != instr->operands.size()) { + aco_ptr old_vec = std::move(instr); + instr.reset(create_instruction(aco_opcode::p_create_vector, Format::PSEUDO, num_ops, 1)); + instr->definitions[0] = old_vec->definitions[0]; + unsigned k = 0; + for (Operand& old_op : old_vec->operands) { + if (old_op.isTemp() && ctx.info[old_op.tempId()].is_vec()) { + for (unsigned j = 0; j < ctx.info[old_op.tempId()].instr->operands.size(); j++) + instr->operands[k++] = ctx.info[old_op.tempId()].instr->operands[j]; + } else { + instr->operands[k++] = old_op; + } + } + assert(k == num_ops); + } + if (instr->operands.size() == 1 && instr->operands[0].isTemp()) + ctx.info[instr->definitions[0].tempId()].set_temp(instr->operands[0].getTemp()); + else if (instr->definitions[0].getTemp().size() == instr->operands.size()) + ctx.info[instr->definitions[0].tempId()].set_vec(instr.get()); + break; + } + case aco_opcode::p_split_vector: { + if (!ctx.info[instr->operands[0].tempId()].is_vec()) + break; + Instruction* vec = ctx.info[instr->operands[0].tempId()].instr; + assert(instr->definitions.size() == vec->operands.size()); + for (unsigned i = 0; i < instr->definitions.size(); i++) { + Operand vec_op = vec->operands[i]; + if (vec_op.isConstant()) { + if (vec_op.isLiteral()) + ctx.info[instr->definitions[i].tempId()].set_literal(vec_op.constantValue()); + else + ctx.info[instr->definitions[i].tempId()].set_constant(vec_op.constantValue()); + } else { + assert(vec_op.isTemp()); + ctx.info[instr->definitions[i].tempId()].set_temp(vec_op.getTemp()); + } + } + break; + } + case aco_opcode::p_extract_vector: { /* mov */ + if (!ctx.info[instr->operands[0].tempId()].is_vec()) + break; + Instruction* vec = ctx.info[instr->operands[0].tempId()].instr; + if (vec->definitions[0].getTemp().size() == vec->operands.size()) { /* TODO: what about 64bit or other combinations? */ + + /* convert this extract into a mov instruction */ + Operand vec_op = vec->operands[instr->operands[1].constantValue()]; + bool is_vgpr = instr->definitions[0].getTemp().type() == RegType::vgpr; + aco_opcode opcode = is_vgpr ? aco_opcode::v_mov_b32 : aco_opcode::s_mov_b32; + Format format = is_vgpr ? Format::VOP1 : Format::SOP1; + instr->opcode = opcode; + instr->format = format; + instr->operands = {instr->operands.begin(), 1 }; + instr->operands[0] = vec_op; + + if (vec_op.isConstant()) { + if (vec_op.isLiteral()) + ctx.info[instr->definitions[0].tempId()].set_literal(vec_op.constantValue()); + else + ctx.info[instr->definitions[0].tempId()].set_constant(vec_op.constantValue()); + } else { + assert(vec_op.isTemp()); + ctx.info[instr->definitions[0].tempId()].set_temp(vec_op.getTemp()); + } + } + break; + } + case aco_opcode::s_mov_b32: /* propagate */ + case aco_opcode::s_mov_b64: + case aco_opcode::v_mov_b32: + case aco_opcode::p_as_uniform: + if (instr->definitions[0].isFixed()) { + /* don't copy-propagate copies into fixed registers */ + } else if (instr->operands[0].isConstant()) { + if (instr->operands[0].isLiteral()) + ctx.info[instr->definitions[0].tempId()].set_literal(instr->operands[0].constantValue()); + else + ctx.info[instr->definitions[0].tempId()].set_constant(instr->operands[0].constantValue()); + } else if (instr->isDPP()) { + // TODO + } else if (instr->operands[0].isTemp()) { + ctx.info[instr->definitions[0].tempId()].set_temp(instr->operands[0].getTemp()); + } else { + assert(instr->operands[0].isFixed()); + } + break; + case aco_opcode::p_is_helper: + if (!ctx.program->needs_wqm) + ctx.info[instr->definitions[0].tempId()].set_constant(0u); + break; + case aco_opcode::s_movk_i32: { + uint32_t v = static_cast(instr.get())->imm; + v = v & 0x8000 ? (v | 0xffff0000) : v; + if (v <= 64 || v >= 0xfffffff0) + ctx.info[instr->definitions[0].tempId()].set_constant(v); + else + ctx.info[instr->definitions[0].tempId()].set_literal(v); + break; + } + case aco_opcode::v_bfrev_b32: + case aco_opcode::s_brev_b32: { + if (instr->operands[0].isConstant()) { + uint32_t v = util_bitreverse(instr->operands[0].constantValue()); + if (v <= 64 || v >= 0xfffffff0) + ctx.info[instr->definitions[0].tempId()].set_constant(v); + else + ctx.info[instr->definitions[0].tempId()].set_literal(v); + } + break; + } + case aco_opcode::s_bfm_b32: { + if (instr->operands[0].isConstant() && instr->operands[1].isConstant()) { + unsigned size = instr->operands[0].constantValue() & 0x1f; + unsigned start = instr->operands[1].constantValue() & 0x1f; + uint32_t v = ((1u << size) - 1u) << start; + if (v <= 64 || v >= 0xfffffff0) + ctx.info[instr->definitions[0].tempId()].set_constant(v); + else + ctx.info[instr->definitions[0].tempId()].set_literal(v); + } + } + case aco_opcode::v_mul_f32: { /* omod */ + /* TODO: try to move the negate/abs modifier to the consumer instead */ + if (instr->isVOP3()) { + VOP3A_instruction *vop3 = static_cast(instr.get()); + if (vop3->abs[0] || vop3->abs[1] || vop3->neg[0] || vop3->neg[1] || vop3->omod || vop3->clamp) + break; + } + + for (unsigned i = 0; i < 2; i++) { + if (instr->operands[!i].isConstant() && instr->operands[i].isTemp()) { + if (instr->operands[!i].constantValue() == 0x40000000) { /* 2.0 */ + ctx.info[instr->operands[i].tempId()].set_omod2(); + } else if (instr->operands[!i].constantValue() == 0x40800000) { /* 4.0 */ + ctx.info[instr->operands[i].tempId()].set_omod4(); + } else if (instr->operands[!i].constantValue() == 0x3f000000) { /* 0.5 */ + ctx.info[instr->operands[i].tempId()].set_omod5(); + } else if (instr->operands[!i].constantValue() == 0x3f800000) { /* 1.0 */ + ctx.info[instr->definitions[0].tempId()].set_temp(instr->operands[i].getTemp()); + } else { + continue; + } + break; + } + } + break; + } + case aco_opcode::v_and_b32: /* abs */ + if (instr->operands[0].constantEquals(0x7FFFFFFF) && instr->operands[1].isTemp()) + ctx.info[instr->definitions[0].tempId()].set_abs(instr->operands[1].getTemp()); + else + ctx.info[instr->definitions[0].tempId()].set_bitwise(instr.get()); + break; + case aco_opcode::v_xor_b32: { /* neg */ + if (instr->operands[0].constantEquals(0x80000000u) && instr->operands[1].isTemp()) { + if (ctx.info[instr->operands[1].tempId()].is_neg()) { + ctx.info[instr->definitions[0].tempId()].set_temp(ctx.info[instr->operands[1].tempId()].temp); + } else { + if (ctx.info[instr->operands[1].tempId()].is_abs()) { /* neg(abs(x)) */ + instr->operands[1].setTemp(ctx.info[instr->operands[1].tempId()].temp); + instr->opcode = aco_opcode::v_or_b32; + ctx.info[instr->definitions[0].tempId()].set_neg_abs(instr->operands[1].getTemp()); + } else { + ctx.info[instr->definitions[0].tempId()].set_neg(instr->operands[1].getTemp()); + } + } + } else { + ctx.info[instr->definitions[0].tempId()].set_bitwise(instr.get()); + } + break; + } + case aco_opcode::v_med3_f32: { /* clamp */ + VOP3A_instruction* vop3 = static_cast(instr.get()); + if (vop3->abs[0] || vop3->neg[0] || vop3->opsel[0] || + vop3->abs[1] || vop3->neg[1] || vop3->opsel[1] || + vop3->abs[2] || vop3->neg[2] || vop3->opsel[2] || + vop3->omod != 0) + break; + + unsigned idx = 0; + bool found_zero = false, found_one = false; + for (unsigned i = 0; i < 3; i++) + { + if (instr->operands[i].constantEquals(0)) + found_zero = true; + else if (instr->operands[i].constantEquals(0x3f800000)) /* 1.0 */ + found_one = true; + else + idx = i; + } + if (found_zero && found_one && instr->operands[idx].isTemp()) { + ctx.info[instr->operands[idx].tempId()].set_clamp(); + } + break; + } + case aco_opcode::v_cndmask_b32: + if (instr->operands[0].constantEquals(0) && + instr->operands[1].constantEquals(0xFFFFFFFF) && + instr->operands[2].isTemp()) + ctx.info[instr->definitions[0].tempId()].set_vcc(instr->operands[2].getTemp()); + else if (instr->operands[0].constantEquals(0) && + instr->operands[1].constantEquals(0x3f800000u) && + instr->operands[2].isTemp()) + ctx.info[instr->definitions[0].tempId()].set_b2f(instr->operands[2].getTemp()); + break; + case aco_opcode::v_cmp_lg_u32: + if (instr->format == Format::VOPC && /* don't optimize VOP3 / SDWA / DPP */ + instr->operands[0].constantEquals(0) && + instr->operands[1].isTemp() && ctx.info[instr->operands[1].tempId()].is_vcc()) + ctx.info[instr->definitions[0].tempId()].set_temp(ctx.info[instr->operands[1].tempId()].temp); + break; + case aco_opcode::p_phi: + case aco_opcode::p_linear_phi: { + /* lower_bool_phis() can create phis like this */ + bool all_same_temp = instr->operands[0].isTemp(); + /* this check is needed when moving uniform loop counters out of a divergent loop */ + if (all_same_temp) + all_same_temp = instr->definitions[0].regClass() == instr->operands[0].regClass(); + for (unsigned i = 1; all_same_temp && (i < instr->operands.size()); i++) { + if (!instr->operands[i].isTemp() || instr->operands[i].tempId() != instr->operands[0].tempId()) + all_same_temp = false; + } + if (all_same_temp) { + ctx.info[instr->definitions[0].tempId()].set_temp(instr->operands[0].getTemp()); + } else { + bool all_undef = instr->operands[0].isUndefined(); + for (unsigned i = 1; all_undef && (i < instr->operands.size()); i++) { + if (!instr->operands[i].isUndefined()) + all_undef = false; + } + if (all_undef) + ctx.info[instr->definitions[0].tempId()].set_undefined(); + } + break; + } + case aco_opcode::v_add_u32: + case aco_opcode::v_add_co_u32: + case aco_opcode::s_add_i32: + case aco_opcode::s_add_u32: + ctx.info[instr->definitions[0].tempId()].set_add_sub(instr.get()); + break; + case aco_opcode::s_not_b32: + case aco_opcode::s_not_b64: + case aco_opcode::s_and_b32: + case aco_opcode::s_and_b64: + case aco_opcode::s_or_b32: + case aco_opcode::s_or_b64: + case aco_opcode::s_xor_b32: + case aco_opcode::s_xor_b64: + case aco_opcode::s_lshl_b32: + case aco_opcode::v_or_b32: + case aco_opcode::v_lshlrev_b32: + ctx.info[instr->definitions[0].tempId()].set_bitwise(instr.get()); + break; + case aco_opcode::v_min_f32: + case aco_opcode::v_min_f16: + case aco_opcode::v_min_u32: + case aco_opcode::v_min_i32: + case aco_opcode::v_min_u16: + case aco_opcode::v_min_i16: + case aco_opcode::v_max_f32: + case aco_opcode::v_max_f16: + case aco_opcode::v_max_u32: + case aco_opcode::v_max_i32: + case aco_opcode::v_max_u16: + case aco_opcode::v_max_i16: + ctx.info[instr->definitions[0].tempId()].set_minmax(instr.get()); + break; + case aco_opcode::v_cmp_lt_f32: + case aco_opcode::v_cmp_eq_f32: + case aco_opcode::v_cmp_le_f32: + case aco_opcode::v_cmp_gt_f32: + case aco_opcode::v_cmp_lg_f32: + case aco_opcode::v_cmp_ge_f32: + case aco_opcode::v_cmp_o_f32: + case aco_opcode::v_cmp_u_f32: + case aco_opcode::v_cmp_nge_f32: + case aco_opcode::v_cmp_nlg_f32: + case aco_opcode::v_cmp_ngt_f32: + case aco_opcode::v_cmp_nle_f32: + case aco_opcode::v_cmp_neq_f32: + case aco_opcode::v_cmp_nlt_f32: + ctx.info[instr->definitions[0].tempId()].set_fcmp(instr.get()); + break; + default: + break; + } +} + +ALWAYS_INLINE bool get_cmp_info(aco_opcode op, aco_opcode *ordered, aco_opcode *unordered, aco_opcode *inverse) +{ + *ordered = *unordered = op; + switch (op) { + #define CMP(ord, unord) \ + case aco_opcode::v_cmp_##ord##_f32:\ + case aco_opcode::v_cmp_n##unord##_f32:\ + *ordered = aco_opcode::v_cmp_##ord##_f32;\ + *unordered = aco_opcode::v_cmp_n##unord##_f32;\ + *inverse = op == aco_opcode::v_cmp_n##unord##_f32 ? aco_opcode::v_cmp_##unord##_f32 : aco_opcode::v_cmp_n##ord##_f32;\ + return true; + CMP(lt, /*n*/ge) + CMP(eq, /*n*/lg) + CMP(le, /*n*/gt) + CMP(gt, /*n*/le) + CMP(lg, /*n*/eq) + CMP(ge, /*n*/lt) + #undef CMP + default: + return false; + } +} + +aco_opcode get_ordered(aco_opcode op) +{ + aco_opcode ordered, unordered, inverse; + return get_cmp_info(op, &ordered, &unordered, &inverse) ? ordered : aco_opcode::last_opcode; +} + +aco_opcode get_unordered(aco_opcode op) +{ + aco_opcode ordered, unordered, inverse; + return get_cmp_info(op, &ordered, &unordered, &inverse) ? unordered : aco_opcode::last_opcode; +} + +aco_opcode get_inverse(aco_opcode op) +{ + aco_opcode ordered, unordered, inverse; + return get_cmp_info(op, &ordered, &unordered, &inverse) ? inverse : aco_opcode::last_opcode; +} + +bool is_cmp(aco_opcode op) +{ + aco_opcode ordered, unordered, inverse; + return get_cmp_info(op, &ordered, &unordered, &inverse); +} + +unsigned original_temp_id(opt_ctx &ctx, Temp tmp) +{ + if (ctx.info[tmp.id()].is_temp()) + return ctx.info[tmp.id()].temp.id(); + else + return tmp.id(); +} + +void decrease_uses(opt_ctx &ctx, Instruction* instr) +{ + if (!--ctx.uses[instr->definitions[0].tempId()]) { + for (const Operand& op : instr->operands) { + if (op.isTemp()) + ctx.uses[op.tempId()]--; + } + } +} + +Instruction *follow_operand(opt_ctx &ctx, Operand op, bool ignore_uses=false) +{ + if (!op.isTemp() || !(ctx.info[op.tempId()].label & instr_labels)) + return nullptr; + if (!ignore_uses && ctx.uses[op.tempId()] > 1) + return nullptr; + + Instruction *instr = ctx.info[op.tempId()].instr; + + if (instr->definitions.size() == 2) { + assert(instr->definitions[0].isTemp() && instr->definitions[0].tempId() == op.tempId()); + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return nullptr; + } + + return instr; +} + +/* s_or_b64(neq(a, a), neq(b, b)) -> v_cmp_u_f32(a, b) + * s_and_b64(eq(a, a), eq(b, b)) -> v_cmp_o_f32(a, b) */ +bool combine_ordering_test(opt_ctx &ctx, aco_ptr& instr) +{ + if (instr->opcode != aco_opcode::s_or_b64 && instr->opcode != aco_opcode::s_and_b64) + return false; + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return false; + + bool neg[2] = {false, false}; + bool abs[2] = {false, false}; + bool opsel[2] = {false, false}; + Instruction *op_instr[2]; + Temp op[2]; + + for (unsigned i = 0; i < 2; i++) { + op_instr[i] = follow_operand(ctx, instr->operands[i], true); + if (!op_instr[i]) + return false; + + aco_opcode expected_cmp = instr->opcode == aco_opcode::s_or_b64 ? + aco_opcode::v_cmp_neq_f32 : aco_opcode::v_cmp_eq_f32; + + if (op_instr[i]->opcode != expected_cmp) + return false; + if (!op_instr[i]->operands[0].isTemp() || !op_instr[i]->operands[1].isTemp()) + return false; + + if (op_instr[i]->isVOP3()) { + VOP3A_instruction *vop3 = static_cast(op_instr[i]); + if (vop3->neg[0] != vop3->neg[1] || vop3->abs[0] != vop3->abs[1] || vop3->opsel[0] != vop3->opsel[1]) + return false; + neg[i] = vop3->neg[0]; + abs[i] = vop3->abs[0]; + opsel[i] = vop3->opsel[0]; + } + + Temp op0 = op_instr[i]->operands[0].getTemp(); + Temp op1 = op_instr[i]->operands[1].getTemp(); + if (original_temp_id(ctx, op0) != original_temp_id(ctx, op1)) + return false; + /* shouldn't happen yet, but best to be safe */ + if (op1.type() != RegType::vgpr) + return false; + + op[i] = op1; + } + + ctx.uses[op[0].id()]++; + ctx.uses[op[1].id()]++; + decrease_uses(ctx, op_instr[0]); + decrease_uses(ctx, op_instr[1]); + + aco_opcode new_op = instr->opcode == aco_opcode::s_or_b64 ? + aco_opcode::v_cmp_u_f32 : aco_opcode::v_cmp_o_f32; + Instruction *new_instr; + if (neg[0] || neg[1] || abs[0] || abs[1] || opsel[0] || opsel[1]) { + VOP3A_instruction *vop3 = create_instruction(new_op, asVOP3(Format::VOPC), 2, 1); + for (unsigned i = 0; i < 2; i++) { + vop3->neg[i] = neg[i]; + vop3->abs[i] = abs[i]; + vop3->opsel[i] = opsel[i]; + } + new_instr = static_cast(vop3); + } else { + new_instr = create_instruction(new_op, Format::VOPC, 2, 1); + } + new_instr->operands[0] = Operand(op[0]); + new_instr->operands[1] = Operand(op[1]); + new_instr->definitions[0] = instr->definitions[0]; + + ctx.info[instr->definitions[0].tempId()].label = 0; + ctx.info[instr->definitions[0].tempId()].set_fcmp(new_instr); + + instr.reset(new_instr); + + return true; +} + +/* s_or_b64(v_cmp_u_f32(a, b), cmp(a, b)) -> get_unordered(cmp)(a, b) + * s_and_b64(v_cmp_o_f32(a, b), cmp(a, b)) -> get_ordered(cmp)(a, b) */ +bool combine_comparison_ordering(opt_ctx &ctx, aco_ptr& instr) +{ + if (instr->opcode != aco_opcode::s_or_b64 && instr->opcode != aco_opcode::s_and_b64) + return false; + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return false; + + aco_opcode expected_nan_test = instr->opcode == aco_opcode::s_or_b64 ? + aco_opcode::v_cmp_u_f32 : aco_opcode::v_cmp_o_f32; + + Instruction *nan_test = follow_operand(ctx, instr->operands[0], true); + Instruction *cmp = follow_operand(ctx, instr->operands[1], true); + if (!nan_test || !cmp) + return false; + + if (cmp->opcode == expected_nan_test) + std::swap(nan_test, cmp); + else if (nan_test->opcode != expected_nan_test) + return false; + + if (!is_cmp(cmp->opcode)) + return false; + + if (!nan_test->operands[0].isTemp() || !nan_test->operands[1].isTemp()) + return false; + if (!cmp->operands[0].isTemp() || !cmp->operands[1].isTemp()) + return false; + + unsigned prop_cmp0 = original_temp_id(ctx, cmp->operands[0].getTemp()); + unsigned prop_cmp1 = original_temp_id(ctx, cmp->operands[1].getTemp()); + unsigned prop_nan0 = original_temp_id(ctx, nan_test->operands[0].getTemp()); + unsigned prop_nan1 = original_temp_id(ctx, nan_test->operands[1].getTemp()); + if (prop_cmp0 != prop_nan0 && prop_cmp0 != prop_nan1) + return false; + if (prop_cmp1 != prop_nan0 && prop_cmp1 != prop_nan1) + return false; + + ctx.uses[cmp->operands[0].tempId()]++; + ctx.uses[cmp->operands[1].tempId()]++; + decrease_uses(ctx, nan_test); + decrease_uses(ctx, cmp); + + aco_opcode new_op = instr->opcode == aco_opcode::s_or_b64 ? + get_unordered(cmp->opcode) : get_ordered(cmp->opcode); + Instruction *new_instr; + if (cmp->isVOP3()) { + VOP3A_instruction *new_vop3 = create_instruction(new_op, asVOP3(Format::VOPC), 2, 1); + VOP3A_instruction *cmp_vop3 = static_cast(cmp); + memcpy(new_vop3->abs, cmp_vop3->abs, sizeof(new_vop3->abs)); + memcpy(new_vop3->opsel, cmp_vop3->opsel, sizeof(new_vop3->opsel)); + memcpy(new_vop3->neg, cmp_vop3->neg, sizeof(new_vop3->neg)); + new_vop3->clamp = cmp_vop3->clamp; + new_vop3->omod = cmp_vop3->omod; + new_instr = new_vop3; + } else { + new_instr = create_instruction(new_op, Format::VOPC, 2, 1); + } + new_instr->operands[0] = cmp->operands[0]; + new_instr->operands[1] = cmp->operands[1]; + new_instr->definitions[0] = instr->definitions[0]; + + ctx.info[instr->definitions[0].tempId()].label = 0; + ctx.info[instr->definitions[0].tempId()].set_fcmp(new_instr); + + instr.reset(new_instr); + + return true; +} + +/* s_or_b64(v_cmp_neq_f32(a, a), cmp(a, #b)) and b is not NaN -> get_unordered(cmp)(a, b) + * s_and_b64(v_cmp_eq_f32(a, a), cmp(a, #b)) and b is not NaN -> get_ordered(cmp)(a, b) */ +bool combine_constant_comparison_ordering(opt_ctx &ctx, aco_ptr& instr) +{ + if (instr->opcode != aco_opcode::s_or_b64 && instr->opcode != aco_opcode::s_and_b64) + return false; + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return false; + + Instruction *nan_test = follow_operand(ctx, instr->operands[0], true); + Instruction *cmp = follow_operand(ctx, instr->operands[1], true); + + if (!nan_test || !cmp) + return false; + + aco_opcode expected_nan_test = instr->opcode == aco_opcode::s_or_b64 ? + aco_opcode::v_cmp_neq_f32 : aco_opcode::v_cmp_eq_f32; + if (cmp->opcode == expected_nan_test) + std::swap(nan_test, cmp); + else if (nan_test->opcode != expected_nan_test) + return false; + + if (!is_cmp(cmp->opcode)) + return false; + + if (!nan_test->operands[0].isTemp() || !nan_test->operands[1].isTemp()) + return false; + if (!cmp->operands[0].isTemp() && !cmp->operands[1].isTemp()) + return false; + + unsigned prop_nan0 = original_temp_id(ctx, nan_test->operands[0].getTemp()); + unsigned prop_nan1 = original_temp_id(ctx, nan_test->operands[1].getTemp()); + if (prop_nan0 != prop_nan1) + return false; + + int constant_operand = -1; + for (unsigned i = 0; i < 2; i++) { + if (cmp->operands[i].isTemp() && original_temp_id(ctx, cmp->operands[i].getTemp()) == prop_nan0) { + constant_operand = !i; + break; + } + } + if (constant_operand == -1) + return false; + + uint32_t constant; + if (cmp->operands[constant_operand].isConstant()) { + constant = cmp->operands[constant_operand].constantValue(); + } else if (cmp->operands[constant_operand].isTemp()) { + unsigned id = cmp->operands[constant_operand].tempId(); + if (!ctx.info[id].is_constant() && !ctx.info[id].is_literal()) + return false; + constant = ctx.info[id].val; + } else { + return false; + } + + float constantf; + memcpy(&constantf, &constant, 4); + if (isnan(constantf)) + return false; + + if (cmp->operands[0].isTemp()) + ctx.uses[cmp->operands[0].tempId()]++; + if (cmp->operands[1].isTemp()) + ctx.uses[cmp->operands[1].tempId()]++; + decrease_uses(ctx, nan_test); + decrease_uses(ctx, cmp); + + aco_opcode new_op = instr->opcode == aco_opcode::s_or_b64 ? + get_unordered(cmp->opcode) : get_ordered(cmp->opcode); + Instruction *new_instr; + if (cmp->isVOP3()) { + VOP3A_instruction *new_vop3 = create_instruction(new_op, asVOP3(Format::VOPC), 2, 1); + VOP3A_instruction *cmp_vop3 = static_cast(cmp); + memcpy(new_vop3->abs, cmp_vop3->abs, sizeof(new_vop3->abs)); + memcpy(new_vop3->opsel, cmp_vop3->opsel, sizeof(new_vop3->opsel)); + memcpy(new_vop3->neg, cmp_vop3->neg, sizeof(new_vop3->neg)); + new_vop3->clamp = cmp_vop3->clamp; + new_vop3->omod = cmp_vop3->omod; + new_instr = new_vop3; + } else { + new_instr = create_instruction(new_op, Format::VOPC, 2, 1); + } + new_instr->operands[0] = cmp->operands[0]; + new_instr->operands[1] = cmp->operands[1]; + new_instr->definitions[0] = instr->definitions[0]; + + ctx.info[instr->definitions[0].tempId()].label = 0; + ctx.info[instr->definitions[0].tempId()].set_fcmp(new_instr); + + instr.reset(new_instr); + + return true; +} + +/* s_not_b64(cmp(a, b) -> get_inverse(cmp)(a, b) */ +bool combine_inverse_comparison(opt_ctx &ctx, aco_ptr& instr) +{ + if (instr->opcode != aco_opcode::s_not_b64) + return false; + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return false; + if (!instr->operands[0].isTemp()) + return false; + + Instruction *cmp = follow_operand(ctx, instr->operands[0]); + if (!cmp) + return false; + + aco_opcode new_opcode = get_inverse(cmp->opcode); + if (new_opcode == aco_opcode::last_opcode) + return false; + + if (cmp->operands[0].isTemp()) + ctx.uses[cmp->operands[0].tempId()]++; + if (cmp->operands[1].isTemp()) + ctx.uses[cmp->operands[1].tempId()]++; + decrease_uses(ctx, cmp); + + Instruction *new_instr; + if (cmp->isVOP3()) { + VOP3A_instruction *new_vop3 = create_instruction(new_opcode, asVOP3(Format::VOPC), 2, 1); + VOP3A_instruction *cmp_vop3 = static_cast(cmp); + memcpy(new_vop3->abs, cmp_vop3->abs, sizeof(new_vop3->abs)); + memcpy(new_vop3->opsel, cmp_vop3->opsel, sizeof(new_vop3->opsel)); + memcpy(new_vop3->neg, cmp_vop3->neg, sizeof(new_vop3->neg)); + new_vop3->clamp = cmp_vop3->clamp; + new_vop3->omod = cmp_vop3->omod; + new_instr = new_vop3; + } else { + new_instr = create_instruction(new_opcode, Format::VOPC, 2, 1); + } + new_instr->operands[0] = cmp->operands[0]; + new_instr->operands[1] = cmp->operands[1]; + new_instr->definitions[0] = instr->definitions[0]; + + ctx.info[instr->definitions[0].tempId()].label = 0; + ctx.info[instr->definitions[0].tempId()].set_fcmp(new_instr); + + instr.reset(new_instr); + + return true; +} + +/* op1(op2(1, 2), 0) if swap = false + * op1(0, op2(1, 2)) if swap = true */ +bool match_op3_for_vop3(opt_ctx &ctx, aco_opcode op1, aco_opcode op2, + Instruction* op1_instr, bool swap, const char *shuffle_str, + Operand operands[3], bool neg[3], bool abs[3], bool opsel[3], + bool *op1_clamp, unsigned *op1_omod, + bool *inbetween_neg, bool *inbetween_abs, bool *inbetween_opsel) +{ + /* checks */ + if (op1_instr->opcode != op1) + return false; + + Instruction *op2_instr = follow_operand(ctx, op1_instr->operands[swap]); + if (!op2_instr || op2_instr->opcode != op2) + return false; + + VOP3A_instruction *op1_vop3 = op1_instr->isVOP3() ? static_cast(op1_instr) : NULL; + VOP3A_instruction *op2_vop3 = op2_instr->isVOP3() ? static_cast(op2_instr) : NULL; + + /* don't support inbetween clamp/omod */ + if (op2_vop3 && (op2_vop3->clamp || op2_vop3->omod)) + return false; + + /* get operands and modifiers and check inbetween modifiers */ + *op1_clamp = op1_vop3 ? op1_vop3->clamp : false; + *op1_omod = op1_vop3 ? op1_vop3->omod : 0u; + + if (inbetween_neg) + *inbetween_neg = op1_vop3 ? op1_vop3->neg[swap] : false; + else if (op1_vop3 && op1_vop3->neg[swap]) + return false; + + if (inbetween_abs) + *inbetween_abs = op1_vop3 ? op1_vop3->abs[swap] : false; + else if (op1_vop3 && op1_vop3->abs[swap]) + return false; + + if (inbetween_opsel) + *inbetween_opsel = op1_vop3 ? op1_vop3->opsel[swap] : false; + else if (op1_vop3 && op1_vop3->opsel[swap]) + return false; + + int shuffle[3]; + shuffle[shuffle_str[0] - '0'] = 0; + shuffle[shuffle_str[1] - '0'] = 1; + shuffle[shuffle_str[2] - '0'] = 2; + + operands[shuffle[0]] = op1_instr->operands[!swap]; + neg[shuffle[0]] = op1_vop3 ? op1_vop3->neg[!swap] : false; + abs[shuffle[0]] = op1_vop3 ? op1_vop3->abs[!swap] : false; + opsel[shuffle[0]] = op1_vop3 ? op1_vop3->opsel[!swap] : false; + + for (unsigned i = 0; i < 2; i++) { + operands[shuffle[i + 1]] = op2_instr->operands[i]; + neg[shuffle[i + 1]] = op2_vop3 ? op2_vop3->neg[i] : false; + abs[shuffle[i + 1]] = op2_vop3 ? op2_vop3->abs[i] : false; + opsel[shuffle[i + 1]] = op2_vop3 ? op2_vop3->opsel[i] : false; + } + + /* check operands */ + unsigned sgpr_id = 0; + for (unsigned i = 0; i < 3; i++) { + Operand op = operands[i]; + if (op.isLiteral()) { + return false; + } else if (op.isTemp() && op.getTemp().type() == RegType::sgpr) { + if (sgpr_id && sgpr_id != op.tempId()) + return false; + sgpr_id = op.tempId(); + } + } + + return true; +} + +void create_vop3_for_op3(opt_ctx& ctx, aco_opcode opcode, aco_ptr& instr, + Operand operands[3], bool neg[3], bool abs[3], bool opsel[3], + bool clamp, unsigned omod) +{ + VOP3A_instruction *new_instr = create_instruction(opcode, Format::VOP3A, 3, 1); + memcpy(new_instr->abs, abs, sizeof(bool[3])); + memcpy(new_instr->opsel, opsel, sizeof(bool[3])); + memcpy(new_instr->neg, neg, sizeof(bool[3])); + new_instr->clamp = clamp; + new_instr->omod = omod; + new_instr->operands[0] = operands[0]; + new_instr->operands[1] = operands[1]; + new_instr->operands[2] = operands[2]; + new_instr->definitions[0] = instr->definitions[0]; + ctx.info[instr->definitions[0].tempId()].label = 0; + + instr.reset(new_instr); +} + +bool combine_minmax3(opt_ctx& ctx, aco_ptr& instr, aco_opcode new_op) +{ + uint32_t omod_clamp = ctx.info[instr->definitions[0].tempId()].label & + (label_omod_success | label_clamp_success); + + for (unsigned swap = 0; swap < 2; swap++) { + Operand operands[3]; + bool neg[3], abs[3], opsel[3], clamp, inbetween_neg, inbetween_abs; + unsigned omod; + if (match_op3_for_vop3(ctx, instr->opcode, instr->opcode, instr.get(), swap, + "012", operands, neg, abs, opsel, + &clamp, &omod, &inbetween_neg, &inbetween_abs, NULL)) { + ctx.uses[instr->operands[swap].tempId()]--; + neg[1] ^= inbetween_neg; + neg[2] ^= inbetween_neg; + abs[1] |= inbetween_abs; + abs[2] |= inbetween_abs; + create_vop3_for_op3(ctx, new_op, instr, operands, neg, abs, opsel, clamp, omod); + if (omod_clamp & label_omod_success) + ctx.info[instr->definitions[0].tempId()].set_omod_success(instr.get()); + if (omod_clamp & label_clamp_success) + ctx.info[instr->definitions[0].tempId()].set_clamp_success(instr.get()); + return true; + } + } + return false; +} + +bool combine_three_valu_op(opt_ctx& ctx, aco_ptr& instr, aco_opcode op2, aco_opcode new_op, const char *shuffle, uint8_t ops) +{ + uint32_t omod_clamp = ctx.info[instr->definitions[0].tempId()].label & + (label_omod_success | label_clamp_success); + + for (unsigned swap = 0; swap < 2; swap++) { + if (!((1 << swap) & ops)) + continue; + + Operand operands[3]; + bool neg[3], abs[3], opsel[3], clamp; + unsigned omod; + if (match_op3_for_vop3(ctx, instr->opcode, op2, + instr.get(), swap, shuffle, + operands, neg, abs, opsel, + &clamp, &omod, NULL, NULL, NULL)) { + ctx.uses[instr->operands[swap].tempId()]--; + create_vop3_for_op3(ctx, new_op, instr, operands, neg, abs, opsel, clamp, omod); + if (omod_clamp & label_omod_success) + ctx.info[instr->definitions[0].tempId()].set_omod_success(instr.get()); + if (omod_clamp & label_clamp_success) + ctx.info[instr->definitions[0].tempId()].set_clamp_success(instr.get()); + return true; + } + } + return false; +} + +/* s_not_b32(s_and_b32(a, b)) -> s_nand_b32(a, b) + * s_not_b32(s_or_b32(a, b)) -> s_nor_b32(a, b) + * s_not_b32(s_xor_b32(a, b)) -> s_xnor_b32(a, b) + * s_not_b64(s_and_b64(a, b)) -> s_nand_b64(a, b) + * s_not_b64(s_or_b64(a, b)) -> s_nor_b64(a, b) + * s_not_b64(s_xor_b64(a, b)) -> s_xnor_b64(a, b) */ +bool combine_salu_not_bitwise(opt_ctx& ctx, aco_ptr& instr) +{ + /* checks */ + if (!instr->operands[0].isTemp()) + return false; + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return false; + + Instruction *op2_instr = follow_operand(ctx, instr->operands[0]); + if (!op2_instr) + return false; + switch (op2_instr->opcode) { + case aco_opcode::s_and_b32: + case aco_opcode::s_or_b32: + case aco_opcode::s_xor_b32: + case aco_opcode::s_and_b64: + case aco_opcode::s_or_b64: + case aco_opcode::s_xor_b64: + break; + default: + return false; + } + + /* create instruction */ + std::swap(instr->definitions[0], op2_instr->definitions[0]); + ctx.uses[instr->operands[0].tempId()]--; + ctx.info[op2_instr->definitions[0].tempId()].label = 0; + + switch (op2_instr->opcode) { + case aco_opcode::s_and_b32: + op2_instr->opcode = aco_opcode::s_nand_b32; + break; + case aco_opcode::s_or_b32: + op2_instr->opcode = aco_opcode::s_nor_b32; + break; + case aco_opcode::s_xor_b32: + op2_instr->opcode = aco_opcode::s_xnor_b32; + break; + case aco_opcode::s_and_b64: + op2_instr->opcode = aco_opcode::s_nand_b64; + break; + case aco_opcode::s_or_b64: + op2_instr->opcode = aco_opcode::s_nor_b64; + break; + case aco_opcode::s_xor_b64: + op2_instr->opcode = aco_opcode::s_xnor_b64; + break; + default: + break; + } + + return true; +} + +/* s_and_b32(a, s_not_b32(b)) -> s_andn2_b32(a, b) + * s_or_b32(a, s_not_b32(b)) -> s_orn2_b32(a, b) + * s_and_b64(a, s_not_b64(b)) -> s_andn2_b64(a, b) + * s_or_b64(a, s_not_b64(b)) -> s_orn2_b64(a, b) */ +bool combine_salu_n2(opt_ctx& ctx, aco_ptr& instr) +{ + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return false; + + for (unsigned i = 0; i < 2; i++) { + Instruction *op2_instr = follow_operand(ctx, instr->operands[i]); + if (!op2_instr || (op2_instr->opcode != aco_opcode::s_not_b32 && op2_instr->opcode != aco_opcode::s_not_b64)) + continue; + + ctx.uses[instr->operands[i].tempId()]--; + instr->operands[0] = instr->operands[!i]; + instr->operands[1] = op2_instr->operands[0]; + ctx.info[instr->definitions[0].tempId()].label = 0; + + switch (instr->opcode) { + case aco_opcode::s_and_b32: + instr->opcode = aco_opcode::s_andn2_b32; + break; + case aco_opcode::s_or_b32: + instr->opcode = aco_opcode::s_orn2_b32; + break; + case aco_opcode::s_and_b64: + instr->opcode = aco_opcode::s_andn2_b64; + break; + case aco_opcode::s_or_b64: + instr->opcode = aco_opcode::s_orn2_b64; + break; + default: + break; + } + + return true; + } + return false; +} + +/* s_add_{i32,u32}(a, s_lshl_b32(b, )) -> s_lshl_add_u32(a, b) */ +bool combine_salu_lshl_add(opt_ctx& ctx, aco_ptr& instr) +{ + if (instr->definitions[1].isTemp() && ctx.uses[instr->definitions[1].tempId()]) + return false; + + for (unsigned i = 0; i < 2; i++) { + Instruction *op2_instr = follow_operand(ctx, instr->operands[i]); + if (!op2_instr || op2_instr->opcode != aco_opcode::s_lshl_b32 || !op2_instr->operands[1].isConstant()) + continue; + + uint32_t shift = op2_instr->operands[1].constantValue(); + if (shift < 1 || shift > 4) + continue; + + ctx.uses[instr->operands[i].tempId()]--; + instr->operands[1] = instr->operands[!i]; + instr->operands[0] = op2_instr->operands[0]; + ctx.info[instr->definitions[0].tempId()].label = 0; + + instr->opcode = ((aco_opcode[]){aco_opcode::s_lshl1_add_u32, + aco_opcode::s_lshl2_add_u32, + aco_opcode::s_lshl3_add_u32, + aco_opcode::s_lshl4_add_u32})[shift - 1]; + + return true; + } + return false; +} + +bool get_minmax_info(aco_opcode op, aco_opcode *min, aco_opcode *max, aco_opcode *min3, aco_opcode *max3, aco_opcode *med3, bool *some_gfx9_only) +{ + switch (op) { + #define MINMAX(type, gfx9) \ + case aco_opcode::v_min_##type:\ + case aco_opcode::v_max_##type:\ + case aco_opcode::v_med3_##type:\ + *min = aco_opcode::v_min_##type;\ + *max = aco_opcode::v_max_##type;\ + *med3 = aco_opcode::v_med3_##type;\ + *min3 = aco_opcode::v_min3_##type;\ + *max3 = aco_opcode::v_max3_##type;\ + *some_gfx9_only = gfx9;\ + return true; + MINMAX(f32, false) + MINMAX(u32, false) + MINMAX(i32, false) + MINMAX(f16, true) + MINMAX(u16, true) + MINMAX(i16, true) + #undef MINMAX + default: + return false; + } +} + +/* v_min_{f,u,i}{16,32}(v_max_{f,u,i}{16,32}(a, lb), ub) -> v_med3_{f,u,i}{16,32}(a, lb, ub) when ub > lb + * v_max_{f,u,i}{16,32}(v_min_{f,u,i}{16,32}(a, ub), lb) -> v_med3_{f,u,i}{16,32}(a, lb, ub) when ub > lb */ +bool combine_clamp(opt_ctx& ctx, aco_ptr& instr, + aco_opcode min, aco_opcode max, aco_opcode med) +{ + aco_opcode other_op; + if (instr->opcode == min) + other_op = max; + else if (instr->opcode == max) + other_op = min; + else + return false; + + uint32_t omod_clamp = ctx.info[instr->definitions[0].tempId()].label & + (label_omod_success | label_clamp_success); + + for (unsigned swap = 0; swap < 2; swap++) { + Operand operands[3]; + bool neg[3], abs[3], opsel[3], clamp, inbetween_neg, inbetween_abs; + unsigned omod; + if (match_op3_for_vop3(ctx, instr->opcode, other_op, instr.get(), swap, + "012", operands, neg, abs, opsel, + &clamp, &omod, &inbetween_neg, &inbetween_abs, NULL)) { + int const0_idx = -1, const1_idx = -1; + uint32_t const0 = 0, const1 = 0; + for (int i = 0; i < 3; i++) { + uint32_t val; + if (operands[i].isConstant()) { + val = operands[i].constantValue(); + } else if (operands[i].isTemp() && ctx.uses[operands[i].tempId()] == 1 && + ctx.info[operands[i].tempId()].is_constant_or_literal()) { + val = ctx.info[operands[i].tempId()].val; + } else { + continue; + } + if (const0_idx >= 0) { + const1_idx = i; + const1 = val; + } else { + const0_idx = i; + const0 = val; + } + } + if (const0_idx < 0 || const1_idx < 0) + continue; + + if (opsel[const0_idx]) + const0 >>= 16; + if (opsel[const1_idx]) + const1 >>= 16; + + int lower_idx = const0_idx; + switch (min) { + case aco_opcode::v_min_f32: + case aco_opcode::v_min_f16: { + float const0_f, const1_f; + if (min == aco_opcode::v_min_f32) { + memcpy(&const0_f, &const0, 4); + memcpy(&const1_f, &const1, 4); + } else { + const0_f = _mesa_half_to_float(const0); + const1_f = _mesa_half_to_float(const1); + } + if (abs[const0_idx]) const0_f = fabsf(const0_f); + if (abs[const1_idx]) const1_f = fabsf(const1_f); + if (neg[const0_idx]) const0_f = -const0_f; + if (neg[const1_idx]) const1_f = -const1_f; + lower_idx = const0_f < const1_f ? const0_idx : const1_idx; + break; + } + case aco_opcode::v_min_u32: { + lower_idx = const0 < const1 ? const0_idx : const1_idx; + break; + } + case aco_opcode::v_min_u16: { + lower_idx = (uint16_t)const0 < (uint16_t)const1 ? const0_idx : const1_idx; + break; + } + case aco_opcode::v_min_i32: { + int32_t const0_i = const0 & 0x80000000u ? -2147483648 + (int32_t)(const0 & 0x7fffffffu) : const0; + int32_t const1_i = const1 & 0x80000000u ? -2147483648 + (int32_t)(const1 & 0x7fffffffu) : const1; + lower_idx = const0_i < const1_i ? const0_idx : const1_idx; + break; + } + case aco_opcode::v_min_i16: { + int16_t const0_i = const0 & 0x8000u ? -32768 + (int16_t)(const0 & 0x7fffu) : const0; + int16_t const1_i = const1 & 0x8000u ? -32768 + (int16_t)(const1 & 0x7fffu) : const1; + lower_idx = const0_i < const1_i ? const0_idx : const1_idx; + break; + } + default: + break; + } + int upper_idx = lower_idx == const0_idx ? const1_idx : const0_idx; + + if (instr->opcode == min) { + if (upper_idx != 0 || lower_idx == 0) + return false; + } else { + if (upper_idx == 0 || lower_idx != 0) + return false; + } + + neg[1] ^= inbetween_neg; + neg[2] ^= inbetween_neg; + abs[1] |= inbetween_abs; + abs[2] |= inbetween_abs; + + ctx.uses[instr->operands[swap].tempId()]--; + create_vop3_for_op3(ctx, med, instr, operands, neg, abs, opsel, clamp, omod); + if (omod_clamp & label_omod_success) + ctx.info[instr->definitions[0].tempId()].set_omod_success(instr.get()); + if (omod_clamp & label_clamp_success) + ctx.info[instr->definitions[0].tempId()].set_clamp_success(instr.get()); + + return true; + } + } + + return false; +} + + +void apply_sgprs(opt_ctx &ctx, aco_ptr& instr) +{ + /* apply sgprs */ + uint32_t sgpr_idx = 0; + uint32_t sgpr_info_id = 0; + bool has_sgpr = false; + uint32_t sgpr_ssa_id = 0; + /* find 'best' possible sgpr */ + for (unsigned i = 0; i < instr->operands.size(); i++) + { + if (instr->operands[i].isLiteral()) { + has_sgpr = true; + break; + } + if (!instr->operands[i].isTemp()) + continue; + if (instr->operands[i].getTemp().type() == RegType::sgpr) { + has_sgpr = true; + sgpr_ssa_id = instr->operands[i].tempId(); + continue; + } + ssa_info& info = ctx.info[instr->operands[i].tempId()]; + if (info.is_temp() && info.temp.type() == RegType::sgpr) { + uint16_t uses = ctx.uses[instr->operands[i].tempId()]; + if (sgpr_info_id == 0 || uses < ctx.uses[sgpr_info_id]) { + sgpr_idx = i; + sgpr_info_id = instr->operands[i].tempId(); + } + } + } + if (!has_sgpr && sgpr_info_id != 0) { + ssa_info& info = ctx.info[sgpr_info_id]; + if (sgpr_idx == 0 || instr->isVOP3()) { + instr->operands[sgpr_idx] = Operand(info.temp); + ctx.uses[sgpr_info_id]--; + ctx.uses[info.temp.id()]++; + } else if (can_swap_operands(instr)) { + instr->operands[sgpr_idx] = instr->operands[0]; + instr->operands[0] = Operand(info.temp); + ctx.uses[sgpr_info_id]--; + ctx.uses[info.temp.id()]++; + } else if (can_use_VOP3(instr)) { + to_VOP3(ctx, instr); + instr->operands[sgpr_idx] = Operand(info.temp); + ctx.uses[sgpr_info_id]--; + ctx.uses[info.temp.id()]++; + } + + /* we can have two sgprs on one instruction if it is the same sgpr! */ + } else if (sgpr_info_id != 0 && + sgpr_ssa_id == sgpr_info_id && + ctx.uses[sgpr_info_id] == 1 && + can_use_VOP3(instr)) { + to_VOP3(ctx, instr); + instr->operands[sgpr_idx] = Operand(ctx.info[sgpr_info_id].temp); + ctx.uses[sgpr_info_id]--; + ctx.uses[ctx.info[sgpr_info_id].temp.id()]++; + } +} + +bool apply_omod_clamp(opt_ctx &ctx, aco_ptr& instr) +{ + /* check if we could apply omod on predecessor */ + if (instr->opcode == aco_opcode::v_mul_f32) { + if (instr->operands[1].isTemp() && ctx.info[instr->operands[1].tempId()].is_omod_success()) { + + /* omod was successfully applied */ + /* if the omod instruction is v_mad, we also have to change the original add */ + if (ctx.info[instr->operands[1].tempId()].is_mad()) { + Instruction* add_instr = ctx.mad_infos[ctx.info[instr->operands[1].tempId()].val].add_instr.get(); + if (ctx.info[instr->definitions[0].tempId()].is_clamp()) + static_cast(add_instr)->clamp = true; + add_instr->definitions[0] = instr->definitions[0]; + } + + Instruction* omod_instr = ctx.info[instr->operands[1].tempId()].instr; + /* check if we have an additional clamp modifier */ + if (ctx.info[instr->definitions[0].tempId()].is_clamp() && ctx.uses[instr->definitions[0].tempId()] == 1) { + static_cast(omod_instr)->clamp = true; + ctx.info[instr->definitions[0].tempId()].set_clamp_success(omod_instr); + } + /* change definition ssa-id of modified instruction */ + omod_instr->definitions[0] = instr->definitions[0]; + + /* change the definition of instr to something unused, e.g. the original omod def */ + instr->definitions[0] = Definition(instr->operands[1].getTemp()); + ctx.uses[instr->definitions[0].tempId()] = 0; + return true; + } + if (!ctx.info[instr->definitions[0].tempId()].label) { + /* in all other cases, label this instruction as option for multiply-add */ + ctx.info[instr->definitions[0].tempId()].set_mul(instr.get()); + } + } + + /* check if we could apply clamp on predecessor */ + if (instr->opcode == aco_opcode::v_med3_f32) { + unsigned idx = 0; + bool found_zero = false, found_one = false; + for (unsigned i = 0; i < 3; i++) + { + if (instr->operands[i].constantEquals(0)) + found_zero = true; + else if (instr->operands[i].constantEquals(0x3f800000)) /* 1.0 */ + found_one = true; + else + idx = i; + } + if (found_zero && found_one && instr->operands[idx].isTemp() && + ctx.info[instr->operands[idx].tempId()].is_clamp_success()) { + /* clamp was successfully applied */ + /* if the clamp instruction is v_mad, we also have to change the original add */ + if (ctx.info[instr->operands[idx].tempId()].is_mad()) { + Instruction* add_instr = ctx.mad_infos[ctx.info[instr->operands[idx].tempId()].val].add_instr.get(); + add_instr->definitions[0] = instr->definitions[0]; + } + Instruction* clamp_instr = ctx.info[instr->operands[idx].tempId()].instr; + /* change definition ssa-id of modified instruction */ + clamp_instr->definitions[0] = instr->definitions[0]; + + /* change the definition of instr to something unused, e.g. the original omod def */ + instr->definitions[0] = Definition(instr->operands[idx].getTemp()); + ctx.uses[instr->definitions[0].tempId()] = 0; + return true; + } + } + + /* apply omod / clamp modifiers if the def is used only once and the instruction can have modifiers */ + if (!instr->definitions.empty() && ctx.uses[instr->definitions[0].tempId()] == 1 && + can_use_VOP3(instr) && instr_info.can_use_output_modifiers[(int)instr->opcode]) { + if(ctx.info[instr->definitions[0].tempId()].is_omod2()) { + to_VOP3(ctx, instr); + static_cast(instr.get())->omod = 1; + ctx.info[instr->definitions[0].tempId()].set_omod_success(instr.get()); + } else if (ctx.info[instr->definitions[0].tempId()].is_omod4()) { + to_VOP3(ctx, instr); + static_cast(instr.get())->omod = 2; + ctx.info[instr->definitions[0].tempId()].set_omod_success(instr.get()); + } else if (ctx.info[instr->definitions[0].tempId()].is_omod5()) { + to_VOP3(ctx, instr); + static_cast(instr.get())->omod = 3; + ctx.info[instr->definitions[0].tempId()].set_omod_success(instr.get()); + } else if (ctx.info[instr->definitions[0].tempId()].is_clamp()) { + to_VOP3(ctx, instr); + static_cast(instr.get())->clamp = true; + ctx.info[instr->definitions[0].tempId()].set_clamp_success(instr.get()); + } + } + + return false; +} + +// TODO: we could possibly move the whole label_instruction pass to combine_instruction: +// this would mean that we'd have to fix the instruction uses while value propagation + +void combine_instruction(opt_ctx &ctx, aco_ptr& instr) +{ + if (instr->definitions.empty() || !ctx.uses[instr->definitions[0].tempId()]) + return; + + if (instr->isVALU()) { + apply_sgprs(ctx, instr); + if (apply_omod_clamp(ctx, instr)) + return; + } + + /* TODO: There are still some peephole optimizations that could be done: + * - abs(a - b) -> s_absdiff_i32 + * - various patterns for s_bitcmp{0,1}_b32 and s_bitset{0,1}_b32 + * - patterns for v_alignbit_b32 and v_alignbyte_b32 + * These aren't probably too interesting though. + * There are also patterns for v_cmp_class_f{16,32,64}. This is difficult but + * probably more useful than the previously mentioned optimizations. + * The various comparison optimizations also currently only work with 32-bit + * floats. */ + + /* neg(mul(a, b)) -> mul(neg(a), b) */ + if (ctx.info[instr->definitions[0].tempId()].is_neg() && ctx.uses[instr->operands[1].tempId()] == 1) { + Temp val = ctx.info[instr->definitions[0].tempId()].temp; + + if (!ctx.info[val.id()].is_mul()) + return; + + Instruction* mul_instr = ctx.info[val.id()].instr; + + if (mul_instr->operands[0].isLiteral()) + return; + if (mul_instr->isVOP3() && static_cast(mul_instr)->clamp) + return; + + /* convert to mul(neg(a), b) */ + ctx.uses[mul_instr->definitions[0].tempId()]--; + Definition def = instr->definitions[0]; + /* neg(abs(mul(a, b))) -> mul(neg(abs(a)), abs(b)) */ + bool is_abs = ctx.info[instr->definitions[0].tempId()].is_abs(); + instr.reset(create_instruction(aco_opcode::v_mul_f32, asVOP3(Format::VOP2), 2, 1)); + instr->operands[0] = mul_instr->operands[0]; + instr->operands[1] = mul_instr->operands[1]; + instr->definitions[0] = def; + VOP3A_instruction* new_mul = static_cast(instr.get()); + if (mul_instr->isVOP3()) { + VOP3A_instruction* mul = static_cast(mul_instr); + new_mul->neg[0] = mul->neg[0] && !is_abs; + new_mul->neg[1] = mul->neg[1] && !is_abs; + new_mul->abs[0] = mul->abs[0] || is_abs; + new_mul->abs[1] = mul->abs[1] || is_abs; + new_mul->omod = mul->omod; + } + new_mul->neg[0] ^= true; + new_mul->clamp = false; + + ctx.info[instr->definitions[0].tempId()].set_mul(instr.get()); + return; + } + /* combine mul+add -> mad */ + else if (instr->opcode == aco_opcode::v_add_f32 || + instr->opcode == aco_opcode::v_sub_f32 || + instr->opcode == aco_opcode::v_subrev_f32) { + + uint32_t uses_src0 = UINT32_MAX; + uint32_t uses_src1 = UINT32_MAX; + Instruction* mul_instr = nullptr; + unsigned add_op_idx; + /* check if any of the operands is a multiplication */ + if (instr->operands[0].isTemp() && ctx.info[instr->operands[0].tempId()].is_mul()) + uses_src0 = ctx.uses[instr->operands[0].tempId()]; + if (instr->operands[1].isTemp() && ctx.info[instr->operands[1].tempId()].is_mul()) + uses_src1 = ctx.uses[instr->operands[1].tempId()]; + + /* find the 'best' mul instruction to combine with the add */ + if (uses_src0 < uses_src1) { + mul_instr = ctx.info[instr->operands[0].tempId()].instr; + add_op_idx = 1; + } else if (uses_src1 < uses_src0) { + mul_instr = ctx.info[instr->operands[1].tempId()].instr; + add_op_idx = 0; + } else if (uses_src0 != UINT32_MAX) { + /* tiebreaker: quite random what to pick */ + if (ctx.info[instr->operands[0].tempId()].instr->operands[0].isLiteral()) { + mul_instr = ctx.info[instr->operands[1].tempId()].instr; + add_op_idx = 0; + } else { + mul_instr = ctx.info[instr->operands[0].tempId()].instr; + add_op_idx = 1; + } + } + if (mul_instr) { + Operand op[3] = {Operand(v1), Operand(v1), Operand(v1)}; + bool neg[3] = {false, false, false}; + bool abs[3] = {false, false, false}; + unsigned omod = 0; + bool clamp = false; + bool need_vop3 = false; + int num_sgpr = 0; + op[0] = mul_instr->operands[0]; + op[1] = mul_instr->operands[1]; + op[2] = instr->operands[add_op_idx]; + for (unsigned i = 0; i < 3; i++) + { + if (op[i].isLiteral()) + return; + if (op[i].isTemp() && op[i].getTemp().type() == RegType::sgpr) + num_sgpr++; + if (!(i == 0 || (op[i].isTemp() && op[i].getTemp().type() == RegType::vgpr))) + need_vop3 = true; + } + // TODO: would be better to check this before selecting a mul instr? + if (num_sgpr > 1) + return; + + if (mul_instr->isVOP3()) { + VOP3A_instruction* vop3 = static_cast (mul_instr); + neg[0] = vop3->neg[0]; + neg[1] = vop3->neg[1]; + abs[0] = vop3->abs[0]; + abs[1] = vop3->abs[1]; + need_vop3 = true; + /* we cannot use these modifiers between mul and add */ + if (vop3->clamp || vop3->omod) + return; + } + + /* convert to mad */ + ctx.uses[mul_instr->definitions[0].tempId()]--; + if (ctx.uses[mul_instr->definitions[0].tempId()]) { + if (op[0].isTemp()) + ctx.uses[op[0].tempId()]++; + if (op[1].isTemp()) + ctx.uses[op[1].tempId()]++; + } + + if (instr->isVOP3()) { + VOP3A_instruction* vop3 = static_cast (instr.get()); + neg[2] = vop3->neg[add_op_idx]; + abs[2] = vop3->abs[add_op_idx]; + omod = vop3->omod; + clamp = vop3->clamp; + /* abs of the multiplication result */ + if (vop3->abs[1 - add_op_idx]) { + neg[0] = false; + neg[1] = false; + abs[0] = true; + abs[1] = true; + } + /* neg of the multiplication result */ + neg[1] = neg[1] ^ vop3->neg[1 - add_op_idx]; + need_vop3 = true; + } + if (instr->opcode == aco_opcode::v_sub_f32) { + neg[1 + add_op_idx] = neg[1 + add_op_idx] ^ true; + need_vop3 = true; + } else if (instr->opcode == aco_opcode::v_subrev_f32) { + neg[2 - add_op_idx] = neg[2 - add_op_idx] ^ true; + need_vop3 = true; + } + + aco_ptr mad{create_instruction(aco_opcode::v_mad_f32, Format::VOP3A, 3, 1)}; + for (unsigned i = 0; i < 3; i++) + { + mad->operands[i] = op[i]; + mad->neg[i] = neg[i]; + mad->abs[i] = abs[i]; + } + mad->omod = omod; + mad->clamp = clamp; + mad->definitions[0] = instr->definitions[0]; + + /* mark this ssa_def to be re-checked for profitability and literals */ + ctx.mad_infos.emplace_back(std::move(instr), mul_instr->definitions[0].tempId(), need_vop3); + ctx.info[mad->definitions[0].tempId()].set_mad(mad.get(), ctx.mad_infos.size() - 1); + instr.reset(mad.release()); + return; + } + } + /* v_mul_f32(v_cndmask_b32(0, 1.0, cond), a) -> v_cndmask_b32(0, a, cond) */ + else if (instr->opcode == aco_opcode::v_mul_f32 && !instr->isVOP3()) { + for (unsigned i = 0; i < 2; i++) { + if (instr->operands[i].isTemp() && ctx.info[instr->operands[i].tempId()].is_b2f() && + ctx.uses[instr->operands[i].tempId()] == 1 && + instr->operands[!i].isTemp() && instr->operands[!i].getTemp().type() == RegType::vgpr) { + ctx.uses[instr->operands[i].tempId()]--; + ctx.uses[ctx.info[instr->operands[i].tempId()].temp.id()]++; + + aco_ptr new_instr{create_instruction(aco_opcode::v_cndmask_b32, Format::VOP2, 3, 1)}; + new_instr->operands[0] = Operand(0u); + new_instr->operands[1] = instr->operands[!i]; + new_instr->operands[2] = Operand(ctx.info[instr->operands[i].tempId()].temp); + new_instr->definitions[0] = instr->definitions[0]; + instr.reset(new_instr.release()); + ctx.info[instr->definitions[0].tempId()].label = 0; + return; + } + } + } else if (instr->opcode == aco_opcode::v_or_b32 && ctx.program->chip_class >= GFX9) { + if (combine_three_valu_op(ctx, instr, aco_opcode::v_or_b32, aco_opcode::v_or3_b32, "012", 1 | 2)) ; + else if (combine_three_valu_op(ctx, instr, aco_opcode::v_and_b32, aco_opcode::v_and_or_b32, "120", 1 | 2)) ; + else combine_three_valu_op(ctx, instr, aco_opcode::v_lshlrev_b32, aco_opcode::v_lshl_or_b32, "210", 1 | 2); + } else if (instr->opcode == aco_opcode::v_add_u32 && ctx.program->chip_class >= GFX9) { + if (combine_three_valu_op(ctx, instr, aco_opcode::v_xor_b32, aco_opcode::v_xad_u32, "120", 1 | 2)) ; + else if (combine_three_valu_op(ctx, instr, aco_opcode::v_add_u32, aco_opcode::v_add3_u32, "012", 1 | 2)) ; + else combine_three_valu_op(ctx, instr, aco_opcode::v_lshlrev_b32, aco_opcode::v_lshl_add_u32, "210", 1 | 2); + } else if (instr->opcode == aco_opcode::v_lshlrev_b32 && ctx.program->chip_class >= GFX9) { + combine_three_valu_op(ctx, instr, aco_opcode::v_add_u32, aco_opcode::v_add_lshl_u32, "120", 2); + } else if ((instr->opcode == aco_opcode::s_add_u32 || instr->opcode == aco_opcode::s_add_i32) && ctx.program->chip_class >= GFX9) { + combine_salu_lshl_add(ctx, instr); + } else if (instr->opcode == aco_opcode::s_not_b32) { + combine_salu_not_bitwise(ctx, instr); + } else if (instr->opcode == aco_opcode::s_not_b64) { + if (combine_inverse_comparison(ctx, instr)) ; + else combine_salu_not_bitwise(ctx, instr); + } else if (instr->opcode == aco_opcode::s_and_b32 || instr->opcode == aco_opcode::s_or_b32) { + combine_salu_n2(ctx, instr); + } else if (instr->opcode == aco_opcode::s_and_b64 || instr->opcode == aco_opcode::s_or_b64) { + if (combine_ordering_test(ctx, instr)) ; + else if (combine_comparison_ordering(ctx, instr)) ; + else if (combine_constant_comparison_ordering(ctx, instr)) ; + else combine_salu_n2(ctx, instr); + } else { + aco_opcode min, max, min3, max3, med3; + bool some_gfx9_only; + if (get_minmax_info(instr->opcode, &min, &max, &min3, &max3, &med3, &some_gfx9_only) && + (!some_gfx9_only || ctx.program->chip_class >= GFX9)) { + if (combine_minmax3(ctx, instr, instr->opcode == min ? min3 : max3)) ; + else combine_clamp(ctx, instr, min, max, med3); + } + } +} + + +void select_instruction(opt_ctx &ctx, aco_ptr& instr) +{ + const uint32_t threshold = 4; + + /* Dead Code Elimination: + * We remove instructions if they define temporaries which all are unused */ + const bool is_used = instr->definitions.empty() || + std::any_of(instr->definitions.begin(), instr->definitions.end(), + [&ctx](const Definition& def) { return ctx.uses[def.tempId()]; }); + if (!is_used) { + instr.reset(); + return; + } + + /* convert split_vector into extract_vector if only one definition is ever used */ + if (instr->opcode == aco_opcode::p_split_vector) { + unsigned num_used = 0; + unsigned idx = 0; + for (unsigned i = 0; i < instr->definitions.size(); i++) { + if (ctx.uses[instr->definitions[i].tempId()]) { + num_used++; + idx = i; + } + } + if (num_used == 1) { + aco_ptr extract{create_instruction(aco_opcode::p_extract_vector, Format::PSEUDO, 2, 1)}; + extract->operands[0] = instr->operands[0]; + extract->operands[1] = Operand((uint32_t) idx); + extract->definitions[0] = instr->definitions[idx]; + instr.reset(extract.release()); + } + } + + /* re-check mad instructions */ + if (instr->opcode == aco_opcode::v_mad_f32 && ctx.info[instr->definitions[0].tempId()].is_mad()) { + mad_info* info = &ctx.mad_infos[ctx.info[instr->definitions[0].tempId()].val]; + /* first, check profitability */ + if (ctx.uses[info->mul_temp_id]) { + ctx.uses[info->mul_temp_id]++; + instr.swap(info->add_instr); + + /* second, check possible literals */ + } else if (!info->needs_vop3) { + uint32_t literal_idx = 0; + uint32_t literal_uses = UINT32_MAX; + for (unsigned i = 0; i < instr->operands.size(); i++) + { + if (!instr->operands[i].isTemp()) + continue; + /* if one of the operands is sgpr, we cannot add a literal somewhere else */ + if (instr->operands[i].getTemp().type() == RegType::sgpr) { + if (ctx.info[instr->operands[i].tempId()].is_literal()) { + literal_uses = ctx.uses[instr->operands[i].tempId()]; + literal_idx = i; + } else { + literal_uses = UINT32_MAX; + } + break; + } + else if (ctx.info[instr->operands[i].tempId()].is_literal() && + ctx.uses[instr->operands[i].tempId()] < literal_uses) { + literal_uses = ctx.uses[instr->operands[i].tempId()]; + literal_idx = i; + } + } + if (literal_uses < threshold) { + ctx.uses[instr->operands[literal_idx].tempId()]--; + info->check_literal = true; + info->literal_idx = literal_idx; + } + } + return; + } + + /* check for literals */ + /* we do not apply the literals yet as we don't know if it is profitable */ + if (instr->isSALU()) { + uint32_t literal_idx = 0; + uint32_t literal_uses = UINT32_MAX; + bool has_literal = false; + for (unsigned i = 0; i < instr->operands.size(); i++) + { + if (instr->operands[i].isLiteral()) { + has_literal = true; + break; + } + if (!instr->operands[i].isTemp()) + continue; + if (ctx.info[instr->operands[i].tempId()].is_literal() && + ctx.uses[instr->operands[i].tempId()] < literal_uses) { + literal_uses = ctx.uses[instr->operands[i].tempId()]; + literal_idx = i; + } + } + if (!has_literal && literal_uses < threshold) { + ctx.uses[instr->operands[literal_idx].tempId()]--; + if (ctx.uses[instr->operands[literal_idx].tempId()] == 0) + instr->operands[literal_idx] = Operand(ctx.info[instr->operands[literal_idx].tempId()].val); + } + } else if (instr->isVALU() && valu_can_accept_literal(ctx, instr) && + instr->operands[0].isTemp() && + ctx.info[instr->operands[0].tempId()].is_literal() && + ctx.uses[instr->operands[0].tempId()] < threshold) { + ctx.uses[instr->operands[0].tempId()]--; + if (ctx.uses[instr->operands[0].tempId()] == 0) + instr->operands[0] = Operand(ctx.info[instr->operands[0].tempId()].val); + } + +} + + +void apply_literals(opt_ctx &ctx, aco_ptr& instr) +{ + /* Cleanup Dead Instructions */ + if (!instr) + return; + + /* apply literals on SALU */ + if (instr->isSALU()) { + for (Operand& op : instr->operands) { + if (!op.isTemp()) + continue; + if (op.isLiteral()) + break; + if (ctx.info[op.tempId()].is_literal() && + ctx.uses[op.tempId()] == 0) + op = Operand(ctx.info[op.tempId()].val); + } + } + + /* apply literals on VALU */ + else if (instr->isVALU() && !instr->isVOP3() && + instr->operands[0].isTemp() && + ctx.info[instr->operands[0].tempId()].is_literal() && + ctx.uses[instr->operands[0].tempId()] == 0) { + instr->operands[0] = Operand(ctx.info[instr->operands[0].tempId()].val); + } + + /* apply literals on MAD */ + else if (instr->opcode == aco_opcode::v_mad_f32 && ctx.info[instr->definitions[0].tempId()].is_mad()) { + mad_info* info = &ctx.mad_infos[ctx.info[instr->definitions[0].tempId()].val]; + aco_ptr new_mad; + if (info->check_literal && ctx.uses[instr->operands[info->literal_idx].tempId()] == 0) { + if (info->literal_idx == 2) { /* add literal -> madak */ + new_mad.reset(create_instruction(aco_opcode::v_madak_f32, Format::VOP2, 3, 1)); + new_mad->operands[0] = instr->operands[0]; + new_mad->operands[1] = instr->operands[1]; + } else { /* mul literal -> madmk */ + new_mad.reset(create_instruction(aco_opcode::v_madmk_f32, Format::VOP2, 3, 1)); + new_mad->operands[0] = instr->operands[1 - info->literal_idx]; + new_mad->operands[1] = instr->operands[2]; + } + new_mad->operands[2] = Operand(ctx.info[instr->operands[info->literal_idx].tempId()].val); + new_mad->definitions[0] = instr->definitions[0]; + instr.swap(new_mad); + } + } + + ctx.instructions.emplace_back(std::move(instr)); +} + + +void optimize(Program* program) +{ + opt_ctx ctx; + ctx.program = program; + std::vector info(program->peekAllocationId()); + ctx.info = info.data(); + + /* 1. Bottom-Up DAG pass (forward) to label all ssa-defs */ + for (Block& block : program->blocks) { + for (aco_ptr& instr : block.instructions) + label_instruction(ctx, instr); + } + + ctx.uses = std::move(dead_code_analysis(program)); + + /* 2. Combine v_mad, omod, clamp and propagate sgpr on VALU instructions */ + for (Block& block : program->blocks) { + for (aco_ptr& instr : block.instructions) + combine_instruction(ctx, instr); + } + + /* 3. Top-Down DAG pass (backward) to select instructions (includes DCE) */ + for (std::vector::reverse_iterator it = program->blocks.rbegin(); it != program->blocks.rend(); ++it) { + Block* block = &(*it); + for (std::vector>::reverse_iterator it = block->instructions.rbegin(); it != block->instructions.rend(); ++it) + select_instruction(ctx, *it); + } + + /* 4. Add literals to instructions */ + for (Block& block : program->blocks) { + ctx.instructions.clear(); + for (aco_ptr& instr : block.instructions) + apply_literals(ctx, instr); + block.instructions.swap(ctx.instructions); + } + +} + +} diff --git a/src/amd/compiler/aco_print_asm.cpp b/src/amd/compiler/aco_print_asm.cpp new file mode 100644 index 00000000000..31079aa1c4a --- /dev/null +++ b/src/amd/compiler/aco_print_asm.cpp @@ -0,0 +1,104 @@ + +#include +#include "aco_ir.h" +#include "llvm-c/Disassembler.h" +#include "ac_llvm_util.h" + +#include + +namespace aco { + +void print_asm(Program *program, std::vector& binary, + unsigned exec_size, enum radeon_family family, std::ostream& out) +{ + std::vector referenced_blocks(program->blocks.size()); + referenced_blocks[0] = true; + for (Block& block : program->blocks) { + for (unsigned succ : block.linear_succs) + referenced_blocks[succ] = true; + } + + std::vector> symbols; + std::vector> block_names; + block_names.reserve(program->blocks.size()); + for (Block& block : program->blocks) { + if (!referenced_blocks[block.index]) + continue; + std::array name; + sprintf(name.data(), "BB%u", block.index); + block_names.push_back(name); + symbols.emplace_back(block.offset * 4, llvm::StringRef(block_names[block_names.size() - 1].data()), 0); + } + + LLVMDisasmContextRef disasm = LLVMCreateDisasmCPU("amdgcn-mesa-mesa3d", + ac_get_llvm_processor_name(family), + &symbols, 0, NULL, NULL); + + char outline[1024]; + size_t pos = 0; + bool invalid = false; + unsigned next_block = 0; + while (pos < exec_size) { + while (next_block < program->blocks.size() && pos == program->blocks[next_block].offset) { + if (referenced_blocks[next_block]) + out << "BB" << std::dec << next_block << ":" << std::endl; + next_block++; + } + + size_t l = LLVMDisasmInstruction(disasm, (uint8_t *) &binary[pos], + (exec_size - pos) * sizeof(uint32_t), pos * 4, + outline, sizeof(outline)); + + size_t new_pos; + const int align_width = 60; + if (program->chip_class == GFX9 && !l && ((binary[pos] & 0xffff8000) == 0xd1348000)) { /* not actually an invalid instruction */ + out << std::left << std::setw(align_width) << std::setfill(' ') << "\tv_add_u32_e64 + clamp"; + new_pos = pos + 2; + } else if (!l) { + out << std::left << std::setw(align_width) << std::setfill(' ') << "(invalid instruction)"; + new_pos = pos + 1; + invalid = true; + } else { + out << std::left << std::setw(align_width) << std::setfill(' ') << outline; + assert(l % 4 == 0); + new_pos = pos + l / 4; + } + out << std::right; + + out << " ;"; + for (; pos < new_pos; pos++) + out << " " << std::setfill('0') << std::setw(8) << std::hex << binary[pos]; + out << std::endl; + } + out << std::setfill(' ') << std::setw(0) << std::dec; + assert(next_block == program->blocks.size()); + + LLVMDisasmDispose(disasm); + + if (program->constant_data.size()) { + out << std::endl << "/* constant data */" << std::endl; + for (unsigned i = 0; i < program->constant_data.size(); i += 32) { + out << '[' << std::setw(6) << std::setfill('0') << std::dec << i << ']'; + unsigned line_size = std::min(program->constant_data.size() - i, 32); + for (unsigned j = 0; j < line_size; j += 4) { + unsigned size = std::min(program->constant_data.size() - (i + j), 4); + uint32_t v = 0; + memcpy(&v, &program->constant_data[i + j], size); + out << " " << std::setw(8) << std::setfill('0') << std::hex << v; + } + out << std::endl; + } + } + + out << std::setfill(' ') << std::setw(0) << std::dec; + + if (invalid) { + /* Invalid instructions usually lead to GPU hangs, which can make + * getting the actual invalid instruction hard. Abort here so that we + * can find the problem. + */ + abort(); + } +} + +} diff --git a/src/amd/compiler/aco_print_ir.cpp b/src/amd/compiler/aco_print_ir.cpp new file mode 100644 index 00000000000..517ddc235ce --- /dev/null +++ b/src/amd/compiler/aco_print_ir.cpp @@ -0,0 +1,575 @@ +#include "aco_ir.h" +#include "aco_builder.h" + +#include "sid.h" + +namespace aco { + +static const char *reduce_ops[] = { + [iadd32] = "iadd32", + [iadd64] = "iadd64", + [imul32] = "imul32", + [imul64] = "imul64", + [fadd32] = "fadd32", + [fadd64] = "fadd64", + [fmul32] = "fmul32", + [fmul64] = "fmul64", + [imin32] = "imin32", + [imin64] = "imin64", + [imax32] = "imax32", + [imax64] = "imax64", + [umin32] = "umin32", + [umin64] = "umin64", + [umax32] = "umax32", + [umax64] = "umax64", + [fmin32] = "fmin32", + [fmin64] = "fmin64", + [fmax32] = "fmax32", + [fmax64] = "fmax64", + [iand32] = "iand32", + [iand64] = "iand64", + [ior32] = "ior32", + [ior64] = "ior64", + [ixor32] = "ixor32", + [ixor64] = "ixor64", +}; + +static void print_reg_class(const RegClass rc, FILE *output) +{ + switch (rc) { + case RegClass::s1: fprintf(output, " s1: "); return; + case RegClass::s2: fprintf(output, " s2: "); return; + case RegClass::s3: fprintf(output, " s3: "); return; + case RegClass::s4: fprintf(output, " s4: "); return; + case RegClass::s6: fprintf(output, " s6: "); return; + case RegClass::s8: fprintf(output, " s8: "); return; + case RegClass::s16: fprintf(output, "s16: "); return; + case RegClass::v1: fprintf(output, " v1: "); return; + case RegClass::v2: fprintf(output, " v2: "); return; + case RegClass::v3: fprintf(output, " v3: "); return; + case RegClass::v4: fprintf(output, " v4: "); return; + case RegClass::v5: fprintf(output, " v5: "); return; + case RegClass::v6: fprintf(output, " v6: "); return; + case RegClass::v7: fprintf(output, " v7: "); return; + case RegClass::v8: fprintf(output, " v8: "); return; + case RegClass::v1_linear: fprintf(output, " v1: "); return; + case RegClass::v2_linear: fprintf(output, " v2: "); return; + } +} + +void print_physReg(unsigned reg, unsigned size, FILE *output) +{ + if (reg == 124) { + fprintf(output, ":m0"); + } else if (reg == 106) { + fprintf(output, ":vcc"); + } else if (reg == 253) { + fprintf(output, ":scc"); + } else if (reg == 126) { + fprintf(output, ":exec"); + } else { + bool is_vgpr = reg / 256; + reg = reg % 256; + fprintf(output, ":%c[%d", is_vgpr ? 'v' : 's', reg); + if (size > 1) + fprintf(output, "-%d]", reg + size -1); + else + fprintf(output, "]"); + } +} + +static void print_constant(uint8_t reg, FILE *output) +{ + if (reg >= 128 && reg <= 192) { + fprintf(output, "%d", reg - 128); + return; + } else if (reg >= 192 && reg <= 208) { + fprintf(output, "%d", 192 - reg); + return; + } + + switch (reg) { + case 240: + fprintf(output, "0.5"); + break; + case 241: + fprintf(output, "-0.5"); + break; + case 242: + fprintf(output, "1.0"); + break; + case 243: + fprintf(output, "-1.0"); + break; + case 244: + fprintf(output, "2.0"); + break; + case 245: + fprintf(output, "-2.0"); + break; + case 246: + fprintf(output, "4.0"); + break; + case 247: + fprintf(output, "-4.0"); + break; + case 248: + fprintf(output, "1/(2*PI)"); + break; + } +} + +static void print_operand(const Operand *operand, FILE *output) +{ + if (operand->isLiteral()) { + fprintf(output, "0x%x", operand->constantValue()); + } else if (operand->isConstant()) { + print_constant(operand->physReg().reg, output); + } else if (operand->isUndefined()) { + print_reg_class(operand->regClass(), output); + fprintf(output, "undef"); + } else { + fprintf(output, "%%%d", operand->tempId()); + + if (operand->isFixed()) + print_physReg(operand->physReg(), operand->size(), output); + } +} + +static void print_definition(const Definition *definition, FILE *output) +{ + print_reg_class(definition->regClass(), output); + fprintf(output, "%%%d", definition->tempId()); + + if (definition->isFixed()) + print_physReg(definition->physReg(), definition->size(), output); +} + +static void print_barrier_reorder(bool can_reorder, barrier_interaction barrier, FILE *output) +{ + if (can_reorder) + fprintf(output, " reorder"); + + if (barrier & barrier_buffer) + fprintf(output, " buffer"); + if (barrier & barrier_image) + fprintf(output, " image"); + if (barrier & barrier_atomic) + fprintf(output, " atomic"); + if (barrier & barrier_shared) + fprintf(output, " shared"); +} + +static void print_instr_format_specific(struct Instruction *instr, FILE *output) +{ + switch (instr->format) { + case Format::SOPK: { + SOPK_instruction* sopk = static_cast(instr); + fprintf(output, " imm:%d", sopk->imm & 0x8000 ? (sopk->imm - 65536) : sopk->imm); + break; + } + case Format::SOPP: { + SOPP_instruction* sopp = static_cast(instr); + uint16_t imm = sopp->imm; + switch (instr->opcode) { + case aco_opcode::s_waitcnt: { + /* we usually should check the chip class for vmcnt/lgkm, but + * insert_waitcnt() should fill it in regardless. */ + unsigned vmcnt = (imm & 0xF) | ((imm & (0x3 << 14)) >> 10); + if (vmcnt != 63) fprintf(output, " vmcnt(%d)", vmcnt); + if (((imm >> 4) & 0x7) < 0x7) fprintf(output, " expcnt(%d)", (imm >> 4) & 0x7); + if (((imm >> 8) & 0x3F) < 0x3F) fprintf(output, " lgkmcnt(%d)", (imm >> 8) & 0x3F); + break; + } + case aco_opcode::s_endpgm: + case aco_opcode::s_endpgm_saved: + case aco_opcode::s_endpgm_ordered_ps_done: + case aco_opcode::s_wakeup: + case aco_opcode::s_barrier: + case aco_opcode::s_icache_inv: + case aco_opcode::s_ttracedata: + case aco_opcode::s_set_gpr_idx_off: { + break; + } + default: { + if (imm) + fprintf(output, " imm:%u", imm); + break; + } + } + if (sopp->block != -1) + fprintf(output, " block:BB%d", sopp->block); + break; + } + case Format::SMEM: { + SMEM_instruction* smem = static_cast(instr); + if (smem->glc) + fprintf(output, " glc"); + if (smem->nv) + fprintf(output, " nv"); + print_barrier_reorder(smem->can_reorder, smem->barrier, output); + break; + } + case Format::VINTRP: { + Interp_instruction* vintrp = static_cast(instr); + fprintf(output, " attr%d.%c", vintrp->attribute, "xyzw"[vintrp->component]); + break; + } + case Format::DS: { + DS_instruction* ds = static_cast(instr); + if (ds->offset0) + fprintf(output, " offset0:%u", ds->offset0); + if (ds->offset1) + fprintf(output, " offset1:%u", ds->offset1); + if (ds->gds) + fprintf(output, " gds"); + break; + } + case Format::MUBUF: { + MUBUF_instruction* mubuf = static_cast(instr); + if (mubuf->offset) + fprintf(output, " offset:%u", mubuf->offset); + if (mubuf->offen) + fprintf(output, " offen"); + if (mubuf->idxen) + fprintf(output, " idxen"); + if (mubuf->glc) + fprintf(output, " glc"); + if (mubuf->slc) + fprintf(output, " slc"); + if (mubuf->tfe) + fprintf(output, " tfe"); + if (mubuf->lds) + fprintf(output, " lds"); + if (mubuf->disable_wqm) + fprintf(output, " disable_wqm"); + print_barrier_reorder(mubuf->can_reorder, mubuf->barrier, output); + break; + } + case Format::MIMG: { + MIMG_instruction* mimg = static_cast(instr); + unsigned identity_dmask = !instr->definitions.empty() ? + (1 << instr->definitions[0].size()) - 1 : + 0xf; + if ((mimg->dmask & identity_dmask) != identity_dmask) + fprintf(output, " dmask:%s%s%s%s", + mimg->dmask & 0x1 ? "x" : "", + mimg->dmask & 0x2 ? "y" : "", + mimg->dmask & 0x4 ? "z" : "", + mimg->dmask & 0x8 ? "w" : ""); + if (mimg->unrm) + fprintf(output, " unrm"); + if (mimg->glc) + fprintf(output, " glc"); + if (mimg->slc) + fprintf(output, " slc"); + if (mimg->tfe) + fprintf(output, " tfe"); + if (mimg->da) + fprintf(output, " da"); + if (mimg->lwe) + fprintf(output, " lwe"); + if (mimg->r128 || mimg->a16) + fprintf(output, " r128/a16"); + if (mimg->d16) + fprintf(output, " d16"); + if (mimg->disable_wqm) + fprintf(output, " disable_wqm"); + print_barrier_reorder(mimg->can_reorder, mimg->barrier, output); + break; + } + case Format::EXP: { + Export_instruction* exp = static_cast(instr); + unsigned identity_mask = exp->compressed ? 0x5 : 0xf; + if ((exp->enabled_mask & identity_mask) != identity_mask) + fprintf(output, " en:%c%c%c%c", + exp->enabled_mask & 0x1 ? 'r' : '*', + exp->enabled_mask & 0x2 ? 'g' : '*', + exp->enabled_mask & 0x4 ? 'b' : '*', + exp->enabled_mask & 0x8 ? 'a' : '*'); + if (exp->compressed) + fprintf(output, " compr"); + if (exp->done) + fprintf(output, " done"); + if (exp->valid_mask) + fprintf(output, " vm"); + + if (exp->dest <= V_008DFC_SQ_EXP_MRT + 7) + fprintf(output, " mrt%d", exp->dest - V_008DFC_SQ_EXP_MRT); + else if (exp->dest == V_008DFC_SQ_EXP_MRTZ) + fprintf(output, " mrtz"); + else if (exp->dest == V_008DFC_SQ_EXP_NULL) + fprintf(output, " null"); + else if (exp->dest >= V_008DFC_SQ_EXP_POS && exp->dest <= V_008DFC_SQ_EXP_POS + 3) + fprintf(output, " pos%d", exp->dest - V_008DFC_SQ_EXP_POS); + else if (exp->dest >= V_008DFC_SQ_EXP_PARAM && exp->dest <= V_008DFC_SQ_EXP_PARAM + 31) + fprintf(output, " param%d", exp->dest - V_008DFC_SQ_EXP_PARAM); + break; + } + case Format::PSEUDO_BRANCH: { + Pseudo_branch_instruction* branch = static_cast(instr); + /* Note: BB0 cannot be a branch target */ + if (branch->target[0] != 0) + fprintf(output, " BB%d", branch->target[0]); + if (branch->target[1] != 0) + fprintf(output, ", BB%d", branch->target[1]); + break; + } + case Format::PSEUDO_REDUCTION: { + Pseudo_reduction_instruction* reduce = static_cast(instr); + fprintf(output, " op:%s", reduce_ops[reduce->reduce_op]); + if (reduce->cluster_size) + fprintf(output, " cluster_size:%u", reduce->cluster_size); + break; + } + case Format::FLAT: + case Format::GLOBAL: + case Format::SCRATCH: { + FLAT_instruction* flat = static_cast(instr); + if (flat->offset) + fprintf(output, " offset:%u", flat->offset); + if (flat->glc) + fprintf(output, " glc"); + if (flat->slc) + fprintf(output, " slc"); + if (flat->lds) + fprintf(output, " lds"); + if (flat->nv) + fprintf(output, " nv"); + break; + } + case Format::MTBUF: { + MTBUF_instruction* mtbuf = static_cast(instr); + fprintf(output, " dfmt:"); + switch (mtbuf->dfmt) { + case V_008F0C_BUF_DATA_FORMAT_8: fprintf(output, "8"); break; + case V_008F0C_BUF_DATA_FORMAT_16: fprintf(output, "16"); break; + case V_008F0C_BUF_DATA_FORMAT_8_8: fprintf(output, "8_8"); break; + case V_008F0C_BUF_DATA_FORMAT_32: fprintf(output, "32"); break; + case V_008F0C_BUF_DATA_FORMAT_16_16: fprintf(output, "16_16"); break; + case V_008F0C_BUF_DATA_FORMAT_10_11_11: fprintf(output, "10_11_11"); break; + case V_008F0C_BUF_DATA_FORMAT_11_11_10: fprintf(output, "11_11_10"); break; + case V_008F0C_BUF_DATA_FORMAT_10_10_10_2: fprintf(output, "10_10_10_2"); break; + case V_008F0C_BUF_DATA_FORMAT_2_10_10_10: fprintf(output, "2_10_10_10"); break; + case V_008F0C_BUF_DATA_FORMAT_8_8_8_8: fprintf(output, "8_8_8_8"); break; + case V_008F0C_BUF_DATA_FORMAT_32_32: fprintf(output, "32_32"); break; + case V_008F0C_BUF_DATA_FORMAT_16_16_16_16: fprintf(output, "16_16_16_16"); break; + case V_008F0C_BUF_DATA_FORMAT_32_32_32: fprintf(output, "32_32_32"); break; + case V_008F0C_BUF_DATA_FORMAT_32_32_32_32: fprintf(output, "32_32_32_32"); break; + case V_008F0C_BUF_DATA_FORMAT_RESERVED_15: fprintf(output, "reserved15"); break; + } + fprintf(output, " nfmt:"); + switch (mtbuf->nfmt) { + case V_008F0C_BUF_NUM_FORMAT_UNORM: fprintf(output, "unorm"); break; + case V_008F0C_BUF_NUM_FORMAT_SNORM: fprintf(output, "snorm"); break; + case V_008F0C_BUF_NUM_FORMAT_USCALED: fprintf(output, "uscaled"); break; + case V_008F0C_BUF_NUM_FORMAT_SSCALED: fprintf(output, "sscaled"); break; + case V_008F0C_BUF_NUM_FORMAT_UINT: fprintf(output, "uint"); break; + case V_008F0C_BUF_NUM_FORMAT_SINT: fprintf(output, "sint"); break; + case V_008F0C_BUF_NUM_FORMAT_SNORM_OGL: fprintf(output, "snorm"); break; + case V_008F0C_BUF_NUM_FORMAT_FLOAT: fprintf(output, "float"); break; + } + if (mtbuf->offset) + fprintf(output, " offset:%u", mtbuf->offset); + if (mtbuf->offen) + fprintf(output, " offen"); + if (mtbuf->idxen) + fprintf(output, " idxen"); + if (mtbuf->glc) + fprintf(output, " glc"); + if (mtbuf->slc) + fprintf(output, " slc"); + if (mtbuf->tfe) + fprintf(output, " tfe"); + if (mtbuf->disable_wqm) + fprintf(output, " disable_wqm"); + print_barrier_reorder(mtbuf->can_reorder, mtbuf->barrier, output); + break; + } + default: { + break; + } + } + if (instr->isVOP3()) { + VOP3A_instruction* vop3 = static_cast(instr); + switch (vop3->omod) { + case 1: + fprintf(output, " *2"); + break; + case 2: + fprintf(output, " *4"); + break; + case 3: + fprintf(output, " *0.5"); + break; + } + if (vop3->clamp) + fprintf(output, " clamp"); + } else if (instr->isDPP()) { + DPP_instruction* dpp = static_cast(instr); + if (dpp->dpp_ctrl <= 0xff) { + fprintf(output, " quad_perm:[%d,%d,%d,%d]", + dpp->dpp_ctrl & 0x3, (dpp->dpp_ctrl >> 2) & 0x3, + (dpp->dpp_ctrl >> 4) & 0x3, (dpp->dpp_ctrl >> 6) & 0x3); + } else if (dpp->dpp_ctrl >= 0x101 && dpp->dpp_ctrl <= 0x10f) { + fprintf(output, " row_shl:%d", dpp->dpp_ctrl & 0xf); + } else if (dpp->dpp_ctrl >= 0x111 && dpp->dpp_ctrl <= 0x11f) { + fprintf(output, " row_shr:%d", dpp->dpp_ctrl & 0xf); + } else if (dpp->dpp_ctrl >= 0x121 && dpp->dpp_ctrl <= 0x12f) { + fprintf(output, " row_ror:%d", dpp->dpp_ctrl & 0xf); + } else if (dpp->dpp_ctrl == dpp_wf_sl1) { + fprintf(output, " wave_shl:1"); + } else if (dpp->dpp_ctrl == dpp_wf_rl1) { + fprintf(output, " wave_rol:1"); + } else if (dpp->dpp_ctrl == dpp_wf_sr1) { + fprintf(output, " wave_shr:1"); + } else if (dpp->dpp_ctrl == dpp_wf_rr1) { + fprintf(output, " wave_ror:1"); + } else if (dpp->dpp_ctrl == dpp_row_mirror) { + fprintf(output, " row_mirror"); + } else if (dpp->dpp_ctrl == dpp_row_half_mirror) { + fprintf(output, " row_half_mirror"); + } else if (dpp->dpp_ctrl == dpp_row_bcast15) { + fprintf(output, " row_bcast:15"); + } else if (dpp->dpp_ctrl == dpp_row_bcast31) { + fprintf(output, " row_bcast:31"); + } else { + fprintf(output, " dpp_ctrl:0x%.3x", dpp->dpp_ctrl); + } + if (dpp->row_mask != 0xf) + fprintf(output, " row_mask:0x%.1x", dpp->row_mask); + if (dpp->bank_mask != 0xf) + fprintf(output, " bank_mask:0x%.1x", dpp->bank_mask); + if (dpp->bound_ctrl) + fprintf(output, " bound_ctrl:1"); + } else if ((int)instr->format & (int)Format::SDWA) { + fprintf(output, " (printing unimplemented)"); + } +} + +void aco_print_instr(struct Instruction *instr, FILE *output) +{ + if (!instr->definitions.empty()) { + for (unsigned i = 0; i < instr->definitions.size(); ++i) { + print_definition(&instr->definitions[i], output); + if (i + 1 != instr->definitions.size()) + fprintf(output, ", "); + } + fprintf(output, " = "); + } + fprintf(output, "%s", instr_info.name[(int)instr->opcode]); + if (instr->operands.size()) { + bool abs[instr->operands.size()]; + bool neg[instr->operands.size()]; + if ((int)instr->format & (int)Format::VOP3A) { + VOP3A_instruction* vop3 = static_cast(instr); + for (unsigned i = 0; i < instr->operands.size(); ++i) { + abs[i] = vop3->abs[i]; + neg[i] = vop3->neg[i]; + } + } else if (instr->isDPP()) { + DPP_instruction* dpp = static_cast(instr); + assert(instr->operands.size() <= 2); + for (unsigned i = 0; i < instr->operands.size(); ++i) { + abs[i] = dpp->abs[i]; + neg[i] = dpp->neg[i]; + } + } else { + for (unsigned i = 0; i < instr->operands.size(); ++i) { + abs[i] = false; + neg[i] = false; + } + } + for (unsigned i = 0; i < instr->operands.size(); ++i) { + if (i) + fprintf(output, ", "); + else + fprintf(output, " "); + + if (neg[i]) + fprintf(output, "-"); + if (abs[i]) + fprintf(output, "|"); + print_operand(&instr->operands[i], output); + if (abs[i]) + fprintf(output, "|"); + } + } + print_instr_format_specific(instr, output); +} + +static void print_block_kind(uint16_t kind, FILE *output) +{ + if (kind & block_kind_uniform) + fprintf(output, "uniform, "); + if (kind & block_kind_top_level) + fprintf(output, "top-level, "); + if (kind & block_kind_loop_preheader) + fprintf(output, "loop-preheader, "); + if (kind & block_kind_loop_header) + fprintf(output, "loop-header, "); + if (kind & block_kind_loop_exit) + fprintf(output, "loop-exit, "); + if (kind & block_kind_continue) + fprintf(output, "continue, "); + if (kind & block_kind_break) + fprintf(output, "break, "); + if (kind & block_kind_continue_or_break) + fprintf(output, "continue_or_break, "); + if (kind & block_kind_discard) + fprintf(output, "discard, "); + if (kind & block_kind_branch) + fprintf(output, "branch, "); + if (kind & block_kind_merge) + fprintf(output, "merge, "); + if (kind & block_kind_invert) + fprintf(output, "invert, "); + if (kind & block_kind_uses_discard_if) + fprintf(output, "discard_if, "); + if (kind & block_kind_needs_lowering) + fprintf(output, "needs_lowering, "); +} + +void aco_print_block(const struct Block* block, FILE *output) +{ + fprintf(output, "BB%d\n", block->index); + fprintf(output, "/* logical preds: "); + for (unsigned pred : block->logical_preds) + fprintf(output, "BB%d, ", pred); + fprintf(output, "/ linear preds: "); + for (unsigned pred : block->linear_preds) + fprintf(output, "BB%d, ", pred); + fprintf(output, "/ kind: "); + print_block_kind(block->kind, output); + fprintf(output, "*/\n"); + for (auto const& instr : block->instructions) { + fprintf(output, "\t"); + aco_print_instr(instr.get(), output); + fprintf(output, "\n"); + } +} + +void aco_print_program(Program *program, FILE *output) +{ + for (Block const& block : program->blocks) + aco_print_block(&block, output); + + if (program->constant_data.size()) { + fprintf(output, "\n/* constant data */\n"); + for (unsigned i = 0; i < program->constant_data.size(); i += 32) { + fprintf(output, "[%06d] ", i); + unsigned line_size = std::min(program->constant_data.size() - i, 32); + for (unsigned j = 0; j < line_size; j += 4) { + unsigned size = std::min(program->constant_data.size() - (i + j), 4); + uint32_t v = 0; + memcpy(&v, &program->constant_data[i + j], size); + fprintf(output, " %08x", v); + } + fprintf(output, "\n"); + } + } + + fprintf(output, "\n"); +} + +} diff --git a/src/amd/compiler/aco_reduce_assign.cpp b/src/amd/compiler/aco_reduce_assign.cpp new file mode 100644 index 00000000000..663a43c539a --- /dev/null +++ b/src/amd/compiler/aco_reduce_assign.cpp @@ -0,0 +1,164 @@ +/* + * Copyright © 2018 Valve Corporation + * Copyright © 2018 Google + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" +#include "aco_builder.h" + +/* + * Insert p_linear_start instructions right before RA to correctly allocate + * temporaries for reductions that have to disrespect EXEC by executing in + * WWM. + */ + +namespace aco { + +void setup_reduce_temp(Program* program) +{ + unsigned last_top_level_block_idx = 0; + unsigned maxSize = 0; + + std::vector hasReductions(program->blocks.size()); + for (Block& block : program->blocks) { + for (aco_ptr& instr : block.instructions) { + if (instr->format != Format::PSEUDO_REDUCTION) + continue; + + maxSize = MAX2(maxSize, instr->operands[0].size()); + hasReductions[block.index] = true; + } + } + + if (maxSize == 0) + return; + + assert(maxSize == 1 || maxSize == 2); + Temp reduceTmp(0, RegClass(RegType::vgpr, maxSize).as_linear()); + Temp vtmp(0, RegClass(RegType::vgpr, maxSize).as_linear()); + int inserted_at = -1; + int vtmp_inserted_at = -1; + bool reduceTmp_in_loop = false; + bool vtmp_in_loop = false; + + for (Block& block : program->blocks) { + + /* insert p_end_linear_vgpr after the outermost loop */ + if (reduceTmp_in_loop && block.loop_nest_depth == 0) { + assert(inserted_at == (int)last_top_level_block_idx); + + aco_ptr end{create_instruction(aco_opcode::p_end_linear_vgpr, Format::PSEUDO, vtmp_in_loop ? 2 : 1, 0)}; + end->operands[0] = Operand(reduceTmp); + if (vtmp_in_loop) + end->operands[1] = Operand(vtmp); + /* insert after the phis of the loop exit block */ + std::vector>::iterator it = block.instructions.begin(); + while ((*it)->opcode == aco_opcode::p_linear_phi || (*it)->opcode == aco_opcode::p_phi) + ++it; + block.instructions.insert(it, std::move(end)); + reduceTmp_in_loop = false; + } + + if (block.kind & block_kind_top_level) + last_top_level_block_idx = block.index; + + if (!hasReductions[block.index]) + continue; + + std::vector>::iterator it; + for (it = block.instructions.begin(); it != block.instructions.end(); ++it) { + Instruction *instr = (*it).get(); + if (instr->format != Format::PSEUDO_REDUCTION) + continue; + + ReduceOp op = static_cast(instr)->reduce_op; + reduceTmp_in_loop |= block.loop_nest_depth > 0; + + if ((int)last_top_level_block_idx != inserted_at) { + reduceTmp = {program->allocateId(), reduceTmp.regClass()}; + aco_ptr create{create_instruction(aco_opcode::p_start_linear_vgpr, Format::PSEUDO, 0, 1)}; + create->definitions[0] = Definition(reduceTmp); + /* find the right place to insert this definition */ + if (last_top_level_block_idx == block.index) { + /* insert right before the current instruction */ + it = block.instructions.insert(it, std::move(create)); + it++; + /* inserted_at is intentionally not updated here, so later blocks + * would insert at the end instead of using this one. */ + } else { + assert(last_top_level_block_idx < block.index); + /* insert before the branch at last top level block */ + std::vector>& instructions = program->blocks[last_top_level_block_idx].instructions; + instructions.insert(std::next(instructions.begin(), instructions.size() - 1), std::move(create)); + inserted_at = last_top_level_block_idx; + } + } + + /* same as before, except for the vector temporary instead of the reduce temporary */ + bool need_vtmp = op == imul32 || op == fadd64 || op == fmul64 || + op == fmin64 || op == fmax64; + + need_vtmp |= static_cast(instr)->cluster_size == 32; + vtmp_in_loop |= need_vtmp && block.loop_nest_depth > 0; + if (need_vtmp && (int)last_top_level_block_idx != vtmp_inserted_at) { + vtmp = {program->allocateId(), vtmp.regClass()}; + aco_ptr create{create_instruction(aco_opcode::p_start_linear_vgpr, Format::PSEUDO, 0, 1)}; + create->definitions[0] = Definition(vtmp); + if (last_top_level_block_idx == block.index) { + it = block.instructions.insert(it, std::move(create)); + it++; + } else { + assert(last_top_level_block_idx < block.index); + std::vector>& instructions = program->blocks[last_top_level_block_idx].instructions; + instructions.insert(std::next(instructions.begin(), instructions.size() - 1), std::move(create)); + vtmp_inserted_at = last_top_level_block_idx; + } + } + + instr->operands[1] = Operand(reduceTmp); + if (need_vtmp) + instr->operands[2] = Operand(vtmp); + + /* scalar temporary */ + Builder bld(program); + instr->definitions[1] = bld.def(s2); + + /* scalar identity temporary */ + if (instr->opcode == aco_opcode::p_exclusive_scan && + (op == imin32 || op == imin64 || + op == imax32 || op == imax64 || + op == fmin32 || op == fmin64 || + op == fmax32 || op == fmax64 || + op == fmul64)) { + instr->definitions[2] = bld.def(RegClass(RegType::sgpr, instr->operands[0].size())); + } + + /* vcc clobber */ + if (op == iadd32 && program->chip_class < GFX9) + instr->definitions[4] = Definition(vcc, s2); + } + } +} + +}; + diff --git a/src/amd/compiler/aco_register_allocation.cpp b/src/amd/compiler/aco_register_allocation.cpp new file mode 100644 index 00000000000..d55f1febc65 --- /dev/null +++ b/src/amd/compiler/aco_register_allocation.cpp @@ -0,0 +1,1924 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Daniel Schürmann (daniel.schuermann@campus.tu-berlin.de) + * Bas Nieuwenhuizen (bas@basnieuwenhuizen.nl) + * + */ + +#include +#include +#include +#include + +#include "aco_ir.h" +#include "sid.h" + +namespace aco { +namespace { + +struct ra_ctx { + std::bitset<512> war_hint; + Program* program; + std::unordered_map> assignments; + std::map orig_names; + unsigned max_used_sgpr = 0; + unsigned max_used_vgpr = 0; + std::bitset<64> defs_done; /* see MAX_ARGS in aco_instruction_selection_setup.cpp */ + + ra_ctx(Program* program) : program(program) {} +}; + + +/* helper function for debugging */ +#if 0 +void print_regs(ra_ctx& ctx, bool vgprs, std::array& reg_file) +{ + unsigned max = vgprs ? ctx.program->max_reg_demand.vgpr : ctx.program->max_reg_demand.sgpr; + unsigned lb = vgprs ? 256 : 0; + unsigned ub = lb + max; + char reg_char = vgprs ? 'v' : 's'; + + /* print markers */ + printf(" "); + for (unsigned i = lb; i < ub; i += 3) { + printf("%.2u ", i - lb); + } + printf("\n"); + + /* print usage */ + printf("%cgprs: ", reg_char); + unsigned free_regs = 0; + unsigned prev = 0; + bool char_select = false; + for (unsigned i = lb; i < ub; i++) { + if (reg_file[i] == 0xFFFF) { + printf("~"); + } else if (reg_file[i]) { + if (reg_file[i] != prev) { + prev = reg_file[i]; + char_select = !char_select; + } + printf(char_select ? "#" : "@"); + } else { + free_regs++; + printf("."); + } + } + printf("\n"); + + printf("%u/%u used, %u/%u free\n", max - free_regs, max, free_regs, max); + + /* print assignments */ + prev = 0; + unsigned size = 0; + for (unsigned i = lb; i < ub; i++) { + if (reg_file[i] != prev) { + if (prev && size > 1) + printf("-%d]\n", i - 1 - lb); + else if (prev) + printf("]\n"); + prev = reg_file[i]; + if (prev && prev != 0xFFFF) { + if (ctx.orig_names.count(reg_file[i]) && ctx.orig_names[reg_file[i]].id() != reg_file[i]) + printf("%%%u (was %%%d) = %c[%d", reg_file[i], ctx.orig_names[reg_file[i]].id(), reg_char, i - lb); + else + printf("%%%u = %c[%d", reg_file[i], reg_char, i - lb); + } + size = 1; + } else { + size++; + } + } + if (prev && size > 1) + printf("-%d]\n", ub - lb - 1); + else if (prev) + printf("]\n"); +} +#endif + + +void adjust_max_used_regs(ra_ctx& ctx, RegClass rc, unsigned reg) +{ + unsigned max_addressible_sgpr = ctx.program->sgpr_limit; + unsigned size = rc.size(); + if (rc.type() == RegType::vgpr) { + assert(reg >= 256); + unsigned hi = reg - 256 + size - 1; + ctx.max_used_vgpr = std::max(ctx.max_used_vgpr, hi); + } else if (reg + rc.size() <= max_addressible_sgpr) { + unsigned hi = reg + size - 1; + ctx.max_used_sgpr = std::max(ctx.max_used_sgpr, std::min(hi, max_addressible_sgpr)); + } +} + + +void update_renames(ra_ctx& ctx, std::array& reg_file, + std::vector>& parallelcopies, + aco_ptr& instr) +{ + /* allocate id's and rename operands: this is done transparently here */ + for (std::pair& copy : parallelcopies) { + /* the definitions with id are not from this function and already handled */ + if (copy.second.isTemp()) + continue; + + // FIXME: if a definition got moved, change the target location and remove the parallelcopy + copy.second.setTemp(Temp(ctx.program->allocateId(), copy.second.regClass())); + ctx.assignments[copy.second.tempId()] = {copy.second.physReg(), copy.second.regClass()}; + for (unsigned i = copy.second.physReg().reg; i < copy.second.physReg() + copy.second.size(); i++) + reg_file[i] = copy.second.tempId(); + /* check if we moved an operand */ + for (Operand& op : instr->operands) { + if (!op.isTemp()) + continue; + if (op.tempId() == copy.first.tempId()) { + bool omit_renaming = instr->opcode == aco_opcode::p_create_vector && !op.isKill(); + for (std::pair& pc : parallelcopies) { + PhysReg def_reg = pc.second.physReg(); + omit_renaming &= def_reg > copy.first.physReg() ? + (copy.first.physReg() + copy.first.size() <= def_reg.reg) : + (def_reg + pc.second.size() <= copy.first.physReg().reg); + } + if (omit_renaming) + continue; + op.setTemp(copy.second.getTemp()); + op.setFixed(copy.second.physReg()); + } + } + } +} + +std::pair get_reg_simple(ra_ctx& ctx, + std::array& reg_file, + uint32_t lb, uint32_t ub, + uint32_t size, uint32_t stride, + RegClass rc) +{ + /* best fit algorithm: find the smallest gap to fit in the variable */ + if (stride == 1) { + unsigned best_pos = 0xFFFF; + unsigned gap_size = 0xFFFF; + unsigned next_pos = 0xFFFF; + + for (unsigned current_reg = lb; current_reg < ub; current_reg++) { + if (reg_file[current_reg] != 0 || ctx.war_hint[current_reg]) { + if (next_pos == 0xFFFF) + continue; + + /* check if the variable fits */ + if (next_pos + size > current_reg) { + next_pos = 0xFFFF; + continue; + } + + /* check if the tested gap is smaller */ + if (current_reg - next_pos < gap_size) { + best_pos = next_pos; + gap_size = current_reg - next_pos; + } + next_pos = 0xFFFF; + continue; + } + + if (next_pos == 0xFFFF) + next_pos = current_reg; + } + + /* final check */ + if (next_pos != 0xFFFF && + next_pos + size <= ub && + ub - next_pos < gap_size) { + best_pos = next_pos; + gap_size = ub - next_pos; + } + if (best_pos != 0xFFFF) { + adjust_max_used_regs(ctx, rc, best_pos); + return {PhysReg{best_pos}, true}; + } + return {{}, false}; + } + + bool found = false; + unsigned reg_lo = lb; + unsigned reg_hi = lb + size - 1; + while (!found && reg_lo + size <= ub) { + if (reg_file[reg_lo] != 0) { + reg_lo += stride; + continue; + } + reg_hi = reg_lo + size - 1; + found = true; + for (unsigned reg = reg_lo + 1; found && reg <= reg_hi; reg++) { + if (reg_file[reg] != 0 || ctx.war_hint[reg]) + found = false; + } + if (found) { + adjust_max_used_regs(ctx, rc, reg_lo); + return {PhysReg{reg_lo}, true}; + } + + reg_lo += stride; + } + + return {{}, false}; +} + +bool get_regs_for_copies(ra_ctx& ctx, + std::array& reg_file, + std::vector>& parallelcopies, + std::set> vars, + uint32_t lb, uint32_t ub, + aco_ptr& instr, + uint32_t def_reg_lo, + uint32_t def_reg_hi) +{ + + /* variables are sorted from small sized to large */ + /* NOTE: variables are also sorted by ID. this only affects a very small number of shaders slightly though. */ + for (std::set>::reverse_iterator it = vars.rbegin(); it != vars.rend(); ++it) { + unsigned id = it->second; + std::pair var = ctx.assignments[id]; + uint32_t size = it->first; + uint32_t stride = 1; + if (var.second.type() == RegType::sgpr) { + if (size == 2) + stride = 2; + if (size > 3) + stride = 4; + } + + /* check if this is a dead operand, then we can re-use the space from the definition */ + bool is_dead_operand = false; + for (unsigned i = 0; !is_phi(instr) && !is_dead_operand && i < instr->operands.size(); i++) { + if (instr->operands[i].isTemp() && instr->operands[i].isKill() && instr->operands[i].tempId() == id) + is_dead_operand = true; + } + + std::pair res; + if (is_dead_operand) { + if (instr->opcode == aco_opcode::p_create_vector) { + for (unsigned i = 0, offset = 0; i < instr->operands.size(); offset += instr->operands[i].size(), i++) { + if (instr->operands[i].isTemp() && instr->operands[i].tempId() == id) { + for (unsigned j = 0; j < size; j++) + assert(reg_file[def_reg_lo + offset + j] == 0); + res = {PhysReg{def_reg_lo + offset}, true}; + break; + } + } + } else { + res = get_reg_simple(ctx, reg_file, def_reg_lo, def_reg_hi + 1, size, stride, var.second); + } + } else { + res = get_reg_simple(ctx, reg_file, lb, def_reg_lo, size, stride, var.second); + if (!res.second) { + unsigned lb = (def_reg_hi + stride) & ~(stride - 1); + res = get_reg_simple(ctx, reg_file, lb, ub, size, stride, var.second); + } + } + + if (res.second) { + /* mark the area as blocked */ + for (unsigned i = res.first.reg; i < res.first + size; i++) + reg_file[i] = 0xFFFFFFFF; + /* create parallelcopy pair (without definition id) */ + Temp tmp = Temp(id, var.second); + Operand pc_op = Operand(tmp); + pc_op.setFixed(var.first); + Definition pc_def = Definition(res.first, pc_op.regClass()); + parallelcopies.emplace_back(pc_op, pc_def); + continue; + } + + unsigned best_pos = lb; + unsigned num_moves = 0xFF; + unsigned num_vars = 0; + + /* we use a sliding window to find potential positions */ + unsigned reg_lo = lb; + unsigned reg_hi = lb + size - 1; + for (reg_lo = lb, reg_hi = lb + size - 1; reg_hi < ub; reg_lo += stride, reg_hi += stride) { + if (!is_dead_operand && ((reg_lo >= def_reg_lo && reg_lo <= def_reg_hi) || + (reg_hi >= def_reg_lo && reg_hi <= def_reg_hi))) + continue; + + /* second, check that we have at most k=num_moves elements in the window + * and no element is larger than the currently processed one */ + unsigned k = 0; + unsigned n = 0; + unsigned last_var = 0; + bool found = true; + for (unsigned j = reg_lo; found && j <= reg_hi; j++) { + if (reg_file[j] == 0 || reg_file[j] == last_var) + continue; + + /* 0xFFFF signals that this area is already blocked! */ + if (reg_file[j] == 0xFFFFFFFF || k > num_moves) { + found = false; + break; + } + /* we cannot split live ranges of linear vgprs */ + if (ctx.assignments[reg_file[j]].second & (1 << 6)) { + found = false; + break; + } + bool is_kill = false; + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isKill() && op.tempId() == reg_file[j]) { + is_kill = true; + break; + } + } + if (!is_kill && ctx.assignments[reg_file[j]].second.size() >= size) { + found = false; + break; + } + + k += ctx.assignments[reg_file[j]].second.size(); + last_var = reg_file[j]; + n++; + if (k > num_moves || (k == num_moves && n <= num_vars)) { + found = false; + break; + } + } + + if (found) { + best_pos = reg_lo; + num_moves = k; + num_vars = n; + } + } + + /* FIXME: we messed up and couldn't find space for the variables to be copied */ + if (num_moves == 0xFF) + return false; + + reg_lo = best_pos; + reg_hi = best_pos + size - 1; + + /* collect variables and block reg file */ + std::set> new_vars; + for (unsigned j = reg_lo; j <= reg_hi; j++) { + if (reg_file[j] != 0) { + unsigned size = ctx.assignments[reg_file[j]].second.size(); + unsigned id = reg_file[j]; + new_vars.emplace(size, id); + for (unsigned k = 0; k < size; k++) + reg_file[ctx.assignments[id].first + k] = 0; + } + } + + /* mark the area as blocked */ + for (unsigned i = reg_lo; i <= reg_hi; i++) + reg_file[i] = 0xFFFFFFFF; + + if (!get_regs_for_copies(ctx, reg_file, parallelcopies, new_vars, lb, ub, instr, def_reg_lo, def_reg_hi)) + return false; + + adjust_max_used_regs(ctx, var.second, reg_lo); + + /* create parallelcopy pair (without definition id) */ + Temp tmp = Temp(id, var.second); + Operand pc_op = Operand(tmp); + pc_op.setFixed(var.first); + Definition pc_def = Definition(PhysReg{reg_lo}, pc_op.regClass()); + parallelcopies.emplace_back(pc_op, pc_def); + } + + return true; +} + + +std::pair get_reg_impl(ra_ctx& ctx, + std::array& reg_file, + std::vector>& parallelcopies, + uint32_t lb, uint32_t ub, + uint32_t size, uint32_t stride, + RegClass rc, + aco_ptr& instr) +{ + unsigned regs_free = 0; + /* check how many free regs we have */ + for (unsigned j = lb; j < ub; j++) { + if (reg_file[j] == 0) + regs_free++; + } + + /* mark and count killed operands */ + unsigned killed_ops = 0; + for (unsigned j = 0; !is_phi(instr) && j < instr->operands.size(); j++) { + if (instr->operands[j].isTemp() && + instr->operands[j].isFirstKill() && + instr->operands[j].physReg() >= lb && + instr->operands[j].physReg() < ub) { + assert(instr->operands[j].isFixed()); + assert(reg_file[instr->operands[j].physReg().reg] == 0); + for (unsigned k = 0; k < instr->operands[j].size(); k++) + reg_file[instr->operands[j].physReg() + k] = 0xFFFFFFFF; + killed_ops += instr->operands[j].getTemp().size(); + } + } + + assert(regs_free >= size); + /* we might have to move dead operands to dst in order to make space */ + unsigned op_moves = 0; + + if (size > (regs_free - killed_ops)) + op_moves = size - (regs_free - killed_ops); + + /* find the best position to place the definition */ + unsigned best_pos = lb; + unsigned num_moves = 0xFF; + unsigned num_vars = 0; + + /* we use a sliding window to check potential positions */ + unsigned reg_lo = lb; + unsigned reg_hi = lb + size - 1; + for (reg_lo = lb, reg_hi = lb + size - 1; reg_hi < ub; reg_lo += stride, reg_hi += stride) { + /* first check the edges: this is what we have to fix to allow for num_moves > size */ + if (reg_lo > lb && reg_file[reg_lo] != 0 && reg_file[reg_lo] == reg_file[reg_lo - 1]) + continue; + if (reg_hi < ub - 1 && reg_file[reg_hi] != 0 && reg_file[reg_hi] == reg_file[reg_hi + 1]) + continue; + + /* second, check that we have at most k=num_moves elements in the window + * and no element is larger than the currently processed one */ + unsigned k = op_moves; + unsigned n = 0; + unsigned remaining_op_moves = op_moves; + unsigned last_var = 0; + bool found = true; + bool aligned = rc == RegClass::v4 && reg_lo % 4 == 0; + for (unsigned j = reg_lo; found && j <= reg_hi; j++) { + if (reg_file[j] == 0 || reg_file[j] == last_var) + continue; + + /* dead operands effectively reduce the number of estimated moves */ + if (remaining_op_moves && reg_file[j] == 0xFFFFFFFF) { + k--; + remaining_op_moves--; + continue; + } + + if (ctx.assignments[reg_file[j]].second.size() >= size) { + found = false; + break; + } + + + /* we cannot split live ranges of linear vgprs */ + if (ctx.assignments[reg_file[j]].second & (1 << 6)) { + found = false; + break; + } + + k += ctx.assignments[reg_file[j]].second.size(); + n++; + last_var = reg_file[j]; + } + + if (!found || k > num_moves) + continue; + if (k == num_moves && n < num_vars) + continue; + if (!aligned && k == num_moves && n == num_vars) + continue; + + if (found) { + best_pos = reg_lo; + num_moves = k; + num_vars = n; + } + } + + if (num_moves == 0xFF) { + /* remove killed operands from reg_file once again */ + for (unsigned i = 0; !is_phi(instr) && i < instr->operands.size(); i++) { + if (instr->operands[i].isTemp() && instr->operands[i].isFirstKill()) { + for (unsigned k = 0; k < instr->operands[i].getTemp().size(); k++) + reg_file[instr->operands[i].physReg() + k] = 0; + } + } + for (unsigned i = 0; i < instr->definitions.size(); i++) { + Definition def = instr->definitions[i]; + if (def.isTemp() && def.isFixed() && ctx.defs_done.test(i)) { + for (unsigned k = 0; k < def.getTemp().size(); k++) + reg_file[def.physReg() + k] = def.tempId(); + } + } + return {{}, false}; + } + + std::array register_file = reg_file; + + /* now, we figured the placement for our definition */ + std::set> vars; + for (unsigned j = best_pos; j < best_pos + size; j++) { + if (reg_file[j] != 0xFFFFFFFF && reg_file[j] != 0) + vars.emplace(ctx.assignments[reg_file[j]].second.size(), reg_file[j]); + reg_file[j] = 0; + } + + if (instr->opcode == aco_opcode::p_create_vector) { + /* move killed operands which aren't yet at the correct position */ + for (unsigned i = 0, offset = 0; i < instr->operands.size(); offset += instr->operands[i].size(), i++) { + if (instr->operands[i].isTemp() && instr->operands[i].isFirstKill() && + instr->operands[i].getTemp().type() == rc.type()) { + + if (instr->operands[i].physReg() != best_pos + offset) { + vars.emplace(instr->operands[i].size(), instr->operands[i].tempId()); + for (unsigned j = 0; j < instr->operands[i].size(); j++) + reg_file[instr->operands[i].physReg() + j] = 0; + } else { + for (unsigned j = 0; j < instr->operands[i].size(); j++) + reg_file[instr->operands[i].physReg() + j] = instr->operands[i].tempId(); + } + } + } + } else { + /* re-enable the killed operands */ + for (unsigned j = 0; !is_phi(instr) && j < instr->operands.size(); j++) { + if (instr->operands[j].isTemp() && instr->operands[j].isFirstKill()) { + for (unsigned k = 0; k < instr->operands[j].getTemp().size(); k++) + reg_file[instr->operands[j].physReg() + k] = instr->operands[j].tempId(); + } + } + } + + std::vector> pc; + if (!get_regs_for_copies(ctx, reg_file, pc, vars, lb, ub, instr, best_pos, best_pos + size - 1)) { + reg_file = std::move(register_file); + /* remove killed operands from reg_file once again */ + if (!is_phi(instr)) { + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) { + for (unsigned k = 0; k < op.getTemp().size(); k++) + reg_file[op.physReg() + k] = 0; + } + } + } + for (unsigned i = 0; i < instr->definitions.size(); i++) { + Definition& def = instr->definitions[i]; + if (def.isTemp() && def.isFixed() && ctx.defs_done.test(i)) { + for (unsigned k = 0; k < def.getTemp().size(); k++) + reg_file[def.physReg() + k] = def.tempId(); + } + } + return {{}, false}; + } + + parallelcopies.insert(parallelcopies.end(), pc.begin(), pc.end()); + + /* we set the definition regs == 0. the actual caller is responsible for correct setting */ + for (unsigned i = 0; i < size; i++) + reg_file[best_pos + i] = 0; + + update_renames(ctx, reg_file, parallelcopies, instr); + + /* remove killed operands from reg_file once again */ + for (unsigned i = 0; !is_phi(instr) && i < instr->operands.size(); i++) { + if (!instr->operands[i].isTemp() || !instr->operands[i].isFixed()) + continue; + assert(!instr->operands[i].isUndefined()); + if (instr->operands[i].isFirstKill()) { + for (unsigned j = 0; j < instr->operands[i].getTemp().size(); j++) + reg_file[instr->operands[i].physReg() + j] = 0; + } + } + for (unsigned i = 0; i < instr->definitions.size(); i++) { + Definition def = instr->definitions[i]; + if (def.isTemp() && def.isFixed() && ctx.defs_done.test(i)) { + for (unsigned k = 0; k < def.getTemp().size(); k++) + reg_file[def.physReg() + k] = def.tempId(); + } + } + + adjust_max_used_regs(ctx, rc, best_pos); + return {PhysReg{best_pos}, true}; +} + +PhysReg get_reg(ra_ctx& ctx, + std::array& reg_file, + RegClass rc, + std::vector>& parallelcopies, + aco_ptr& instr) +{ + uint32_t size = rc.size(); + uint32_t stride = 1; + uint32_t lb, ub; + if (rc.type() == RegType::vgpr) { + lb = 256; + ub = 256 + ctx.program->max_reg_demand.vgpr; + } else { + lb = 0; + ub = ctx.program->max_reg_demand.sgpr; + if (size == 2) + stride = 2; + else if (size >= 4) + stride = 4; + } + + std::pair res = {{}, false}; + /* try to find space without live-range splits */ + if (rc.type() == RegType::vgpr && (size == 4 || size == 8)) + res = get_reg_simple(ctx, reg_file, lb, ub, size, 4, rc); + if (!res.second) + res = get_reg_simple(ctx, reg_file, lb, ub, size, stride, rc); + if (res.second) + return res.first; + + /* try to find space with live-range splits */ + res = get_reg_impl(ctx, reg_file, parallelcopies, lb, ub, size, stride, rc, instr); + + if (res.second) + return res.first; + + unsigned regs_free = 0; + for (unsigned i = lb; i < ub; i++) { + if (!reg_file[i]) + regs_free++; + } + + /* We should only fail here because keeping under the limit would require + * too many moves. */ + assert(regs_free >= size); + + /* try using more registers */ + uint16_t max_addressible_sgpr = ctx.program->sgpr_limit; + if (rc.type() == RegType::vgpr && ctx.program->max_reg_demand.vgpr < 256) { + update_vgpr_sgpr_demand(ctx.program, RegisterDemand(ctx.program->max_reg_demand.vgpr + 1, ctx.program->max_reg_demand.sgpr)); + return get_reg(ctx, reg_file, rc, parallelcopies, instr); + } else if (rc.type() == RegType::sgpr && ctx.program->max_reg_demand.sgpr < max_addressible_sgpr) { + update_vgpr_sgpr_demand(ctx.program, RegisterDemand(ctx.program->max_reg_demand.vgpr, ctx.program->max_reg_demand.sgpr + 1)); + return get_reg(ctx, reg_file, rc, parallelcopies, instr); + } + + //FIXME: if nothing helps, shift-rotate the registers to make space + + unreachable("did not find a register"); +} + + +std::pair get_reg_vec(ra_ctx& ctx, + std::array& reg_file, + RegClass rc) +{ + uint32_t size = rc.size(); + uint32_t stride = 1; + uint32_t lb, ub; + if (rc.type() == RegType::vgpr) { + lb = 256; + ub = 256 + ctx.program->max_reg_demand.vgpr; + } else { + lb = 0; + ub = ctx.program->max_reg_demand.sgpr; + if (size == 2) + stride = 2; + else if (size >= 4) + stride = 4; + } + return get_reg_simple(ctx, reg_file, lb, ub, size, stride, rc); +} + + +PhysReg get_reg_create_vector(ra_ctx& ctx, + std::array& reg_file, + RegClass rc, + std::vector>& parallelcopies, + aco_ptr& instr) +{ + /* create_vector instructions have different costs w.r.t. register coalescing */ + uint32_t size = rc.size(); + uint32_t stride = 1; + uint32_t lb, ub; + if (rc.type() == RegType::vgpr) { + lb = 256; + ub = 256 + ctx.program->max_reg_demand.vgpr; + } else { + lb = 0; + ub = ctx.program->max_reg_demand.sgpr; + if (size == 2) + stride = 2; + else if (size >= 4) + stride = 4; + } + + unsigned best_pos = -1; + unsigned num_moves = 0xFF; + bool best_war_hint = true; + + /* test for each operand which definition placement causes the least shuffle instructions */ + for (unsigned i = 0, offset = 0; i < instr->operands.size(); offset += instr->operands[i].size(), i++) { + // TODO: think about, if we can alias live operands on the same register + if (!instr->operands[i].isTemp() || !instr->operands[i].isKill() || instr->operands[i].getTemp().type() != rc.type()) + continue; + + if (offset > instr->operands[i].physReg()) + continue; + + unsigned reg_lo = instr->operands[i].physReg() - offset; + unsigned reg_hi = reg_lo + size - 1; + unsigned k = 0; + + /* no need to check multiple times */ + if (reg_lo == best_pos) + continue; + + /* check borders */ + // TODO: this can be improved */ + if (reg_lo < lb || reg_hi >= ub || reg_lo % stride != 0) + continue; + if (reg_lo > lb && reg_file[reg_lo] != 0 && reg_file[reg_lo] == reg_file[reg_lo - 1]) + continue; + if (reg_hi < ub - 1 && reg_file[reg_hi] != 0 && reg_file[reg_hi] == reg_file[reg_hi + 1]) + continue; + + /* count variables to be moved and check war_hint */ + bool war_hint = false; + for (unsigned j = reg_lo; j <= reg_hi; j++) { + if (reg_file[j] != 0) + k++; + war_hint |= ctx.war_hint[j]; + } + + /* count operands in wrong positions */ + for (unsigned j = 0, offset = 0; j < instr->operands.size(); offset += instr->operands[j].size(), j++) { + if (j == i || + !instr->operands[j].isTemp() || + instr->operands[j].getTemp().type() != rc.type()) + continue; + if (instr->operands[j].physReg() != reg_lo + offset) + k += instr->operands[j].size(); + } + bool aligned = rc == RegClass::v4 && reg_lo % 4 == 0; + if (k > num_moves || (!aligned && k == num_moves) || (war_hint && !best_war_hint)) + continue; + + best_pos = reg_lo; + num_moves = k; + best_war_hint = war_hint; + } + + if (num_moves >= size) + return get_reg(ctx, reg_file, rc, parallelcopies, instr); + + /* collect variables to be moved */ + std::set> vars; + for (unsigned i = best_pos; i < best_pos + size; i++) { + if (reg_file[i] != 0) + vars.emplace(ctx.assignments[reg_file[i]].second.size(), reg_file[i]); + reg_file[i] = 0; + } + + /* move killed operands which aren't yet at the correct position */ + for (unsigned i = 0, offset = 0; i < instr->operands.size(); offset += instr->operands[i].size(), i++) { + if (instr->operands[i].isTemp() && instr->operands[i].isFirstKill() && instr->operands[i].getTemp().type() == rc.type()) { + if (instr->operands[i].physReg() != best_pos + offset) { + vars.emplace(instr->operands[i].size(), instr->operands[i].tempId()); + } else { + for (unsigned j = 0; j < instr->operands[i].size(); j++) + reg_file[instr->operands[i].physReg() + j] = instr->operands[i].tempId(); + } + } + } + + ASSERTED bool success = false; + success = get_regs_for_copies(ctx, reg_file, parallelcopies, vars, lb, ub, instr, best_pos, best_pos + size - 1); + assert(success); + + update_renames(ctx, reg_file, parallelcopies, instr); + adjust_max_used_regs(ctx, rc, best_pos); + return PhysReg{best_pos}; +} + +bool get_reg_specified(ra_ctx& ctx, + std::array& reg_file, + RegClass rc, + std::vector>& parallelcopies, + aco_ptr& instr, + PhysReg reg) +{ + uint32_t size = rc.size(); + uint32_t stride = 1; + uint32_t lb, ub; + + if (rc.type() == RegType::vgpr) { + lb = 256; + ub = 256 + ctx.program->max_reg_demand.vgpr; + } else { + if (size == 2) + stride = 2; + else if (size >= 4) + stride = 4; + if (reg % stride != 0) + return false; + lb = 0; + ub = ctx.program->max_reg_demand.sgpr; + } + + uint32_t reg_lo = reg.reg; + uint32_t reg_hi = reg + (size - 1); + + if (reg_lo < lb || reg_hi >= ub || reg_lo > reg_hi) + return false; + + for (unsigned i = reg_lo; i <= reg_hi; i++) { + if (reg_file[i] != 0) + return false; + } + adjust_max_used_regs(ctx, rc, reg_lo); + return true; +} + +void handle_pseudo(ra_ctx& ctx, + const std::array& reg_file, + Instruction* instr) +{ + if (instr->format != Format::PSEUDO) + return; + + /* all instructions which use handle_operands() need this information */ + switch (instr->opcode) { + case aco_opcode::p_extract_vector: + case aco_opcode::p_create_vector: + case aco_opcode::p_split_vector: + case aco_opcode::p_parallelcopy: + case aco_opcode::p_wqm: + break; + default: + return; + } + + /* if all definitions are vgpr, no need to care for SCC */ + bool writes_sgpr = false; + for (Definition& def : instr->definitions) { + if (def.getTemp().type() == RegType::sgpr) { + writes_sgpr = true; + break; + } + } + if (!writes_sgpr) + return; + + Pseudo_instruction *pi = (Pseudo_instruction *)instr; + if (reg_file[scc.reg]) { + pi->tmp_in_scc = true; + + int reg = ctx.max_used_sgpr; + for (; reg >= 0 && reg_file[reg]; reg--) + ; + if (reg < 0) { + reg = ctx.max_used_sgpr + 1; + for (; reg < ctx.program->max_reg_demand.sgpr && reg_file[reg]; reg++) + ; + assert(reg < ctx.program->max_reg_demand.sgpr); + } + + adjust_max_used_regs(ctx, s1, reg); + pi->scratch_sgpr = PhysReg{(unsigned)reg}; + } else { + pi->tmp_in_scc = false; + } +} + +bool operand_can_use_reg(aco_ptr& instr, unsigned idx, PhysReg reg) +{ + switch (instr->format) { + case Format::SMEM: + return reg != scc && + reg != exec && + (reg != m0 || idx == 1 || idx == 3) && /* offset can be m0 */ + (reg != vcc || (instr->definitions.empty() && idx == 2)); /* sdata can be vcc */ + default: + // TODO: there are more instructions with restrictions on registers + return true; + } +} + +} /* end namespace */ + + +void register_allocation(Program *program, std::vector> live_out_per_block) +{ + ra_ctx ctx(program); + + std::vector> renames(program->blocks.size()); + + struct phi_info { + Instruction* phi; + unsigned block_idx; + std::set uses; + }; + + bool filled[program->blocks.size()]; + bool sealed[program->blocks.size()]; + memset(filled, 0, sizeof filled); + memset(sealed, 0, sizeof sealed); + std::vector> incomplete_phis(program->blocks.size()); + std::map phi_map; + std::map affinities; + std::function read_variable; + std::function handle_live_in; + std::function::iterator)> try_remove_trivial_phi; + + read_variable = [&](Temp val, unsigned block_idx) -> Temp { + std::unordered_map::iterator it = renames[block_idx].find(val.id()); + assert(it != renames[block_idx].end()); + return it->second; + }; + + handle_live_in = [&](Temp val, Block *block) -> Temp { + std::vector& preds = val.is_linear() ? block->linear_preds : block->logical_preds; + if (preds.size() == 0 && block->index != 0) { + renames[block->index][val.id()] = val; + return val; + } + assert(preds.size() > 0); + + Temp new_val; + if (!sealed[block->index]) { + /* consider rename from already processed predecessor */ + Temp tmp = read_variable(val, preds[0]); + + /* if the block is not sealed yet, we create an incomplete phi (which might later get removed again) */ + new_val = Temp{program->allocateId(), val.regClass()}; + aco_opcode opcode = val.is_linear() ? aco_opcode::p_linear_phi : aco_opcode::p_phi; + aco_ptr phi{create_instruction(opcode, Format::PSEUDO, preds.size(), 1)}; + phi->definitions[0] = Definition(new_val); + for (unsigned i = 0; i < preds.size(); i++) + phi->operands[i] = Operand(val); + if (tmp.regClass() == new_val.regClass()) + affinities[new_val.id()] = tmp.id(); + + phi_map.emplace(new_val.id(), phi_info{phi.get(), block->index}); + incomplete_phis[block->index].emplace_back(phi.get()); + block->instructions.insert(block->instructions.begin(), std::move(phi)); + + } else if (preds.size() == 1) { + /* if the block has only one predecessor, just look there for the name */ + new_val = read_variable(val, preds[0]); + } else { + /* there are multiple predecessors and the block is sealed */ + Temp ops[preds.size()]; + + /* we start assuming that the name is the same from all predecessors */ + renames[block->index][val.id()] = val; + bool needs_phi = false; + + /* get the rename from each predecessor and check if they are the same */ + for (unsigned i = 0; i < preds.size(); i++) { + ops[i] = read_variable(val, preds[i]); + if (i == 0) + new_val = ops[i]; + else + needs_phi |= !(new_val == ops[i]); + } + + if (needs_phi) { + /* the variable has been renamed differently in the predecessors: we need to insert a phi */ + aco_opcode opcode = val.is_linear() ? aco_opcode::p_linear_phi : aco_opcode::p_phi; + aco_ptr phi{create_instruction(opcode, Format::PSEUDO, preds.size(), 1)}; + new_val = Temp{program->allocateId(), val.regClass()}; + phi->definitions[0] = Definition(new_val); + for (unsigned i = 0; i < preds.size(); i++) { + phi->operands[i] = Operand(ops[i]); + phi->operands[i].setFixed(ctx.assignments[ops[i].id()].first); + if (ops[i].regClass() == new_val.regClass()) + affinities[new_val.id()] = ops[i].id(); + } + phi_map.emplace(new_val.id(), phi_info{phi.get(), block->index}); + block->instructions.insert(block->instructions.begin(), std::move(phi)); + } + } + + renames[block->index][val.id()] = new_val; + renames[block->index][new_val.id()] = new_val; + ctx.orig_names[new_val.id()] = val; + return new_val; + }; + + try_remove_trivial_phi = [&] (std::map::iterator info) -> Temp { + assert(info->second.block_idx != 0); + Instruction* phi = info->second.phi; + Temp same = Temp(); + + Definition def = phi->definitions[0]; + /* a phi node is trivial if all operands are the same as the definition of the phi */ + for (const Operand& op : phi->operands) { + const Temp t = op.getTemp(); + if (t == same || t == def.getTemp()) + continue; + if (!(same == Temp()) || !(op.physReg() == def.physReg())) { + /* phi is not trivial */ + return def.getTemp(); + } + same = t; + } + assert(!(same == Temp() || same == def.getTemp())); + + /* reroute all uses to same and remove phi */ + std::vector::iterator> phi_users; + std::map::iterator same_phi_info = phi_map.find(same.id()); + for (Instruction* instr : info->second.uses) { + assert(phi != instr); + /* recursively try to remove trivial phis */ + if (is_phi(instr)) { + /* ignore if the phi was already flagged trivial */ + if (instr->definitions.empty()) + continue; + + std::map::iterator it = phi_map.find(instr->definitions[0].tempId()); + if (it != phi_map.end() && it != info) + phi_users.emplace_back(it); + } + for (Operand& op : instr->operands) { + if (op.isTemp() && op.tempId() == def.tempId()) { + op.setTemp(same); + if (same_phi_info != phi_map.end()) + same_phi_info->second.uses.emplace(instr); + } + } + } + + auto it = ctx.orig_names.find(same.id()); + unsigned orig_var = it != ctx.orig_names.end() ? it->second.id() : same.id(); + for (unsigned i = 0; i < program->blocks.size(); i++) { + auto it = renames[i].find(orig_var); + if (it != renames[i].end() && it->second == def.getTemp()) + renames[i][orig_var] = same; + } + + unsigned block_idx = info->second.block_idx; + phi->definitions.clear(); /* this indicates that the phi can be removed */ + phi_map.erase(info); + for (auto it : phi_users) { + if (sealed[it->second.block_idx]) + try_remove_trivial_phi(it); + } + + /* due to the removal of other phis, the name might have changed once again! */ + return renames[block_idx][orig_var]; + }; + + std::map vectors; + std::vector> phi_ressources; + std::map temp_to_phi_ressources; + + for (std::vector::reverse_iterator it = program->blocks.rbegin(); it != program->blocks.rend(); it++) { + Block& block = *it; + + /* first, compute the death points of all live vars within the block */ + std::set& live = live_out_per_block[block.index]; + + std::vector>::reverse_iterator rit; + for (rit = block.instructions.rbegin(); rit != block.instructions.rend(); ++rit) { + aco_ptr& instr = *rit; + if (!is_phi(instr)) { + for (const Operand& op : instr->operands) { + if (op.isTemp()) + live.emplace(op.getTemp()); + } + if (instr->opcode == aco_opcode::p_create_vector) { + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.getTemp().type() == instr->definitions[0].getTemp().type()) + vectors[op.tempId()] = instr.get(); + } + } + } else if (!instr->definitions[0].isKill() && !instr->definitions[0].isFixed()) { + /* collect information about affinity-related temporaries */ + std::vector affinity_related; + /* affinity_related[0] is the last seen affinity-related temp */ + affinity_related.emplace_back(instr->definitions[0].getTemp()); + affinity_related.emplace_back(instr->definitions[0].getTemp()); + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.regClass() == instr->definitions[0].regClass()) { + affinity_related.emplace_back(op.getTemp()); + temp_to_phi_ressources[op.tempId()] = phi_ressources.size(); + } + } + phi_ressources.emplace_back(std::move(affinity_related)); + } + + /* erase from live */ + for (const Definition& def : instr->definitions) { + if (def.isTemp()) { + live.erase(def.getTemp()); + std::map::iterator it = temp_to_phi_ressources.find(def.tempId()); + if (it != temp_to_phi_ressources.end() && def.regClass() == phi_ressources[it->second][0].regClass()) + phi_ressources[it->second][0] = def.getTemp(); + } + } + } + } + /* create affinities */ + for (std::vector& vec : phi_ressources) { + assert(vec.size() > 1); + for (unsigned i = 1; i < vec.size(); i++) + if (vec[i].id() != vec[0].id()) + affinities[vec[i].id()] = vec[0].id(); + } + + /* state of register file after phis */ + std::vector> sgpr_live_in(program->blocks.size()); + + for (Block& block : program->blocks) { + std::set& live = live_out_per_block[block.index]; + /* initialize register file */ + assert(block.index != 0 || live.empty()); + std::array register_file = {0}; + ctx.war_hint.reset(); + + for (Temp t : live) { + Temp renamed = handle_live_in(t, &block); + if (ctx.assignments.find(renamed.id()) != ctx.assignments.end()) { + for (unsigned i = 0; i < t.size(); i++) + register_file[ctx.assignments[renamed.id()].first + i] = renamed.id(); + } + } + + std::vector> instructions; + std::vector>::iterator it; + + /* this is a slight adjustment from the paper as we already have phi nodes: + * We consider them incomplete phis and only handle the definition. */ + + /* handle fixed phi definitions */ + for (it = block.instructions.begin(); it != block.instructions.end(); ++it) { + aco_ptr& phi = *it; + if (!is_phi(phi)) + break; + Definition& definition = phi->definitions[0]; + if (!definition.isFixed()) + continue; + + /* check if a dead exec mask phi is needed */ + if (definition.isKill()) { + for (Operand& op : phi->operands) { + assert(op.isTemp()); + if (ctx.assignments.find(op.tempId()) == ctx.assignments.end() || + ctx.assignments[op.tempId()].first != exec) { + definition.setKill(false); + break; + } + } + } + + if (definition.isKill()) + continue; + + assert(definition.physReg() == exec); + for (unsigned i = 0; i < definition.size(); i++) { + assert(register_file[definition.physReg() + i] == 0); + register_file[definition.physReg() + i] = definition.tempId(); + } + ctx.assignments[definition.tempId()] = {definition.physReg(), definition.regClass()}; + } + + /* look up the affinities */ + for (it = block.instructions.begin(); it != block.instructions.end(); ++it) { + aco_ptr& phi = *it; + if (!is_phi(phi)) + break; + Definition& definition = phi->definitions[0]; + if (definition.isKill() || definition.isFixed()) + continue; + + if (affinities.find(definition.tempId()) != affinities.end() && + ctx.assignments.find(affinities[definition.tempId()]) != ctx.assignments.end()) { + assert(ctx.assignments[affinities[definition.tempId()]].second == definition.regClass()); + PhysReg reg = ctx.assignments[affinities[definition.tempId()]].first; + bool try_use_special_reg = reg == scc || reg == exec; + if (try_use_special_reg) { + for (const Operand& op : phi->operands) { + if (!op.isTemp() || + ctx.assignments.find(op.tempId()) == ctx.assignments.end() || + !(ctx.assignments[op.tempId()].first == reg)) { + try_use_special_reg = false; + break; + } + } + if (!try_use_special_reg) + continue; + } + bool reg_free = true; + for (unsigned i = reg.reg; reg_free && i < reg + definition.size(); i++) { + if (register_file[i] != 0) + reg_free = false; + } + /* only assign if register is still free */ + if (reg_free) { + definition.setFixed(reg); + for (unsigned i = 0; i < definition.size(); i++) + register_file[definition.physReg() + i] = definition.tempId(); + ctx.assignments[definition.tempId()] = {definition.physReg(), definition.regClass()}; + } + } + } + + /* find registers for phis without affinity or where the register was blocked */ + for (it = block.instructions.begin();it != block.instructions.end(); ++it) { + aco_ptr& phi = *it; + if (!is_phi(phi)) + break; + + Definition& definition = phi->definitions[0]; + if (definition.isKill()) + continue; + + renames[block.index][definition.tempId()] = definition.getTemp(); + + if (!definition.isFixed()) { + std::vector> parallelcopy; + /* try to find a register that is used by at least one operand */ + for (const Operand& op : phi->operands) { + if (!op.isTemp() || + ctx.assignments.find(op.tempId()) == ctx.assignments.end()) + continue; + PhysReg reg = ctx.assignments[op.tempId()].first; + /* we tried this already on the previous loop */ + if (reg == scc || reg == exec) + continue; + if (get_reg_specified(ctx, register_file, definition.regClass(), parallelcopy, phi, reg)) { + definition.setFixed(reg); + break; + } + } + if (!definition.isFixed()) + definition.setFixed(get_reg(ctx, register_file, definition.regClass(), parallelcopy, phi)); + + /* process parallelcopy */ + for (std::pair pc : parallelcopy) { + /* rename */ + std::map::iterator orig_it = ctx.orig_names.find(pc.first.tempId()); + Temp orig = pc.first.getTemp(); + if (orig_it != ctx.orig_names.end()) + orig = orig_it->second; + else + ctx.orig_names[pc.second.tempId()] = orig; + renames[block.index][orig.id()] = pc.second.getTemp(); + renames[block.index][pc.second.tempId()] = pc.second.getTemp(); + + /* see if it's a copy from a previous phi */ + //TODO: prefer moving some previous phis over live-ins + //TODO: somehow prevent phis fixed before the RA from being updated (shouldn't be a problem in practice since they can only be fixed to exec) + Instruction *prev_phi = NULL; + for (auto it2 = instructions.begin(); it2 != instructions.end(); ++it2) { + if ((*it2)->definitions[0].tempId() == pc.first.tempId()) + prev_phi = it2->get(); + } + if (prev_phi) { + /* if so, just update that phi */ + prev_phi->definitions[0] = pc.second; + continue; + } + + /* otherwise, this is a live-in and we need to create a new phi + * to move it in this block's predecessors */ + aco_opcode opcode = pc.first.getTemp().is_linear() ? aco_opcode::p_linear_phi : aco_opcode::p_phi; + std::vector& preds = pc.first.getTemp().is_linear() ? block.linear_preds : block.logical_preds; + aco_ptr new_phi{create_instruction(opcode, Format::PSEUDO, preds.size(), 1)}; + new_phi->definitions[0] = pc.second; + for (unsigned i = 0; i < preds.size(); i++) + new_phi->operands[i] = Operand(pc.first); + instructions.emplace_back(std::move(new_phi)); + } + + for (unsigned i = 0; i < definition.size(); i++) + register_file[definition.physReg() + i] = definition.tempId(); + ctx.assignments[definition.tempId()] = {definition.physReg(), definition.regClass()}; + } + live.emplace(definition.getTemp()); + + /* update phi affinities */ + for (const Operand& op : phi->operands) { + if (op.isTemp() && op.regClass() == phi->definitions[0].regClass()) + affinities[op.tempId()] = definition.tempId(); + } + + instructions.emplace_back(std::move(*it)); + } + + /* fill in sgpr_live_in */ + for (unsigned i = 0; i < ctx.max_used_sgpr; i++) + sgpr_live_in[block.index][i] = register_file[i]; + sgpr_live_in[block.index][127] = register_file[scc.reg]; + + /* Handle all other instructions of the block */ + for (; it != block.instructions.end(); ++it) { + aco_ptr& instr = *it; + + /* parallelcopies from p_phi are inserted here which means + * live ranges of killed operands end here as well */ + if (instr->opcode == aco_opcode::p_logical_end) { + /* no need to process this instruction any further */ + if (block.logical_succs.size() != 1) { + instructions.emplace_back(std::move(instr)); + continue; + } + + Block& succ = program->blocks[block.logical_succs[0]]; + unsigned idx = 0; + for (; idx < succ.logical_preds.size(); idx++) { + if (succ.logical_preds[idx] == block.index) + break; + } + for (aco_ptr& phi : succ.instructions) { + if (phi->opcode == aco_opcode::p_phi) { + if (phi->operands[idx].isTemp() && + phi->operands[idx].getTemp().type() == RegType::sgpr && + phi->operands[idx].isFirstKill()) { + Temp phi_op = read_variable(phi->operands[idx].getTemp(), block.index); + PhysReg reg = ctx.assignments[phi_op.id()].first; + assert(register_file[reg] == phi_op.id()); + register_file[reg] = 0; + } + } else if (phi->opcode != aco_opcode::p_linear_phi) { + break; + } + } + instructions.emplace_back(std::move(instr)); + continue; + } + + std::vector> parallelcopy; + + assert(!is_phi(instr)); + + /* handle operands */ + for (unsigned i = 0; i < instr->operands.size(); ++i) { + auto& operand = instr->operands[i]; + if (!operand.isTemp()) + continue; + + /* rename operands */ + operand.setTemp(read_variable(operand.getTemp(), block.index)); + + /* check if the operand is fixed */ + if (operand.isFixed()) { + + if (operand.physReg() == ctx.assignments[operand.tempId()].first) { + /* we are fine: the operand is already assigned the correct reg */ + + } else { + /* check if target reg is blocked, and move away the blocking var */ + if (register_file[operand.physReg().reg]) { + uint32_t blocking_id = register_file[operand.physReg().reg]; + RegClass rc = ctx.assignments[blocking_id].second; + Operand pc_op = Operand(Temp{blocking_id, rc}); + pc_op.setFixed(operand.physReg()); + Definition pc_def = Definition(Temp{program->allocateId(), pc_op.regClass()}); + /* find free reg */ + PhysReg reg = get_reg(ctx, register_file, pc_op.regClass(), parallelcopy, instr); + pc_def.setFixed(reg); + ctx.assignments[pc_def.tempId()] = {reg, pc_def.regClass()}; + for (unsigned i = 0; i < operand.size(); i++) { + register_file[pc_op.physReg() + i] = 0; + register_file[pc_def.physReg() + i] = pc_def.tempId(); + } + parallelcopy.emplace_back(pc_op, pc_def); + + /* handle renames of previous operands */ + for (unsigned j = 0; j < i; j++) { + Operand& op = instr->operands[j]; + if (op.isTemp() && op.tempId() == blocking_id) { + op = Operand(pc_def.getTemp()); + op.setFixed(reg); + } + } + } + /* move operand to fixed reg and create parallelcopy pair */ + Operand pc_op = operand; + Temp tmp = Temp{program->allocateId(), operand.regClass()}; + Definition pc_def = Definition(tmp); + pc_def.setFixed(operand.physReg()); + pc_op.setFixed(ctx.assignments[operand.tempId()].first); + operand.setTemp(tmp); + ctx.assignments[tmp.id()] = {pc_def.physReg(), pc_def.regClass()}; + operand.setFixed(pc_def.physReg()); + for (unsigned i = 0; i < operand.size(); i++) { + register_file[pc_op.physReg() + i] = 0; + register_file[pc_def.physReg() + i] = tmp.id(); + } + parallelcopy.emplace_back(pc_op, pc_def); + } + } else { + assert(ctx.assignments.find(operand.tempId()) != ctx.assignments.end()); + PhysReg reg = ctx.assignments[operand.tempId()].first; + + if (operand_can_use_reg(instr, i, reg)) { + operand.setFixed(ctx.assignments[operand.tempId()].first); + } else { + Operand pc_op = operand; + pc_op.setFixed(reg); + PhysReg new_reg = get_reg(ctx, register_file, operand.regClass(), parallelcopy, instr); + Definition pc_def = Definition(program->allocateId(), new_reg, pc_op.regClass()); + ctx.assignments[pc_def.tempId()] = {reg, pc_def.regClass()}; + for (unsigned i = 0; i < operand.size(); i++) { + register_file[pc_op.physReg() + i] = 0; + register_file[pc_def.physReg() + i] = pc_def.tempId(); + } + parallelcopy.emplace_back(pc_op, pc_def); + operand.setFixed(new_reg); + } + + if (instr->format == Format::EXP || + (instr->isVMEM() && i == 3 && program->chip_class == GFX6) || + (instr->format == Format::DS && static_cast(instr.get())->gds)) { + for (unsigned j = 0; j < operand.size(); j++) + ctx.war_hint.set(operand.physReg().reg + j); + } + } + std::map::iterator phi = phi_map.find(operand.getTemp().id()); + if (phi != phi_map.end()) + phi->second.uses.emplace(instr.get()); + + } + /* remove dead vars from register file */ + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) + for (unsigned j = 0; j < op.size(); j++) + register_file[op.physReg() + j] = 0; + } + + /* try to optimize v_mad_f32 -> v_mac_f32 */ + if (instr->opcode == aco_opcode::v_mad_f32 && + instr->operands[2].isTemp() && + instr->operands[2].isKill() && + instr->operands[2].getTemp().type() == RegType::vgpr && + instr->operands[1].isTemp() && + instr->operands[1].getTemp().type() == RegType::vgpr) { /* TODO: swap src0 and src1 in this case */ + VOP3A_instruction* vop3 = static_cast(instr.get()); + bool can_use_mac = !(vop3->abs[0] || vop3->abs[1] || vop3->abs[2] || + vop3->opsel[0] || vop3->opsel[1] || vop3->opsel[2] || + vop3->neg[0] || vop3->neg[1] || vop3->neg[2] || + vop3->clamp || vop3->omod); + if (can_use_mac) { + instr->format = Format::VOP2; + instr->opcode = aco_opcode::v_mac_f32; + } + } + + /* handle definitions which must have the same register as an operand */ + if (instr->opcode == aco_opcode::v_interp_p2_f32 || + instr->opcode == aco_opcode::v_mac_f32 || + instr->opcode == aco_opcode::v_writelane_b32) { + instr->definitions[0].setFixed(instr->operands[2].physReg()); + } else if (instr->opcode == aco_opcode::s_addk_i32 || + instr->opcode == aco_opcode::s_mulk_i32) { + instr->definitions[0].setFixed(instr->operands[0].physReg()); + } else if ((instr->format == Format::MUBUF || + instr->format == Format::MIMG) && + instr->definitions.size() == 1 && + instr->operands.size() == 4) { + instr->definitions[0].setFixed(instr->operands[3].physReg()); + } + + ctx.defs_done.reset(); + + /* handle fixed definitions first */ + for (unsigned i = 0; i < instr->definitions.size(); ++i) { + auto& definition = instr->definitions[i]; + if (!definition.isFixed()) + continue; + + adjust_max_used_regs(ctx, definition.regClass(), definition.physReg()); + /* check if the target register is blocked */ + if (register_file[definition.physReg().reg] != 0) { + /* create parallelcopy pair to move blocking var */ + Temp tmp = {register_file[definition.physReg()], ctx.assignments[register_file[definition.physReg()]].second}; + Operand pc_op = Operand(tmp); + pc_op.setFixed(ctx.assignments[register_file[definition.physReg().reg]].first); + RegClass rc = pc_op.regClass(); + tmp = Temp{program->allocateId(), rc}; + Definition pc_def = Definition(tmp); + + /* re-enable the killed operands, so that we don't move the blocking var there */ + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) + for (unsigned j = 0; j < op.size(); j++) + register_file[op.physReg() + j] = 0xFFFF; + } + + /* find a new register for the blocking variable */ + PhysReg reg = get_reg(ctx, register_file, rc, parallelcopy, instr); + /* once again, disable killed operands */ + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) + for (unsigned j = 0; j < op.size(); j++) + register_file[op.physReg() + j] = 0; + } + for (unsigned k = 0; k < i; k++) { + if (instr->definitions[k].isTemp() && ctx.defs_done.test(k) && !instr->definitions[k].isKill()) + for (unsigned j = 0; j < instr->definitions[k].size(); j++) + register_file[instr->definitions[k].physReg() + j] = instr->definitions[k].tempId(); + } + pc_def.setFixed(reg); + + /* finish assignment of parallelcopy */ + ctx.assignments[pc_def.tempId()] = {reg, pc_def.regClass()}; + parallelcopy.emplace_back(pc_op, pc_def); + + /* add changes to reg_file */ + for (unsigned i = 0; i < pc_op.size(); i++) { + register_file[pc_op.physReg() + i] = 0x0; + register_file[pc_def.physReg() + i] = pc_def.tempId(); + } + } + ctx.defs_done.set(i); + + if (!definition.isTemp()) + continue; + + /* set live if it has a kill point */ + if (!definition.isKill()) + live.emplace(definition.getTemp()); + + ctx.assignments[definition.tempId()] = {definition.physReg(), definition.regClass()}; + renames[block.index][definition.tempId()] = definition.getTemp(); + for (unsigned j = 0; j < definition.size(); j++) + register_file[definition.physReg() + j] = definition.tempId(); + } + + /* handle all other definitions */ + for (unsigned i = 0; i < instr->definitions.size(); ++i) { + auto& definition = instr->definitions[i]; + + if (definition.isFixed() || !definition.isTemp()) + continue; + + /* find free reg */ + if (definition.hasHint() && register_file[definition.physReg().reg] == 0) + definition.setFixed(definition.physReg()); + else if (instr->opcode == aco_opcode::p_split_vector) { + PhysReg reg = PhysReg{instr->operands[0].physReg() + i * definition.size()}; + if (!get_reg_specified(ctx, register_file, definition.regClass(), parallelcopy, instr, reg)) + reg = get_reg(ctx, register_file, definition.regClass(), parallelcopy, instr); + definition.setFixed(reg); + } else if (instr->opcode == aco_opcode::p_wqm) { + PhysReg reg; + if (instr->operands[0].isKill() && instr->operands[0].getTemp().type() == definition.getTemp().type()) { + reg = instr->operands[0].physReg(); + assert(register_file[reg.reg] == 0); + } else { + reg = get_reg(ctx, register_file, definition.regClass(), parallelcopy, instr); + } + definition.setFixed(reg); + } else if (instr->opcode == aco_opcode::p_extract_vector) { + PhysReg reg; + if (instr->operands[0].isKill() && + instr->operands[0].getTemp().type() == definition.getTemp().type()) { + reg = instr->operands[0].physReg(); + reg.reg += definition.size() * instr->operands[1].constantValue(); + assert(register_file[reg.reg] == 0); + } else { + reg = get_reg(ctx, register_file, definition.regClass(), parallelcopy, instr); + } + definition.setFixed(reg); + } else if (instr->opcode == aco_opcode::p_create_vector) { + PhysReg reg = get_reg_create_vector(ctx, register_file, definition.regClass(), + parallelcopy, instr); + definition.setFixed(reg); + } else if (affinities.find(definition.tempId()) != affinities.end() && + ctx.assignments.find(affinities[definition.tempId()]) != ctx.assignments.end()) { + PhysReg reg = ctx.assignments[affinities[definition.tempId()]].first; + if (get_reg_specified(ctx, register_file, definition.regClass(), parallelcopy, instr, reg)) + definition.setFixed(reg); + else + definition.setFixed(get_reg(ctx, register_file, definition.regClass(), parallelcopy, instr)); + + } else if (vectors.find(definition.tempId()) != vectors.end()) { + Instruction* vec = vectors[definition.tempId()]; + unsigned offset = 0; + for (const Operand& op : vec->operands) { + if (op.isTemp() && op.tempId() == definition.tempId()) + break; + else + offset += op.size(); + } + unsigned k = 0; + for (const Operand& op : vec->operands) { + if (op.isTemp() && + op.tempId() != definition.tempId() && + op.getTemp().type() == definition.getTemp().type() && + ctx.assignments.find(op.tempId()) != ctx.assignments.end()) { + PhysReg reg = ctx.assignments[op.tempId()].first; + reg.reg = reg - k + offset; + if (get_reg_specified(ctx, register_file, definition.regClass(), parallelcopy, instr, reg)) { + definition.setFixed(reg); + break; + } + } + k += op.size(); + } + if (!definition.isFixed()) { + std::pair res = get_reg_vec(ctx, register_file, vec->definitions[0].regClass()); + PhysReg reg = res.first; + if (res.second) { + reg.reg += offset; + } else { + reg = get_reg(ctx, register_file, definition.regClass(), parallelcopy, instr); + } + definition.setFixed(reg); + } + } else + definition.setFixed(get_reg(ctx, register_file, definition.regClass(), parallelcopy, instr)); + + assert(definition.isFixed() && ((definition.getTemp().type() == RegType::vgpr && definition.physReg() >= 256) || + (definition.getTemp().type() != RegType::vgpr && definition.physReg() < 256))); + ctx.defs_done.set(i); + + /* set live if it has a kill point */ + if (!definition.isKill()) + live.emplace(definition.getTemp()); + + ctx.assignments[definition.tempId()] = {definition.physReg(), definition.regClass()}; + renames[block.index][definition.tempId()] = definition.getTemp(); + for (unsigned j = 0; j < definition.size(); j++) + register_file[definition.physReg() + j] = definition.tempId(); + } + + handle_pseudo(ctx, register_file, instr.get()); + + /* kill definitions */ + for (const Definition& def : instr->definitions) { + if (def.isTemp() && def.isKill()) { + for (unsigned j = 0; j < def.size(); j++) { + register_file[def.physReg() + j] = 0; + } + } + } + + /* emit parallelcopy */ + if (!parallelcopy.empty()) { + aco_ptr pc; + pc.reset(create_instruction(aco_opcode::p_parallelcopy, Format::PSEUDO, parallelcopy.size(), parallelcopy.size())); + bool temp_in_scc = register_file[scc.reg]; + bool sgpr_operands_alias_defs = false; + uint64_t sgpr_operands[4] = {0, 0, 0, 0}; + for (unsigned i = 0; i < parallelcopy.size(); i++) { + if (temp_in_scc && parallelcopy[i].first.isTemp() && parallelcopy[i].first.getTemp().type() == RegType::sgpr) { + if (!sgpr_operands_alias_defs) { + unsigned reg = parallelcopy[i].first.physReg().reg; + unsigned size = parallelcopy[i].first.getTemp().size(); + sgpr_operands[reg / 64u] |= ((1u << size) - 1) << (reg % 64u); + + reg = parallelcopy[i].second.physReg().reg; + size = parallelcopy[i].second.getTemp().size(); + if (sgpr_operands[reg / 64u] & ((1u << size) - 1) << (reg % 64u)) + sgpr_operands_alias_defs = true; + } + } + + pc->operands[i] = parallelcopy[i].first; + pc->definitions[i] = parallelcopy[i].second; + + /* it might happen that the operand is already renamed. we have to restore the original name. */ + std::map::iterator it = ctx.orig_names.find(pc->operands[i].tempId()); + if (it != ctx.orig_names.end()) + pc->operands[i].setTemp(it->second); + unsigned orig_id = pc->operands[i].tempId(); + ctx.orig_names[pc->definitions[i].tempId()] = pc->operands[i].getTemp(); + + pc->operands[i].setTemp(read_variable(pc->operands[i].getTemp(), block.index)); + renames[block.index][orig_id] = pc->definitions[i].getTemp(); + renames[block.index][pc->definitions[i].tempId()] = pc->definitions[i].getTemp(); + std::map::iterator phi = phi_map.find(pc->operands[i].tempId()); + if (phi != phi_map.end()) + phi->second.uses.emplace(pc.get()); + } + + if (temp_in_scc && sgpr_operands_alias_defs) { + /* disable definitions and re-enable operands */ + for (const Definition& def : instr->definitions) { + if (def.isTemp() && !def.isKill()) { + for (unsigned j = 0; j < def.size(); j++) { + register_file[def.physReg() + j] = 0x0; + } + } + } + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) { + for (unsigned j = 0; j < op.size(); j++) + register_file[op.physReg() + j] = 0xFFFF; + } + } + + handle_pseudo(ctx, register_file, pc.get()); + + /* re-enable live vars */ + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) + for (unsigned j = 0; j < op.size(); j++) + register_file[op.physReg() + j] = 0x0; + } + for (const Definition& def : instr->definitions) { + if (def.isTemp() && !def.isKill()) { + for (unsigned j = 0; j < def.size(); j++) { + register_file[def.physReg() + j] = def.tempId(); + } + } + } + } else { + pc->tmp_in_scc = false; + } + + instructions.emplace_back(std::move(pc)); + } + + /* some instructions need VOP3 encoding if operand/definition is not assigned to VCC */ + bool instr_needs_vop3 = !instr->isVOP3() && + ((instr->format == Format::VOPC && !(instr->definitions[0].physReg() == vcc)) || + (instr->opcode == aco_opcode::v_cndmask_b32 && !(instr->operands[2].physReg() == vcc)) || + ((instr->opcode == aco_opcode::v_add_co_u32 || + instr->opcode == aco_opcode::v_addc_co_u32 || + instr->opcode == aco_opcode::v_sub_co_u32 || + instr->opcode == aco_opcode::v_subb_co_u32 || + instr->opcode == aco_opcode::v_subrev_co_u32 || + instr->opcode == aco_opcode::v_subbrev_co_u32) && + !(instr->definitions[1].physReg() == vcc)) || + ((instr->opcode == aco_opcode::v_addc_co_u32 || + instr->opcode == aco_opcode::v_subb_co_u32 || + instr->opcode == aco_opcode::v_subbrev_co_u32) && + !(instr->operands[2].physReg() == vcc))); + if (instr_needs_vop3) { + + /* if the first operand is a literal, we have to move it to a reg */ + if (instr->operands.size() && instr->operands[0].isLiteral()) { + bool can_sgpr = true; + /* check, if we have to move to vgpr */ + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.getTemp().type() == RegType::sgpr) { + can_sgpr = false; + break; + } + } + aco_ptr mov; + if (can_sgpr) + mov.reset(create_instruction(aco_opcode::s_mov_b32, Format::SOP1, 1, 1)); + else + mov.reset(create_instruction(aco_opcode::v_mov_b32, Format::VOP1, 1, 1)); + mov->operands[0] = instr->operands[0]; + Temp tmp = {program->allocateId(), can_sgpr ? s1 : v1}; + mov->definitions[0] = Definition(tmp); + /* disable definitions and re-enable operands */ + for (const Definition& def : instr->definitions) { + for (unsigned j = 0; j < def.size(); j++) { + register_file[def.physReg() + j] = 0x0; + } + } + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) { + for (unsigned j = 0; j < op.size(); j++) + register_file[op.physReg() + j] = 0xFFFF; + } + } + mov->definitions[0].setFixed(get_reg(ctx, register_file, tmp.regClass(), parallelcopy, mov)); + instr->operands[0] = Operand(tmp); + instr->operands[0].setFixed(mov->definitions[0].physReg()); + instructions.emplace_back(std::move(mov)); + /* re-enable live vars */ + for (const Operand& op : instr->operands) { + if (op.isTemp() && op.isFirstKill()) + for (unsigned j = 0; j < op.size(); j++) + register_file[op.physReg() + j] = 0x0; + } + for (const Definition& def : instr->definitions) { + if (def.isTemp() && !def.isKill()) { + for (unsigned j = 0; j < def.size(); j++) { + register_file[def.physReg() + j] = def.tempId(); + } + } + } + } + + /* change the instruction to VOP3 to enable an arbitrary register pair as dst */ + aco_ptr tmp = std::move(instr); + Format format = asVOP3(tmp->format); + instr.reset(create_instruction(tmp->opcode, format, tmp->operands.size(), tmp->definitions.size())); + for (unsigned i = 0; i < instr->operands.size(); i++) { + Operand& operand = tmp->operands[i]; + instr->operands[i] = operand; + /* keep phi_map up to date */ + if (operand.isTemp()) { + std::map::iterator phi = phi_map.find(operand.tempId()); + if (phi != phi_map.end()) { + phi->second.uses.erase(tmp.get()); + phi->second.uses.emplace(instr.get()); + } + } + } + std::copy(tmp->definitions.begin(), tmp->definitions.end(), instr->definitions.begin()); + } + instructions.emplace_back(std::move(*it)); + + } /* end for Instr */ + + block.instructions = std::move(instructions); + + filled[block.index] = true; + for (unsigned succ_idx : block.linear_succs) { + Block& succ = program->blocks[succ_idx]; + /* seal block if all predecessors are filled */ + bool all_filled = true; + for (unsigned pred_idx : succ.linear_preds) { + if (!filled[pred_idx]) { + all_filled = false; + break; + } + } + if (all_filled) { + /* finish incomplete phis and check if they became trivial */ + for (Instruction* phi : incomplete_phis[succ_idx]) { + std::vector preds = phi->definitions[0].getTemp().is_linear() ? succ.linear_preds : succ.logical_preds; + for (unsigned i = 0; i < phi->operands.size(); i++) { + phi->operands[i].setTemp(read_variable(phi->operands[i].getTemp(), preds[i])); + phi->operands[i].setFixed(ctx.assignments[phi->operands[i].tempId()].first); + } + try_remove_trivial_phi(phi_map.find(phi->definitions[0].tempId())); + } + /* complete the original phi nodes, but no need to check triviality */ + for (aco_ptr& instr : succ.instructions) { + if (!is_phi(instr)) + break; + std::vector preds = instr->opcode == aco_opcode::p_phi ? succ.logical_preds : succ.linear_preds; + + for (unsigned i = 0; i < instr->operands.size(); i++) { + auto& operand = instr->operands[i]; + if (!operand.isTemp()) + continue; + operand.setTemp(read_variable(operand.getTemp(), preds[i])); + operand.setFixed(ctx.assignments[operand.tempId()].first); + std::map::iterator phi = phi_map.find(operand.getTemp().id()); + if (phi != phi_map.end()) + phi->second.uses.emplace(instr.get()); + } + } + sealed[succ_idx] = true; + } + } + } /* end for BB */ + + /* remove trivial phis */ + for (Block& block : program->blocks) { + auto end = std::find_if(block.instructions.begin(), block.instructions.end(), + [](aco_ptr& instr) { return !is_phi(instr);}); + auto middle = std::remove_if(block.instructions.begin(), end, + [](const aco_ptr& instr) { return instr->definitions.empty();}); + block.instructions.erase(middle, end); + } + + /* find scc spill registers which may be needed for parallelcopies created by phis */ + for (Block& block : program->blocks) { + if (block.linear_preds.size() <= 1) + continue; + + std::bitset<128> regs = sgpr_live_in[block.index]; + if (!regs[127]) + continue; + + /* choose a register */ + int16_t reg = 0; + for (; reg < ctx.program->max_reg_demand.sgpr && regs[reg]; reg++) + ; + assert(reg < ctx.program->max_reg_demand.sgpr); + adjust_max_used_regs(ctx, s1, reg); + + /* update predecessors */ + for (unsigned& pred_index : block.linear_preds) { + Block& pred = program->blocks[pred_index]; + pred.scc_live_out = true; + pred.scratch_sgpr = PhysReg{(uint16_t)reg}; + } + } + + /* num_gpr = rnd_up(max_used_gpr + 1) */ + program->config->num_vgprs = (ctx.max_used_vgpr + 1 + 3) & ~3; + if (program->family == CHIP_TONGA || program->family == CHIP_ICELAND) { + assert(ctx.max_used_sgpr <= 93); + ctx.max_used_sgpr = 93; /* workaround hardware bug */ + } + program->config->num_sgprs = (ctx.max_used_sgpr + 1 + 2 + 7) & ~7; /* + 2 sgprs for vcc */ +} + +} diff --git a/src/amd/compiler/aco_scheduler.cpp b/src/amd/compiler/aco_scheduler.cpp new file mode 100644 index 00000000000..0cd67a979e0 --- /dev/null +++ b/src/amd/compiler/aco_scheduler.cpp @@ -0,0 +1,835 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" +#include +#include + +#include "vulkan/radv_shader.h" // for radv_nir_compiler_options +#include "amdgfxregs.h" + +#define SMEM_WINDOW_SIZE (350 - ctx.num_waves * 35) +#define VMEM_WINDOW_SIZE (1024 - ctx.num_waves * 64) +#define POS_EXP_WINDOW_SIZE 512 +#define SMEM_MAX_MOVES (80 - ctx.num_waves * 8) +#define VMEM_MAX_MOVES (128 - ctx.num_waves * 4) +#define POS_EXP_MAX_MOVES 512 + +namespace aco { + +struct sched_ctx { + std::vector depends_on; + std::vector RAR_dependencies; + RegisterDemand max_registers; + int16_t num_waves; + int16_t last_SMEM_stall; + int last_SMEM_dep_idx; +}; + +/* This scheduler is a simple bottom-up pass based on ideas from + * "A Novel Lightweight Instruction Scheduling Algorithm for Just-In-Time Compiler" + * from Xiaohua Shi and Peng Guo. + * The basic approach is to iterate over all instructions. When a memory instruction + * is encountered it tries to move independent instructions from above and below + * between the memory instruction and it's first user. + * The novelty is that this scheduler cares for the current register pressure: + * Instructions will only be moved if the register pressure won't exceed a certain bound. + */ + +template +void move_element(T& list, size_t idx, size_t before) { + if (idx < before) { + auto begin = std::next(list.begin(), idx); + auto end = std::next(list.begin(), before); + std::rotate(begin, begin + 1, end); + } else if (idx > before) { + auto begin = std::next(list.begin(), before); + auto end = std::next(list.begin(), idx + 1); + std::rotate(begin, end - 1, end); + } +} + +static RegisterDemand getLiveChanges(aco_ptr& instr) +{ + RegisterDemand changes; + for (const Definition& def : instr->definitions) { + if (!def.isTemp() || def.isKill()) + continue; + changes += def.getTemp(); + } + + for (const Operand& op : instr->operands) { + if (!op.isTemp() || !op.isFirstKill()) + continue; + changes -= op.getTemp(); + } + + return changes; +} + +static RegisterDemand getTempRegisters(aco_ptr& instr) +{ + RegisterDemand temp_registers; + for (const Definition& def : instr->definitions) { + if (!def.isTemp() || !def.isKill()) + continue; + temp_registers += def.getTemp(); + } + return temp_registers; +} + +static bool is_spill_reload(aco_ptr& instr) +{ + return instr->opcode == aco_opcode::p_spill || instr->opcode == aco_opcode::p_reload; +} + +bool can_move_instr(aco_ptr& instr, Instruction* current, int moving_interaction) +{ + /* don't move exports so that they stay closer together */ + if (instr->format == Format::EXP) + return false; + + /* handle barriers */ + + /* TODO: instead of stopping, maybe try to move the barriers and any + * instructions interacting with them instead? */ + if (instr->format != Format::PSEUDO_BARRIER) { + if (instr->opcode == aco_opcode::s_barrier) { + bool can_reorder = false; + switch (current->format) { + case Format::SMEM: + can_reorder = static_cast(current)->can_reorder; + break; + case Format::MUBUF: + can_reorder = static_cast(current)->can_reorder; + break; + case Format::MIMG: + can_reorder = static_cast(current)->can_reorder; + break; + default: + break; + } + return can_reorder && moving_interaction == barrier_none; + } else { + return true; + } + } + + int interaction = get_barrier_interaction(current); + interaction |= moving_interaction; + + switch (instr->opcode) { + case aco_opcode::p_memory_barrier_atomic: + return !(interaction & barrier_atomic); + /* For now, buffer and image barriers are treated the same. this is because of + * dEQP-VK.memory_model.message_passing.core11.u32.coherent.fence_fence.atomicwrite.device.payload_nonlocal.buffer.guard_nonlocal.image.comp + * which seems to use an image load to determine if the result of a buffer load is valid. So the ordering of the two loads is important. + * I /think/ we should probably eventually expand the meaning of a buffer barrier so that all buffer operations before it, must stay before it + * and that both image and buffer operations after it, must stay after it. We should also do the same for image barriers. + * Or perhaps the problem is that we don't have a combined barrier instruction for both buffers and images, but the CTS test expects us to? + * Either way, this solution should work. */ + case aco_opcode::p_memory_barrier_buffer: + case aco_opcode::p_memory_barrier_image: + return !(interaction & (barrier_image | barrier_buffer)); + case aco_opcode::p_memory_barrier_shared: + return !(interaction & barrier_shared); + case aco_opcode::p_memory_barrier_all: + return interaction == barrier_none; + default: + return false; + } +} + +bool can_reorder(Instruction* candidate, bool allow_smem) +{ + switch (candidate->format) { + case Format::SMEM: + return allow_smem || static_cast(candidate)->can_reorder; + case Format::MUBUF: + return static_cast(candidate)->can_reorder; + case Format::MIMG: + return static_cast(candidate)->can_reorder; + case Format::MTBUF: + return static_cast(candidate)->can_reorder; + case Format::FLAT: + case Format::GLOBAL: + case Format::SCRATCH: + return false; + default: + return true; + } +} + +void schedule_SMEM(sched_ctx& ctx, Block* block, + std::vector& register_demand, + Instruction* current, int idx) +{ + assert(idx != 0); + int window_size = SMEM_WINDOW_SIZE; + int max_moves = SMEM_MAX_MOVES; + int16_t k = 0; + bool can_reorder_cur = can_reorder(current, false); + + /* create the initial set of values which current depends on */ + std::fill(ctx.depends_on.begin(), ctx.depends_on.end(), false); + for (const Operand& op : current->operands) { + if (op.isTemp()) + ctx.depends_on[op.tempId()] = true; + } + + /* maintain how many registers remain free when moving instructions */ + RegisterDemand register_pressure = register_demand[idx]; + + /* first, check if we have instructions before current to move down */ + int insert_idx = idx + 1; + int moving_interaction = barrier_none; + bool moving_spill = false; + + for (int candidate_idx = idx - 1; k < max_moves && candidate_idx > (int) idx - window_size; candidate_idx--) { + assert(candidate_idx >= 0); + aco_ptr& candidate = block->instructions[candidate_idx]; + + /* break if we'd make the previous SMEM instruction stall */ + bool can_stall_prev_smem = idx <= ctx.last_SMEM_dep_idx && candidate_idx < ctx.last_SMEM_dep_idx; + if (can_stall_prev_smem && ctx.last_SMEM_stall >= 0) + break; + + /* break when encountering another MEM instruction, logical_start or barriers */ + if (!can_reorder(candidate.get(), false) && !can_reorder_cur) + break; + if (candidate->opcode == aco_opcode::p_logical_start) + break; + if (!can_move_instr(candidate, current, moving_interaction)) + break; + register_pressure.update(register_demand[candidate_idx]); + + /* if current depends on candidate, add additional dependencies and continue */ + bool can_move_down = true; + bool writes_exec = false; + for (const Definition& def : candidate->definitions) { + if (def.isTemp() && ctx.depends_on[def.tempId()]) + can_move_down = false; + if (def.isFixed() && def.physReg() == exec) + writes_exec = true; + } + if (writes_exec) + break; + + if (moving_spill && is_spill_reload(candidate)) + can_move_down = false; + if ((moving_interaction & barrier_shared) && candidate->format == Format::DS) + can_move_down = false; + moving_interaction |= get_barrier_interaction(candidate.get()); + moving_spill |= is_spill_reload(candidate); + if (!can_move_down) { + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.depends_on[op.tempId()] = true; + } + continue; + } + + bool register_pressure_unknown = false; + /* check if one of candidate's operands is killed by depending instruction */ + for (const Operand& op : candidate->operands) { + if (op.isTemp() && ctx.depends_on[op.tempId()]) { + // FIXME: account for difference in register pressure + register_pressure_unknown = true; + } + } + if (register_pressure_unknown) { + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.depends_on[op.tempId()] = true; + } + continue; + } + + /* check if register pressure is low enough: the diff is negative if register pressure is decreased */ + const RegisterDemand candidate_diff = getLiveChanges(candidate); + const RegisterDemand tempDemand = getTempRegisters(candidate); + if (RegisterDemand(register_pressure - candidate_diff).exceeds(ctx.max_registers)) + break; + const RegisterDemand tempDemand2 = getTempRegisters(block->instructions[insert_idx - 1]); + const RegisterDemand new_demand = register_demand[insert_idx - 1] - tempDemand2 + tempDemand; + if (new_demand.exceeds(ctx.max_registers)) + break; + // TODO: we might want to look further to find a sequence of instructions to move down which doesn't exceed reg pressure + + /* move the candidate below the memory load */ + move_element(block->instructions, candidate_idx, insert_idx); + + /* update register pressure */ + move_element(register_demand, candidate_idx, insert_idx); + for (int i = candidate_idx; i < insert_idx - 1; i++) { + register_demand[i] -= candidate_diff; + } + register_demand[insert_idx - 1] = new_demand; + register_pressure -= candidate_diff; + + if (candidate_idx < ctx.last_SMEM_dep_idx) + ctx.last_SMEM_stall++; + insert_idx--; + k++; + } + + /* create the initial set of values which depend on current */ + std::fill(ctx.depends_on.begin(), ctx.depends_on.end(), false); + std::fill(ctx.RAR_dependencies.begin(), ctx.RAR_dependencies.end(), false); + for (const Definition& def : current->definitions) { + if (def.isTemp()) + ctx.depends_on[def.tempId()] = true; + } + + /* find the first instruction depending on current or find another MEM */ + insert_idx = idx + 1; + moving_interaction = barrier_none; + moving_spill = false; + + bool found_dependency = false; + /* second, check if we have instructions after current to move up */ + for (int candidate_idx = idx + 1; k < max_moves && candidate_idx < (int) idx + window_size; candidate_idx++) { + assert(candidate_idx < (int) block->instructions.size()); + aco_ptr& candidate = block->instructions[candidate_idx]; + + if (candidate->opcode == aco_opcode::p_logical_end) + break; + if (!can_move_instr(candidate, current, moving_interaction)) + break; + + const bool writes_exec = std::any_of(candidate->definitions.begin(), candidate->definitions.end(), + [](const Definition& def) { return def.isFixed() && def.physReg() == exec;}); + if (writes_exec) + break; + + /* check if candidate depends on current */ + bool is_dependency = std::any_of(candidate->operands.begin(), candidate->operands.end(), + [&ctx](const Operand& op) { return op.isTemp() && ctx.depends_on[op.tempId()];}); + if (moving_spill && is_spill_reload(candidate)) + is_dependency = true; + if ((moving_interaction & barrier_shared) && candidate->format == Format::DS) + is_dependency = true; + moving_interaction |= get_barrier_interaction(candidate.get()); + moving_spill |= is_spill_reload(candidate); + if (is_dependency) { + for (const Definition& def : candidate->definitions) { + if (def.isTemp()) + ctx.depends_on[def.tempId()] = true; + } + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.RAR_dependencies[op.tempId()] = true; + } + if (!found_dependency) { + insert_idx = candidate_idx; + found_dependency = true; + /* init register pressure */ + register_pressure = register_demand[insert_idx - 1]; + } + } + + if (!can_reorder(candidate.get(), false) && !can_reorder_cur) + break; + + if (!found_dependency) { + k++; + continue; + } + + /* update register pressure */ + register_pressure.update(register_demand[candidate_idx - 1]); + + if (is_dependency) + continue; + assert(insert_idx != idx); + + // TODO: correctly calculate register pressure for this case + bool register_pressure_unknown = false; + /* check if candidate uses/kills an operand which is used by a dependency */ + for (const Operand& op : candidate->operands) { + if (op.isTemp() && ctx.RAR_dependencies[op.tempId()]) + register_pressure_unknown = true; + } + if (register_pressure_unknown) { + for (const Definition& def : candidate->definitions) { + if (def.isTemp()) + ctx.RAR_dependencies[def.tempId()] = true; + } + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.RAR_dependencies[op.tempId()] = true; + } + continue; + } + + /* check if register pressure is low enough: the diff is negative if register pressure is decreased */ + const RegisterDemand candidate_diff = getLiveChanges(candidate); + const RegisterDemand temp = getTempRegisters(candidate); + if (RegisterDemand(register_pressure + candidate_diff).exceeds(ctx.max_registers)) + break; + const RegisterDemand temp2 = getTempRegisters(block->instructions[insert_idx - 1]); + const RegisterDemand new_demand = register_demand[insert_idx - 1] - temp2 + candidate_diff + temp; + if (new_demand.exceeds(ctx.max_registers)) + break; + + /* move the candidate above the insert_idx */ + move_element(block->instructions, candidate_idx, insert_idx); + + /* update register pressure */ + move_element(register_demand, candidate_idx, insert_idx); + for (int i = insert_idx + 1; i <= candidate_idx; i++) { + register_demand[i] += candidate_diff; + } + register_demand[insert_idx] = new_demand; + register_pressure += candidate_diff; + insert_idx++; + k++; + } + + ctx.last_SMEM_dep_idx = found_dependency ? insert_idx : 0; + ctx.last_SMEM_stall = 10 - ctx.num_waves - k; +} + +void schedule_VMEM(sched_ctx& ctx, Block* block, + std::vector& register_demand, + Instruction* current, int idx) +{ + assert(idx != 0); + int window_size = VMEM_WINDOW_SIZE; + int max_moves = VMEM_MAX_MOVES; + int16_t k = 0; + bool can_reorder_cur = can_reorder(current, false); + + /* create the initial set of values which current depends on */ + std::fill(ctx.depends_on.begin(), ctx.depends_on.end(), false); + for (const Operand& op : current->operands) { + if (op.isTemp()) + ctx.depends_on[op.tempId()] = true; + } + + /* maintain how many registers remain free when moving instructions */ + RegisterDemand register_pressure = register_demand[idx]; + + /* first, check if we have instructions before current to move down */ + int insert_idx = idx + 1; + int moving_interaction = barrier_none; + bool moving_spill = false; + + for (int candidate_idx = idx - 1; k < max_moves && candidate_idx > (int) idx - window_size; candidate_idx--) { + assert(candidate_idx >= 0); + aco_ptr& candidate = block->instructions[candidate_idx]; + + /* break when encountering another VMEM instruction, logical_start or barriers */ + if (!can_reorder(candidate.get(), true) && !can_reorder_cur) + break; + if (candidate->opcode == aco_opcode::p_logical_start) + break; + if (!can_move_instr(candidate, current, moving_interaction)) + break; + + /* break if we'd make the previous SMEM instruction stall */ + bool can_stall_prev_smem = idx <= ctx.last_SMEM_dep_idx && candidate_idx < ctx.last_SMEM_dep_idx; + if (can_stall_prev_smem && ctx.last_SMEM_stall >= 0) + break; + register_pressure.update(register_demand[candidate_idx]); + + /* if current depends on candidate, add additional dependencies and continue */ + bool can_move_down = true; + bool writes_exec = false; + for (const Definition& def : candidate->definitions) { + if (def.isTemp() && ctx.depends_on[def.tempId()]) + can_move_down = false; + if (def.isFixed() && def.physReg() == exec) + writes_exec = true; + } + if (writes_exec) + break; + + if (moving_spill && is_spill_reload(candidate)) + can_move_down = false; + if ((moving_interaction & barrier_shared) && candidate->format == Format::DS) + can_move_down = false; + moving_interaction |= get_barrier_interaction(candidate.get()); + moving_spill |= is_spill_reload(candidate); + if (!can_move_down) { + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.depends_on[op.tempId()] = true; + } + continue; + } + + bool register_pressure_unknown = false; + /* check if one of candidate's operands is killed by depending instruction */ + for (const Operand& op : candidate->operands) { + if (op.isTemp() && ctx.depends_on[op.tempId()]) { + // FIXME: account for difference in register pressure + register_pressure_unknown = true; + } + } + if (register_pressure_unknown) { + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.depends_on[op.tempId()] = true; + } + continue; + } + + /* check if register pressure is low enough: the diff is negative if register pressure is decreased */ + const RegisterDemand candidate_diff = getLiveChanges(candidate); + const RegisterDemand temp = getTempRegisters(candidate);; + if (RegisterDemand(register_pressure - candidate_diff).exceeds(ctx.max_registers)) + break; + const RegisterDemand temp2 = getTempRegisters(block->instructions[insert_idx - 1]); + const RegisterDemand new_demand = register_demand[insert_idx - 1] - temp2 + temp; + if (new_demand.exceeds(ctx.max_registers)) + break; + // TODO: we might want to look further to find a sequence of instructions to move down which doesn't exceed reg pressure + + /* move the candidate below the memory load */ + move_element(block->instructions, candidate_idx, insert_idx); + + /* update register pressure */ + move_element(register_demand, candidate_idx, insert_idx); + for (int i = candidate_idx; i < insert_idx - 1; i++) { + register_demand[i] -= candidate_diff; + } + register_demand[insert_idx - 1] = new_demand; + register_pressure -= candidate_diff; + insert_idx--; + k++; + if (candidate_idx < ctx.last_SMEM_dep_idx) + ctx.last_SMEM_stall++; + } + + /* create the initial set of values which depend on current */ + std::fill(ctx.depends_on.begin(), ctx.depends_on.end(), false); + std::fill(ctx.RAR_dependencies.begin(), ctx.RAR_dependencies.end(), false); + for (const Definition& def : current->definitions) { + if (def.isTemp()) + ctx.depends_on[def.tempId()] = true; + } + + /* find the first instruction depending on current or find another VMEM */ + insert_idx = idx; + moving_interaction = barrier_none; + moving_spill = false; + + bool found_dependency = false; + /* second, check if we have instructions after current to move up */ + for (int candidate_idx = idx + 1; k < max_moves && candidate_idx < (int) idx + window_size; candidate_idx++) { + assert(candidate_idx < (int) block->instructions.size()); + aco_ptr& candidate = block->instructions[candidate_idx]; + + if (candidate->opcode == aco_opcode::p_logical_end) + break; + if (!can_move_instr(candidate, current, moving_interaction)) + break; + + const bool writes_exec = std::any_of(candidate->definitions.begin(), candidate->definitions.end(), + [](const Definition& def) {return def.isFixed() && def.physReg() == exec; }); + if (writes_exec) + break; + + /* check if candidate depends on current */ + bool is_dependency = !can_reorder(candidate.get(), true) && !can_reorder_cur; + for (const Operand& op : candidate->operands) { + if (op.isTemp() && ctx.depends_on[op.tempId()]) { + is_dependency = true; + break; + } + } + if (moving_spill && is_spill_reload(candidate)) + is_dependency = true; + if ((moving_interaction & barrier_shared) && candidate->format == Format::DS) + is_dependency = true; + moving_interaction |= get_barrier_interaction(candidate.get()); + moving_spill |= is_spill_reload(candidate); + if (is_dependency) { + for (const Definition& def : candidate->definitions) { + if (def.isTemp()) + ctx.depends_on[def.tempId()] = true; + } + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.RAR_dependencies[op.tempId()] = true; + } + if (!found_dependency) { + insert_idx = candidate_idx; + found_dependency = true; + /* init register pressure */ + register_pressure = register_demand[insert_idx - 1]; + continue; + } + } + + /* update register pressure */ + register_pressure.update(register_demand[candidate_idx - 1]); + + if (is_dependency || !found_dependency) + continue; + assert(insert_idx != idx); + + bool register_pressure_unknown = false; + /* check if candidate uses/kills an operand which is used by a dependency */ + for (const Operand& op : candidate->operands) { + if (op.isTemp() && ctx.RAR_dependencies[op.tempId()]) + register_pressure_unknown = true; + } + if (register_pressure_unknown) { + for (const Definition& def : candidate->definitions) { + if (def.isTemp()) + ctx.RAR_dependencies[def.tempId()] = true; + } + for (const Operand& op : candidate->operands) { + if (op.isTemp()) + ctx.RAR_dependencies[op.tempId()] = true; + } + continue; + } + + /* check if register pressure is low enough: the diff is negative if register pressure is decreased */ + const RegisterDemand candidate_diff = getLiveChanges(candidate); + const RegisterDemand temp = getTempRegisters(candidate); + if (RegisterDemand(register_pressure + candidate_diff).exceeds(ctx.max_registers)) + break; + const RegisterDemand temp2 = getTempRegisters(block->instructions[insert_idx - 1]); + const RegisterDemand new_demand = register_demand[insert_idx - 1] - temp2 + candidate_diff + temp; + if (new_demand.exceeds(ctx.max_registers)) + break; + + /* move the candidate above the insert_idx */ + move_element(block->instructions, candidate_idx, insert_idx); + + /* update register pressure */ + move_element(register_demand, candidate_idx, insert_idx); + for (int i = insert_idx + 1; i <= candidate_idx; i++) { + register_demand[i] += candidate_diff; + } + register_demand[insert_idx] = new_demand; + register_pressure += candidate_diff; + insert_idx++; + k++; + } +} + +void schedule_position_export(sched_ctx& ctx, Block* block, + std::vector& register_demand, + Instruction* current, int idx) +{ + assert(idx != 0); + int window_size = POS_EXP_WINDOW_SIZE; + int max_moves = POS_EXP_MAX_MOVES; + int16_t k = 0; + + /* create the initial set of values which current depends on */ + std::fill(ctx.depends_on.begin(), ctx.depends_on.end(), false); + for (unsigned i = 0; i < current->operands.size(); i++) { + if (current->operands[i].isTemp()) + ctx.depends_on[current->operands[i].tempId()] = true; + } + + /* maintain how many registers remain free when moving instructions */ + RegisterDemand register_pressure = register_demand[idx]; + + /* first, check if we have instructions before current to move down */ + int insert_idx = idx + 1; + int moving_interaction = barrier_none; + bool moving_spill = false; + + for (int candidate_idx = idx - 1; k < max_moves && candidate_idx > (int) idx - window_size; candidate_idx--) { + assert(candidate_idx >= 0); + aco_ptr& candidate = block->instructions[candidate_idx]; + + /* break when encountering logical_start or barriers */ + if (candidate->opcode == aco_opcode::p_logical_start) + break; + if (candidate->isVMEM() || candidate->format == Format::SMEM) + break; + if (!can_move_instr(candidate, current, moving_interaction)) + break; + + register_pressure.update(register_demand[candidate_idx]); + + /* if current depends on candidate, add additional dependencies and continue */ + bool can_move_down = true; + bool writes_exec = false; + for (unsigned i = 0; i < candidate->definitions.size(); i++) { + if (candidate->definitions[i].isTemp() && ctx.depends_on[candidate->definitions[i].tempId()]) + can_move_down = false; + if (candidate->definitions[i].isFixed() && candidate->definitions[i].physReg() == exec) + writes_exec = true; + } + if (writes_exec) + break; + + if (moving_spill && is_spill_reload(candidate)) + can_move_down = false; + if ((moving_interaction & barrier_shared) && candidate->format == Format::DS) + can_move_down = false; + moving_interaction |= get_barrier_interaction(candidate.get()); + moving_spill |= is_spill_reload(candidate); + if (!can_move_down) { + for (unsigned i = 0; i < candidate->operands.size(); i++) { + if (candidate->operands[i].isTemp()) + ctx.depends_on[candidate->operands[i].tempId()] = true; + } + continue; + } + + bool register_pressure_unknown = false; + /* check if one of candidate's operands is killed by depending instruction */ + for (unsigned i = 0; i < candidate->operands.size(); i++) { + if (candidate->operands[i].isTemp() && ctx.depends_on[candidate->operands[i].tempId()]) { + // FIXME: account for difference in register pressure + register_pressure_unknown = true; + } + } + if (register_pressure_unknown) { + for (unsigned i = 0; i < candidate->operands.size(); i++) { + if (candidate->operands[i].isTemp()) + ctx.depends_on[candidate->operands[i].tempId()] = true; + } + continue; + } + + /* check if register pressure is low enough: the diff is negative if register pressure is decreased */ + const RegisterDemand candidate_diff = getLiveChanges(candidate); + const RegisterDemand temp = getTempRegisters(candidate);; + if (RegisterDemand(register_pressure - candidate_diff).exceeds(ctx.max_registers)) + break; + const RegisterDemand temp2 = getTempRegisters(block->instructions[insert_idx - 1]); + const RegisterDemand new_demand = register_demand[insert_idx - 1] - temp2 + temp; + if (new_demand.exceeds(ctx.max_registers)) + break; + // TODO: we might want to look further to find a sequence of instructions to move down which doesn't exceed reg pressure + + /* move the candidate below the export */ + move_element(block->instructions, candidate_idx, insert_idx); + + /* update register pressure */ + move_element(register_demand, candidate_idx, insert_idx); + for (int i = candidate_idx; i < insert_idx - 1; i++) { + register_demand[i] -= candidate_diff; + } + register_demand[insert_idx - 1] = new_demand; + register_pressure -= candidate_diff; + insert_idx--; + k++; + } +} + +void schedule_block(sched_ctx& ctx, Program *program, Block* block, live& live_vars) +{ + ctx.last_SMEM_dep_idx = 0; + ctx.last_SMEM_stall = INT16_MIN; + + /* go through all instructions and find memory loads */ + for (unsigned idx = 0; idx < block->instructions.size(); idx++) { + Instruction* current = block->instructions[idx].get(); + + if (current->definitions.empty()) + continue; + + if (current->isVMEM()) + schedule_VMEM(ctx, block, live_vars.register_demand[block->index], current, idx); + if (current->format == Format::SMEM) + schedule_SMEM(ctx, block, live_vars.register_demand[block->index], current, idx); + } + + if ((program->stage & hw_vs) && block->index == program->blocks.size() - 1) { + /* Try to move position exports as far up as possible, to reduce register + * usage and because ISA reference guides say so. */ + for (unsigned idx = 0; idx < block->instructions.size(); idx++) { + Instruction* current = block->instructions[idx].get(); + + if (current->format == Format::EXP) { + unsigned target = static_cast(current)->dest; + if (target >= V_008DFC_SQ_EXP_POS && target < V_008DFC_SQ_EXP_PARAM) + schedule_position_export(ctx, block, live_vars.register_demand[block->index], current, idx); + } + } + } + + /* resummarize the block's register demand */ + block->register_demand = RegisterDemand(); + for (unsigned idx = 0; idx < block->instructions.size(); idx++) { + block->register_demand.update(live_vars.register_demand[block->index][idx]); + } +} + + +void schedule_program(Program *program, live& live_vars) +{ + sched_ctx ctx; + ctx.depends_on.resize(program->peekAllocationId()); + ctx.RAR_dependencies.resize(program->peekAllocationId()); + /* Allowing the scheduler to reduce the number of waves to as low as 5 + * improves performance of Thrones of Britannia significantly and doesn't + * seem to hurt anything else. */ + //TODO: maybe use some sort of heuristic instead + //TODO: this also increases window-size/max-moves? did I realize that at the time? + ctx.num_waves = std::min(program->num_waves, 5); + assert(ctx.num_waves); + uint16_t total_sgpr_regs = program->chip_class >= GFX8 ? 800 : 512; + uint16_t max_addressible_sgpr = program->sgpr_limit; + ctx.max_registers = { int16_t(((256 / ctx.num_waves) & ~3) - 2), std::min(((total_sgpr_regs / ctx.num_waves) & ~7) - 2, max_addressible_sgpr)}; + + for (Block& block : program->blocks) + schedule_block(ctx, program, &block, live_vars); + + /* update max_reg_demand and num_waves */ + RegisterDemand new_demand; + for (Block& block : program->blocks) { + new_demand.update(block.register_demand); + } + update_vgpr_sgpr_demand(program, new_demand); + + /* if enabled, this code asserts that register_demand is updated correctly */ + #if 0 + int prev_num_waves = program->num_waves; + const RegisterDemand prev_max_demand = program->max_reg_demand; + + std::vector demands(program->blocks.size()); + for (unsigned j = 0; j < program->blocks.size(); j++) { + demands[j] = program->blocks[j].register_demand; + } + + struct radv_nir_compiler_options options; + options.chip_class = program->chip_class; + live live_vars2 = aco::live_var_analysis(program, &options); + + for (unsigned j = 0; j < program->blocks.size(); j++) { + Block &b = program->blocks[j]; + for (unsigned i = 0; i < b.instructions.size(); i++) + assert(live_vars.register_demand[b.index][i] == live_vars2.register_demand[b.index][i]); + assert(b.register_demand == demands[j]); + } + + assert(program->max_reg_demand == prev_max_demand); + assert(program->num_waves == prev_num_waves); + #endif +} + +} diff --git a/src/amd/compiler/aco_spill.cpp b/src/amd/compiler/aco_spill.cpp new file mode 100644 index 00000000000..92a23bb355c --- /dev/null +++ b/src/amd/compiler/aco_spill.cpp @@ -0,0 +1,1630 @@ +/* + * Copyright © 2018 Valve Corporation + * Copyright © 2018 Google + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" +#include +#include +#include "vulkan/radv_shader.h" + + +/* + * Implements the spilling algorithm on SSA-form from + * "Register Spilling and Live-Range Splitting for SSA-Form Programs" + * by Matthias Braun and Sebastian Hack. + */ + +namespace aco { + +namespace { + +struct remat_info { + Instruction *instr; +}; + +struct spill_ctx { + RegisterDemand target_pressure; + Program* program; + std::vector> register_demand; + std::vector> renames; + std::vector> spills_entry; + std::vector> spills_exit; + std::vector processed; + std::stack loop_header; + std::vector>> next_use_distances_start; + std::vector>> next_use_distances_end; + std::vector>> interferences; + std::vector> affinities; + std::vector is_reloaded; + std::map remat; + std::map remat_used; + + spill_ctx(const RegisterDemand target_pressure, Program* program, + std::vector> register_demand) + : target_pressure(target_pressure), program(program), + register_demand(register_demand), renames(program->blocks.size()), + spills_entry(program->blocks.size()), spills_exit(program->blocks.size()), + processed(program->blocks.size(), false) {} + + uint32_t allocate_spill_id(RegClass rc) + { + interferences.emplace_back(rc, std::set()); + is_reloaded.push_back(false); + return next_spill_id++; + } + + uint32_t next_spill_id = 0; +}; + +int32_t get_dominator(int idx_a, int idx_b, Program* program, bool is_linear) +{ + + if (idx_a == -1) + return idx_b; + if (idx_b == -1) + return idx_a; + if (is_linear) { + while (idx_a != idx_b) { + if (idx_a > idx_b) + idx_a = program->blocks[idx_a].linear_idom; + else + idx_b = program->blocks[idx_b].linear_idom; + } + } else { + while (idx_a != idx_b) { + if (idx_a > idx_b) + idx_a = program->blocks[idx_a].logical_idom; + else + idx_b = program->blocks[idx_b].logical_idom; + } + } + assert(idx_a != -1); + return idx_a; +} + +void next_uses_per_block(spill_ctx& ctx, unsigned block_idx, std::set& worklist) +{ + Block* block = &ctx.program->blocks[block_idx]; + std::map> next_uses = ctx.next_use_distances_end[block_idx]; + + /* to compute the next use distance at the beginning of the block, we have to add the block's size */ + for (std::map>::iterator it = next_uses.begin(); it != next_uses.end();) { + it->second.second = it->second.second + block->instructions.size(); + + /* remove the live out exec mask as we really don't want to spill it */ + if (it->first == block->live_out_exec) + it = next_uses.erase(it); + else + ++it; + } + + int idx = block->instructions.size() - 1; + while (idx >= 0) { + aco_ptr& instr = block->instructions[idx]; + + if (instr->opcode == aco_opcode::p_linear_phi || + instr->opcode == aco_opcode::p_phi) + break; + + for (const Definition& def : instr->definitions) { + if (def.isTemp()) + next_uses.erase(def.getTemp()); + } + + for (const Operand& op : instr->operands) { + /* omit exec mask */ + if (op.isFixed() && op.physReg() == exec) + continue; + if (op.isTemp()) + next_uses[op.getTemp()] = {block_idx, idx}; + } + idx--; + } + + assert(block_idx != 0 || next_uses.empty()); + ctx.next_use_distances_start[block_idx] = next_uses; + while (idx >= 0) { + aco_ptr& instr = block->instructions[idx]; + assert(instr->opcode == aco_opcode::p_linear_phi || instr->opcode == aco_opcode::p_phi); + + for (unsigned i = 0; i < instr->operands.size(); i++) { + unsigned pred_idx = instr->opcode == aco_opcode::p_phi ? + block->logical_preds[i] : + block->linear_preds[i]; + if (instr->operands[i].isTemp()) { + if (ctx.next_use_distances_end[pred_idx].find(instr->operands[i].getTemp()) == ctx.next_use_distances_end[pred_idx].end() || + ctx.next_use_distances_end[pred_idx][instr->operands[i].getTemp()] != std::pair{block_idx, 0}) + worklist.insert(pred_idx); + ctx.next_use_distances_end[pred_idx][instr->operands[i].getTemp()] = {block_idx, 0}; + } + } + next_uses.erase(instr->definitions[0].getTemp()); + idx--; + } + + /* all remaining live vars must be live-out at the predecessors */ + for (std::pair> pair : next_uses) { + Temp temp = pair.first; + uint32_t distance = pair.second.second; + uint32_t dom = pair.second.first; + std::vector& preds = temp.is_linear() ? block->linear_preds : block->logical_preds; + for (unsigned pred_idx : preds) { + if (ctx.program->blocks[pred_idx].loop_nest_depth > block->loop_nest_depth) + distance += 0xFFFF; + if (ctx.next_use_distances_end[pred_idx].find(temp) != ctx.next_use_distances_end[pred_idx].end()) { + dom = get_dominator(dom, ctx.next_use_distances_end[pred_idx][temp].first, ctx.program, temp.is_linear()); + distance = std::min(ctx.next_use_distances_end[pred_idx][temp].second, distance); + } + if (ctx.next_use_distances_end[pred_idx][temp] != std::pair{dom, distance}) + worklist.insert(pred_idx); + ctx.next_use_distances_end[pred_idx][temp] = {dom, distance}; + } + } + +} + +void compute_global_next_uses(spill_ctx& ctx, std::vector>& live_out) +{ + ctx.next_use_distances_start.resize(ctx.program->blocks.size()); + ctx.next_use_distances_end.resize(ctx.program->blocks.size()); + std::set worklist; + for (Block& block : ctx.program->blocks) + worklist.insert(block.index); + + while (!worklist.empty()) { + std::set::reverse_iterator b_it = worklist.rbegin(); + unsigned block_idx = *b_it; + worklist.erase(block_idx); + next_uses_per_block(ctx, block_idx, worklist); + } +} + +bool should_rematerialize(aco_ptr& instr) +{ + /* TODO: rematerialization is only supported for VOP1, SOP1 and PSEUDO */ + if (instr->format != Format::VOP1 && instr->format != Format::SOP1 && instr->format != Format::PSEUDO) + return false; + /* TODO: pseudo-instruction rematerialization is only supported for p_create_vector */ + if (instr->format == Format::PSEUDO && instr->opcode != aco_opcode::p_create_vector) + return false; + + for (const Operand& op : instr->operands) { + /* TODO: rematerialization using temporaries isn't yet supported */ + if (op.isTemp()) + return false; + } + + /* TODO: rematerialization with multiple definitions isn't yet supported */ + if (instr->definitions.size() > 1) + return false; + + return true; +} + +aco_ptr do_reload(spill_ctx& ctx, Temp tmp, Temp new_name, uint32_t spill_id) +{ + std::map::iterator remat = ctx.remat.find(tmp); + if (remat != ctx.remat.end()) { + Instruction *instr = remat->second.instr; + assert((instr->format == Format::VOP1 || instr->format == Format::SOP1 || instr->format == Format::PSEUDO) && "unsupported"); + assert((instr->format != Format::PSEUDO || instr->opcode == aco_opcode::p_create_vector) && "unsupported"); + assert(instr->definitions.size() == 1 && "unsupported"); + + aco_ptr res; + if (instr->format == Format::VOP1) { + res.reset(create_instruction(instr->opcode, instr->format, instr->operands.size(), instr->definitions.size())); + } else if (instr->format == Format::SOP1) { + res.reset(create_instruction(instr->opcode, instr->format, instr->operands.size(), instr->definitions.size())); + } else if (instr->format == Format::PSEUDO) { + res.reset(create_instruction(instr->opcode, instr->format, instr->operands.size(), instr->definitions.size())); + } + for (unsigned i = 0; i < instr->operands.size(); i++) { + res->operands[i] = instr->operands[i]; + if (instr->operands[i].isTemp()) { + assert(false && "unsupported"); + if (ctx.remat.count(instr->operands[i].getTemp())) + ctx.remat_used[ctx.remat[instr->operands[i].getTemp()].instr] = true; + } + } + res->definitions[0] = Definition(new_name); + return res; + } else { + aco_ptr reload{create_instruction(aco_opcode::p_reload, Format::PSEUDO, 1, 1)}; + reload->operands[0] = Operand(spill_id); + reload->definitions[0] = Definition(new_name); + ctx.is_reloaded[spill_id] = true; + return reload; + } +} + +void get_rematerialize_info(spill_ctx& ctx) +{ + for (Block& block : ctx.program->blocks) { + bool logical = false; + for (aco_ptr& instr : block.instructions) { + if (instr->opcode == aco_opcode::p_logical_start) + logical = true; + else if (instr->opcode == aco_opcode::p_logical_end) + logical = false; + if (logical && should_rematerialize(instr)) { + for (const Definition& def : instr->definitions) { + if (def.isTemp()) { + ctx.remat[def.getTemp()] = (remat_info){instr.get()}; + ctx.remat_used[instr.get()] = false; + } + } + } + } + } +} + +std::vector> local_next_uses(spill_ctx& ctx, Block* block) +{ + std::vector> local_next_uses(block->instructions.size()); + + std::map next_uses; + for (std::pair> pair : ctx.next_use_distances_end[block->index]) { + /* omit live out exec mask */ + if (pair.first == block->live_out_exec) + continue; + + next_uses[pair.first] = pair.second.second + block->instructions.size(); + } + + for (int idx = block->instructions.size() - 1; idx >= 0; idx--) { + aco_ptr& instr = block->instructions[idx]; + if (!instr) + break; + if (instr->opcode == aco_opcode::p_phi || instr->opcode == aco_opcode::p_linear_phi) + break; + + for (const Operand& op : instr->operands) { + if (op.isFixed() && op.physReg() == exec) + continue; + if (op.isTemp()) + next_uses[op.getTemp()] = idx; + } + for (const Definition& def : instr->definitions) { + if (def.isTemp()) + next_uses.erase(def.getTemp()); + } + local_next_uses[idx] = next_uses; + } + return local_next_uses; +} + + +RegisterDemand init_live_in_vars(spill_ctx& ctx, Block* block, unsigned block_idx) +{ + RegisterDemand spilled_registers; + + /* first block, nothing was spilled before */ + if (block_idx == 0) + return {0, 0}; + + /* loop header block */ + if (block->loop_nest_depth > ctx.program->blocks[block_idx - 1].loop_nest_depth) { + assert(block->linear_preds[0] == block_idx - 1); + assert(block->logical_preds[0] == block_idx - 1); + + /* create new loop_info */ + ctx.loop_header.emplace(block); + + /* check how many live-through variables should be spilled */ + RegisterDemand new_demand; + unsigned i = block_idx; + while (ctx.program->blocks[i].loop_nest_depth >= block->loop_nest_depth) { + assert(ctx.program->blocks.size() > i); + new_demand.update(ctx.program->blocks[i].register_demand); + i++; + } + unsigned loop_end = i; + + /* select live-through vgpr variables */ + while (new_demand.vgpr - spilled_registers.vgpr > ctx.target_pressure.vgpr) { + unsigned distance = 0; + Temp to_spill; + for (std::pair> pair : ctx.next_use_distances_end[block_idx - 1]) { + if (pair.first.type() == RegType::vgpr && + pair.second.first >= loop_end && + pair.second.second > distance && + ctx.spills_entry[block_idx].find(pair.first) == ctx.spills_entry[block_idx].end()) { + to_spill = pair.first; + distance = pair.second.second; + } + } + if (distance == 0) + break; + + uint32_t spill_id; + if (ctx.spills_exit[block_idx - 1].find(to_spill) == ctx.spills_exit[block_idx - 1].end()) { + spill_id = ctx.allocate_spill_id(to_spill.regClass()); + } else { + spill_id = ctx.spills_exit[block_idx - 1][to_spill]; + } + + ctx.spills_entry[block_idx][to_spill] = spill_id; + spilled_registers.vgpr += to_spill.size(); + } + + /* select live-through sgpr variables */ + while (new_demand.sgpr - spilled_registers.sgpr > ctx.target_pressure.sgpr) { + unsigned distance = 0; + Temp to_spill; + for (std::pair> pair : ctx.next_use_distances_end[block_idx - 1]) { + if (pair.first.type() == RegType::sgpr && + pair.second.first >= loop_end && + pair.second.second > distance && + ctx.spills_entry[block_idx].find(pair.first) == ctx.spills_entry[block_idx].end()) { + to_spill = pair.first; + distance = pair.second.second; + } + } + if (distance == 0) + break; + + uint32_t spill_id; + if (ctx.spills_exit[block_idx - 1].find(to_spill) == ctx.spills_exit[block_idx - 1].end()) { + spill_id = ctx.allocate_spill_id(to_spill.regClass()); + } else { + spill_id = ctx.spills_exit[block_idx - 1][to_spill]; + } + + ctx.spills_entry[block_idx][to_spill] = spill_id; + spilled_registers.sgpr += to_spill.size(); + } + + + + /* shortcut */ + if (!RegisterDemand(new_demand - spilled_registers).exceeds(ctx.target_pressure)) + return spilled_registers; + + /* if reg pressure is too high at beginning of loop, add variables with furthest use */ + unsigned idx = 0; + while (block->instructions[idx]->opcode == aco_opcode::p_phi || block->instructions[idx]->opcode == aco_opcode::p_linear_phi) + idx++; + + assert(idx != 0 && "loop without phis: TODO"); + idx--; + RegisterDemand reg_pressure = ctx.register_demand[block_idx][idx] - spilled_registers; + while (reg_pressure.sgpr > ctx.target_pressure.sgpr) { + unsigned distance = 0; + Temp to_spill; + for (std::pair> pair : ctx.next_use_distances_start[block_idx]) { + if (pair.first.type() == RegType::sgpr && + pair.second.second > distance && + ctx.spills_entry[block_idx].find(pair.first) == ctx.spills_entry[block_idx].end()) { + to_spill = pair.first; + distance = pair.second.second; + } + } + assert(distance != 0); + + ctx.spills_entry[block_idx][to_spill] = ctx.allocate_spill_id(to_spill.regClass()); + spilled_registers.sgpr += to_spill.size(); + reg_pressure.sgpr -= to_spill.size(); + } + while (reg_pressure.vgpr > ctx.target_pressure.vgpr) { + unsigned distance = 0; + Temp to_spill; + for (std::pair> pair : ctx.next_use_distances_start[block_idx]) { + if (pair.first.type() == RegType::vgpr && + pair.second.second > distance && + ctx.spills_entry[block_idx].find(pair.first) == ctx.spills_entry[block_idx].end()) { + to_spill = pair.first; + distance = pair.second.second; + } + } + assert(distance != 0); + ctx.spills_entry[block_idx][to_spill] = ctx.allocate_spill_id(to_spill.regClass()); + spilled_registers.vgpr += to_spill.size(); + reg_pressure.vgpr -= to_spill.size(); + } + + return spilled_registers; + } + + /* branch block */ + if (block->linear_preds.size() == 1) { + /* keep variables spilled if they are alive and not used in the current block */ + unsigned pred_idx = block->linear_preds[0]; + for (std::pair pair : ctx.spills_exit[pred_idx]) { + if (pair.first.type() == RegType::sgpr && + ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() && + ctx.next_use_distances_start[block_idx][pair.first].second > block_idx) { + ctx.spills_entry[block_idx].insert(pair); + spilled_registers.sgpr += pair.first.size(); + } + } + if (block->logical_preds.size() == 1) { + pred_idx = block->logical_preds[0]; + for (std::pair pair : ctx.spills_exit[pred_idx]) { + if (pair.first.type() == RegType::vgpr && + ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() && + ctx.next_use_distances_end[pred_idx][pair.first].second > block_idx) { + ctx.spills_entry[block_idx].insert(pair); + spilled_registers.vgpr += pair.first.size(); + } + } + } + + /* if register demand is still too high, we just keep all spilled live vars and process the block */ + if (block->register_demand.sgpr - spilled_registers.sgpr > ctx.target_pressure.sgpr) { + pred_idx = block->linear_preds[0]; + for (std::pair pair : ctx.spills_exit[pred_idx]) { + if (pair.first.type() == RegType::sgpr && + ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() && + ctx.spills_entry[block_idx].insert(pair).second) { + spilled_registers.sgpr += pair.first.size(); + } + } + } + if (block->register_demand.vgpr - spilled_registers.vgpr > ctx.target_pressure.vgpr && block->logical_preds.size() == 1) { + pred_idx = block->logical_preds[0]; + for (std::pair pair : ctx.spills_exit[pred_idx]) { + if (pair.first.type() == RegType::vgpr && + ctx.next_use_distances_start[block_idx].find(pair.first) != ctx.next_use_distances_start[block_idx].end() && + ctx.spills_entry[block_idx].insert(pair).second) { + spilled_registers.vgpr += pair.first.size(); + } + } + } + + return spilled_registers; + } + + /* else: merge block */ + std::set partial_spills; + + /* keep variables spilled on all incoming paths */ + for (std::pair> pair : ctx.next_use_distances_start[block_idx]) { + std::vector& preds = pair.first.type() == RegType::vgpr ? block->logical_preds : block->linear_preds; + /* If it can be rematerialized, keep the variable spilled if all predecessors do not reload it. + * Otherwise, if any predecessor reloads it, ensure it's reloaded on all other predecessors. + * The idea is that it's better in practice to rematerialize redundantly than to create lots of phis. */ + /* TODO: test this idea with more than Dawn of War III shaders (the current pipeline-db doesn't seem to exercise this path much) */ + bool remat = ctx.remat.count(pair.first); + bool spill = !remat; + uint32_t spill_id = 0; + for (unsigned pred_idx : preds) { + /* variable is not even live at the predecessor: probably from a phi */ + if (ctx.next_use_distances_end[pred_idx].find(pair.first) == ctx.next_use_distances_end[pred_idx].end()) { + spill = false; + break; + } + if (ctx.spills_exit[pred_idx].find(pair.first) == ctx.spills_exit[pred_idx].end()) { + if (!remat) + spill = false; + } else { + partial_spills.insert(pair.first); + /* it might be that on one incoming path, the variable has a different spill_id, but add_couple_code() will take care of that. */ + spill_id = ctx.spills_exit[pred_idx][pair.first]; + if (remat) + spill = true; + } + } + if (spill) { + ctx.spills_entry[block_idx][pair.first] = spill_id; + partial_spills.erase(pair.first); + spilled_registers += pair.first; + } + } + + /* same for phis */ + unsigned idx = 0; + while (block->instructions[idx]->opcode == aco_opcode::p_linear_phi || + block->instructions[idx]->opcode == aco_opcode::p_phi) { + aco_ptr& phi = block->instructions[idx]; + std::vector& preds = phi->opcode == aco_opcode::p_phi ? block->logical_preds : block->linear_preds; + bool spill = true; + + for (unsigned i = 0; i < phi->operands.size(); i++) { + if (!phi->operands[i].isTemp()) + spill = false; + else if (ctx.spills_exit[preds[i]].find(phi->operands[i].getTemp()) == ctx.spills_exit[preds[i]].end()) + spill = false; + else + partial_spills.insert(phi->definitions[0].getTemp()); + } + if (spill) { + ctx.spills_entry[block_idx][phi->definitions[0].getTemp()] = ctx.allocate_spill_id(phi->definitions[0].regClass()); + partial_spills.erase(phi->definitions[0].getTemp()); + spilled_registers += phi->definitions[0].getTemp(); + } + + idx++; + } + + /* if reg pressure at first instruction is still too high, add partially spilled variables */ + RegisterDemand reg_pressure; + if (idx == 0) { + for (const Definition& def : block->instructions[idx]->definitions) { + if (def.isTemp()) { + reg_pressure -= def.getTemp(); + } + } + for (const Operand& op : block->instructions[idx]->operands) { + if (op.isTemp() && op.isFirstKill()) { + reg_pressure += op.getTemp(); + } + } + } else { + idx--; + } + reg_pressure += ctx.register_demand[block_idx][idx] - spilled_registers; + + while (reg_pressure.sgpr > ctx.target_pressure.sgpr) { + assert(!partial_spills.empty()); + + std::set::iterator it = partial_spills.begin(); + Temp to_spill = *it; + unsigned distance = ctx.next_use_distances_start[block_idx][*it].second; + while (it != partial_spills.end()) { + assert(ctx.spills_entry[block_idx].find(*it) == ctx.spills_entry[block_idx].end()); + + if (it->type() == RegType::sgpr && ctx.next_use_distances_start[block_idx][*it].second > distance) { + distance = ctx.next_use_distances_start[block_idx][*it].second; + to_spill = *it; + } + ++it; + } + assert(distance != 0); + + ctx.spills_entry[block_idx][to_spill] = ctx.allocate_spill_id(to_spill.regClass()); + partial_spills.erase(to_spill); + spilled_registers.sgpr += to_spill.size(); + reg_pressure.sgpr -= to_spill.size(); + } + + while (reg_pressure.vgpr > ctx.target_pressure.vgpr) { + assert(!partial_spills.empty()); + + std::set::iterator it = partial_spills.begin(); + Temp to_spill = *it; + unsigned distance = ctx.next_use_distances_start[block_idx][*it].second; + while (it != partial_spills.end()) { + assert(ctx.spills_entry[block_idx].find(*it) == ctx.spills_entry[block_idx].end()); + + if (it->type() == RegType::vgpr && ctx.next_use_distances_start[block_idx][*it].second > distance) { + distance = ctx.next_use_distances_start[block_idx][*it].second; + to_spill = *it; + } + ++it; + } + assert(distance != 0); + + ctx.spills_entry[block_idx][to_spill] = ctx.allocate_spill_id(to_spill.regClass()); + partial_spills.erase(to_spill); + spilled_registers.vgpr += to_spill.size(); + reg_pressure.vgpr -= to_spill.size(); + } + + return spilled_registers; +} + + +void add_coupling_code(spill_ctx& ctx, Block* block, unsigned block_idx) +{ + /* no coupling code necessary */ + if (block->linear_preds.size() == 0) + return; + + std::vector> instructions; + /* branch block: TODO take other branch into consideration */ + if (block->linear_preds.size() == 1) { + assert(ctx.processed[block->linear_preds[0]]); + + if (block->logical_preds.size() == 1) { + unsigned pred_idx = block->logical_preds[0]; + for (std::pair> live : ctx.next_use_distances_start[block_idx]) { + if (live.first.type() == RegType::sgpr) + continue; + /* still spilled */ + if (ctx.spills_entry[block_idx].find(live.first) != ctx.spills_entry[block_idx].end()) + continue; + + /* in register at end of predecessor */ + if (ctx.spills_exit[pred_idx].find(live.first) == ctx.spills_exit[pred_idx].end()) { + std::map::iterator it = ctx.renames[pred_idx].find(live.first); + if (it != ctx.renames[pred_idx].end()) + ctx.renames[block_idx].insert(*it); + continue; + } + + /* variable is spilled at predecessor and live at current block: create reload instruction */ + Temp new_name = {ctx.program->allocateId(), live.first.regClass()}; + aco_ptr reload = do_reload(ctx, live.first, new_name, ctx.spills_exit[pred_idx][live.first]); + instructions.emplace_back(std::move(reload)); + ctx.renames[block_idx][live.first] = new_name; + } + } + + unsigned pred_idx = block->linear_preds[0]; + for (std::pair> live : ctx.next_use_distances_start[block_idx]) { + if (live.first.type() == RegType::vgpr) + continue; + /* still spilled */ + if (ctx.spills_entry[block_idx].find(live.first) != ctx.spills_entry[block_idx].end()) + continue; + + /* in register at end of predecessor */ + if (ctx.spills_exit[pred_idx].find(live.first) == ctx.spills_exit[pred_idx].end()) { + std::map::iterator it = ctx.renames[pred_idx].find(live.first); + if (it != ctx.renames[pred_idx].end()) + ctx.renames[block_idx].insert(*it); + continue; + } + + /* variable is spilled at predecessor and live at current block: create reload instruction */ + Temp new_name = {ctx.program->allocateId(), live.first.regClass()}; + aco_ptr reload = do_reload(ctx, live.first, new_name, ctx.spills_exit[pred_idx][live.first]); + instructions.emplace_back(std::move(reload)); + ctx.renames[block_idx][live.first] = new_name; + } + + /* combine new reload instructions with original block */ + if (!instructions.empty()) { + unsigned insert_idx = 0; + while (block->instructions[insert_idx]->opcode == aco_opcode::p_phi || + block->instructions[insert_idx]->opcode == aco_opcode::p_linear_phi) { + insert_idx++; + } + ctx.register_demand[block->index].insert(std::next(ctx.register_demand[block->index].begin(), insert_idx), + instructions.size(), RegisterDemand()); + block->instructions.insert(std::next(block->instructions.begin(), insert_idx), + std::move_iterator>::iterator>(instructions.begin()), + std::move_iterator>::iterator>(instructions.end())); + } + return; + } + + /* loop header and merge blocks: check if all (linear) predecessors have been processed */ + for (ASSERTED unsigned pred : block->linear_preds) + assert(ctx.processed[pred]); + + /* iterate the phi nodes for which operands to spill at the predecessor */ + for (aco_ptr& phi : block->instructions) { + if (phi->opcode != aco_opcode::p_phi && + phi->opcode != aco_opcode::p_linear_phi) + break; + + /* if the phi is not spilled, add to instructions */ + if (ctx.spills_entry[block_idx].find(phi->definitions[0].getTemp()) == ctx.spills_entry[block_idx].end()) { + instructions.emplace_back(std::move(phi)); + continue; + } + + std::vector& preds = phi->opcode == aco_opcode::p_phi ? block->logical_preds : block->linear_preds; + uint32_t def_spill_id = ctx.spills_entry[block_idx][phi->definitions[0].getTemp()]; + + for (unsigned i = 0; i < phi->operands.size(); i++) { + unsigned pred_idx = preds[i]; + + /* we have to spill constants to the same memory address */ + if (phi->operands[i].isConstant()) { + uint32_t spill_id = ctx.allocate_spill_id(phi->definitions[0].regClass()); + for (std::pair pair : ctx.spills_exit[pred_idx]) { + ctx.interferences[def_spill_id].second.emplace(pair.second); + ctx.interferences[pair.second].second.emplace(def_spill_id); + } + ctx.affinities.emplace_back(std::pair{def_spill_id, spill_id}); + + aco_ptr spill{create_instruction(aco_opcode::p_spill, Format::PSEUDO, 2, 0)}; + spill->operands[0] = phi->operands[i]; + spill->operands[1] = Operand(spill_id); + Block& pred = ctx.program->blocks[pred_idx]; + unsigned idx = pred.instructions.size(); + do { + assert(idx != 0); + idx--; + } while (phi->opcode == aco_opcode::p_phi && pred.instructions[idx]->opcode != aco_opcode::p_logical_end); + std::vector>::iterator it = std::next(pred.instructions.begin(), idx); + pred.instructions.insert(it, std::move(spill)); + continue; + } + if (!phi->operands[i].isTemp()) + continue; + + /* build interferences between the phi def and all spilled variables at the predecessor blocks */ + for (std::pair pair : ctx.spills_exit[pred_idx]) { + if (phi->operands[i].getTemp() == pair.first) + continue; + ctx.interferences[def_spill_id].second.emplace(pair.second); + ctx.interferences[pair.second].second.emplace(def_spill_id); + } + + /* variable is already spilled at predecessor */ + std::map::iterator spilled = ctx.spills_exit[pred_idx].find(phi->operands[i].getTemp()); + if (spilled != ctx.spills_exit[pred_idx].end()) { + if (spilled->second != def_spill_id) + ctx.affinities.emplace_back(std::pair{def_spill_id, spilled->second}); + continue; + } + + /* rename if necessary */ + Temp var = phi->operands[i].getTemp(); + std::map::iterator rename_it = ctx.renames[pred_idx].find(var); + if (rename_it != ctx.renames[pred_idx].end()) { + var = rename_it->second; + ctx.renames[pred_idx].erase(rename_it); + } + + uint32_t spill_id = ctx.allocate_spill_id(phi->definitions[0].regClass()); + ctx.affinities.emplace_back(std::pair{def_spill_id, spill_id}); + aco_ptr spill{create_instruction(aco_opcode::p_spill, Format::PSEUDO, 2, 0)}; + spill->operands[0] = Operand(var); + spill->operands[1] = Operand(spill_id); + Block& pred = ctx.program->blocks[pred_idx]; + unsigned idx = pred.instructions.size(); + do { + assert(idx != 0); + idx--; + } while (phi->opcode == aco_opcode::p_phi && pred.instructions[idx]->opcode != aco_opcode::p_logical_end); + std::vector>::iterator it = std::next(pred.instructions.begin(), idx); + pred.instructions.insert(it, std::move(spill)); + ctx.spills_exit[pred_idx][phi->operands[i].getTemp()] = spill_id; + } + + /* remove phi from instructions */ + phi.reset(); + } + + /* iterate all (other) spilled variables for which to spill at the predecessor */ + // TODO: would be better to have them sorted: first vgprs and first with longest distance + for (std::pair pair : ctx.spills_entry[block_idx]) { + std::vector preds = pair.first.type() == RegType::vgpr ? block->logical_preds : block->linear_preds; + + for (unsigned pred_idx : preds) { + /* add interferences between spilled variable and predecessors exit spills */ + for (std::pair exit_spill : ctx.spills_exit[pred_idx]) { + if (exit_spill.first == pair.first) + continue; + ctx.interferences[exit_spill.second].second.emplace(pair.second); + ctx.interferences[pair.second].second.emplace(exit_spill.second); + } + + /* variable is already spilled at predecessor */ + std::map::iterator spilled = ctx.spills_exit[pred_idx].find(pair.first); + if (spilled != ctx.spills_exit[pred_idx].end()) { + if (spilled->second != pair.second) + ctx.affinities.emplace_back(std::pair{pair.second, spilled->second}); + continue; + } + + /* variable is dead at predecessor, it must be from a phi: this works because of CSSA form */ // FIXME: lower_to_cssa() + if (ctx.next_use_distances_end[pred_idx].find(pair.first) == ctx.next_use_distances_end[pred_idx].end()) + continue; + + /* variable is in register at predecessor and has to be spilled */ + /* rename if necessary */ + Temp var = pair.first; + std::map::iterator rename_it = ctx.renames[pred_idx].find(var); + if (rename_it != ctx.renames[pred_idx].end()) { + var = rename_it->second; + ctx.renames[pred_idx].erase(rename_it); + } + + aco_ptr spill{create_instruction(aco_opcode::p_spill, Format::PSEUDO, 2, 0)}; + spill->operands[0] = Operand(var); + spill->operands[1] = Operand(pair.second); + Block& pred = ctx.program->blocks[pred_idx]; + unsigned idx = pred.instructions.size(); + do { + assert(idx != 0); + idx--; + } while (pair.first.type() == RegType::vgpr && pred.instructions[idx]->opcode != aco_opcode::p_logical_end); + std::vector>::iterator it = std::next(pred.instructions.begin(), idx); + pred.instructions.insert(it, std::move(spill)); + ctx.spills_exit[pred.index][pair.first] = pair.second; + } + } + + /* iterate phis for which operands to reload */ + for (aco_ptr& phi : instructions) { + assert(phi->opcode == aco_opcode::p_phi || phi->opcode == aco_opcode::p_linear_phi); + assert(ctx.spills_entry[block_idx].find(phi->definitions[0].getTemp()) == ctx.spills_entry[block_idx].end()); + + std::vector& preds = phi->opcode == aco_opcode::p_phi ? block->logical_preds : block->linear_preds; + for (unsigned i = 0; i < phi->operands.size(); i++) { + if (!phi->operands[i].isTemp()) + continue; + unsigned pred_idx = preds[i]; + + /* rename operand */ + if (ctx.spills_exit[pred_idx].find(phi->operands[i].getTemp()) == ctx.spills_exit[pred_idx].end()) { + std::map::iterator it = ctx.renames[pred_idx].find(phi->operands[i].getTemp()); + if (it != ctx.renames[pred_idx].end()) + phi->operands[i].setTemp(it->second); + continue; + } + + Temp tmp = phi->operands[i].getTemp(); + + /* reload phi operand at end of predecessor block */ + Temp new_name = {ctx.program->allocateId(), tmp.regClass()}; + Block& pred = ctx.program->blocks[pred_idx]; + unsigned idx = pred.instructions.size(); + do { + assert(idx != 0); + idx--; + } while (phi->opcode == aco_opcode::p_phi && pred.instructions[idx]->opcode != aco_opcode::p_logical_end); + std::vector>::iterator it = std::next(pred.instructions.begin(), idx); + + aco_ptr reload = do_reload(ctx, tmp, new_name, ctx.spills_exit[pred_idx][tmp]); + pred.instructions.insert(it, std::move(reload)); + + ctx.spills_exit[pred_idx].erase(tmp); + ctx.renames[pred_idx][tmp] = new_name; + phi->operands[i].setTemp(new_name); + } + } + + /* iterate live variables for which to reload */ + // TODO: reload at current block if variable is spilled on all predecessors + for (std::pair> pair : ctx.next_use_distances_start[block_idx]) { + /* skip spilled variables */ + if (ctx.spills_entry[block_idx].find(pair.first) != ctx.spills_entry[block_idx].end()) + continue; + std::vector preds = pair.first.type() == RegType::vgpr ? block->logical_preds : block->linear_preds; + + /* variable is dead at predecessor, it must be from a phi */ + bool is_dead = false; + for (unsigned pred_idx : preds) { + if (ctx.next_use_distances_end[pred_idx].find(pair.first) == ctx.next_use_distances_end[pred_idx].end()) + is_dead = true; + } + if (is_dead) + continue; + for (unsigned pred_idx : preds) { + /* the variable is not spilled at the predecessor */ + if (ctx.spills_exit[pred_idx].find(pair.first) == ctx.spills_exit[pred_idx].end()) + continue; + + /* variable is spilled at predecessor and has to be reloaded */ + Temp new_name = {ctx.program->allocateId(), pair.first.regClass()}; + Block& pred = ctx.program->blocks[pred_idx]; + unsigned idx = pred.instructions.size(); + do { + assert(idx != 0); + idx--; + } while (pair.first.type() == RegType::vgpr && pred.instructions[idx]->opcode != aco_opcode::p_logical_end); + std::vector>::iterator it = std::next(pred.instructions.begin(), idx); + + aco_ptr reload = do_reload(ctx, pair.first, new_name, ctx.spills_exit[pred.index][pair.first]); + pred.instructions.insert(it, std::move(reload)); + + ctx.spills_exit[pred.index].erase(pair.first); + ctx.renames[pred.index][pair.first] = new_name; + } + + /* check if we have to create a new phi for this variable */ + Temp rename = Temp(); + bool is_same = true; + for (unsigned pred_idx : preds) { + if (ctx.renames[pred_idx].find(pair.first) == ctx.renames[pred_idx].end()) { + if (rename == Temp()) + rename = pair.first; + else + is_same = rename == pair.first; + } else { + if (rename == Temp()) + rename = ctx.renames[pred_idx][pair.first]; + else + is_same = rename == ctx.renames[pred_idx][pair.first]; + } + + if (!is_same) + break; + } + + if (!is_same) { + /* the variable was renamed differently in the predecessors: we have to create a phi */ + aco_opcode opcode = pair.first.type() == RegType::vgpr ? aco_opcode::p_phi : aco_opcode::p_linear_phi; + aco_ptr phi{create_instruction(opcode, Format::PSEUDO, preds.size(), 1)}; + rename = {ctx.program->allocateId(), pair.first.regClass()}; + for (unsigned i = 0; i < phi->operands.size(); i++) { + Temp tmp; + if (ctx.renames[preds[i]].find(pair.first) != ctx.renames[preds[i]].end()) + tmp = ctx.renames[preds[i]][pair.first]; + else if (preds[i] >= block_idx) + tmp = rename; + else + tmp = pair.first; + phi->operands[i] = Operand(tmp); + } + phi->definitions[0] = Definition(rename); + instructions.emplace_back(std::move(phi)); + } + + /* the variable was renamed: add new name to renames */ + if (!(rename == Temp() || rename == pair.first)) + ctx.renames[block_idx][pair.first] = rename; + } + + /* combine phis with instructions */ + unsigned idx = 0; + while (!block->instructions[idx]) { + idx++; + } + + ctx.register_demand[block->index].erase(ctx.register_demand[block->index].begin(), ctx.register_demand[block->index].begin() + idx); + ctx.register_demand[block->index].insert(ctx.register_demand[block->index].begin(), instructions.size(), RegisterDemand()); + + std::vector>::iterator start = std::next(block->instructions.begin(), idx); + instructions.insert(instructions.end(), std::move_iterator>::iterator>(start), + std::move_iterator>::iterator>(block->instructions.end())); + block->instructions = std::move(instructions); +} + +void process_block(spill_ctx& ctx, unsigned block_idx, Block* block, + std::map ¤t_spills, RegisterDemand spilled_registers) +{ + std::vector> local_next_use_distance; + std::vector> instructions; + unsigned idx = 0; + + /* phis are handled separetely */ + while (block->instructions[idx]->opcode == aco_opcode::p_phi || + block->instructions[idx]->opcode == aco_opcode::p_linear_phi) { + aco_ptr& instr = block->instructions[idx]; + for (const Operand& op : instr->operands) { + /* prevent it's definining instruction from being DCE'd if it could be rematerialized */ + if (op.isTemp() && ctx.remat.count(op.getTemp())) + ctx.remat_used[ctx.remat[op.getTemp()].instr] = true; + } + instructions.emplace_back(std::move(instr)); + idx++; + } + + if (block->register_demand.exceeds(ctx.target_pressure)) + local_next_use_distance = local_next_uses(ctx, block); + + while (idx < block->instructions.size()) { + aco_ptr& instr = block->instructions[idx]; + + std::map> reloads; + std::map spills; + /* rename and reload operands */ + for (Operand& op : instr->operands) { + if (!op.isTemp()) + continue; + if (current_spills.find(op.getTemp()) == current_spills.end()) { + /* the Operand is in register: check if it was renamed */ + if (ctx.renames[block_idx].find(op.getTemp()) != ctx.renames[block_idx].end()) + op.setTemp(ctx.renames[block_idx][op.getTemp()]); + /* prevent it's definining instruction from being DCE'd if it could be rematerialized */ + if (ctx.remat.count(op.getTemp())) + ctx.remat_used[ctx.remat[op.getTemp()].instr] = true; + continue; + } + /* the Operand is spilled: add it to reloads */ + Temp new_tmp = {ctx.program->allocateId(), op.regClass()}; + ctx.renames[block_idx][op.getTemp()] = new_tmp; + reloads[new_tmp] = std::make_pair(op.getTemp(), current_spills[op.getTemp()]); + current_spills.erase(op.getTemp()); + op.setTemp(new_tmp); + spilled_registers -= new_tmp; + } + + /* check if register demand is low enough before and after the current instruction */ + if (block->register_demand.exceeds(ctx.target_pressure)) { + + RegisterDemand new_demand = ctx.register_demand[block_idx][idx]; + if (idx == 0) { + for (const Definition& def : instr->definitions) { + if (!def.isTemp()) + continue; + new_demand += def.getTemp(); + } + } else { + new_demand.update(ctx.register_demand[block_idx][idx - 1]); + } + + assert(!local_next_use_distance.empty()); + + /* if reg pressure is too high, spill variable with furthest next use */ + while (RegisterDemand(new_demand - spilled_registers).exceeds(ctx.target_pressure)) { + unsigned distance = 0; + Temp to_spill; + bool do_rematerialize = false; + if (new_demand.vgpr - spilled_registers.vgpr > ctx.target_pressure.vgpr) { + for (std::pair pair : local_next_use_distance[idx]) { + bool can_rematerialize = ctx.remat.count(pair.first); + if (pair.first.type() == RegType::vgpr && + ((pair.second > distance && can_rematerialize == do_rematerialize) || + (can_rematerialize && !do_rematerialize && pair.second > idx)) && + current_spills.find(pair.first) == current_spills.end() && + ctx.spills_exit[block_idx].find(pair.first) == ctx.spills_exit[block_idx].end()) { + to_spill = pair.first; + distance = pair.second; + do_rematerialize = can_rematerialize; + } + } + } else { + for (std::pair pair : local_next_use_distance[idx]) { + bool can_rematerialize = ctx.remat.count(pair.first); + if (pair.first.type() == RegType::sgpr && + ((pair.second > distance && can_rematerialize == do_rematerialize) || + (can_rematerialize && !do_rematerialize && pair.second > idx)) && + current_spills.find(pair.first) == current_spills.end() && + ctx.spills_exit[block_idx].find(pair.first) == ctx.spills_exit[block_idx].end()) { + to_spill = pair.first; + distance = pair.second; + do_rematerialize = can_rematerialize; + } + } + } + + assert(distance != 0 && distance > idx); + uint32_t spill_id = ctx.allocate_spill_id(to_spill.regClass()); + + /* add interferences with currently spilled variables */ + for (std::pair pair : current_spills) { + ctx.interferences[spill_id].second.emplace(pair.second); + ctx.interferences[pair.second].second.emplace(spill_id); + } + for (std::pair> pair : reloads) { + ctx.interferences[spill_id].second.emplace(pair.second.second); + ctx.interferences[pair.second.second].second.emplace(spill_id); + } + + current_spills[to_spill] = spill_id; + spilled_registers += to_spill; + + /* rename if necessary */ + if (ctx.renames[block_idx].find(to_spill) != ctx.renames[block_idx].end()) { + to_spill = ctx.renames[block_idx][to_spill]; + } + + /* add spill to new instructions */ + aco_ptr spill{create_instruction(aco_opcode::p_spill, Format::PSEUDO, 2, 0)}; + spill->operands[0] = Operand(to_spill); + spill->operands[1] = Operand(spill_id); + instructions.emplace_back(std::move(spill)); + } + } + + /* add reloads and instruction to new instructions */ + for (std::pair> pair : reloads) { + aco_ptr reload = do_reload(ctx, pair.second.first, pair.first, pair.second.second); + instructions.emplace_back(std::move(reload)); + } + instructions.emplace_back(std::move(instr)); + idx++; + } + + block->instructions = std::move(instructions); + ctx.spills_exit[block_idx].insert(current_spills.begin(), current_spills.end()); +} + +void spill_block(spill_ctx& ctx, unsigned block_idx) +{ + Block* block = &ctx.program->blocks[block_idx]; + ctx.processed[block_idx] = true; + + /* determine set of variables which are spilled at the beginning of the block */ + RegisterDemand spilled_registers = init_live_in_vars(ctx, block, block_idx); + + /* add interferences for spilled variables */ + for (std::pair x : ctx.spills_entry[block_idx]) { + for (std::pair y : ctx.spills_entry[block_idx]) + if (x.second != y.second) + ctx.interferences[x.second].second.emplace(y.second); + } + + bool is_loop_header = block->loop_nest_depth && ctx.loop_header.top()->index == block_idx; + if (!is_loop_header) { + /* add spill/reload code on incoming control flow edges */ + add_coupling_code(ctx, block, block_idx); + } + + std::map current_spills = ctx.spills_entry[block_idx]; + + /* check conditions to process this block */ + bool process = RegisterDemand(block->register_demand - spilled_registers).exceeds(ctx.target_pressure) || + !ctx.renames[block_idx].empty() || + ctx.remat_used.size(); + + std::map::iterator it = current_spills.begin(); + while (!process && it != current_spills.end()) { + if (ctx.next_use_distances_start[block_idx][it->first].first == block_idx) + process = true; + ++it; + } + + if (process) + process_block(ctx, block_idx, block, current_spills, spilled_registers); + else + ctx.spills_exit[block_idx].insert(current_spills.begin(), current_spills.end()); + + /* check if the next block leaves the current loop */ + if (block->loop_nest_depth == 0 || ctx.program->blocks[block_idx + 1].loop_nest_depth >= block->loop_nest_depth) + return; + + Block* loop_header = ctx.loop_header.top(); + + /* preserve original renames at end of loop header block */ + std::map renames = std::move(ctx.renames[loop_header->index]); + + /* add coupling code to all loop header predecessors */ + add_coupling_code(ctx, loop_header, loop_header->index); + + /* update remat_used for phis added in add_coupling_code() */ + for (aco_ptr& instr : loop_header->instructions) { + if (!is_phi(instr)) + break; + for (const Operand& op : instr->operands) { + if (op.isTemp() && ctx.remat.count(op.getTemp())) + ctx.remat_used[ctx.remat[op.getTemp()].instr] = true; + } + } + + /* propagate new renames through loop: i.e. repair the SSA */ + renames.swap(ctx.renames[loop_header->index]); + for (std::pair rename : renames) { + for (unsigned idx = loop_header->index; idx <= block_idx; idx++) { + Block& current = ctx.program->blocks[idx]; + std::vector>::iterator instr_it = current.instructions.begin(); + + /* first rename phis */ + while (instr_it != current.instructions.end()) { + aco_ptr& phi = *instr_it; + if (phi->opcode != aco_opcode::p_phi && phi->opcode != aco_opcode::p_linear_phi) + break; + /* no need to rename the loop header phis once again. this happened in add_coupling_code() */ + if (idx == loop_header->index) { + instr_it++; + continue; + } + + for (Operand& op : phi->operands) { + if (!op.isTemp()) + continue; + if (op.getTemp() == rename.first) + op.setTemp(rename.second); + } + instr_it++; + } + + std::map>::iterator it = ctx.next_use_distances_start[idx].find(rename.first); + + /* variable is not live at beginning of this block */ + if (it == ctx.next_use_distances_start[idx].end()) + continue; + + /* if the variable is live at the block's exit, add rename */ + if (ctx.next_use_distances_end[idx].find(rename.first) != ctx.next_use_distances_end[idx].end()) + ctx.renames[idx].insert(rename); + + /* rename all uses in this block */ + bool renamed = false; + while (!renamed && instr_it != current.instructions.end()) { + aco_ptr& instr = *instr_it; + for (Operand& op : instr->operands) { + if (!op.isTemp()) + continue; + if (op.getTemp() == rename.first) { + op.setTemp(rename.second); + /* we can stop with this block as soon as the variable is spilled */ + if (instr->opcode == aco_opcode::p_spill) + renamed = true; + } + } + instr_it++; + } + } + } + + /* remove loop header info from stack */ + ctx.loop_header.pop(); +} + +void assign_spill_slots(spill_ctx& ctx, unsigned spills_to_vgpr) { + std::map sgpr_slot; + std::map vgpr_slot; + std::vector is_assigned(ctx.interferences.size()); + + /* first, handle affinities: just merge all interferences into both spill ids */ + for (std::pair pair : ctx.affinities) { + assert(pair.first != pair.second); + for (uint32_t id : ctx.interferences[pair.first].second) + ctx.interferences[id].second.insert(pair.second); + for (uint32_t id : ctx.interferences[pair.second].second) + ctx.interferences[id].second.insert(pair.first); + ctx.interferences[pair.first].second.insert(ctx.interferences[pair.second].second.begin(), ctx.interferences[pair.second].second.end()); + ctx.interferences[pair.second].second.insert(ctx.interferences[pair.first].second.begin(), ctx.interferences[pair.first].second.end()); + + bool reloaded = ctx.is_reloaded[pair.first] || ctx.is_reloaded[pair.second]; + ctx.is_reloaded[pair.first] = ctx.is_reloaded[pair.second] = reloaded; + } + for (ASSERTED uint32_t i = 0; i < ctx.interferences.size(); i++) + for (ASSERTED uint32_t id : ctx.interferences[i].second) + assert(i != id); + + /* for each spill slot, assign as many spill ids as possible */ + std::vector> spill_slot_interferences; + unsigned slot_idx = 0; + bool done = false; + + /* assign sgpr spill slots */ + while (!done) { + done = true; + for (unsigned id = 0; id < ctx.interferences.size(); id++) { + if (is_assigned[id] || !ctx.is_reloaded[id]) + continue; + if (ctx.interferences[id].first.type() != RegType::sgpr) + continue; + + /* check interferences */ + bool interferes = false; + for (unsigned i = slot_idx; i < slot_idx + ctx.interferences[id].first.size(); i++) { + if (i == spill_slot_interferences.size()) + spill_slot_interferences.emplace_back(std::set()); + if (spill_slot_interferences[i].find(id) != spill_slot_interferences[i].end() || i / 64 != slot_idx / 64) { + interferes = true; + break; + } + } + if (interferes) { + done = false; + continue; + } + + /* we found a spill id which can be assigned to current spill slot */ + sgpr_slot[id] = slot_idx; + is_assigned[id] = true; + for (unsigned i = slot_idx; i < slot_idx + ctx.interferences[id].first.size(); i++) + spill_slot_interferences[i].insert(ctx.interferences[id].second.begin(), ctx.interferences[id].second.end()); + } + slot_idx++; + } + + slot_idx = 0; + done = false; + + /* assign vgpr spill slots */ + while (!done) { + done = true; + for (unsigned id = 0; id < ctx.interferences.size(); id++) { + if (is_assigned[id] || !ctx.is_reloaded[id]) + continue; + if (ctx.interferences[id].first.type() != RegType::vgpr) + continue; + + /* check interferences */ + bool interferes = false; + for (unsigned i = slot_idx; i < slot_idx + ctx.interferences[id].first.size(); i++) { + if (i == spill_slot_interferences.size()) + spill_slot_interferences.emplace_back(std::set()); + /* check for interference and ensure that vector regs are stored next to each other */ + if (spill_slot_interferences[i].find(id) != spill_slot_interferences[i].end() || i / 64 != slot_idx / 64) { + interferes = true; + break; + } + } + if (interferes) { + done = false; + continue; + } + + /* we found a spill id which can be assigned to current spill slot */ + vgpr_slot[id] = slot_idx; + is_assigned[id] = true; + for (unsigned i = slot_idx; i < slot_idx + ctx.interferences[id].first.size(); i++) + spill_slot_interferences[i].insert(ctx.interferences[id].second.begin(), ctx.interferences[id].second.end()); + } + slot_idx++; + } + + for (unsigned id = 0; id < is_assigned.size(); id++) + assert(is_assigned[id] || !ctx.is_reloaded[id]); + + for (std::pair pair : ctx.affinities) { + assert(is_assigned[pair.first] == is_assigned[pair.second]); + if (!is_assigned[pair.first]) + continue; + assert(ctx.is_reloaded[pair.first] == ctx.is_reloaded[pair.second]); + assert(ctx.interferences[pair.first].first.type() == ctx.interferences[pair.second].first.type()); + if (ctx.interferences[pair.first].first.type() == RegType::sgpr) + assert(sgpr_slot[pair.first] == sgpr_slot[pair.second]); + else + assert(vgpr_slot[pair.first] == vgpr_slot[pair.second]); + } + + /* hope, we didn't mess up */ + std::vector vgpr_spill_temps((spill_slot_interferences.size() + 63) / 64); + assert(vgpr_spill_temps.size() <= spills_to_vgpr); + + /* replace pseudo instructions with actual hardware instructions */ + unsigned last_top_level_block_idx = 0; + std::vector reload_in_loop(vgpr_spill_temps.size()); + for (Block& block : ctx.program->blocks) { + + /* after loops, we insert a user if there was a reload inside the loop */ + if (block.loop_nest_depth == 0) { + int end_vgprs = 0; + for (unsigned i = 0; i < vgpr_spill_temps.size(); i++) { + if (reload_in_loop[i]) + end_vgprs++; + } + + if (end_vgprs > 0) { + aco_ptr destr{create_instruction(aco_opcode::p_end_linear_vgpr, Format::PSEUDO, end_vgprs, 0)}; + int k = 0; + for (unsigned i = 0; i < vgpr_spill_temps.size(); i++) { + if (reload_in_loop[i]) + destr->operands[k++] = Operand(vgpr_spill_temps[i]); + reload_in_loop[i] = false; + } + /* find insertion point */ + std::vector>::iterator it = block.instructions.begin(); + while ((*it)->opcode == aco_opcode::p_linear_phi || (*it)->opcode == aco_opcode::p_phi) + ++it; + block.instructions.insert(it, std::move(destr)); + } + } + + if (block.kind & block_kind_top_level && !block.linear_preds.empty()) { + last_top_level_block_idx = block.index; + + /* check if any spilled variables use a created linear vgpr, otherwise destroy them */ + for (unsigned i = 0; i < vgpr_spill_temps.size(); i++) { + if (vgpr_spill_temps[i] == Temp()) + continue; + + bool can_destroy = true; + for (std::pair pair : ctx.spills_exit[block.linear_preds[0]]) { + + if (sgpr_slot.find(pair.second) != sgpr_slot.end() && + sgpr_slot[pair.second] / 64 == i) { + can_destroy = false; + break; + } + } + if (can_destroy) + vgpr_spill_temps[i] = Temp(); + } + } + + std::vector>::iterator it; + std::vector> instructions; + instructions.reserve(block.instructions.size()); + for (it = block.instructions.begin(); it != block.instructions.end(); ++it) { + + if ((*it)->opcode == aco_opcode::p_spill) { + uint32_t spill_id = (*it)->operands[1].constantValue(); + + if (!ctx.is_reloaded[spill_id]) { + /* never reloaded, so don't spill */ + } else if (vgpr_slot.find(spill_id) != vgpr_slot.end()) { + /* spill vgpr */ + ctx.program->config->spilled_vgprs += (*it)->operands[0].size(); + + assert(false && "vgpr spilling not yet implemented."); + } else if (sgpr_slot.find(spill_id) != sgpr_slot.end()) { + ctx.program->config->spilled_sgprs += (*it)->operands[0].size(); + + uint32_t spill_slot = sgpr_slot[spill_id]; + + /* check if the linear vgpr already exists */ + if (vgpr_spill_temps[spill_slot / 64] == Temp()) { + Temp linear_vgpr = {ctx.program->allocateId(), v1.as_linear()}; + vgpr_spill_temps[spill_slot / 64] = linear_vgpr; + aco_ptr create{create_instruction(aco_opcode::p_start_linear_vgpr, Format::PSEUDO, 0, 1)}; + create->definitions[0] = Definition(linear_vgpr); + /* find the right place to insert this definition */ + if (last_top_level_block_idx == block.index) { + /* insert right before the current instruction */ + instructions.emplace_back(std::move(create)); + } else { + assert(last_top_level_block_idx < block.index); + /* insert before the branch at last top level block */ + std::vector>& instructions = ctx.program->blocks[last_top_level_block_idx].instructions; + instructions.insert(std::next(instructions.begin(), instructions.size() - 1), std::move(create)); + } + } + + /* spill sgpr: just add the vgpr temp to operands */ + Pseudo_instruction* spill = create_instruction(aco_opcode::p_spill, Format::PSEUDO, 3, 0); + spill->operands[0] = Operand(vgpr_spill_temps[spill_slot / 64]); + spill->operands[1] = Operand(spill_slot % 64); + spill->operands[2] = (*it)->operands[0]; + instructions.emplace_back(aco_ptr(spill)); + } else { + unreachable("No spill slot assigned for spill id"); + } + + } else if ((*it)->opcode == aco_opcode::p_reload) { + uint32_t spill_id = (*it)->operands[0].constantValue(); + assert(ctx.is_reloaded[spill_id]); + + if (vgpr_slot.find(spill_id) != vgpr_slot.end()) { + /* reload vgpr */ + assert(false && "vgpr spilling not yet implemented."); + + } else if (sgpr_slot.find(spill_id) != sgpr_slot.end()) { + uint32_t spill_slot = sgpr_slot[spill_id]; + reload_in_loop[spill_slot / 64] = block.loop_nest_depth > 0; + + /* check if the linear vgpr already exists */ + if (vgpr_spill_temps[spill_slot / 64] == Temp()) { + Temp linear_vgpr = {ctx.program->allocateId(), v1.as_linear()}; + vgpr_spill_temps[spill_slot / 64] = linear_vgpr; + aco_ptr create{create_instruction(aco_opcode::p_start_linear_vgpr, Format::PSEUDO, 0, 1)}; + create->definitions[0] = Definition(linear_vgpr); + /* find the right place to insert this definition */ + if (last_top_level_block_idx == block.index) { + /* insert right before the current instruction */ + instructions.emplace_back(std::move(create)); + } else { + assert(last_top_level_block_idx < block.index); + /* insert before the branch at last top level block */ + std::vector>& instructions = ctx.program->blocks[last_top_level_block_idx].instructions; + instructions.insert(std::next(instructions.begin(), instructions.size() - 1), std::move(create)); + } + } + + /* reload sgpr: just add the vgpr temp to operands */ + Pseudo_instruction* reload = create_instruction(aco_opcode::p_reload, Format::PSEUDO, 2, 1); + reload->operands[0] = Operand(vgpr_spill_temps[spill_slot / 64]); + reload->operands[1] = Operand(spill_slot % 64); + reload->definitions[0] = (*it)->definitions[0]; + instructions.emplace_back(aco_ptr(reload)); + } else { + unreachable("No spill slot assigned for spill id"); + } + } else if (!ctx.remat_used.count(it->get()) || ctx.remat_used[it->get()]) { + instructions.emplace_back(std::move(*it)); + } + + } + block.instructions = std::move(instructions); + } + + /* SSA elimination inserts copies for logical phis right before p_logical_end + * So if a linear vgpr is used between that p_logical_end and the branch, + * we need to ensure logical phis don't choose a definition which aliases + * the linear vgpr. + * TODO: Moving the spills and reloads to before p_logical_end might produce + * slightly better code. */ + for (Block& block : ctx.program->blocks) { + /* loops exits are already handled */ + if (block.logical_preds.size() <= 1) + continue; + + bool has_logical_phis = false; + for (aco_ptr& instr : block.instructions) { + if (instr->opcode == aco_opcode::p_phi) { + has_logical_phis = true; + break; + } else if (instr->opcode != aco_opcode::p_linear_phi) { + break; + } + } + if (!has_logical_phis) + continue; + + std::set vgprs; + for (unsigned pred_idx : block.logical_preds) { + Block& pred = ctx.program->blocks[pred_idx]; + for (int i = pred.instructions.size() - 1; i >= 0; i--) { + aco_ptr& pred_instr = pred.instructions[i]; + if (pred_instr->opcode == aco_opcode::p_logical_end) { + break; + } else if (pred_instr->opcode == aco_opcode::p_spill || + pred_instr->opcode == aco_opcode::p_reload) { + vgprs.insert(pred_instr->operands[0].getTemp()); + } + } + } + if (!vgprs.size()) + continue; + + aco_ptr destr{create_instruction(aco_opcode::p_end_linear_vgpr, Format::PSEUDO, vgprs.size(), 0)}; + int k = 0; + for (Temp tmp : vgprs) { + destr->operands[k++] = Operand(tmp); + } + /* find insertion point */ + std::vector>::iterator it = block.instructions.begin(); + while ((*it)->opcode == aco_opcode::p_linear_phi || (*it)->opcode == aco_opcode::p_phi) + ++it; + block.instructions.insert(it, std::move(destr)); + } +} + +} /* end namespace */ + + +void spill(Program* program, live& live_vars, const struct radv_nir_compiler_options *options) +{ + program->config->spilled_vgprs = 0; + program->config->spilled_sgprs = 0; + + /* no spilling when wave count is already high */ + if (program->num_waves >= 6) + return; + + /* else, we check if we can improve things a bit */ + uint16_t total_sgpr_regs = options->chip_class >= GFX8 ? 800 : 512; + uint16_t max_addressible_sgpr = program->sgpr_limit; + + /* calculate target register demand */ + RegisterDemand max_reg_demand; + for (Block& block : program->blocks) { + max_reg_demand.update(block.register_demand); + } + + RegisterDemand target_pressure = {256, int16_t(max_addressible_sgpr)}; + unsigned num_waves = 1; + int spills_to_vgpr = (max_reg_demand.sgpr - max_addressible_sgpr + 63) / 64; + + /* test if it possible to increase occupancy with little spilling */ + for (unsigned num_waves_next = 2; num_waves_next <= 8; num_waves_next++) { + RegisterDemand target_pressure_next = {int16_t((256 / num_waves_next) & ~3), + int16_t(std::min(((total_sgpr_regs / num_waves_next) & ~7) - 2, max_addressible_sgpr))}; + + /* Currently no vgpr spilling supported. + * Spill as many sgprs as necessary to not hinder occupancy */ + if (max_reg_demand.vgpr > target_pressure_next.vgpr) + break; + /* check that we have enough free vgprs to spill sgprs to */ + if (max_reg_demand.sgpr > target_pressure_next.sgpr) { + /* add some buffer in case graph coloring is not perfect ... */ + const int spills_to_vgpr_next = (max_reg_demand.sgpr - target_pressure_next.sgpr + 63 + 32) / 64; + if (spills_to_vgpr_next + max_reg_demand.vgpr > target_pressure_next.vgpr) + break; + spills_to_vgpr = spills_to_vgpr_next; + } + + target_pressure = target_pressure_next; + num_waves = num_waves_next; + } + + assert(max_reg_demand.vgpr <= target_pressure.vgpr && "VGPR spilling not yet supported."); + /* nothing to do */ + if (num_waves == program->num_waves) + return; + + /* initialize ctx */ + spill_ctx ctx(target_pressure, program, live_vars.register_demand); + compute_global_next_uses(ctx, live_vars.live_out); + get_rematerialize_info(ctx); + + /* create spills and reloads */ + for (unsigned i = 0; i < program->blocks.size(); i++) + spill_block(ctx, i); + + /* assign spill slots and DCE rematerialized code */ + assign_spill_slots(ctx, spills_to_vgpr); + + /* update live variable information */ + live_vars = live_var_analysis(program, options); + + assert(program->num_waves >= num_waves); +} + +} + diff --git a/src/amd/compiler/aco_ssa_elimination.cpp b/src/amd/compiler/aco_ssa_elimination.cpp new file mode 100644 index 00000000000..3d76dcd8867 --- /dev/null +++ b/src/amd/compiler/aco_ssa_elimination.cpp @@ -0,0 +1,291 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + + +#include "aco_ir.h" + +#include + +namespace aco { +namespace { + +/* map: block-id -> pair (dest, src) to store phi information */ +typedef std::map>> phi_info; + +struct ssa_elimination_ctx { + phi_info logical_phi_info; + phi_info linear_phi_info; + std::vector empty_blocks; + Program* program; + + ssa_elimination_ctx(Program* program) : empty_blocks(program->blocks.size(), true), program(program) {} +}; + +void collect_phi_info(ssa_elimination_ctx& ctx) +{ + for (Block& block : ctx.program->blocks) { + for (aco_ptr& phi : block.instructions) { + if (phi->opcode != aco_opcode::p_phi && phi->opcode != aco_opcode::p_linear_phi) + break; + + for (unsigned i = 0; i < phi->operands.size(); i++) { + if (phi->operands[i].isUndefined()) + continue; + if (phi->operands[i].isTemp() && phi->operands[i].physReg() == phi->definitions[0].physReg()) + continue; + + std::vector& preds = phi->opcode == aco_opcode::p_phi ? block.logical_preds : block.linear_preds; + phi_info& info = phi->opcode == aco_opcode::p_phi ? ctx.logical_phi_info : ctx.linear_phi_info; + const auto result = info.emplace(preds[i], std::vector>()); + result.first->second.emplace_back(phi->definitions[0], phi->operands[i]); + ctx.empty_blocks[preds[i]] = false; + } + } + } +} + +void insert_parallelcopies(ssa_elimination_ctx& ctx) +{ + /* insert the parallelcopies from logical phis before p_logical_end */ + for (auto&& entry : ctx.logical_phi_info) { + Block& block = ctx.program->blocks[entry.first]; + unsigned idx = block.instructions.size() - 1; + while (block.instructions[idx]->opcode != aco_opcode::p_logical_end) { + assert(idx > 0); + idx--; + } + + std::vector>::iterator it = std::next(block.instructions.begin(), idx); + aco_ptr pc{create_instruction(aco_opcode::p_parallelcopy, Format::PSEUDO, entry.second.size(), entry.second.size())}; + unsigned i = 0; + for (std::pair& pair : entry.second) + { + pc->definitions[i] = pair.first; + pc->operands[i] = pair.second; + i++; + } + /* this shouldn't be needed since we're only copying vgprs */ + pc->tmp_in_scc = false; + block.instructions.insert(it, std::move(pc)); + } + + /* insert parallelcopies for the linear phis at the end of blocks just before the branch */ + for (auto&& entry : ctx.linear_phi_info) { + Block& block = ctx.program->blocks[entry.first]; + std::vector>::iterator it = block.instructions.end(); + --it; + assert((*it)->format == Format::PSEUDO_BRANCH); + aco_ptr pc{create_instruction(aco_opcode::p_parallelcopy, Format::PSEUDO, entry.second.size(), entry.second.size())}; + unsigned i = 0; + for (std::pair& pair : entry.second) + { + pc->definitions[i] = pair.first; + pc->operands[i] = pair.second; + i++; + } + pc->tmp_in_scc = block.scc_live_out; + pc->scratch_sgpr = block.scratch_sgpr; + block.instructions.insert(it, std::move(pc)); + } +} + + +void try_remove_merge_block(ssa_elimination_ctx& ctx, Block* block) +{ + /* check if the successor is another merge block which restores exec */ + // TODO: divergent loops also restore exec + if (block->linear_succs.size() != 1 || + !(ctx.program->blocks[block->linear_succs[0]].kind & block_kind_merge)) + return; + + /* check if this block is empty and the exec mask is not needed */ + for (aco_ptr& instr : block->instructions) { + if (instr->opcode == aco_opcode::p_parallelcopy) { + if (instr->definitions[0].physReg() == exec) + continue; + else + return; + } + + if (instr->opcode != aco_opcode::p_linear_phi && + instr->opcode != aco_opcode::p_phi && + instr->opcode != aco_opcode::p_logical_start && + instr->opcode != aco_opcode::p_logical_end && + instr->opcode != aco_opcode::p_branch) + return; + } + + /* keep the branch instruction and remove the rest */ + aco_ptr branch = std::move(block->instructions.back()); + block->instructions.clear(); + block->instructions.emplace_back(std::move(branch)); +} + +void try_remove_invert_block(ssa_elimination_ctx& ctx, Block* block) +{ + assert(block->linear_succs.size() == 2); + if (block->linear_succs[0] != block->linear_succs[1]) + return; + + /* check if we can remove this block */ + for (aco_ptr& instr : block->instructions) { + if (instr->opcode != aco_opcode::p_linear_phi && + instr->opcode != aco_opcode::p_phi && + instr->opcode != aco_opcode::s_andn2_b64 && + instr->opcode != aco_opcode::p_branch) + return; + } + + unsigned succ_idx = block->linear_succs[0]; + assert(block->linear_preds.size() == 2); + for (unsigned i = 0; i < 2; i++) { + Block *pred = &ctx.program->blocks[block->linear_preds[i]]; + pred->linear_succs[0] = succ_idx; + ctx.program->blocks[succ_idx].linear_preds[i] = pred->index; + + Pseudo_branch_instruction *branch = static_cast(pred->instructions.back().get()); + assert(branch->format == Format::PSEUDO_BRANCH); + branch->target[0] = succ_idx; + branch->target[1] = succ_idx; + } + + block->instructions.clear(); + block->linear_preds.clear(); + block->linear_succs.clear(); +} + +void try_remove_simple_block(ssa_elimination_ctx& ctx, Block* block) +{ + for (aco_ptr& instr : block->instructions) { + if (instr->opcode != aco_opcode::p_logical_start && + instr->opcode != aco_opcode::p_logical_end && + instr->opcode != aco_opcode::p_branch) + return; + } + + Block& pred = ctx.program->blocks[block->linear_preds[0]]; + Block& succ = ctx.program->blocks[block->linear_succs[0]]; + Pseudo_branch_instruction* branch = static_cast(pred.instructions.back().get()); + if (branch->opcode == aco_opcode::p_branch) { + branch->target[0] = succ.index; + branch->target[1] = succ.index; + } else if (branch->target[0] == block->index) { + branch->target[0] = succ.index; + } else if (branch->target[0] == succ.index) { + assert(branch->target[1] == block->index); + branch->target[1] = succ.index; + branch->opcode = aco_opcode::p_branch; + } else if (branch->target[1] == block->index) { + /* check if there is a fall-through path from block to succ */ + bool falls_through = true; + for (unsigned j = block->index + 1; falls_through && j < succ.index; j++) { + assert(ctx.program->blocks[j].index == j); + if (!ctx.program->blocks[j].instructions.empty()) + falls_through = false; + } + if (falls_through) { + branch->target[1] = succ.index; + } else { + /* check if there is a fall-through path for the alternative target */ + for (unsigned j = block->index + 1; j < branch->target[0]; j++) { + if (!ctx.program->blocks[j].instructions.empty()) + return; + } + + /* This is a (uniform) break or continue block. The branch condition has to be inverted. */ + if (branch->opcode == aco_opcode::p_cbranch_z) + branch->opcode = aco_opcode::p_cbranch_nz; + else if (branch->opcode == aco_opcode::p_cbranch_nz) + branch->opcode = aco_opcode::p_cbranch_z; + else + assert(false); + /* also invert the linear successors */ + pred.linear_succs[0] = pred.linear_succs[1]; + pred.linear_succs[1] = succ.index; + branch->target[1] = branch->target[0]; + branch->target[0] = succ.index; + } + } else { + assert(false); + } + + if (branch->target[0] == branch->target[1]) + branch->opcode = aco_opcode::p_branch; + + for (unsigned i = 0; i < pred.linear_succs.size(); i++) + if (pred.linear_succs[i] == block->index) + pred.linear_succs[i] = succ.index; + + for (unsigned i = 0; i < succ.linear_preds.size(); i++) + if (succ.linear_preds[i] == block->index) + succ.linear_preds[i] = pred.index; + + block->instructions.clear(); + block->linear_preds.clear(); + block->linear_succs.clear(); +} + +void jump_threading(ssa_elimination_ctx& ctx) +{ + for (int i = ctx.program->blocks.size() - 1; i >= 0; i--) { + Block* block = &ctx.program->blocks[i]; + + if (!ctx.empty_blocks[i]) + continue; + + if (block->kind & block_kind_invert) { + try_remove_invert_block(ctx, block); + continue; + } + + if (block->linear_succs.size() > 1) + continue; + + if (block->kind & block_kind_merge || + block->kind & block_kind_loop_exit) + try_remove_merge_block(ctx, block); + + if (block->linear_preds.size() == 1) + try_remove_simple_block(ctx, block); + } +} + +} /* end namespace */ + + +void ssa_elimination(Program* program) +{ + ssa_elimination_ctx ctx(program); + + /* Collect information about every phi-instruction */ + collect_phi_info(ctx); + + /* eliminate empty blocks */ + jump_threading(ctx); + + /* insert parallelcopies from SSA elimination */ + insert_parallelcopies(ctx); + +} +} diff --git a/src/amd/compiler/aco_util.h b/src/amd/compiler/aco_util.h new file mode 100644 index 00000000000..ec77ba55716 --- /dev/null +++ b/src/amd/compiler/aco_util.h @@ -0,0 +1,233 @@ +/* + * Copyright Michael Schellenberger Costa + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#ifndef ACO_UTIL_H +#define ACO_UTIL_H + +#include +#include + +namespace aco { + +/*! \brief Definition of a span object +* +* \details A "span" is an "array view" type for holding a view of contiguous +* data. The "span" object does not own the data itself. +*/ +template +class span { +public: + using value_type = T; + using pointer = value_type*; + using const_pointer = const value_type*; + using reference = value_type&; + using const_reference = const value_type&; + using iterator = pointer; + using const_iterator = const_pointer; + using reverse_iterator = std::reverse_iterator; + using const_reverse_iterator = std::reverse_iterator; + using size_type = std::size_t; + using difference_type = std::ptrdiff_t; + + /*! \brief Compiler generated default constructor + */ + constexpr span() = default; + + /*! \brief Constructor taking a pointer and the length of the span + * \param[in] data Pointer to the underlying data array + * \param[in] length The size of the span + */ + constexpr span(pointer data, const size_type length) + : data{ data } , length{ length } {} + + /*! \brief Returns an iterator to the begin of the span + * \return data + */ + constexpr iterator begin() noexcept { + return data; + } + + /*! \brief Returns a const_iterator to the begin of the span + * \return data + */ + constexpr const_iterator begin() const noexcept { + return data; + } + + /*! \brief Returns an iterator to the end of the span + * \return data + length + */ + constexpr iterator end() noexcept { + return std::next(data, length); + } + + /*! \brief Returns a const_iterator to the end of the span + * \return data + length + */ + constexpr const_iterator end() const noexcept { + return std::next(data, length); + } + + /*! \brief Returns a const_iterator to the begin of the span + * \return data + */ + constexpr const_iterator cbegin() const noexcept { + return data; + } + + /*! \brief Returns a const_iterator to the end of the span + * \return data + length + */ + constexpr const_iterator cend() const noexcept { + return std::next(data, length); + } + + /*! \brief Returns a reverse_iterator to the end of the span + * \return reverse_iterator(end()) + */ + constexpr reverse_iterator rbegin() noexcept { + return reverse_iterator(end()); + } + + /*! \brief Returns a const_reverse_iterator to the end of the span + * \return reverse_iterator(end()) + */ + constexpr const_reverse_iterator rbegin() const noexcept { + return const_reverse_iterator(end()); + } + + /*! \brief Returns a reverse_iterator to the begin of the span + * \return reverse_iterator(begin()) + */ + constexpr reverse_iterator rend() noexcept { + return reverse_iterator(begin()); + } + + /*! \brief Returns a const_reverse_iterator to the begin of the span + * \return reverse_iterator(begin()) + */ + constexpr const_reverse_iterator rend() const noexcept { + return const_reverse_iterator(begin()); + } + + /*! \brief Returns a const_reverse_iterator to the end of the span + * \return rbegin() + */ + constexpr const_reverse_iterator crbegin() const noexcept { + return const_reverse_iterator(cend()); + } + + /*! \brief Returns a const_reverse_iterator to the begin of the span + * \return rend() + */ + constexpr const_reverse_iterator crend() const noexcept { + return const_reverse_iterator(cbegin()); + } + + /*! \brief Unchecked access operator + * \param[in] index Index of the element we want to access + * \return *(std::next(data, index)) + */ + constexpr reference operator[](const size_type index) noexcept { + assert(length > index); + return *(std::next(data, index)); + } + + /*! \brief Unchecked const access operator + * \param[in] index Index of the element we want to access + * \return *(std::next(data, index)) + */ + constexpr const_reference operator[](const size_type index) const noexcept { + assert(length > index); + return *(std::next(data, index)); + } + + /*! \brief Returns a reference to the last element of the span + * \return *(std::next(data, length - 1)) + */ + constexpr reference back() noexcept { + assert(length > 0); + return *(std::next(data, length - 1)); + } + + /*! \brief Returns a const_reference to the last element of the span + * \return *(std::next(data, length - 1)) + */ + constexpr const_reference back() const noexcept { + assert(length > 0); + return *(std::next(data, length - 1)); + } + + /*! \brief Returns a reference to the first element of the span + * \return *begin() + */ + constexpr reference front() noexcept { + assert(length > 0); + return *begin(); + } + + /*! \brief Returns a const_reference to the first element of the span + * \return *cbegin() + */ + constexpr const_reference front() const noexcept { + assert(length > 0); + return *cbegin(); + } + + /*! \brief Returns true if the span is empty + * \return length == 0 + */ + constexpr bool empty() const noexcept { + return length == 0; + } + + /*! \brief Returns the size of the span + * \return length == 0 + */ + constexpr size_type size() const noexcept { + return length; + } + + /*! \brief Decreases the size of the span by 1 + */ + constexpr void pop_back() noexcept { + assert(length > 0); + --length; + } + + /*! \brief Clears the span + */ + constexpr void clear() noexcept { + data = nullptr; + length = 0; + } + +private: + pointer data{ nullptr }; //!> Pointer to the underlying data array + size_type length{ 0 }; //!> Size of the span +}; + +} // namespace aco + +#endif // ACO_UTIL_H \ No newline at end of file diff --git a/src/amd/compiler/aco_validate.cpp b/src/amd/compiler/aco_validate.cpp new file mode 100644 index 00000000000..0988d66df3a --- /dev/null +++ b/src/amd/compiler/aco_validate.cpp @@ -0,0 +1,460 @@ +/* + * Copyright © 2018 Valve Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "aco_ir.h" + +#include + +namespace aco { + +#ifndef NDEBUG +void perfwarn(bool cond, const char *msg, Instruction *instr) +{ + if (cond) { + fprintf(stderr, "ACO performance warning: %s\n", msg); + if (instr) { + fprintf(stderr, "instruction: "); + aco_print_instr(instr, stderr); + fprintf(stderr, "\n"); + } + + if (debug_flags & DEBUG_PERFWARN) + exit(1); + } +} +#endif + +void validate(Program* program, FILE * output) +{ + if (!(debug_flags & DEBUG_VALIDATE)) + return; + + bool is_valid = true; + auto check = [&output, &is_valid](bool check, const char * msg, aco::Instruction * instr) -> void { + if (!check) { + fprintf(output, "%s: ", msg); + aco_print_instr(instr, output); + fprintf(output, "\n"); + is_valid = false; + } + }; + + for (Block& block : program->blocks) { + for (aco_ptr& instr : block.instructions) { + + /* check base format */ + Format base_format = instr->format; + base_format = (Format)((uint32_t)base_format & ~(uint32_t)Format::SDWA); + base_format = (Format)((uint32_t)base_format & ~(uint32_t)Format::DPP); + if ((uint32_t)base_format & (uint32_t)Format::VOP1) + base_format = Format::VOP1; + else if ((uint32_t)base_format & (uint32_t)Format::VOP2) + base_format = Format::VOP2; + else if ((uint32_t)base_format & (uint32_t)Format::VOPC) + base_format = Format::VOPC; + else if ((uint32_t)base_format & (uint32_t)Format::VINTRP) + base_format = Format::VINTRP; + check(base_format == instr_info.format[(int)instr->opcode], "Wrong base format for instruction", instr.get()); + + /* check VOP3 modifiers */ + if (((uint32_t)instr->format & (uint32_t)Format::VOP3) && instr->format != Format::VOP3) { + check(base_format == Format::VOP2 || + base_format == Format::VOP1 || + base_format == Format::VOPC || + base_format == Format::VINTRP, + "Format cannot have VOP3A/VOP3B applied", instr.get()); + } + + /* check for undefs */ + for (unsigned i = 0; i < instr->operands.size(); i++) { + if (instr->operands[i].isUndefined()) { + bool flat = instr->format == Format::FLAT || instr->format == Format::SCRATCH || instr->format == Format::GLOBAL; + bool can_be_undef = is_phi(instr) || instr->format == Format::EXP || + instr->format == Format::PSEUDO_REDUCTION || + (flat && i == 1) || (instr->format == Format::MIMG && i == 2) || + ((instr->format == Format::MUBUF || instr->format == Format::MTBUF) && i == 0); + check(can_be_undef, "Undefs can only be used in certain operands", instr.get()); + } + } + + /* check num literals */ + if (instr->isSALU() || instr->isVALU()) { + unsigned num_literals = 0; + for (unsigned i = 0; i < instr->operands.size(); i++) + { + if (instr->operands[i].isLiteral()) { + check(instr->format == Format::SOP1 || + instr->format == Format::SOP2 || + instr->format == Format::SOPC || + instr->format == Format::VOP1 || + instr->format == Format::VOP2 || + instr->format == Format::VOPC, + "Literal applied on wrong instruction format", instr.get()); + + num_literals++; + check(!instr->isVALU() || i == 0 || i == 2, "Wrong source position for Literal argument", instr.get()); + } + } + check(num_literals <= 1, "Only 1 Literal allowed", instr.get()); + + /* check num sgprs for VALU */ + if (instr->isVALU()) { + check(instr->definitions[0].getTemp().type() == RegType::vgpr || + (int) instr->format & (int) Format::VOPC || + instr->opcode == aco_opcode::v_readfirstlane_b32 || + instr->opcode == aco_opcode::v_readlane_b32, + "Wrong Definition type for VALU instruction", instr.get()); + unsigned num_sgpr = 0; + unsigned sgpr_idx = instr->operands.size(); + for (unsigned i = 0; i < instr->operands.size(); i++) + { + if (instr->operands[i].isTemp() && instr->operands[i].regClass().type() == RegType::sgpr) { + check(i != 1 || (int) instr->format & (int) Format::VOP3A, "Wrong source position for SGPR argument", instr.get()); + + if (sgpr_idx == instr->operands.size() || instr->operands[sgpr_idx].tempId() != instr->operands[i].tempId()) + num_sgpr++; + sgpr_idx = i; + } + + if (instr->operands[i].isConstant() && !instr->operands[i].isLiteral()) + check(i == 0 || (int) instr->format & (int) Format::VOP3A, "Wrong source position for constant argument", instr.get()); + } + check(num_sgpr + num_literals <= 1, "Only 1 Literal OR 1 SGPR allowed", instr.get()); + } + + if (instr->format == Format::SOP1 || instr->format == Format::SOP2) { + check(instr->definitions[0].getTemp().type() == RegType::sgpr, "Wrong Definition type for SALU instruction", instr.get()); + for (const Operand& op : instr->operands) { + check(op.isConstant() || op.regClass().type() <= RegType::sgpr, + "Wrong Operand type for SALU instruction", instr.get()); + } + } + } + + switch (instr->format) { + case Format::PSEUDO: { + if (instr->opcode == aco_opcode::p_create_vector) { + unsigned size = 0; + for (const Operand& op : instr->operands) { + size += op.size(); + } + check(size == instr->definitions[0].size(), "Definition size does not match operand sizes", instr.get()); + if (instr->definitions[0].getTemp().type() == RegType::sgpr) { + for (const Operand& op : instr->operands) { + check(op.isConstant() || op.regClass().type() == RegType::sgpr, + "Wrong Operand type for scalar vector", instr.get()); + } + } + } else if (instr->opcode == aco_opcode::p_extract_vector) { + check((instr->operands[0].isTemp()) && instr->operands[1].isConstant(), "Wrong Operand types", instr.get()); + check(instr->operands[1].constantValue() < instr->operands[0].size(), "Index out of range", instr.get()); + check(instr->definitions[0].getTemp().type() == RegType::vgpr || instr->operands[0].regClass().type() == RegType::sgpr, + "Cannot extract SGPR value from VGPR vector", instr.get()); + } else if (instr->opcode == aco_opcode::p_parallelcopy) { + check(instr->definitions.size() == instr->operands.size(), "Number of Operands does not match number of Definitions", instr.get()); + for (unsigned i = 0; i < instr->operands.size(); i++) { + if (instr->operands[i].isTemp()) + check((instr->definitions[i].getTemp().type() == instr->operands[i].regClass().type()) || + (instr->definitions[i].getTemp().type() == RegType::vgpr && instr->operands[i].regClass().type() == RegType::sgpr), + "Operand and Definition types do not match", instr.get()); + } + } else if (instr->opcode == aco_opcode::p_phi) { + check(instr->operands.size() == block.logical_preds.size(), "Number of Operands does not match number of predecessors", instr.get()); + check(instr->definitions[0].getTemp().type() == RegType::vgpr || instr->definitions[0].getTemp().regClass() == s2, "Logical Phi Definition must be vgpr or divergent boolean", instr.get()); + } else if (instr->opcode == aco_opcode::p_linear_phi) { + for (const Operand& op : instr->operands) + check(!op.isTemp() || op.getTemp().is_linear(), "Wrong Operand type", instr.get()); + check(instr->operands.size() == block.linear_preds.size(), "Number of Operands does not match number of predecessors", instr.get()); + } + break; + } + case Format::SMEM: { + if (instr->operands.size() >= 1) + check(instr->operands[0].isTemp() && instr->operands[0].regClass().type() == RegType::sgpr, "SMEM operands must be sgpr", instr.get()); + if (instr->operands.size() >= 2) + check(instr->operands[1].isConstant() || (instr->operands[1].isTemp() && instr->operands[1].regClass().type() == RegType::sgpr), + "SMEM offset must be constant or sgpr", instr.get()); + if (!instr->definitions.empty()) + check(instr->definitions[0].getTemp().type() == RegType::sgpr, "SMEM result must be sgpr", instr.get()); + break; + } + case Format::MTBUF: + case Format::MUBUF: + case Format::MIMG: { + check(instr->operands.size() > 1, "VMEM instructions must have at least one operand", instr.get()); + check(instr->operands[0].hasRegClass() && instr->operands[0].regClass().type() == RegType::vgpr, + "VADDR must be in vgpr for VMEM instructions", instr.get()); + check(instr->operands[1].isTemp() && instr->operands[1].regClass().type() == RegType::sgpr, "VMEM resource constant must be sgpr", instr.get()); + check(instr->operands.size() < 4 || (instr->operands[3].isTemp() && instr->operands[3].regClass().type() == RegType::vgpr), "VMEM write data must be vgpr", instr.get()); + break; + } + case Format::DS: { + for (const Operand& op : instr->operands) { + check((op.isTemp() && op.regClass().type() == RegType::vgpr) || op.physReg() == m0, + "Only VGPRs are valid DS instruction operands", instr.get()); + } + if (!instr->definitions.empty()) + check(instr->definitions[0].getTemp().type() == RegType::vgpr, "DS instruction must return VGPR", instr.get()); + break; + } + case Format::EXP: { + for (unsigned i = 0; i < 4; i++) + check(instr->operands[i].hasRegClass() && instr->operands[i].regClass().type() == RegType::vgpr, + "Only VGPRs are valid Export arguments", instr.get()); + break; + } + case Format::FLAT: + check(instr->operands[1].isUndefined(), "Flat instructions don't support SADDR", instr.get()); + /* fallthrough */ + case Format::GLOBAL: + case Format::SCRATCH: { + check(instr->operands[0].isTemp() && instr->operands[0].regClass().type() == RegType::vgpr, "FLAT/GLOBAL/SCRATCH address must be vgpr", instr.get()); + check(instr->operands[1].hasRegClass() && instr->operands[1].regClass().type() == RegType::sgpr, + "FLAT/GLOBAL/SCRATCH sgpr address must be undefined or sgpr", instr.get()); + if (!instr->definitions.empty()) + check(instr->definitions[0].getTemp().type() == RegType::vgpr, "FLAT/GLOBAL/SCRATCH result must be vgpr", instr.get()); + else + check(instr->operands[2].regClass().type() == RegType::vgpr, "FLAT/GLOBAL/SCRATCH data must be vgpr", instr.get()); + break; + } + default: + break; + } + } + } + assert(is_valid); +} + +/* RA validation */ +namespace { + +struct Location { + Location() : block(NULL), instr(NULL) {} + + Block *block; + Instruction *instr; //NULL if it's the block's live-in +}; + +struct Assignment { + Location defloc; + Location firstloc; + PhysReg reg; +}; + +bool ra_fail(FILE *output, Location loc, Location loc2, const char *fmt, ...) { + va_list args; + va_start(args, fmt); + char msg[1024]; + vsprintf(msg, fmt, args); + va_end(args); + + fprintf(stderr, "RA error found at instruction in BB%d:\n", loc.block->index); + if (loc.instr) { + aco_print_instr(loc.instr, stderr); + fprintf(stderr, "\n%s", msg); + } else { + fprintf(stderr, "%s", msg); + } + if (loc2.block) { + fprintf(stderr, " in BB%d:\n", loc2.block->index); + aco_print_instr(loc2.instr, stderr); + } + fprintf(stderr, "\n\n"); + + return true; +} + +} /* end namespace */ + +bool validate_ra(Program *program, const struct radv_nir_compiler_options *options, FILE *output) { + if (!(debug_flags & DEBUG_VALIDATE_RA)) + return false; + + bool err = false; + aco::live live_vars = aco::live_var_analysis(program, options); + std::vector> phi_sgpr_ops(program->blocks.size()); + + std::map assignments; + for (Block& block : program->blocks) { + Location loc; + loc.block = █ + for (aco_ptr& instr : block.instructions) { + if (instr->opcode == aco_opcode::p_phi) { + for (unsigned i = 0; i < instr->operands.size(); i++) { + if (instr->operands[i].isTemp() && + instr->operands[i].getTemp().type() == RegType::sgpr && + instr->operands[i].isFirstKill()) + phi_sgpr_ops[block.logical_preds[i]].emplace_back(instr->operands[i].getTemp()); + } + } + + loc.instr = instr.get(); + for (unsigned i = 0; i < instr->operands.size(); i++) { + Operand& op = instr->operands[i]; + if (!op.isTemp()) + continue; + if (!op.isFixed()) + err |= ra_fail(output, loc, Location(), "Operand %d is not assigned a register", i); + if (assignments.count(op.tempId()) && assignments[op.tempId()].reg != op.physReg()) + err |= ra_fail(output, loc, assignments.at(op.tempId()).firstloc, "Operand %d has an inconsistent register assignment with instruction", i); + if ((op.getTemp().type() == RegType::vgpr && op.physReg() + op.size() > 256 + program->config->num_vgprs) || + (op.getTemp().type() == RegType::sgpr && op.physReg() + op.size() > program->config->num_sgprs && op.physReg() < program->sgpr_limit)) + err |= ra_fail(output, loc, assignments.at(op.tempId()).firstloc, "Operand %d has an out-of-bounds register assignment", i); + if (!assignments[op.tempId()].firstloc.block) + assignments[op.tempId()].firstloc = loc; + if (!assignments[op.tempId()].defloc.block) + assignments[op.tempId()].reg = op.physReg(); + } + + for (unsigned i = 0; i < instr->definitions.size(); i++) { + Definition& def = instr->definitions[i]; + if (!def.isTemp()) + continue; + if (!def.isFixed()) + err |= ra_fail(output, loc, Location(), "Definition %d is not assigned a register", i); + if (assignments[def.tempId()].defloc.block) + err |= ra_fail(output, loc, assignments.at(def.tempId()).defloc, "Temporary %%%d also defined by instruction", def.tempId()); + if ((def.getTemp().type() == RegType::vgpr && def.physReg() + def.size() > 256 + program->config->num_vgprs) || + (def.getTemp().type() == RegType::sgpr && def.physReg() + def.size() > program->config->num_sgprs && def.physReg() < program->sgpr_limit)) + err |= ra_fail(output, loc, assignments.at(def.tempId()).firstloc, "Definition %d has an out-of-bounds register assignment", i); + if (!assignments[def.tempId()].firstloc.block) + assignments[def.tempId()].firstloc = loc; + assignments[def.tempId()].defloc = loc; + assignments[def.tempId()].reg = def.physReg(); + } + } + } + + for (Block& block : program->blocks) { + Location loc; + loc.block = █ + + std::array regs; + regs.fill(0); + + std::set live; + live.insert(live_vars.live_out[block.index].begin(), live_vars.live_out[block.index].end()); + /* remove killed p_phi sgpr operands */ + for (Temp tmp : phi_sgpr_ops[block.index]) + live.erase(tmp); + + /* check live out */ + for (Temp tmp : live) { + PhysReg reg = assignments.at(tmp.id()).reg; + for (unsigned i = 0; i < tmp.size(); i++) { + if (regs[reg + i]) { + err |= ra_fail(output, loc, Location(), "Assignment of element %d of %%%d already taken by %%%d in live-out", i, tmp.id(), regs[reg + i]); + } + regs[reg + i] = tmp.id(); + } + } + regs.fill(0); + + for (auto it = block.instructions.rbegin(); it != block.instructions.rend(); ++it) { + aco_ptr& instr = *it; + + /* check killed p_phi sgpr operands */ + if (instr->opcode == aco_opcode::p_logical_end) { + for (Temp tmp : phi_sgpr_ops[block.index]) { + PhysReg reg = assignments.at(tmp.id()).reg; + for (unsigned i = 0; i < tmp.size(); i++) { + if (regs[reg + i]) + err |= ra_fail(output, loc, Location(), "Assignment of element %d of %%%d already taken by %%%d in live-out", i, tmp.id(), regs[reg + i]); + } + live.emplace(tmp); + } + } + + for (const Definition& def : instr->definitions) { + if (!def.isTemp()) + continue; + live.erase(def.getTemp()); + } + + /* don't count phi operands as live-in, since they are actually + * killed when they are copied at the predecessor */ + if (instr->opcode != aco_opcode::p_phi && instr->opcode != aco_opcode::p_linear_phi) { + for (const Operand& op : instr->operands) { + if (!op.isTemp()) + continue; + live.insert(op.getTemp()); + } + } + } + + for (Temp tmp : live) { + PhysReg reg = assignments.at(tmp.id()).reg; + for (unsigned i = 0; i < tmp.size(); i++) + regs[reg + i] = tmp.id(); + } + + for (aco_ptr& instr : block.instructions) { + loc.instr = instr.get(); + + /* remove killed p_phi operands from regs */ + if (instr->opcode == aco_opcode::p_logical_end) { + for (Temp tmp : phi_sgpr_ops[block.index]) { + PhysReg reg = assignments.at(tmp.id()).reg; + regs[reg] = 0; + } + } + + if (instr->opcode != aco_opcode::p_phi && instr->opcode != aco_opcode::p_linear_phi) { + for (const Operand& op : instr->operands) { + if (!op.isTemp()) + continue; + if (op.isFirstKill()) { + for (unsigned j = 0; j < op.getTemp().size(); j++) + regs[op.physReg() + j] = 0; + } + } + } + + for (unsigned i = 0; i < instr->definitions.size(); i++) { + Definition& def = instr->definitions[i]; + if (!def.isTemp()) + continue; + Temp tmp = def.getTemp(); + PhysReg reg = assignments.at(tmp.id()).reg; + for (unsigned j = 0; j < tmp.size(); j++) { + if (regs[reg + j]) + err |= ra_fail(output, loc, assignments.at(regs[reg + i]).defloc, "Assignment of element %d of %%%d already taken by %%%d from instruction", i, tmp.id(), regs[reg + j]); + regs[reg + j] = tmp.id(); + } + } + + for (const Definition& def : instr->definitions) { + if (!def.isTemp()) + continue; + if (def.isKill()) { + for (unsigned j = 0; j < def.getTemp().size(); j++) + regs[def.physReg() + j] = 0; + } + } + } + } + + return err; +} +} diff --git a/src/amd/compiler/meson.build b/src/amd/compiler/meson.build new file mode 100644 index 00000000000..73151cad6eb --- /dev/null +++ b/src/amd/compiler/meson.build @@ -0,0 +1,103 @@ +# Copyright © 2018 Valve Corporation + +# Permission is hereby granted, free of charge, to any person obtaining a copy +# of this software and associated documentation files (the "Software"), to deal +# in the Software without restriction, including without limitation the rights +# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +# copies of the Software, and to permit persons to whom the Software is +# furnished to do so, subject to the following conditions: + +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. + +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +# SOFTWARE. + +aco_depends = files('aco_opcodes.py') + +aco_opcodes_h = custom_target( + 'aco_opcodes.h', + input : 'aco_opcodes_h.py', + output : 'aco_opcodes.h', + command : [prog_python, '@INPUT@'], + capture : true, + depend_files : aco_depends, +) + +aco_opcodes_c = custom_target( + 'aco_opcodes.cpp', + input : 'aco_opcodes_cpp.py', + output : 'aco_opcodes.cpp', + command : [prog_python, '@INPUT@'], + capture : true, + depend_files : aco_depends, +) + +aco_builder_h = custom_target( + 'aco_builder.h', + input : 'aco_builder_h.py', + output : 'aco_builder.h', + command : [prog_python, '@INPUT@'], + capture : true, + depend_files : aco_depends, +) + +# Headers-only dependency +idep_aco_headers = declare_dependency( + sources : [aco_opcodes_h], + include_directories : include_directories('.'), +) + +libaco_files = files( + 'aco_dead_code_analysis.cpp', + 'aco_dominance.cpp', + 'aco_instruction_selection.cpp', + 'aco_instruction_selection_setup.cpp', + 'aco_interface.cpp', + 'aco_interface.h', + 'aco_ir.h', + 'aco_assembler.cpp', + 'aco_insert_exec_mask.cpp', + 'aco_insert_NOPs.cpp', + 'aco_insert_waitcnt.cpp', + 'aco_reduce_assign.cpp', + 'aco_register_allocation.cpp', + 'aco_live_var_analysis.cpp', + 'aco_lower_bool_phis.cpp', + 'aco_lower_to_hw_instr.cpp', + 'aco_optimizer.cpp', + 'aco_opt_value_numbering.cpp', + 'aco_print_asm.cpp', + 'aco_print_ir.cpp', + 'aco_scheduler.cpp', + 'aco_ssa_elimination.cpp', + 'aco_spill.cpp', + 'aco_util.h', + 'aco_validate.cpp', +) + +_libaco = static_library( + 'aco', + [libaco_files, aco_opcodes_c, aco_opcodes_h, aco_builder_h], + include_directories : [ + inc_common, inc_compiler, inc_mesa, inc_mapi, inc_amd, inc_amd_common, + ], + dependencies : [ + dep_llvm, dep_thread, dep_elf, dep_libdrm_amdgpu, dep_valgrind, + idep_nir_headers, idep_amdgfxregs_h, + ], + c_args : [c_vis_args], + cpp_args : [cpp_vis_args], + build_by_default : true, +) + +# Also link with aco +idep_aco = declare_dependency( + dependencies : idep_aco_headers, + link_with : _libaco, +) -- 2.30.2