pan/midgard: Add perspective projection recombine pass

We don't use it yet, since it's actually a shader-db regression. This is
primarily helpful as an intermediate step for attaching projection to
varyings.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>

diff --git a/src/panfrost/midgard/compiler.h b/src/panfrost/midgard/compiler.h
index 804180e571a18bb9cebf809cedb6ad607039ce70..1daf876a0f453ad3446b7fefe953f79f18ead745 100644 (file)
--- a/src/panfrost/midgard/compiler.h
+++ b/src/panfrost/midgard/compiler.h
@@ -498,5 +498,6 @@ nir_clamp_psiz(nir_shader *shader, float min_size, float max_size);
 /* Optimizations */
 
 bool midgard_opt_copy_prop(compiler_context *ctx, midgard_block *block);
+bool midgard_opt_combine_projection(compiler_context *ctx, midgard_block *block);
 
 #endif

diff --git a/src/panfrost/midgard/meson.build b/src/panfrost/midgard/meson.build
index ead81cf2a5b4269ff365d1e46ab1e5fb4ca1ce68..da3ea5dd4dfc91e704b0ca0c23900f919eb18c7b 100644 (file)
--- a/src/panfrost/midgard/meson.build
+++ b/src/panfrost/midgard/meson.build
@@ -31,6 +31,7 @@ libpanfrost_midgard_files = files(
   'midgard_ops.c',
   'mir_promote_uniforms.c',
   'midgard_opt_copy_prop.c',
+  'midgard_opt_perspective.c',
   'cppwrap.cpp',
   'disassemble.c',
 )

diff --git a/src/panfrost/midgard/midgard_opt_perspective.c b/src/panfrost/midgard/midgard_opt_perspective.c
new file mode 100644 (file)
index 0000000..7abfd15
--- /dev/null
+++ b/src/panfrost/midgard/midgard_opt_perspective.c
@@ -0,0 +1,115 @@
+/*
+ * Copyright (C) 2019 Collabora, Ltd.
+ *
+ * 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.
+ */
+
+/* Midgard has some accelerated support for perspective projection on the
+ * load/store pipes. So the first perspective projection pass looks for
+ * lowered/open-coded perspective projection of the form "fmul (A.xyz,
+ * frcp(A.w))" or "fmul (A.xy, frcp(A.z))" and rewrite with a native
+ * perspective division opcode (on the load/store pipe).
+ *
+ * Caveats apply: the frcp should be used only once to make this optimization
+ * worthwhile.
+ */
+
+#include "compiler.h"
+
+bool
+midgard_opt_combine_projection(compiler_context *ctx, midgard_block *block)
+{
+        bool progress = false;
+
+        mir_foreach_instr_in_block_safe(block, ins) {
+                /* First search for fmul */
+                if (ins->type != TAG_ALU_4) continue;
+                if (ins->alu.op != midgard_alu_op_fmul) continue;
+
+                /* TODO: Flip */
+
+                /* Check the swizzles */
+                
+                midgard_vector_alu_src src1 =
+                        vector_alu_from_unsigned(ins->alu.src1);
+
+                midgard_vector_alu_src src2 =
+                        vector_alu_from_unsigned(ins->alu.src2);
+
+                if (!mir_is_simple_swizzle(src1.swizzle, ins->mask)) continue;
+                if (src2.swizzle != SWIZZLE_XXXX) continue;
+
+                /* Awesome, we're the right form. Now check where src2 is from */
+                unsigned frcp = ins->ssa_args.src1;
+                unsigned to = ins->ssa_args.dest;
+
+                if (frcp >= ctx->func->impl->ssa_alloc) continue;
+                if (to >= ctx->func->impl->ssa_alloc) continue;
+
+                bool frcp_found = false;
+                unsigned frcp_component = 0;
+                unsigned frcp_from = 0;
+
+                mir_foreach_instr_in_block_safe(block, sub) {
+                        if (sub->ssa_args.dest != frcp) continue;
+
+                        midgard_vector_alu_src s =
+                                vector_alu_from_unsigned(sub->alu.src1);
+
+                        frcp_component = s.swizzle & 3;
+                        frcp_from = sub->ssa_args.src0;
+
+                        frcp_found =
+                                (sub->type == TAG_ALU_4) &&
+                                (sub->alu.op == midgard_alu_op_frcp);
+                        break;
+                }
+
+                if (!frcp_found) continue;
+                if (frcp_component != COMPONENT_W && frcp_component != COMPONENT_Z) continue;
+                if (!mir_single_use(ctx, frcp)) continue;
+
+                /* Nice, we got the form spot on. Let's convert! */
+
+                midgard_instruction accel = {
+                        .type = TAG_LOAD_STORE_4,
+                        .mask = ins->mask,
+                        .ssa_args = {
+                                .dest = to,
+                                .src0 = frcp_from,
+                                .src1 = -1
+                        },
+                        .load_store = {
+                                .op = frcp_component == COMPONENT_W ?
+                                        midgard_op_ldst_perspective_division_w : 
+                                        midgard_op_ldst_perspective_division_z,
+                                .swizzle = SWIZZLE_XYZW,
+                                .unknown = 0x24,
+                        }
+                };
+
+                mir_insert_instruction_before(ins, accel);
+                mir_remove_instruction(ins);
+
+                progress |= true;
+        }
+
+        return progress;
+}