From ea35fb0fbead2902b1ba37e7cdb1523853fabd8b Mon Sep 17 00:00:00 2001 From: Jason Ekstrand Date: Wed, 30 Sep 2015 10:31:03 -0700 Subject: [PATCH] i965/vec4: Use the actual channels used in pack_uniform_registers Previously, pack_uniform_registers worked based on the size of the uniform as given to us when we initially set up the uniforms. However, we have to walk through the uniforms and figure out liveness anyway, so we migh as well record the number of channels used as we go. This may also allow us to pack things tighter in a few cases. Reviewed-by: Iago Toral Quiroga --- src/mesa/drivers/dri/i965/brw_vec4.cpp | 51 +++++++++++++++++++------- 1 file changed, 37 insertions(+), 14 deletions(-) diff --git a/src/mesa/drivers/dri/i965/brw_vec4.cpp b/src/mesa/drivers/dri/i965/brw_vec4.cpp index 407698f5573..057537ecae3 100644 --- a/src/mesa/drivers/dri/i965/brw_vec4.cpp +++ b/src/mesa/drivers/dri/i965/brw_vec4.cpp @@ -518,11 +518,11 @@ vec4_visitor::split_uniform_registers() void vec4_visitor::pack_uniform_registers() { - bool uniform_used[this->uniforms]; + uint8_t chans_used[this->uniforms]; int new_loc[this->uniforms]; int new_chan[this->uniforms]; - memset(uniform_used, 0, sizeof(uniform_used)); + memset(chans_used, 0, sizeof(chans_used)); memset(new_loc, 0, sizeof(new_loc)); memset(new_chan, 0, sizeof(new_chan)); @@ -531,11 +531,36 @@ vec4_visitor::pack_uniform_registers() * to pull constants, and from some GLSL code generators like wine. */ foreach_block_and_inst(block, vec4_instruction, inst, cfg) { + unsigned readmask; + switch (inst->opcode) { + case VEC4_OPCODE_PACK_BYTES: + case BRW_OPCODE_DP4: + case BRW_OPCODE_DPH: + readmask = 0xf; + break; + case BRW_OPCODE_DP3: + readmask = 0x7; + break; + case BRW_OPCODE_DP2: + readmask = 0x3; + break; + default: + readmask = inst->dst.writemask; + break; + } + for (int i = 0 ; i < 3; i++) { - if (inst->src[i].file != UNIFORM) - continue; + if (inst->src[i].file != UNIFORM) + continue; - uniform_used[inst->src[i].reg] = true; + int reg = inst->src[i].reg; + for (int c = 0; c < 4; c++) { + if (!(readmask & (1 << c))) + continue; + + chans_used[reg] = MAX2(chans_used[reg], + BRW_GET_SWZ(inst->src[i].swizzle, c) + 1); + } } } @@ -546,17 +571,15 @@ vec4_visitor::pack_uniform_registers() */ for (int src = 0; src < uniforms; src++) { assert(src < uniform_array_size); - int size = this->uniform_vector_size[src]; + int size = chans_used[src]; - if (!uniform_used[src]) { - this->uniform_vector_size[src] = 0; - continue; - } + if (size == 0) + continue; int dst; /* Find the lowest place we can slot this uniform in. */ for (dst = 0; dst < src; dst++) { - if (this->uniform_vector_size[dst] + size <= 4) + if (chans_used[dst] + size <= 4) break; } @@ -565,7 +588,7 @@ vec4_visitor::pack_uniform_registers() new_chan[src] = 0; } else { new_loc[src] = dst; - new_chan[src] = this->uniform_vector_size[dst]; + new_chan[src] = chans_used[dst]; /* Move the references to the data */ for (int j = 0; j < size; j++) { @@ -573,8 +596,8 @@ vec4_visitor::pack_uniform_registers() stage_prog_data->param[src * 4 + j]; } - this->uniform_vector_size[dst] += size; - this->uniform_vector_size[src] = 0; + chans_used[dst] += size; + chans_used[src] = 0; } new_uniform_count = MAX2(new_uniform_count, dst + 1); -- 2.30.2