From ace3833293e5aa49bb76e11aa96ad0a01e9538bf Mon Sep 17 00:00:00 2001 From: =?utf8?q?Timur=20Krist=C3=B3f?= Date: Thu, 12 Mar 2020 16:30:58 +0100 Subject: [PATCH] aco: Allow combining TCS output VMEM stores. MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit Some copypasta may have stuck in the code. This was left on false by mistake. Totals (GFX10): Code Size: 254939248 -> 254896008 (-0.02 %) bytes Totals from affected shaders (GFX10): VGPRS: 16196 -> 16212 (0.10 %) Code Size: 1126332 -> 1083092 (-3.84 %) bytes Max Waves: 2336 -> 2334 (-0.09 %) Signed-off-by: Timur Kristóf Reviewed-by: Rhys Perry Part-of: --- src/amd/compiler/aco_instruction_selection.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/amd/compiler/aco_instruction_selection.cpp b/src/amd/compiler/aco_instruction_selection.cpp index e03dbcee9fd..44728127332 100644 --- a/src/amd/compiler/aco_instruction_selection.cpp +++ b/src/amd/compiler/aco_instruction_selection.cpp @@ -3382,7 +3382,7 @@ void visit_store_tcs_output(isel_context *ctx, nir_intrinsic_instr *instr, bool Temp hs_ring_tess_offchip = bld.smem(aco_opcode::s_load_dwordx4, bld.def(s4), ctx->program->private_segment_buffer, Operand(RING_HS_TESS_OFFCHIP * 16u)); Temp oc_lds = get_arg(ctx, ctx->args->oc_lds); - store_vmem_mubuf(ctx, store_val, hs_ring_tess_offchip, vmem_offs.first, oc_lds, vmem_offs.second, elem_size_bytes, write_mask, false, false); + store_vmem_mubuf(ctx, store_val, hs_ring_tess_offchip, vmem_offs.first, oc_lds, vmem_offs.second, elem_size_bytes, write_mask, true, false); } if (write_to_lds) { -- 2.30.2