From 18372a710028fcbe1ff74f2f727e986c223957ba Mon Sep 17 00:00:00 2001 From: Matt Turner Date: Thu, 17 Apr 2014 15:13:00 -0700 Subject: [PATCH] i965/fs: Copy propagate from load_payload. But only into non-load_payload instructions. Otherwise we would prevent register coalescing from combining identical payloads. --- .../dri/i965/brw_fs_copy_propagation.cpp | 22 +++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp b/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp index 158d0bad507..cc6e86f55e4 100644 --- a/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp +++ b/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp @@ -42,6 +42,7 @@ namespace { /* avoid conflict with opt_copy_propagation_elements */ struct acp_entry : public exec_node { fs_reg dst; fs_reg src; + enum opcode opcode; }; struct block_data { @@ -287,6 +288,10 @@ fs_visitor::try_copy_propagate(fs_inst *inst, int arg, acp_entry *entry) if (entry->src.file == IMM) return false; + if (entry->opcode == SHADER_OPCODE_LOAD_PAYLOAD && + inst->opcode == SHADER_OPCODE_LOAD_PAYLOAD) + return false; + /* Bail if inst is reading more than entry is writing. */ if ((inst->regs_read(this, arg) * inst->src[arg].stride * type_sz(inst->src[arg].type)) > type_sz(entry->dst.type)) @@ -569,7 +574,24 @@ fs_visitor::opt_copy_propagate_local(void *copy_prop_ctx, bblock_t *block, acp_entry *entry = ralloc(copy_prop_ctx, acp_entry); entry->dst = inst->dst; entry->src = inst->src[0]; + entry->opcode = inst->opcode; acp[entry->dst.reg % ACP_HASH_SIZE].push_tail(entry); + } else if (inst->opcode == SHADER_OPCODE_LOAD_PAYLOAD && + inst->dst.file == GRF) { + for (int i = 0; i < inst->sources; i++) { + if (inst->src[i].file == GRF) { + acp_entry *entry = ralloc(copy_prop_ctx, acp_entry); + entry->dst = inst->dst; + entry->dst.reg_offset = i; + entry->src = inst->src[i]; + entry->opcode = inst->opcode; + if (!entry->dst.equals(inst->src[i])) { + acp[entry->dst.reg % ACP_HASH_SIZE].push_tail(entry); + } else { + ralloc_free(entry); + } + } + } } } -- 2.30.2