<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Mon, Sep 11, 2017 at 8:00 AM, Jose Maria Casanova Crespo <<a href="mailto:jmcasanova@igalia.com" target="_blank">jmcasanova@igalia.com</a>> wrote: <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">From: Alejandro Piñeiro <<a href="mailto:apinheiro@igalia.com">apinheiro@igalia.com</a>> Although from SPIR-V point of view, rounding modes are attached to the operation/destination, on i965 it is a status, so we don't need to explicitly set the rounding mode if the one we want is already set. Taking into account that the default mode is RTE, one possible optimization would be optimize out the first RTE set for each block. For in order to work, we would need to take into account block interrelationships. At this point, it is not worth to complicate the optimization for such small gain. v2: Use a single SHADER_OPCODE_RND_MODE opcode taking an immediate with the rounding mode (Curro) v3: Reset optimization for every block. (Jason Ekstrand) Signed-off-by: Jose Maria Casanova Crespo <<a href="mailto:jmcasanova@igalia.com">jmcasanova@igalia.com</a>> Signed-off-by: Alejandro Piñeiro <<a href="mailto:apinheiro@igalia.com">apinheiro@igalia.com</a>> --- src/intel/compiler/brw_eu_defines.h | 3 ++- src/intel/compiler/brw_fs.cpp | 37 +++++++++++++++++++++++++++++++++++++ src/intel/compiler/brw_fs.h | 1 + 3 files changed, 40 insertions(+), 1 deletion(-) diff --git a/src/intel/compiler/brw_eu_defines.h b/src/intel/compiler/brw_eu_defines.h index 72f773352e..aea58284bc 100644 --- a/src/intel/compiler/brw_eu_defines.h +++ b/src/intel/compiler/brw_eu_defines.h @@ -1227,7 +1227,8 @@ enum PACKED brw_rnd_mode { BRW_RND_MODE_RTNE = 0, /* Round to Nearest or Even */ BRW_RND_MODE_RU = 1, /* Round Up, toward +inf */ BRW_RND_MODE_RD = 2, /* Round Down, toward -inf */ - BRW_RND_MODE_RTZ = 3 /* Round Toward Zero */ + BRW_RND_MODE_RTZ = 3, /* Round Toward Zero */ + BRW_RND_MODE_UNSPECIFIED /* Unspecified rounding mode */ </blockquote><div> </div><div>In the future, please end enum lists with commas. The C grammar allows it and it means you don't have an extra +- line if you add an entry to the end.</div><div> </div><div>Reviewed-by: Jason Ekstrand <<a href="mailto:jason@jlekstrand.net">jason@jlekstrand.net</a>> </div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> }; #endif /* BRW_EU_DEFINES_H */ diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp index 45236e3cab..9b33a32917 100644 --- a/src/intel/compiler/brw_fs.cpp +++ b/src/intel/compiler/brw_fs.cpp @@ -3071,6 +3071,42 @@ fs_visitor::remove_duplicate_mrf_writes() return progress; } +/** + * Rounding modes for conversion instructions are included for each + * conversion, but right now it is a state. So once it is set, + * we don't need to call it again for subsequent calls. + * + * This is useful for vector/matrices conversions, as setting the + * mode once is enough for the full vector/matrix + */ +bool +fs_visitor::remove_extra_rounding_modes() +{ + bool progress = false; + + foreach_block (block, cfg) { + brw_rnd_mode prev_mode = BRW_RND_MODE_UNSPECIFIED; + + foreach_inst_in_block_safe (fs_inst, inst, block) { + if (inst->opcode == SHADER_OPCODE_RND_MODE) { + assert(inst->src[0].file == BRW_IMMEDIATE_VALUE); + const brw_rnd_mode mode = (brw_rnd_mode) inst->src[0].d; + if (mode == prev_mode) { + inst->remove(block); + progress = true; + } else { + prev_mode = mode; + } + } + } + } + + if (progress) + invalidate_live_intervals(); + + return progress; +} + static void clear_deps_for_inst_src(fs_inst *inst, bool *deps, int first_grf, int grf_len) { @@ -5748,6 +5784,7 @@ fs_visitor::optimize() int pass_num = 0; OPT(opt_drop_redundant_mov_to_flags); + OPT(remove_extra_rounding_modes); do { progress = false; diff --git a/src/intel/compiler/brw_fs.h b/src/intel/compiler/brw_fs.h index f1ba193de7..b9476e69ed 100644 --- a/src/intel/compiler/brw_fs.h +++ b/src/intel/compiler/brw_fs.h @@ -150,6 +150,7 @@ public: bool eliminate_find_live_channel(); bool dead_code_eliminate(); bool remove_duplicate_mrf_writes(); + bool remove_extra_rounding_modes(); bool opt_sampler_eot(); bool virtual_grf_interferes(int a, int b); -- 2.13.5 </blockquote></div> </div></div>