diff options
author | Jason Ekstrand <[email protected]> | 2019-05-07 20:11:08 -0500 |
---|---|---|
committer | Jason Ekstrand <[email protected]> | 2019-05-14 12:30:22 -0500 |
commit | 472ef2f98d32c931a15495a344371ea60a6b633e (patch) | |
tree | b35781aef5db5794a0c9e4a722c086957048a360 /src | |
parent | 096ad8a8099cbcb3c868c08814fbe14ac79ca680 (diff) |
intel/fs/ra: Initialize grf_used with first_non_payload_grf
There's no reason why we need to use the calculated payload_node_count
value which is just first_non_payload_grf aligned up. The grf_used
value will be aligned up to 16 anyway (which is a much bigger alignment)
before being handed off to hardware.
Reviewed-by: Kenneth Graunke <[email protected]>
Diffstat (limited to 'src')
-rw-r--r-- | src/intel/compiler/brw_fs_reg_allocate.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/intel/compiler/brw_fs_reg_allocate.cpp b/src/intel/compiler/brw_fs_reg_allocate.cpp index 35903c4030e..2debd36f231 100644 --- a/src/intel/compiler/brw_fs_reg_allocate.cpp +++ b/src/intel/compiler/brw_fs_reg_allocate.cpp @@ -749,7 +749,7 @@ fs_visitor::assign_regs(bool allow_spilling, bool spill_all) * regs in the register classes back down to real hardware reg * numbers. */ - this->grf_used = payload_node_count; + this->grf_used = this->first_non_payload_grf; for (unsigned i = 0; i < this->alloc.count; i++) { int reg = ra_get_node_reg(g, i); |