fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-22 13:30:12 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

intel/brw: Use size_written for NoMask instructions in is_partial_write

Browse source 
The intention of inst->is_partial_write() is that it should return true
when any REG_SIZE (32B) chunk of inst's destination is written but not
fully overwritten.  This can be used to tell whether inst combines new
data with existing data, or screens off any previous writes, so the old
values are no longer required.

The existing (exec_size * brw_type_size_bytes(this->dst.type) < 32)
check doesn't work in a number of cases.  For example, LSC block loads
have exec_size == 1 and force_writemask_all set, but may write multiple
full registers of data.  (Currently, we only see them with exec_size 1
after logical-send-lowering, so our SHADER_OPCODE_SEND special case
was covering those.)  We had also special cased UNDEF.

Instead, we can simply check:

   1. Predication
   2. !inst->dst.contiguous()
   3. inst->dst.offset % REG_SIZE != 0
   4. inst->size_written % REG_SIZE != 0

We had the first three already, but #4 is new.  If either #3 or #4
are true, then that implies there is a REG_SIZE chunk of the destination
which is written, but not entirely written, so it's a partial write.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30828>
This commit is contained in:
Kenneth Graunke 2024-08-26 15:34:21 -07:00 • committed by Marge Bot
parent ab0b9b6792
commit e8883bd40b
1 changed files with 4 additions and 17 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

21
src/intel/compiler/brw_fs.cpp
View file

@ -617,26 +617,13 @@ fs_inst::is_partial_write() const
this->opcode != BRW_OPCODE_SEL) this->opcode != BRW_OPCODE_SEL)
return true; return true;
if (!this->dst.is_contiguous())
return true;
if (this->dst.offset % REG_SIZE != 0) if (this->dst.offset % REG_SIZE != 0)
return true; return true;
/* SEND instructions always write whole registers */ return this->size_written % REG_SIZE != 0;
if (this->opcode == SHADER_OPCODE_SEND)
return false;
/* Special case UNDEF since a lot of places in the backend do things like this :
*
* fs_builder ubld = bld.exec_all().group(1, 0);
* brw_reg tmp = ubld.vgrf(BRW_TYPE_UD);
* ubld.UNDEF(tmp); <- partial write, even if the whole register is concerned
*/
if (this->opcode == SHADER_OPCODE_UNDEF) {
assert(this->dst.is_contiguous());
return this->size_written < 32;
}
return this->exec_size * brw_type_size_bytes(this->dst.type) < 32 ||
!this->dst.is_contiguous();
} }
unsigned unsigned