fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-28 14:40:10 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

i965/fs: extend exec_size halving in the generator

Browse source 
The HW has a restriction that only vertical stride may cross register
boundaries. Previously, this only mattered for SIMD16 instructions where
we needed to use the same regioning parameters as the equivalent SIMD8
instruction but double the exec size. But we need to do the same
splitting for 64-bit instructions as well as instructions with a stride
of 2 (which effectively consume 64 bits per element). Fix up the code to
do the right thing instead of special-casing SIMD16.

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This commit is contained in:
Connor Abbott 2015-08-03 17:44:08 -07:00 • committed by Samuel Iglesias Gonsálvez
parent 4f3888c1ca
commit a5d7e144ea
1 changed files with 10 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
src/mesa/drivers/dri/i965/brw_fs_generator.cpp
View file

@ -65,8 +65,9 @@ brw_reg_from_fs_reg(fs_inst *inst, fs_reg *reg, unsigned gen)
case VGRF:
if (reg->stride == 0) {
brw_reg = brw_vec1_reg(brw_file_from_reg(reg), reg->nr, 0);
} else if (inst->exec_size < 8) {
brw_reg = brw_vec8_reg(brw_file_from_reg(reg), reg->nr, 0);
} else if (inst->exec_size * reg->stride * type_sz(reg->type) <= 32) {
brw_reg = brw_vecn_reg(inst->exec_size, brw_file_from_reg(reg),
reg->nr, 0);
brw_reg = stride(brw_reg, inst->exec_size * reg->stride,
inst->exec_size, reg->stride);
} else {
@ -76,11 +77,14 @@ brw_reg_from_fs_reg(fs_inst *inst, fs_reg *reg, unsigned gen)
* rule implies that elements within a 'Width' cannot cross GRF
* boundaries.
*
* So, for registers with width > 8, we have to use a width of 8
* and trust the compression state to sort out the exec size.
* So, for registers that are large enough, we have to split the exec
* size in two and trust the compression state to sort it out.
*/
brw_reg = brw_vec8_reg(brw_file_from_reg(reg), reg->nr, 0);
brw_reg = stride(brw_reg, 8 * reg->stride, 8, reg->stride);
assert(inst->exec_size / 2 * reg->stride * type_sz(reg->type) <= 32);
brw_reg = brw_vecn_reg(inst->exec_size / 2, brw_file_from_reg(reg),
reg->nr, 0);
brw_reg = stride(brw_reg, inst->exec_size / 2 * reg->stride,
inst->exec_size / 2, reg->stride);
}
brw_reg = retype(brw_reg, reg->type);