fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-03-21 03:10:35 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

gallivm: handle u8/u16 const loads properly on big-endian.
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
Details
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Details

Browse source 
Turns out just putting the u32 in doesn't work on big endian, so
put the proper u8/u16 values in.

Got a report that since the loop limiter got removed, a gtk4 blur
shader was looping forever on s390x. Turns out it was using a 16-bit
loop variable (because why wouldn't you), and the loop counter was
just staying at 0 all the time.

Cc: mesa-stable
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37593>
This commit is contained in:
Dave Airlie 2025-09-26 12:21:24 +10:00
parent f90e0f0797
commit e28cfb2bad
1 changed files with 4 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
src/gallium/auxiliary/gallivm/lp_bld_nir_soa.c
View file

@ -1267,8 +1267,10 @@ emit_load_const(struct lp_build_nir_soa_context *bld,
for (unsigned i = 0; i < instr->def.num_components; i++) { for (unsigned i = 0; i < instr->def.num_components; i++) {
outval[i] = lp_build_const_int_vec(bld->base.gallivm, int_bld->type, outval[i] = lp_build_const_int_vec(bld->base.gallivm, int_bld->type,
bits == 32 ? instr->value[i].u32 bits == 8 ? instr->value[i].u8 :
: instr->value[i].u64); bits == 16 ? instr->value[i].u32 :
bits == 32 ? instr->value[i].u32 :
instr->value[i].u64);
} }
for (unsigned i = instr->def.num_components; i < NIR_MAX_VEC_COMPONENTS; i++) { for (unsigned i = instr->def.num_components; i < NIR_MAX_VEC_COMPONENTS; i++) {
outval[i] = NULL; outval[i] = NULL;