It is more efficient to compute the child index of the current node
inside the parent node and write the bounds when available. The previous
code could load up to 16 AABBs to compute the new ones. The new code
also only needs 1/7 of the previously used scratch memory. The new code
seems to be around 30% faster (0.5ms) in GOTG on a 6700XT.
Reviewed-by: Natalie Vock <natalie.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39139>
Extracting from an SGPR could cause a wrong RegClass on
the operand which could later lead to selecting VOPD
instructions which falsely operate on the corresponding
VGPR.
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39107>
There is no users for that function, is_volatile is only used in
brw_opt_cse.cpp is_expression() but it access the information using brw_send_inst
struct.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39104>
RGP expects linear indexing across the number of counters/components.
Because a component can be used with different counters, its index
should be re-used when needed.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39100>
This reverts commit dcd9b90aff.
The game falls back to software rendering if any of the enumerated GPUs
have invalid vendorID. This fix causes a regression for systems with an
AMD GPU where the game works perfectly fine, that also happen to have an
Intel iGPU.
The Intel iGPU will be enumerated with vendorID=-1 and this will cause
the game to not use the AMD GPU either. Clearly an issue with the game
but since this workaround is causing more issues than it is solving,
remove the config.
From what I can tell, this fix _did_ work at the time of committing, but
as this is a live service game it gets regularly updated and this buggy
behavior with choosing the device has been introduced recently.
Fixes: dcd9b90aff ("drirc/anv: force_vk_vendor=-1 for Wuthering Waves")
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Acked-by: Taras Pisetskyi <taras.pisetskyi@globallogic.com>
Signed-off-by: llyyr <llyyr.public@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39089>
(('fadd', a, 0.0), a) worked for both a + 0 and a + -0 before, but now
requires explicit pattern after 0d255011ae
Just noticed because it pushed few dEQPs over the instruction limit.
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39151>
Fix compiler error:
../src/util/blob.c:344:8: error: assigning to 'uint8_t *' from
'const void *' discards qualifiers
[-Werror,-Wincompatible-pointer-types-discards-qualifiers]
glibc now provides C23-style type-generic string functions. memchr
returns const void * when passed a const void * argument. Update nul
declaration to const since it's only used to find the null terminator
position and calculate a size.
Fixes: 1c9877327e ("glsl: Add blob.c---a simple interface for serializing data")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39048>
Add astc hdr (float) formats, those get treated identically as the ldr
formats as the blocks have enough metadata to be decoded as float.
Reviewed-by: Rob Clark <rob.clark@oss.qualcomm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38859>
The wayland helper will use uninitialized data, causing undefined
behaviour.
This was because the feedback data was allocated, but not fully
initialized.
After the fix, SDL's vulkantest runs cleanly through valgrind. Before,
it would not.
Signed-off-by: Bram Stolk <b.stolk@gmail.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39133>
Implement a mitigation for VM faults caused by SMEM reading
out of bounds when using robust buffer access.
- Pad uniform and storage buffer allocations with a readonly VM page
- Clamp SMEM offsets that can potentially read past the next page
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38769>
