AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
193,069 Commits 1 Branch 0 Tags
0c024bbe641b092bbbc751baae54a37642794de0
Commit Graph

193069 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
David Rosca
0c024bbe64 radeonsi/vcn: Add decode DPB buffers as CS dependency 
This is needed to ensure correct synchronization in kernel eg. when it
moves the buffers between VRAM and GTT.

Backport-to: 24.2
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3437
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11624
Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30510>
 2024-08-06 14:09:50 +00:00
Surafel Assefa
979dc41558 vulkan: MESA_VK_ENABLE_SUBMIT_THREAD=0 disables threaded submit 
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30492>
 2024-08-06 13:19:40 +00:00
Iago Toral Quiroga
d58f7a24d1 v3d: do not expose EXT_float_blend 
This extension is all about exposing blending with 32-bit floating
point formats, which V3D doesn't support at all so we should not
be exposing it.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30512>
 2024-08-06 10:33:10 +00:00
Alvin Wong
0413e1f7dc hasvk: Conditionally expose VK_KHR_present_wait 
Gate it behind driconf query for now.

Co-authored-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Acked-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30480>
 2024-08-06 11:39:38 +08:00
Kenneth Graunke
c19e5a0a75 intel/brw: Replace predicated break optimization with a simple peephole 
We can achieve most of what brw_fs_opt_predicated_break() does with
simple peepholes at NIR -> BRW conversion time.

For predicated break and continue, we can simply look at an IF ... ENDIF
sequence after emitting it.  If there's a single instruction between the
two, and it's a BREAK or CONTINUE, then we can move the predicate from
the IF onto the jump, and delete the IF/ENDIF.  Because we haven't built
the CFG at this stage, we only need to remove them from the linked list
of instructions, which is trivial to do.

For the predicated while optimization, we can rely on the fact that we
already did the predicated break optimization, and simply look for a
predicated BREAK just before the WHILE.  If so, we move the predicate
onto the WHILE, invert it, and remove the BREAK.

There are a few cases where this approach does a worse job than the old
one: nir_convert_from_ssa may introduce load_reg and store_reg in blocks
containing break, and nir_trivialize_registers may decide it needs to
insert movs into those blocks.  So, at NIR -> BRW time, we'll actually
emit some MOVs there, which might have been possible to copy propagate
out after later optimizations.

However, the fossil-db results show that it's still pretty competitive.
For instructions, 1017 shaders were helped (average -1.87 instructions),
while only 62 were hurt (average +2.19 instructions).  In affected
shaders, it was -0.08% for instructions.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30498>
 2024-08-05 19:17:55 -07:00
Kenneth Graunke
fad63d6483 intel/brw: Delete the brw_fs_opt_dead_control_flow_eliminate() pass 
With the select peephole gone, this no longer does much of anything.

No instruction changes in fossil-db on Alchemist.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30498>
 2024-08-05 19:17:55 -07:00
Kenneth Graunke
06e8335e11 intel/brw: Delete the brw_fs_opt_peephole_select() pass 
Now that we can handle load_ubo in NIR's peephole select pass, the
backend pass isn't really useful anymore.

fossil-db results on Alchemist show almost no impact:

   Totals:
   Instrs: 150646561 -> 150647106 (+0.00%); split: -0.00%, +0.00%
   Cycles: 12633748945 -> 12633760459 (+0.00%)

   Totals from 261 (0.04% of 630008) affected shaders:
   Instrs: 404946 -> 405491 (+0.13%); split: -0.00%, +0.14%
   Cycles: 23947172 -> 23958686 (+0.05%)

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30498>
 2024-08-05 19:17:55 -07:00
Kenneth Graunke
7c579f448f intel/brw: Mark all UBO access with a direct buffer index as speculative 
UBO loads with a non-indirect buffer index should be safe to perform
speculatively.  With a direct offset, we may sometimes turn them into
push constants, at which point it's just reading a register with no
cost at all.  Otherwise, we access them via messages that use surface
state, and automatically perform bounds checking.  So we shouldn't have
any issues with reading out of bounds and page faulting, for example.

This allows nir_opt_peephole_sel() to operate on load_ubo intrinsics,
so we can turn simple if's with loads on both sides to bcsels.  In some
cases this can collapse a surprising amount of control flow, allowing
other optimizations to work better.

The i965 OpenGL driver used load_uniform intrinsics, which are allowed
in NIR's peephole select pass.  But iris uses the Gallium NIR pass that
translates uniforms to loads from UBO 0, so we haven't been able to take
advantage of NIR's peephole select pass there.  The backend pass was
still able to handle this to some extent, however.

fossil-db results on Alchemist:

   Totals:
   Instrs: 150656329 -> 150645307 (-0.01%); split: -0.01%, +0.00%
   Cycles: 12635230179 -> 12633696811 (-0.01%); split: -0.02%, +0.00%
   Send messages: 7416330 -> 7416261 (-0.00%)
   Spill count: 52471 -> 52473 (+0.00%)
   Fill count: 100818 -> 100803 (-0.01%); split: -0.02%, +0.00%
   Scratch Memory Size: 3197952 -> 3198976 (+0.03%)

   Totals from 1848 (0.29% of 630003) affected shaders:
   Instrs: 1412300 -> 1401278 (-0.78%); split: -0.80%, +0.02%
   Cycles: 1809789567 -> 1808256199 (-0.08%); split: -0.11%, +0.03%
   Send messages: 59829 -> 59760 (-0.12%)
   Spill count: 3870 -> 3872 (+0.05%)
   Fill count: 9693 -> 9678 (-0.15%); split: -0.18%, +0.02%
   Scratch Memory Size: 174080 -> 175104 (+0.59%)

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30498>
 2024-08-05 19:17:55 -07:00
Felix DeGrood
0eed818588 anv/measure: ignore events from reused command buffers 
INTEL_MEASURE currently does not support measuring events in
parallel from reused command buffers. When this case is detected,
warn user and disallow. Fixes observed segfaults in such apps.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30523>
 2024-08-05 23:45:41 +00:00
Timothy Arceri
140ca7e5d7 glsl: fix glsl to nir support for lower precision builtins 
When we switch to the full nir based glsl linker in an upcomming merge
request this is required for existing tests from 8fcf8e7fd4 to
continue to pass, this is because they never exercised glsl to nir so
the missing support went unnoticed.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11456
Fixes: 8fcf8e7fd4 ("glsl: lower builtins to mediump that ignore precision of certain parameters ")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30395>
 2024-08-05 22:59:59 +00:00
Karol Herbst
50ce777edd rusticl/device: Make supported 1Dbuffer formats a strict subset of 1D 
This works around a bug in the OpenCL CTS, which a few drivers here run
into.

Fixes a couple of tests with agx.

Reviewed-by: @LingMan
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30514>
 2024-08-05 22:17:43 +00:00
Iván Briano
f8553f56ac intel/rt: fix terminateOnFirstHit handling 
If TraceRay() is called with the TerminateOnFirstHit flag, we need to
terminate the ray on the first confirmed intersection. This is handled
by the lowering of accept_ray_intersection and it's working fine for the
case of multiple instances of the intersection shader being called.

But if the shader calls reportIntersection() more than once, we were
handling them all and accepting the closest one regardless of the flag.

Check for the flag on every confirmed intersection and, if set, accept
it right there. The subsequent lowering will take care of terminating
handling the ray termination if necessary.

Fixes new test dEQP-VK.ray_tracing_pipeline.amber.flags-accept-first

Cc: mesa-stable

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30418>
 2024-08-05 21:43:36 +00:00
Lionel Landwerlin
c6bf1f02c4 anv: reuse object string for RMV token 
The current code is not handling the potential NULL pointer in
VkDebugUtilsObjectNameInfoEXT::pObjectName

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: e1b9a6e4f3 ("anv: initial RMV support")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30516>
 2024-08-05 21:12:59 +00:00
Lionel Landwerlin
ae9a249dfe vulkan/runtime: allow null/empty debug names 
VkDebugUtilsObjectNameInfoEXT::pObjectName can be NULL [1] :

   "Applications may change the name associated with an object simply
    by calling vkSetDebugUtilsObjectNameEXT again with a new string. If
    pObjectName is either NULL or an empty string, then any previously
    set name is removed."

The current code will segfault.

[1] : https://registry.khronos.org/vulkan/specs/1.3-extensions/html/chap50.html#VkDebugUtilsObjectNameInfoEXT

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 3b361b234a ("vulkan: Implement VK_EXT_debug_utils")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30516>
 2024-08-05 21:12:59 +00:00
Mike Blumenkrantz
33a336309e egl: simplify multibuffers check 
this is all x11 now, so all the extra checks are unnecessary

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
f5a340058d egl: move multibuffers check into platform_x11 
there are no longer any extensions to bind, so this function has no
other purpose

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
8407e674cc egl: delete DRI_MUTABLE_RENDER_BUFFER_DRIVER remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
79b8ea9b76 egl/android: replace DRI_MUTABLE_RENDER_BUFFER_DRIVER check with non-kms_swrast check 
this should be roughly equivalent

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
7452a1d853 egl: delete DRI2_BLOB remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
591bf3c153 egl: inline DRI2_BLOB 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
101e78152e dri: make DRI2_BLOB public 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
24a1651910 egl: delete DRI2_INTEROP remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
ccd0a4c1fe egl: always advertise MESA_gl_interop, inline calls 
this matches GLX since the inner functions return support correctly

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
1d31a15b24 egl: delete DRI2_FENCE remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
3c692131ff egl: inline DRI2_FENCE 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
a19839e25d egl: don't check for create_fence_fd method when importing a fence 
this is always supported

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
50b4be8d48 egl: simplify native fence check 
this is the same conditional during init

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
b23acb801e egl: delete checks for DRI2_FENCE 
this is supported everywhere

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
a19ba7d9f4 dri: make DRI2_FENCE public 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
c684a9b918 egl: delete DRI_TEX_BUFFER remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
f6c649a0ef egl: inline DRI_TEX_BUFFER 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
cb018b6883 egl: delete DRI_CONFIG_QUERY remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
d723991ded egl: inline DRI2_CONFIG_QUERY 
this is always supported

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
197e4f2ae2 egl: delete DRI_FLUSH_CONTROL remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
6b7e6e3bdc egl: always advertise KHR_context_flush_control 
this is supported by all dri interfaces

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
0ae2958452 egl: delete DRI_BUFFER_DAMAGE remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
591bc52e3e egl: inline DRI_BUFFER_DAMAGE 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
77e85377ae egl: use pipe_screen check for damage region, simplify buffer_damage checks 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
e7b303c680 egl: add a util to get the pipe_screen 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
9d41f1f552 dri: make set_damage_region public 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
85d459fcec egl: simplify DRI_BUFFER_DAMAGE checks 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:15 +00:00
Mike Blumenkrantz
f6afe277d3 dri: only add DRI_BUFFER_DAMAGE if the driver supports set_damage_region 
100% of callers check both

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
f8fbec8e63 egl: delete DRI2_FLUSH remnants 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
fd17c018e9 egl/wayland: replace a couple DRI2_FLUSH checks with non-kms_swrast checks 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
a98434d257 egl/drm: replace a DRI2_FLUSH check with a non-kms_swarst check 
only (non-kms) swrast doesn't support this

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
b1f1046857 egl: replace a DRI2_FLUSH check with swrast check in x11_copy_buffers 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
1d41299109 egl: replace another DRI2_FLUSH check with a non-kms_swrast check in surface query 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
3b050af311 egl: inline DRI2_FLUSH usage 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
09b52c61bf egl: replace another DRI2_FLUSH check with non-kms_swrast check in dri2_wait_client 
only (non-kms) swrast doesn't support this

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00
Mike Blumenkrantz
e5f9d004bb egl: delete a DRI2_FLUSH check in dri2 swapbuffers 
this is always true here

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30494>
 2024-08-05 20:33:14 +00:00