fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-27 10:08:13 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 2
170477 commits 66 branches 936 tags 952 MiB
Commit graph

157727 commits

Author SHA1 Message Date
Antonio Gomes
2b01934bc8 rusticl: Move nir compilation to Program 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22434>
 2023-04-26 20:49:42 +00:00
Mike Blumenkrantz
feb4103272 zink: prune some validation errors from ci 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22715>
 2023-04-26 19:37:58 +00:00
Alyssa Rosenzweig
1ede7c053b pan/bi: Don't set has_fsub 
All this option controls is fusing fadd(x, fneg(y)) into a single instruction,
but our modifier propagation can already do that without the special cases.

shader-db changes are mostly noise.

total instructions in shared programs: 2698039 -> 2697812 (<.01%)
instructions in affected programs: 156742 -> 156515 (-0.14%)
helped: 280
HURT: 109
helped stats (abs) min: 1.0 max: 12.0 x̄: 1.69 x̃: 1
helped stats (rel) min: 0.08% max: 10.71% x̄: 0.65% x̃: 0.32%
HURT stats (abs)   min: 1.0 max: 8.0 x̄: 2.26 x̃: 2
HURT stats (rel)   min: 0.15% max: 5.77% x̄: 2.08% x̃: 2.22%
95% mean confidence interval for instructions value: -0.80 -0.36
95% mean confidence interval for instructions %-change: -0.05% 0.28%
Inconclusive result (%-change mean confidence interval includes 0).

total cycles in shared programs: 143679.94 -> 143676.33 (<.01%)
cycles in affected programs: 617.98 -> 614.38 (-0.58%)
helped: 148
HURT: 3
helped stats (abs) min: 0.015625 max: 0.1875 x̄: 0.02 x̃: 0
helped stats (rel) min: 0.17% max: 3.33% x̄: 0.71% x̃: 0.57%
HURT stats (abs)   min: 0.015625 max: 0.046875 x̄: 0.03 x̃: 0
HURT stats (rel)   min: 0.66% max: 2.27% x̄: 1.46% x̃: 1.45%
95% mean confidence interval for cycles value: -0.03 -0.02
95% mean confidence interval for cycles %-change: -0.77% -0.56%
Cycles are helped.

total fma in shared programs: 22102.44 -> 22098.33 (-0.02%)
fma in affected programs: 395.98 -> 391.88 (-1.04%)
helped: 123
HURT: 3
helped stats (abs) min: 0.015625 max: 0.1875 x̄: 0.03 x̃: 0
helped stats (rel) min: 0.43% max: 18.75% x̄: 1.81% x̃: 0.89%
HURT stats (abs)   min: 0.015625 max: 0.046875 x̄: 0.03 x̃: 0
HURT stats (rel)   min: 0.66% max: 4.17% x̄: 2.09% x̃: 1.45%
95% mean confidence interval for fma value: -0.04 -0.03
95% mean confidence interval for fma %-change: -2.17% -1.27%
Fma are helped.

total cvt in shared programs: 14735.23 -> 14735.80 (<.01%)
cvt in affected programs: 969.64 -> 970.20 (0.06%)
helped: 179
HURT: 121
helped stats (abs) min: 0.015625 max: 0.09375 x̄: 0.02 x̃: 0
helped stats (rel) min: 0.17% max: 46.15% x̄: 1.13% x̃: 0.31%
HURT stats (abs)   min: 0.015625 max: 0.125 x̄: 0.03 x̃: 0
HURT stats (rel)   min: 0.42% max: 28.57% x̄: 6.62% x̃: 6.25%
95% mean confidence interval for cvt value: -0.00 0.01
95% mean confidence interval for cvt %-change: 1.32% 2.68%
Inconclusive result (value mean confidence interval includes 0).

total quadwords in shared programs: 1460952 -> 1460680 (-0.02%)
quadwords in affected programs: 10216 -> 9944 (-2.66%)
helped: 39
HURT: 5
helped stats (abs) min: 8.0 max: 8.0 x̄: 8.00 x̃: 8
helped stats (rel) min: 0.97% max: 11.11% x̄: 4.53% x̃: 3.45%
HURT stats (abs)   min: 8.0 max: 8.0 x̄: 8.00 x̃: 8
HURT stats (rel)   min: 12.50% max: 25.00% x̄: 15.71% x̃: 14.29%
95% mean confidence interval for quadwords value: -7.74 -4.62
95% mean confidence interval for quadwords %-change: -4.43% -0.04%
Quadwords are helped.

total threads in shared programs: 53538 -> 53537 (<.01%)
threads in affected programs: 2 -> 1 (-50.00%)
helped: 0
HURT: 1

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22687>
 2023-04-26 17:23:27 +00:00
Adam Jackson
218aa77941 egl: Clear EGL_WINDOW_BIT for non-double-buffered EGLConfigs 
EGL windows are not required to support single-buffered rendering,
and it's awful, so let's not.

Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22685>
 2023-04-26 15:44:31 +00:00
Samuel Pitoiset
3f14fd8578 radv: reserve command buffer index for SQTT 
These indexes are used to match command buffers with queue events.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22670>
 2023-04-26 13:57:39 +00:00
Samuel Pitoiset
9c33c4dcbe ac/sqtt: add a helper to get cmdbuf IDs per queue 
These will be used by RADV to implement queue event timings.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22670>
 2023-04-26 13:57:39 +00:00
Samuel Pitoiset
a7b4573c11 ac/sqtt: add rgp_sqtt_marker_cb_id definition 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22670>
 2023-04-26 13:57:39 +00:00
Rhys Perry
5e20fbd424 aco: don't move exec reads around exec writes 
Fixes flickering and blocky plants in Jedi: Fallen Order.

Also fixes flickering squares in The Last of Us Part 1.

fossil-db (navi21):
Totals from 92 (0.07% of 135636) affected shaders:
Instrs: 35324 -> 35354 (+0.08%); split: -0.03%, +0.11%
CodeSize: 189568 -> 189668 (+0.05%); split: -0.03%, +0.08%
Latency: 345305 -> 346529 (+0.35%); split: -0.02%, +0.37%
InvThroughput: 78632 -> 78625 (-0.01%)
SClause: 1955 -> 1972 (+0.87%); split: -0.61%, +1.48%
Copies: 1311 -> 1304 (-0.53%); split: -0.69%, +0.15%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8883
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8878
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22696>
 2023-04-26 13:16:00 +00:00
Mike Blumenkrantz
2c77344048 bump VVL to 1.3.248 
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22672>
 2023-04-26 11:45:28 +00:00
Lionel Landwerlin
5c214117ad intel/tools: add ability to dump out raw kernels data 
Useful for debug.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22703>
 2023-04-26 10:00:54 +00:00
Pavel Ondračka
ca634459c7 r300: add CI list of known rv370 dEQP failures 
We don't have CI, but its still convenient for local testing.

Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22705>
 2023-04-26 09:42:38 +00:00
Pavel Ondračka
db6c3cd13d r300: fix unconditional KIL on R300/R400 
0: KIL -none.1111

Negate is not allowed for texturing opcodes, so the incorrect swizzle
was detected, however later optimization, where we try to rewrite incorrect
swizzles from constant (immediate) registers by adding a new ones with
correct order was interfering and not handling this correctly, so we
ended with

CONST[0] = {    -1.0000    -1.0000    -1.0000    -1.0000 }
  0: KIL const[0].xyz-w;

Even if it would get the swizzle right, texturing opcodes can't read from
constant registers, so just skip it and let this be handled by a later
part which inserts an extra mov instead.

Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Fixes: a8e1e5b5c2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22704>
 2023-04-26 09:25:13 +00:00
Martin Roukala (né Peres)
7d4eb78e1b radv/ci: document another vkcts flake on vega10 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22702>
 2023-04-26 08:32:56 +00:00
Erik Faye-Lund
b9ba2c272a glsl: remove ir_state_slot::swizzle 
Same story as with the NIR counterpart in the previous commit.

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22620>
 2023-04-26 05:51:39 +00:00
Erik Faye-Lund
4e8b532db3 nir: remove nir_state_slot::swizzle 
This is only ever written to, never read from. Let's just get rid of it!

This also saves us a few needless includes.

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22620>
 2023-04-26 05:51:39 +00:00
Mike Blumenkrantz
4c47d83051 zink: use EXT_shader_object to implement generic separate shader precompile 
this adds precompile for all separate shader stages (+tcs,tes,geom)
using separate shaders, which should eliminate stuttering for games
using it (e.g., Tomb Raider)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
234f9953a2 zink: fix longstanding TODO for generated tcs 
with dynamic pcp this doesn't matter, and this should only be reached
in async mode if dynamic pcp is available

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
d5a3e2db89 zink: handle all stages in fixup_io_locations() 
this makes the handling a bit more complex, as both input and output
need to be handled for most stages, and also the per-component handling
gets trickier

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
7a83d6289e zink: move separate shader creation to shader CSO creation 
this is a more logical place for it and also enables u_blitter
shaders to be fast-linked

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
0d448d441c zink: use a more standardized loop for initing separate shader program descriptors 
this should be identical to previous behavior

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
01694e5bf3 zink: assign separate shader prog stages from ctx->shader_stages 
this is functionally equivalent given the checks above which already
restrict which stages can be passed

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
7415627853 zink: move some shader CSO functions around 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
fe095fcdcd zink: switch to a regular loop to wait on precompile shader fences 
even if these aren't done yet, it'll still be faster to wait than
to start compiling new pipelines now

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
ab8499fe00 zink: streamline separate shader descriptor update 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
991d9d9924 zink: simplify separate shader prog init a little 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
38d149ffc7 zink: use intermediate variable for separate shader db resize check 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Mike Blumenkrantz
f46e5f2c0c zink: use intermediate variable for separate shader descriptor update loop 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>
 2023-04-26 05:12:25 +00:00
Qiang Yu
5c287290d8 aco,radv: remove unused aco compile options 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
a4b60295a7 aco,ac/llvm,radv,radeonsi: handle ps bc optimization in nir for radv 
The side effect is removing the aco/llvm backend bc optimization code
and linear/persp_centroid variable.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
df74919bc2 ac/nir/ps: remove used nir_variable if created 
RADV won't do this, so remove them at last.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
bfcf03872e radv: implement nir_load_barycentric_optimize_amd 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
33d683bf09 ac/llvm: remove output variable declaration for radv ps 
radv ps does not support epilog when llvm, so outputs will always
be lowered to exports in nir.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
290c3d360e aco,radv: lower outputs to exports when nir for monolithic ps 
Remove the compiler backend code for outputs to exports.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
d3611af389 aco: support nir_export_amd with ps targets 
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
471418077a ac/nir/ps: add no_color_export option 
For radv which always do ps lower but may use epilog or not.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
c877d26454 ac/nir/ps: use nir_export_dual_src_blend_amd when aco 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
9763b6e0da aco: implement nir_export_dual_src_blend_amd 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
583402a332 aco: move create_fs_dual_src_export_gfx11 above 
Will be used in visit_intrinsic(), content is not changed.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Qiang Yu
eb1fe8c32f nir: add nir_export_dual_src_blend_amd intrinsic 
For GFX11 export dual source blend outputs when ACO.
ACO need a pseudo instruction to emit a block of
code which can't be done in nir currently.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Rhys Perry
01f4addc18 ac/nir/ps: fix null export write mask miss set to 0xf 
Fixes: c182154456 ("ac/nir: add ac_nir_lower_ps")
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>
 2023-04-26 03:27:26 +00:00
Daniel Schürmann
1080ff3971 radv/rt: remove merged VkRayTracingShaderGroupCreateInfoKHR 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22686>
 2023-04-26 02:48:29 +00:00
Daniel Schürmann
b72c50a885 radv/rt: replace uses of pGroups with radv_ray_tracing_group 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22686>
 2023-04-26 02:48:29 +00:00
Friedrich Vock
c809c05f4f radv: Hash pipeline libraries separately 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22686>
 2023-04-26 02:48:29 +00:00
Daniel Schürmann
a98b44cd34 radv/rt: add shader stage indices to radv_ray_tracing_group 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22686>
 2023-04-26 02:48:29 +00:00
Daniel Schürmann
1a0ae06091 radv/rt: rename radv_ray_tracing_module -> radv_ray_tracing_group 
This name better reflects the purpose and content of this struct.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22686>
 2023-04-26 02:48:29 +00:00
Charmaine Lee
c661f38342 svga: set PIPE_CAP_VERTEX_ATTRIB_ELEMENT_ALIGNED_ONLY for VGPU10 device 
Instead of forcing vertex buffer stride to be 4 byte aligned only,
DX10 actually allows the stride to be non 4-byte aligned but the
alignment of an element must be the nearest power of 2 greater or equal to the
width of the element's format, or 4, whichever is less.  So the requirement is
better met with PIPE_CAP_VERTEX_ATTRIB_ELEMENT_ALIGNED_ONLY which if set to
TRUE, the sum of vertex element offset + vertex buffer offset + vertex buffer
stride must be aligned to the vertex attributes component size.
Note: PIPE_CAP_VERTEX_ATTRIB_ELEMENT_ALIGNED_ONLY cannot be set
with other alignment-requiring CAPs, so we have to return 0 for all the
other alignement CAPs.

This avoids some unnecessary software vertex translate fallback.

cc: mesa-stable

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22689>
 2023-04-26 02:29:34 +00:00
Mark Janes
acb2a7d2ec intel/dev: report stepping for TGL systems 
Workaround 14010672564 requires a check for the TGL B0 stepping.

Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22458>
 2023-04-26 02:00:17 +00:00
Mark Janes
47ac056d0f intel/dev: update mesa_defs.json from defect database 
These modifications represent:

 * changes to defects made since Feb 16, 2023
 * changes to automated processing of defect state

Reviewed-by: Rohan Garg <rohan.garg@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22458>
 2023-04-26 02:00:17 +00:00
Illia Polishchuk
45ea17d244 glx: add fail check for current context in another thread 
The GLX spec for glXMakeCurrent (3.3):
"If ctx is current to some other thread, then glXMakeCurrent will generate
a BadAccess error"

The GLX spec for glXCopyContext (3.3):
"If the destination context is current for some thread then a BadAccess
error is generated"

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7961
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Illia Polishchuk <illia.a.polishchuk@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22546>
 2023-04-26 00:20:08 +00:00
Daniel Schürmann
2795cf7422 radv/rt: properly destroy radv_ray_tracing_lib_pipeline on error 
Also return the correct error code.

Fixes: 4dafb69d61 ('radv/rt: defer library_pipeline allocation')
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22684>
 2023-04-25 23:52:49 +00:00