Italo Nicola [Fri, 9 Jul 2021 10:27:01 +0000 (07:27 -0300)]
virgl: flush cmd buffer when flushing frontbuffer
When a resource is multisampled, we usually submit a multisampling
resolving blit before we present it or use it in some other way, but
currently we don't always flush the cmd buffer before flushing the
frontbuffer, this commit fixes that.
Fixes piglit's glx/glx-copy-sub-buffer MSAA cases on vtest, in
conjunction with other commits of this series.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11714>
Italo Nicola [Mon, 5 Jul 2021 09:22:16 +0000 (06:22 -0300)]
virgl/vtest: implement resource_create_front
This is required for glXCopySubBufferMESA to work.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11714>
Italo Nicola [Fri, 9 Jul 2021 10:47:29 +0000 (07:47 -0300)]
virgl/vtest: use correct resource stride in flush_frontbuffer
The displaytarget's resource stride is alignment is currently 64-bytes,
where the shared resource stride is unaligned.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11714>
Caio Oliveira [Sat, 4 Dec 2021 00:58:47 +0000 (16:58 -0800)]
util: Use ralloc for strings in cache test
Also avoid warnings about asprintf result not being checked.
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14054>
Caio Oliveira [Fri, 3 Dec 2021 21:34:08 +0000 (13:34 -0800)]
util: Convert cache test to use gtest
Replace a bunch of helper functions for checking results with ones
from GTest.
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14054>
Jason Ekstrand [Wed, 3 Nov 2021 13:59:53 +0000 (08:59 -0500)]
intel/dev: Add gtt_size to devinfo
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13647>
Jason Ekstrand [Thu, 2 Dec 2021 20:42:16 +0000 (14:42 -0600)]
anv: Stop doing too much per-sample shading
We were setting anv_pipeline::sample_shading_enable based on
sampleShadingEnable without looking at minSampleShading. We would then
pass this value into nir_lower_wpos_center which would add sample_pos to
frag_coord. Then the back-end compiler picks up on the existence of
sample_pos and forces persample dispatch. This leads to doing
per-sample dispatch whenever sampleShadingEnable = VK_TRUE regardless of
the value of minSampleShading. This is almost certainly costing us
perf somewhere.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14022>
Nanley Chery [Fri, 6 Aug 2021 22:37:21 +0000 (15:37 -0700)]
iris: Update the initial CCS state on XeHP
We can't map the CCS on this platform to initialize it into the
PASS_THROUGH state. This can cause issues with optimizations in the
driver that rely on this state.
For example, after rendering to a surface with AUX_NONE, we can then
render to it with AUX_CCS_E without an ambiguate in between (if the CCS
in the PASS_THROUGH state). If that state was incorrect and the aux was
actually compressed, there can be rendering corruption because the
contents may be misinterpreted on the second render.
Use a more accurate initial aux state to avoid these issues.
One notable change in behavior here is that aux surfaces can be created
with fast-cleared blocks even though the caller may specify a modifier
that doesn't support fast clears. This should be fine, so long as all HW
units that can access these surfaces can handle that bit-pattern. We
haven't seen an applicable restriction yet.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Thu, 21 Oct 2021 22:49:11 +0000 (15:49 -0700)]
iris: Modify the comment about zeroing CCS
Among other changes, we highlight the fact that we'll map the CCS -
something we can't do on XeHP.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Fri, 22 Oct 2021 16:36:49 +0000 (09:36 -0700)]
iris: Don't assert a NULL aux BO during aux config
The assert was introduced in a function that allocated an auxiliary
surface BO, iris_resource_alloc_aux. After refactors, the function it's
in now, iris_resource_configure_aux, no longer does this allocation.
Drop the assert because its purpose is unclear and it's no longer
relevant for CCS on XeHP.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Wed, 20 Oct 2021 20:30:33 +0000 (13:30 -0700)]
iris: Don't allocate and initialize CCS on XeHP
The memory for CCS on XeHP can't be mapped by the CPU.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Mon, 25 Oct 2021 20:39:34 +0000 (13:39 -0700)]
iris: Drop row pitch param from iris_get_ccs_surf
This parameter won't be used for XeHP, because we can't directly control
the row pitch of the CCS independently from the main surface.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Mon, 18 Oct 2021 19:38:46 +0000 (12:38 -0700)]
iris: Don't allocate a clear color BO for some Z/S
The only depth/stencil aux usage that can actually use the BO is
ISL_AUX_USAGE_HIZ_CCS_WT. Even with that aux usage, iris may disable
sampling depending on the surface configuration.
Allocate the clear color BO when it'd be usable, not just when the
auxiliary surface size is non-zero on ICL+. This prepares for CCS on
XeHP, which won't have an auxiliary surface.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Mon, 18 Oct 2021 19:02:28 +0000 (12:02 -0700)]
iris: Simplify iris_get_aux_clear_color_state_size
isl_dev.ss.clear_color_state_size is already zero on BDW and SKL. Drop
the redundant platform check and return the field directly.
We're going to have this function return zero more often and it will do
so uniformly using if-statements. We choose to remove the redundant
expression instead of adding a redundant if-statement.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Wed, 20 Oct 2021 00:26:34 +0000 (17:26 -0700)]
iris: Move some BO setup to iris_resource_init_aux_buf
To ease verification, place the assignment and reference of the aux BO
right before the same operations are done for the clear color BO. Also,
move the call to map_aux_addresses that's in the same if-block.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Tue, 19 Oct 2021 15:12:30 +0000 (08:12 -0700)]
iris: Use the aux BO and surf less during init
res->aux.bo and res->aux.surf will be NULL and zeroed, respectively, for
CCS on XeHP. Move and modify iris_resource_init_aux_buf to support this.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Tue, 19 Oct 2021 15:12:30 +0000 (08:12 -0700)]
iris: Change a param of iris_resource_init_aux_buf
Have iris_resource_init_aux_buf compute the clear color state size
(with an iris_screen struct) instead of passing it in directly.
We're going to move the function call soon. This keeps us from having to
move a passed in variable along with it.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Mon, 25 Oct 2021 18:18:24 +0000 (11:18 -0700)]
intel/blorp: Modify get_fast_clear_rect for XeHP
The alignment and scale down values have changed on this platform.
To support drivers that won't use a CCS surface on this platform, this
patch computes the CCS fast clear rectangle using the main surface.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Sat, 23 Oct 2021 00:13:13 +0000 (17:13 -0700)]
intel/blorp: Modify the SKL+ CCS resolve rectangle
According to Bspec 2424, "Render Target Resolve":
The Resolve Rectangle size is same as Clear Rectangle size from SKL+.
Use get_fast_clear_rect in blorp_ccs_resolve for SKL+.
Note that the Bspec differs from Vol7 of the Sky Lake PRM, which only
specifies aligning by the scaledown factors.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Nanley Chery [Tue, 26 Oct 2021 17:03:15 +0000 (10:03 -0700)]
intel/isl: Require aux map for some 64K alignment
The comment states that 64K alignment of surfaces is required when an
aux map is present on the platform. However, the code checks for GFX12
instead of dev->info->has_aux_map. Update the code to match the comment.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13555>
Jesse Natalie [Fri, 10 Dec 2021 18:42:28 +0000 (10:42 -0800)]
ci/windows: Remove line numbers from assertions in spirv2dxil tests
Reviewed-by: Enrico Galli <enrico.galli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14156>
Lucas Stach [Wed, 6 Oct 2021 14:26:57 +0000 (16:26 +0200)]
etnaviv: fix alpha blend with dither on older GPUs
While setting up DITHER_MODE allows alpha blending to work properly
together with dithering on new GPUs (those with PE_DITHER_FIX), older
cores still change the render target. As dithering is optional and
implementation defined we can simply disable it on the affected GPUs,
when alpha blending is enabled to work around this bug.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13396>
Emma Anholt [Thu, 9 Dec 2021 21:09:35 +0000 (13:09 -0800)]
nir/nir_opt_move,sink: Include load_ubo_vec4 as a load_ubo instr.
We weren't doing much motion in nir-to-tgsi because we considered all our
lowered load-ubos as unmovable.
softpipe shader-db:
total temps in shared programs: 563942 -> 563136 (-0.14%)
temps in affected programs: 9833 -> 9027 (-8.20%)
r300 shader-db:
instructions in affected programs: 22858 -> 23575 (3.14%)
temps in affected programs: 2039 -> 1813 (-11.08%)
(NIR had given r300 -19% instrs for +40% temps, so this feels like a
worthwhile trade back).
Reivewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14138>
Erico Nunes [Fri, 26 Nov 2021 18:59:29 +0000 (19:59 +0100)]
mesa: fix GL_MAX_SAMPLES with GLES2
EXT_multisampled_render_to_texture on GLES2 allows the
GL_MAX_SAMPLES_EXT enum to be used.
Move the condition from the GLES3 section to the GLES2 one so
that it stops returning GL_INVALID_ENUM in that case.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13967>
Silvestrs Timofejevs [Tue, 2 Apr 2019 15:36:22 +0000 (16:36 +0100)]
egl: add config debug printout
Feature to print out EGL returned configs for debug purposes.
'eglChooseConfig' and 'eglGetConfigs' debug information printout is
enabled when the log level equals '_EGL_DEBUG'. The configs are
printed, and if any of them are "chosen" they are marked with their
index in the chosen configs array.
v2:
a) re-factor the code in line with review comments
b) rename function _snprintfStrcat, split it out and put into the
src/util/u_string.h, make it a separate patch.
v3:
remove unnecessary 'const' qualifiers from the function prototype
v4:
a) re-factor the code in line with review comments
b) move util_strnappend from utils back to eglconfigdebug.c. In my
opinion this function is the best way of achieving the desired
result, so it still used but made private to eglconfigdebug.c.
v5:
a) drop unused parameter from function signature
b) more const annotations
c) directly access config parameters instead of going
through _eglGetConfigKey
Signed-off-by: Silvestrs Timofejevs <silvestrs.timofejevs@imgtec.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13705>
Silvestrs Timofejevs [Tue, 2 Apr 2019 15:36:21 +0000 (16:36 +0100)]
egl: introduce a log level getter function
Being able to retrieve the log level can be useful to enable/disable
debug code. The alternative, which is calling 'getenv' function every
time to retrieve the log level, is more "expensive".
Signed-off-by: Silvestrs Timofejevs <silvestrs.timofejevs@imgtec.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13705>
Jordan Justen [Thu, 9 Dec 2021 18:11:37 +0000 (10:11 -0800)]
intel/l3: Make DG1 urb-size exception more generic
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14136>
Rhys Perry [Fri, 3 Dec 2021 12:46:26 +0000 (12:46 +0000)]
aco: improve clrx disassembly
- remove uninteresting lines of output
- remove binary offset before instructions, for easier diffing
- replace generated labels with block numbers
- add encoded instructions at the end of lines, like LLVM dissaembly
- print constant data instead of trying to disassemble it
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14042>
Jesse Natalie [Fri, 10 Dec 2021 05:12:47 +0000 (21:12 -0800)]
microsoft/compiler: Remove algebaric pass for inot
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14140>
Jesse Natalie [Fri, 10 Dec 2021 00:27:39 +0000 (16:27 -0800)]
microsoft/compiler: Implement inot
Fixes:
cb283616 ("nir/algebraic: Small optimizations for SpvOpFOrdNotEqual and SpvOpFUnordEqual")
Reviewed-by: Enrico Galli <enrico.galli@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14140>
Khem Raj [Tue, 7 Dec 2021 19:13:27 +0000 (11:13 -0800)]
v3dv: account for 64bit time_t on 32bit arches
This makes is a bit more portable, especially on 32bit architectures
with 64bit time_t defaults. Especially on musl its a must.
Fixes
../mesa-21.3.0/src/broadcom/vulkan/v3dv_bo.c:71:15: error: format specifies type 'long' but the argument has type 'time_t' (aka 'long long') [-Werror,-Wformat]
time.tv_sec);
^~~~~~~~~~~
Also reported here [1]
[1] https://github.com/agherzan/meta-raspberrypi/issues/969
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14118>
Samuel Pitoiset [Tue, 7 Dec 2021 16:33:55 +0000 (17:33 +0100)]
radv: do not perform depth/stencil resolves for suspended render pass
From the Vulkan spec:
"Store and resolve operations are only performed at the end of a
render pass instance that does not specify the
VK_RENDERING_SUSPENDING_BIT_KHR flag."
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14112>
Samuel Pitoiset [Thu, 9 Dec 2021 07:32:58 +0000 (08:32 +0100)]
Revert "radv: Add bufferDeviceAddressMultiDevice support."
This was a workaround for fixing a crash with Baldur's Gate 3 at start
but the game fixed it since.
This reverts commit
1fe375e7cf8da6d0313b7954ae76120cde92db14.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14134>
Jason Ekstrand [Sat, 4 Dec 2021 04:34:39 +0000 (22:34 -0600)]
intel/fs: Drop high_quality_derivatives
We've never bothered to hook it up in crocus or iris. If we do in the
future, it should probably be a NIR pasa anyway.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14056>
Jason Ekstrand [Wed, 10 Nov 2021 17:42:03 +0000 (11:42 -0600)]
intel/compiler: Get rid of wm_prog_key::frag_coord_adds_sample_pos
This hasn't actually done anything for a while.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14056>
Jason Ekstrand [Sat, 4 Dec 2021 04:20:30 +0000 (22:20 -0600)]
intel/fs,vec4: Drop prog_data binding tables
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14056>
Jason Ekstrand [Sat, 4 Dec 2021 04:20:36 +0000 (22:20 -0600)]
intel/blorp: Stop depending on prog_data binding tables
Instead, set BLORP_TEXTURE_BT_INDEX on the texture instructions
directly.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14056>
Jason Ekstrand [Sat, 4 Dec 2021 03:55:56 +0000 (21:55 -0600)]
intel/fs,vec4: Drop support for shader time
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14056>
Jason Ekstrand [Sat, 4 Dec 2021 03:34:06 +0000 (21:34 -0600)]
intel/fs,vec4: Drop uniform compaction and pull constant support
The only driver using these was i965 and it's gone now. This is all
dead code.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14056>
Jason Ekstrand [Fri, 10 Dec 2021 00:20:55 +0000 (18:20 -0600)]
crocus: wm_prog_key::key_alpha_test uses GL enums
Fixes:
f3630548f1da ("crocus: initial gallium driver for Intel gfx 4-7")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14141>
Danylo Piliaiev [Fri, 10 Dec 2021 15:15:39 +0000 (17:15 +0200)]
turnip: Fix operator precedence in address calculation macros for queries
Fixes crash in Oblivion, Skyrim, Crysis running through DXVK on 32b
systems.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5723
Fixes:
937dd76426b2b372a18be35e1416eed291524af7 "turnip: Implement VK_KHR_performance_query"
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14148>
Emma Anholt [Wed, 8 Dec 2021 19:33:28 +0000 (11:33 -0800)]
nir_to_tgsi: Enable nir_opt_move.
This moves some ops down to when they're needed, generally reducing the
number of temps in use. It's not always a win -- sometimes you can end up
moving a generator of a component used by a nir_op_vec down, which means
that op's sources stay live while the vec (whose register likely gets
coalesced with the ops creating it) is also live. But it's generally
good.
softpipe results:
temps in affected programs: 18115 -> 18026 (-0.49%)
imm in affected programs: 19 -> 22 (15.79%)
r300 results:
instructions in affected programs: 174 -> 178 (2.30%)
vinst in affected programs: 156 -> 160 (2.56%)
sinst in affected programs: 54 -> 50 (-7.41%)
temps in affected programs: 2634 -> 2169 (-17.65%)
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14096>
Emma Anholt [Mon, 6 Dec 2021 20:11:43 +0000 (12:11 -0800)]
r300: Request NIR shaders from mesa/st and use NIR-to-TGSI.
This brings us into parity on state tracker paths with most other
supported drivers, and a lot of additional optimization on our shaders.
Results on a subset of shader-db that doesn't crash:
instructions in affected programs: 59502 -> 47991 (-19.35%)
vinst in affected programs: 17633 -> 15197 (-13.82%)
sinst in affected programs: 9296 -> 7319 (-21.27%)
flowcontrol in affected programs: 627 -> 310 (-50.56%)
presub in affected programs: 4220 -> 1554 (-63.18%)
temps in affected programs: 5775 -> 8570 (48.40%)
lits in affected programs: 215 -> 37 (-82.79%)
The temps (register usage) increase is unfortunate, but it seems that
instruction counts tend to be our limit before reg counts are.
Fixes: #3354
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14096>
Emma Anholt [Tue, 7 Dec 2021 05:43:26 +0000 (21:43 -0800)]
r300: Disable loop unrolling on r500.
It's buggy, and we should just trust GLSL or NIR to do unrolling for us.
Cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14096>
Emma Anholt [Tue, 7 Dec 2021 23:12:14 +0000 (15:12 -0800)]
nir_to_tgsi: Make !native_integers front face input match glsl_to_tgsi.
Avoids regression on r300, which has 0.0 vs 1.0 frontface despite what
tgsi.rst says.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14096>
Emma Anholt [Mon, 6 Dec 2021 20:25:02 +0000 (12:25 -0800)]
nir/nir_to_tgsi: Add support for "if" statements with !native_integers
Previously we've only used this on HW that had all ifs lowered.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14096>
Emma Anholt [Tue, 7 Dec 2021 22:35:36 +0000 (14:35 -0800)]
r300/ci: Add some piglit expectations.
Not a full run, but a bit of sanity-check for the NTT change.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14096>
Adam Jackson [Wed, 8 Dec 2021 15:12:11 +0000 (10:12 -0500)]
docs: Remove no-longer-accurate text about the xlib driver
Mesa hasn't supported color-index rendering in a long time, and
gallium's xlib target doesn't respect MESA_GAMMA.
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14125>
Ian Romanick [Mon, 6 Dec 2021 21:04:41 +0000 (13:04 -0800)]
intel/stub: Implement I915_PARAM_HAS_USERPTR_PROBE
Just say no for now.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14132>
Ian Romanick [Mon, 6 Dec 2021 20:56:01 +0000 (12:56 -0800)]
intel/stub: Implement DRM_I915_QUERY_MEMORY_REGIONS
Borrowed from sim-drm.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14132>
Ian Romanick [Mon, 6 Dec 2021 20:37:42 +0000 (12:37 -0800)]
intel/stub: Implement DRM_I915_QUERY_ENGINE_INFO
Borrowed from sim-drm.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14132>
Ian Romanick [Mon, 6 Dec 2021 20:23:55 +0000 (12:23 -0800)]
intel/stub: Suppress warnings about DRM_I915_QUERY_PERF_CONFIG
There's not a useful way to implement this, so just silence the warning
to cleanup shader-db runs.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14132>
Rhys Perry [Fri, 3 Dec 2021 13:42:25 +0000 (13:42 +0000)]
aco: don't create unnecessary addition in indirect get_sampler_desc()
I don't think this has any effect on GFX9+ because the addition is
combined into the load.
fossil-db (polaris10):
Totals from 12595 (9.29% of 135627) affected shaders:
SGPRs: 1054348 -> 1054860 (+0.05%); split: -0.02%, +0.07%
VGPRs: 667240 -> 667320 (+0.01%); split: -0.01%, +0.02%
CodeSize:
82761508 ->
82512816 (-0.30%); split: -0.30%, +0.00%
MaxWaves: 62182 -> 62181 (-0.00%)
Instrs:
16072934 ->
16010764 (-0.39%); split: -0.39%, +0.00%
Latency:
582819635 ->
582287964 (-0.09%); split: -0.13%, +0.04%
InvThroughput:
276460536 ->
276417613 (-0.02%); split: -0.06%, +0.05%
VClause: 261656 -> 261654 (-0.00%); split: -0.01%, +0.01%
SClause: 680952 -> 680854 (-0.01%); split: -0.05%, +0.04%
Copies: 1727202 -> 1727742 (+0.03%); split: -0.12%, +0.15%
Branches: 547050 -> 547033 (-0.00%); split: -0.01%, +0.00%
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14043>
Timur Kristóf [Tue, 23 Nov 2021 15:50:20 +0000 (16:50 +0100)]
aco: Clean up and fix quad group instructions with WQM.
According to the Vulkan spec chapter 9.25 Helper Invocations,
quad group operations have to be executed by helper invocations.
This commit cleans up the code for quad group instructions by
unifying the code path of quad broadcast with the others, and then
calling emit_wqm just once at the end.
Fixes:
93c8ebfa780ebd1495095e794731881aef29e7d3
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5570
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13929>
Emma Anholt [Thu, 9 Dec 2021 00:36:59 +0000 (16:36 -0800)]
loader: Restore i915g support.
The cleanup of i915c cleaned up our PCI ID list.
Fixes:
0cad451f007f ("classic/i915: Remove driver")
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14131>
Qiang Yu [Wed, 8 Dec 2021 02:57:45 +0000 (10:57 +0800)]
glx: fix regression for drawable type detection
Newer version of XServer supporting GLX_DRAWABLE_TYPE query also
support query with raw X11 window ID besides GLXWindow ID. So we
should not limit the suppported type to GLXPbuffer when query
success.
Otherwise can't start GLX application on newer XServer with:
libGL error: GLX drawable type is not supported
libGL error: GLX drawable type is not supported
X Error of failed request: GLXBadContext
Major opcode of failed request: 149 (GLX)
Minor opcode of failed request: 5 (X_GLXMakeCurrent)
Serial number of failed request: 35
Current serial number in output stream: 35
Fixes:
6625c960c58 ("glx: check drawable type before create drawble")
Tested-by: Mike Lothian <mike@fireburn.co.uk>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14120>
Ian Romanick [Fri, 8 Oct 2021 19:09:04 +0000 (12:09 -0700)]
intel/compiler: Don't store "scalar stage" bits on Gfx8 or Gfx9
Since
1d71b1a311239, only Gfx7 and earlier have any vec4 stages ever.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14128>
Ian Romanick [Mon, 20 Sep 2021 23:21:45 +0000 (16:21 -0700)]
intel/compiler: Don't predicate a WHILE if there is a CONT
Previously a predicated BREAK that appeared immediately before the WHILE
would get merged into the WHILE. This doesn't work if other flow
control (e.g., a CONT) can transfer directly to the WHILE.
On Intel platforms, this fixes the CTS test
dEQP-VK.graphicsfuzz.stable-binarysearch-tree-nested-if-and-conditional.
No shader-db changes on any Intel platform.
When this commit was first created (over a month before it is going to
land), there were some regressions that were prevented by other commits
in MR !13095. That does not appear to be the case now, so I don't know
what changed. Basically, the treatment of discard as a combination of
demote and terminate causes additional continues in some loops, and
those continues trigger this bug. The other commits from that MR
prevent those continues from being generated in the first place.
All Intel platforms had simlar fossil-db results. (Ice Lake shown)
Instructions in all programs:
144419989 ->
144419995 (+0.0%)
SENDs in all programs: 6947332 -> 6947332 (+0.0%)
Loops in all programs: 38277 -> 38277 (+0.0%)
Spills in all programs: 204075 -> 204075 (+0.0%)
Fills in all programs: 319480 -> 319480 (+0.0%)
A few shaders in Doom 2016 were hurt by one instruction each. It seems
likely that these shaders would have experienced at least some
mis-rendering.
Closes: #4213
Fixes:
d13bcdb3a9f ("i965/fs: Extend predicated break pass to predicate WHILE.")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14128>
Dave Airlie [Tue, 7 Dec 2021 07:20:31 +0000 (17:20 +1000)]
treewide: drop mtypes/macros includes from main
These aren't required in lots of places, so remove them.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14127>
Roman Stratiienko [Fri, 3 Dec 2021 16:38:27 +0000 (18:38 +0200)]
v3dv: Fix dEQP-VK.info#instance_extensions test
When mesa3d is built without VK_USE_PLATFORM_DISPLAY_KHR definition,
dEQP test fails:
dEQP : Test case 'dEQP-VK.info.instance_extensions'..
dEQP : Fail (Extension VK_KHR_get_display_properties2 is missing
dependency: VK_KHR_display)
dEQP : DONE!
Enable KHR_get_display_properties2 only if VK_USE_PLATFORM_DISPLAY_KHR
is enabled.
Fixes:
f884c2e3be36 ("v3dv: implement VK_KHR_get_display_properties2")
Signed-off-by: Roman Stratiienko <roman.o.stratiienko@globallogic.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14047>
Chia-I Wu [Tue, 7 Dec 2021 23:04:16 +0000 (15:04 -0800)]
venus: prefer VIRTGPU_BLOB_MEM_HOST3D for shmems
They are logically contiguously in the host. More importantly, they
enable host process isolation.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13800>
Jesse Natalie [Fri, 24 Sep 2021 15:47:32 +0000 (08:47 -0700)]
d3d12: Use overall resource format + plane format to get format info
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14123>
Jesse Natalie [Thu, 23 Sep 2021 15:37:59 +0000 (08:37 -0700)]
d3d12: Allow creating planar resources
Also handle opening planar resources with a single handle, instead
of per-plane handles.
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14123>
Jesse Natalie [Thu, 23 Sep 2021 14:42:04 +0000 (07:42 -0700)]
d3d12: Handle opening planar resources
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14123>
Jesse Natalie [Sun, 26 Sep 2021 15:24:10 +0000 (08:24 -0700)]
d3d12: Force emulation of all YUV formats using per-plane formats
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14123>
Bas Nieuwenhuizen [Mon, 6 Dec 2021 22:41:16 +0000 (23:41 +0100)]
radv: Use the winsys 0 cmdbuffer submission support.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14097>
Bas Nieuwenhuizen [Mon, 6 Dec 2021 22:40:10 +0000 (23:40 +0100)]
radv/amdgpu: Add support for submitting 0 commandbuffers.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14097>
Bas Nieuwenhuizen [Mon, 6 Dec 2021 22:14:16 +0000 (23:14 +0100)]
radv/winsys: Add queue family param to submit.
Extracting it from the first CS will not go over well once we try
submitting 0 of them.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14097>
Bas Nieuwenhuizen [Mon, 6 Dec 2021 22:03:52 +0000 (23:03 +0100)]
radv/amdgpu: Add a syncobj per queue.
For merging our own dependencies in without submitting.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14097>
Dave Airlie [Tue, 7 Dec 2021 03:53:48 +0000 (13:53 +1000)]
mesa/st: drop Draw from dd function table.
Draw is only called in the feedback paths now, and the only thing
it is set to is the st feedback draw path.
So just always have the fallback call the draw feedback path and
get rid of the draw entry.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 03:47:07 +0000 (13:47 +1000)]
mesa/st: move draw indirect and xfb to direct calls.
These don't get used any other way.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 03:01:37 +0000 (13:01 +1000)]
mesa/st: move compute to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 02:55:27 +0000 (12:55 +1000)]
mesa/st: move msaa functions to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 02:52:42 +0000 (12:52 +1000)]
mesa/st: convert DrawTex to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 02:50:13 +0000 (12:50 +1000)]
mesa/st: convert the non-optional egl image to direct calls
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 02:43:29 +0000 (12:43 +1000)]
mesa/st: move blit function to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 02:39:50 +0000 (12:39 +1000)]
mesa/st: replace most of buffer funcs with direct calls.
This leaves invalidate buffer for later
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 04:46:02 +0000 (14:46 +1000)]
mesa/st: move program calls to direct call
Can't move NewProgram yet as the GLSL standalone compiler relies
on it, can probably fix that up later.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 01:52:47 +0000 (11:52 +1000)]
mesa/st: move copy image sub data to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 01:48:58 +0000 (11:48 +1000)]
mesa/st: move viewport to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 01:42:56 +0000 (11:42 +1000)]
mesa/st: move some context functions to direct calls
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 01:37:07 +0000 (11:37 +1000)]
mesa/st: move clear/flush/finish to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 01:20:39 +0000 (11:20 +1000)]
mesa/st: move pixel/bitmap functions to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 00:54:42 +0000 (10:54 +1000)]
mesa/st: move Clear to new direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Tue, 7 Dec 2021 00:51:49 +0000 (10:51 +1000)]
mesa/st: move texture APIs to direct st calls
This is a bit more spreadout, but is still pretty straightforward
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Mon, 6 Dec 2021 23:39:05 +0000 (09:39 +1000)]
mesa/st: move fbo code to direct calling
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Mon, 6 Dec 2021 07:58:45 +0000 (17:58 +1000)]
mesa/dd: drop purgeable interface
This whole extension can likely be dropped, just remove dd.h
bits for now.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Mon, 6 Dec 2021 07:56:59 +0000 (17:56 +1000)]
mesa/st: move perfomance monitor to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Mon, 6 Dec 2021 07:48:53 +0000 (17:48 +1000)]
mesa/st: move perf query to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Mon, 6 Dec 2021 07:41:07 +0000 (17:41 +1000)]
mesa/st: move query memory info to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Mon, 6 Dec 2021 07:39:39 +0000 (17:39 +1000)]
mesa/st: move Enable to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
Dave Airlie [Mon, 6 Dec 2021 07:36:11 +0000 (17:36 +1000)]
mesa/st: move rendermode to direct call
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14100>
mwezdeck [Fri, 5 Nov 2021 12:13:01 +0000 (13:13 +0100)]
virgl/drm: New optimization for uploading textures
1. We can create resource with size of "1" on drm, because size
is not passed to the renderer.
2. We can't create resource with size of "1" on vtest, because shmem
is created based on that.
3. If renderer supports copy_transfer_from_host, then use staging
buffer for transfer in both ways to and from host.
This will allow to reduce memory consumption in the guest.
v2:
- add inline function for checking if we can use this optimization
- add check in readback path. If renderer doesn't support
copy transfer from host, then we need to go with previous
path in readback (through transfer_get ioctl)
v3:
- fix logic for readback
v4:
- refactor the implementation to integrate it more to
existing code base
v5:
- reuse COPY_TRANSFER3D in both directions
v6:
- encode direction in COPY_TRANSFER3D if host supports it
v7:
- renamed cap bit
- introduced COPY_TRANSFER3D_SIZE_LEGACY define
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13689>
Rhys Perry [Tue, 7 Dec 2021 11:57:34 +0000 (11:57 +0000)]
radv: initialize workgroup_size in radv_meta_init_shader
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14087>
Rhys Perry [Mon, 6 Dec 2021 19:11:06 +0000 (19:11 +0000)]
radv: clone shader in radv_shader_compile_to_nir
This way, radv_shader_compile_to_nir doesn't alter the NIR.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14087>
Rhys Perry [Mon, 6 Dec 2021 18:21:11 +0000 (18:21 +0000)]
radv: include RT shaders in RADV_DEBUG=shaders,shaderstats
Instead of using module->nir or nir->info->name to determine if it's a
meta shader, use nir->info->internal.
This also has an effect of disabling printing of meta shaders with
NIR_DEBUG=print.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14087>
Rhys Perry [Mon, 6 Dec 2021 18:21:11 +0000 (18:21 +0000)]
radv: add radv_meta_init_shader
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14087>
James Jones [Fri, 19 Nov 2021 19:50:18 +0000 (11:50 -0800)]
gbm: Don't pass default usage flags on ABIs < 1
Older drivers will not expect any flags from the
GBM front-end when modifiers are in use, and will
likely fail the allocation or handle them
incorrectly as a result. Only specify usage flags
when allocating from a backend with an ABI >= 1,
as that's the ABI version that added support for
specifying usage flags along with modifiers.
Fixes:
ad50b47a14e9 ("gbm: assume USE_SCANOUT in create_with_modifiers")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5709
Signed-off-by: James Jones <jajones@nvidia.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Tested-by: Olivier Fourdan <ofourdan@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14049>
Samuel Pitoiset [Mon, 6 Dec 2021 12:48:25 +0000 (13:48 +0100)]
radv: mark GFX10.3 (aka RDNA2) as conformant products with CTS 1.2.7.1
https://www.khronos.org/conformance/adopters/conformant-products#submission_599
Also bump the CTS version for GFX8, GFX9 and GFX10 because they also
pass 1.2.7.1.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14076>
Samuel Pitoiset [Tue, 7 Dec 2021 16:54:56 +0000 (17:54 +0100)]
radv: fix resume/suspend render pass with depth/stencil attachment
Shouldn't clear on resume.
This fixes dEQP-VK.dynamic_rendering.*resuming.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14113>