AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
135,727 Commits 1 Branch 0 Tags
fbabd2163d5237c50798852fc5f3f3f0bfa55893
Commit Graph

135727 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Timothy Arceri fbabd2163d util/disk_cache: fix crash in fossilize_db 
We don't need to close the file here check_files_opened_successfully()
already does that for us if needed.

This fixes a crash when invalid filename/paths are provided to the
MESA_DISK_CACHE_READ_ONLY_FOZ_DBS environment variable.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9251>
 2021-02-25 03:27:14 +00:00
Mike Blumenkrantz ba4bc54dea zink: rename zink_context::buffers -> vertex_buffers (and usage mask) 
this is more descriptive

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9272>
 2021-02-24 21:41:05 -05:00
Mike Blumenkrantz cfc669585b zink: use dynamic vertex buffer strides 
this removes another case of needing to rehash/create a new pipeline

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9272>
 2021-02-24 21:41:05 -05:00
Mike Blumenkrantz 9d0434bf64 zink: make dynamic state usage in pipeline creation more explicit/flexible 
we'll be expanding this, so we want to have it very easy to update and read

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9272>
 2021-02-24 21:41:05 -05:00
Mike Blumenkrantz 51e527a9ba zink: setup CmdBindVertexBuffers2EXT member in screen for dynamic state 
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9272>
 2021-02-24 21:41:05 -05:00
Marek Olšák f827b29234 nouveau_vieux: use align_calloc for the context to fix m32 crashes 
Fixes: 3175b63a0d "mesa: don't allocate matrices with malloc"

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Zoltán Böszörményi <zboszor@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9220>
 2021-02-25 02:33:12 +00:00
Marek Olšák 81cf4bef7e radeon,r200: use align_calloc for the context to fix m32 crashes 
Fixes: 3175b63a0d "mesa: don't allocate matrices with malloc"

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Zoltán Böszörményi <zboszor@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9220>
 2021-02-25 02:33:12 +00:00
Marek Olšák 75d5c1229e i915: use align_calloc for the context to fix m32 crashes 
Fixes: 3175b63a0d "mesa: don't allocate matrices with malloc"
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4295

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Zoltán Böszörményi <zboszor@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9220>
 2021-02-25 02:33:12 +00:00
Mike Blumenkrantz 22fc869187 zink: set VK_IMAGE_LAYOUT_PRESENT_SRC_KHR on fb resources at eof flush 
this should maybe fix garbled images on amd?

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Mike Blumenkrantz 55f05638f5 zink: handle VK_IMAGE_LAYOUT_PRESENT_SRC_KHR barriers 
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Mike Blumenkrantz c5292710c4 zink: handle PIPE_FLUSH_DEFERRED 
these flushes queue the fence to be submitted on the next finish call, so
we have to store the context to the fence to be able to handle threaded calls

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Mike Blumenkrantz c271ac0066 zink: add util function for returning previous batch 
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Mike Blumenkrantz 24b237d3d4 zink: add a define for compute batch count 
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Mike Blumenkrantz ef3dc995a9 zink: only flush batches in pipe_context::flush if they actually have work 
no need to submit command buffers to the queue which have no commands, so we
can just proceed to give back an inactive fence object that always returns success

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Antonio Caggiano 9dac191f6e zink: fix destroy batch 
Extract batch destruction in its own function, then make sure to
release objects.

Signed-off-by: Antonio Caggiano <antonio.caggiano@collabora.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Mike Blumenkrantz da2d8f1078 zink: create a single fence per batch on startup and then reuse 
previously we would be creating a new fence per batch for every submit,
but we can just reset and reuse them if we're careful with it

based on original patch by Antonio Caggiano

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9243>
 2021-02-24 21:09:52 -05:00
Mike Blumenkrantz a27570326f zink: reorder zink_bind_vertex_buffers() 
we should be making this a no-op if we have no bindings (somehow)

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9241>
 2021-02-24 20:47:51 -05:00
Mike Blumenkrantz 381e92e811 zink: move samplerview referencing around in update_descriptors() 
no functional changes

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9241>
 2021-02-24 20:47:51 -05:00
Mike Blumenkrantz 679dd69741 zink: avoid memset during update_descriptors() for resources refs 
we can simplify this a bit to further reduce cpu usage here

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9241>
 2021-02-24 20:47:51 -05:00
Dave Airlie 7826648e14 lavapipe: fix depth texturing swizzle 
This is pretty much derived from what works and passes
both GL via zink and VK-GL-CTS test suites.

The zink multisample regressions tests rendered black, so
"passed" before, now they render more junk.

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9260>
 2021-02-25 01:32:33 +00:00
Eric Anholt 6d2b2515c4 freedreno/a5xx+: Drop the unused no_decode_srgb flag. 
This was a flag used for a4xx GMEM draws.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9267>
 2021-02-25 00:38:11 +00:00
Eric Anholt 9522087b75 freedreno/a5xx+: Stop recompiling on texture samples changes. 
The shader only looks at this bit of the key on a3xx, and a4xx already
didn't set it.  This will help give us precompile hits for MSAA texture
sampling.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9267>
 2021-02-25 00:38:11 +00:00
Eric Anholt c93fd1046a freedreno: Use the mesa/st frontend lowering of GL_CLAMP. 
350 lines of code for this stupid feature, and we weren't even doing it
right for CS/GS/tess.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9267>
 2021-02-25 00:38:11 +00:00
Mike Blumenkrantz 05d0514a8f zink: ci updates 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9115>
 2021-02-24 23:25:01 +00:00
Mike Blumenkrantz cf8ca77be1 zink: handle point sprite 
this needs custom yinvert handling and also prim mode detection, which
gallium can't do on its own

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9115>
 2021-02-24 23:25:01 +00:00
Mike Blumenkrantz b9676976d0 zink: store prim mode to context during draw 
we need to be able to access this for shader rewrites

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9115>
 2021-02-24 23:25:01 +00:00
Mike Blumenkrantz 2e60929b47 nir/texcoord_replace: add a yinvert param 
vulkan needs to invert the y coord in order to handle PIPE_SPRITE_COORD_LOWER_LEFT

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9115>
 2021-02-24 23:25:01 +00:00
Mike Blumenkrantz 707dc04b78 zink: no-op descriptor updating for draws without descriptors 
this is a valid case that we can trivially shortcut to cut down on calls

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9155>
 2021-02-24 22:44:50 +00:00
Eric Anholt 5fa27e6670 freedreno: Drop custom driver lowering of GL's color clamping. 
The mesa/st frontend can do it for us now that we don't need to worry
about breaking precompiles.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8997>
 2021-02-24 21:48:54 +00:00
Eric Anholt 3b9f6af1a9 freedreno: Drop custom driver lowering of two-sided color. 
The GL frontend can do it for us now, so just use their code instead of
our own shader variants.  In the past we had to do hide the GL shader
variants in the driver to get precompiles from st, but no longer as of
!8601.

I tested with drawoverhead -test 6 (shader program change, n=30) and -test
1 (no statechanges, n=43) and saw no change in driver overhead.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8997>
 2021-02-24 21:48:54 +00:00
Eric Anholt de17b4aab5 freedreno: Remove uniform variables after finalizing NIR. 
mesa/st optimizes the uniform storage if you have the finalize hook in
place, causing the uniforms declared to potentially not have storage in
the ParameterValues list any more.  If you leave your uniforms around in
the NIR, then a later finalization after variant creation will re-add the
uniforms to parameters, defeating the optimization and likely reallocating
the uniform storage (causing use-after-free).  So, we have to do this
before we can start using variants in mesa/st.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8997>
 2021-02-24 21:48:54 +00:00
Eric Anholt 9fcc369770 mesa: Always make sure uniform storage doesn't get reallocated. 
Every caller of associate_uniform_storage was doing this to safety-check
that the uniform storage didn't get reallocated, except for
st_deserialise_ir_program().  This ended up leaving an opening for
use-after-free without hitting the assert in the hot-cache case (and I
found it on freedreno).  Having added it, it also reveals use-after-frees
in the drawpixels shader variant cases on llvmpipe and zink.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8997>
 2021-02-24 21:48:54 +00:00
Mauro Rossi 3468c0dd4d android: mesa: Move the FXT1 compressor/decompressor to util/ 
After commit a46b73e in order to build mesa for Android
we need to generate $(intermediates)/util/format/u_format_pack.h
instead of $(intermediates)/format/u_format_pack.h

Fixes the following building error in Android build:

external/mesa/src/util/format/u_format_fxt1.c:28:10: fatal error: 'util/format/u_format_pack.h' file not found
         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.

Fixes: a46b73e ("mesa: Move the FXT1 compressor/decompressor to util/")
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9240>
 2021-02-24 21:29:14 +00:00
Jose Fonseca a4ffa25ee8 scons: Add u_format_pack.h include path. 
Fixes MSVC SCons build.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9263>
 2021-02-24 20:48:47 +00:00
Mike Blumenkrantz a6c72af908 zink: use 0 as default for spec constants 
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9258>
 2021-02-24 19:15:48 +00:00
Eric Anholt 978d244114 ci/iris: Move the traces yml file to the driver-specific dir. 
Since the job is manual, I missed it in the move and it got dropped from
the artifacts.

Fixes: 60d413b894 ("ci: Move the piglit expectations lists to the per-driver CI dirs.")
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9226>
 2021-02-24 18:55:02 +00:00
Eric Anholt bcea453d4a ci/piglit: Stop including the test counts at the end of expectations. 
It's just a ton of fuss for driver developers fixing piglit tests.  This
makes the trace expectation files pretty silly (empty expectation, but
you'll get a diff to a non-empty result when something fails)

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Andres Gomez <agomez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9226>
 2021-02-24 18:55:02 +00:00
Dave Airlie 1121ea8192 lavapipe: fix descriptor set layout freeing. 
Have to store the used allocator otherwise ends up being free wrong.

Fix for
dEQP-VK.api.object_management.alloc_callback_fail.descriptor_set_layout*

Fixes: f94a5f30e0 ("lavapipe: add reference counting to descriptor set layout")

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9247>
 2021-02-25 04:38:44 +10:00
Kenneth Graunke b9133e48a6 iris: Pin surface state buffers after possibly updating the clear color 
On Gen8, updating the clear color will end up allocating new
SURFACE_STATE entries.  These might end up living in a different BO
than the original copies, which means that we have to pin _after_
updating the clear color, not before.

Found by inspection.

Reviewed-by: Zoltán Böszörményi <zboszor@gmail.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9257>
 2021-02-24 18:32:29 +00:00
Eric Anholt 60573b443b v3d: Replace driver lowering of GL_CLAMP with mesa/st's. 
Mesa core can do this logic for us now.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9228>
 2021-02-24 18:03:46 +00:00
Rob Clark e6bacd3e9a v3d: Drop foreach_bit() macro 
Now that we have a global one.

Signed-off-by: Rob Clark <robdclark@chromium.org>

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Rob Clark 0ca5b1a6b9 etnaviv: Drop foreach_bit() macro 
Now that we have a global one.

Signed-off-by: Rob Clark <robdclark@chromium.org>

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Rob Clark 2b020e84e7 freedreno: Drop foreach_bit() macro 
Signed-off-by: Rob Clark <robdclark@chromium.org>

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Rob Clark e5a64e34d8 freedreno/ir3: Drop foreach_bit() macro 
Now that there is a global one in util/bitscan.h

Note this version had an extra assert which is not really suitable to a
generic foreach_bit().. just move the assert to the two usages of the
iterator macro.

Signed-off-by: Rob Clark <robdclark@chromium.org>

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Mike Blumenkrantz 025b57524f anv: for_each_bit -> foreach_bit 
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Mike Blumenkrantz 77cba4b9f2 freedreno/vulkan: for_each_bit -> foreach_bit 
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Mike Blumenkrantz 6875e10350 radv: for_each_bit -> foreach_bit 
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Mike Blumenkrantz e89f158b82 v3dv: remove for_each_bit() macro 
this was unused

Reviewed-by: Rob Clark <robclark@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Mike Blumenkrantz e7c7150d63 util/bitscan: add u_foreach_bit macros 
this is a standardized (and very slightly improved for usability) version
of the macro that has been copied into every vulkan driver

includes fixup from Rob Clark <robclark@freedesktop.org>

Reviewed-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9191>
 2021-02-24 17:11:44 +00:00
Mike Blumenkrantz 618f46ee02 zink: add more ci flakes 
wtf

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9242>
 2021-02-24 17:02:45 +00:00