AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
153,204 Commits 1 Branch 0 Tags
e4884c35d6b45bbdd357b2706d65e7ae1c1e6095
Commit Graph

153204 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Mike Blumenkrantz e4884c35d6 llvmpipe: remove misleading query comment 
u_blitter already handles this

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16278>
 2022-05-03 13:00:50 +00:00
Mike Blumenkrantz a36c3f6f6a llvmpipe: disable statistic queries during draw if queries are disabled 
cc: mesa-stable

fixes (zink):
spec@ext_transform_feedback@generatemipmap prims_generated

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16278>
 2022-05-03 13:00:50 +00:00
Pierre-Eric Pelloux-Prayer 6ee9e6ce87 radeonsi: generate unique shader name in si_get_nir_shader 
This function modifies the NIR shader, so when using NIR_DEBUG=print and
nir_viewer all the variants are displayed using the original name.

With this change, we get the original shader (eg: GLSL3), and then each
variant gets its own name (GLSL3-xxxxxxxx) - and is displayed in its
own tab in nir_viewer.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16247>
 2022-05-03 10:39:51 +00:00
Pierre-Eric Pelloux-Prayer f1f2d407e7 radeonsi: don't always strip NIR debug infos 
When using NIR_DEBUG=print, always having shader names help debugging.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16247>
 2022-05-03 10:39:51 +00:00
Joshua Ashton ce7966fcb4 aco: Use movk for AddressHi bits in vertex prolog 
Eliminates a useless extra dword by transforming
        s_mov_b32 s47, 0xffff8000                                   ; beaf03ff ffff8000
to
        s_movk_i32 s47, 0x8000                                      ; b02f8000

which does the same thing.

Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15415>
 2022-05-03 10:15:36 +00:00
Daniel Schürmann 58bd9a379e aco/ra: fix live-range splits of phi definitions 
It could happen that at the time of a live-range split,
a phi was not yet placed in the new instruction vector,
and thus, instead of renamed, a new phi was created.

Fixes: dEQP-VK.subgroups.ballot_broadcast.compute.subgroupbroadcast_i8vec2
Cc: mesa-stable

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16248>
 2022-05-03 09:54:30 +00:00
Konstantin Seurer d639608b8b ac/nir: Do not set cursor in try_extract_additions 
Fixes: 61ac5ac ("radv,ac/nir: lower global access to _amd global access intrinsics")
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16270>
 2022-05-03 09:23:49 +00:00
Samuel Pitoiset 3829981eb7 radv: use 3D views for 3D internal operations on GFX6-8 
Instead of binding 2D images views and relying on BASE_ARRAY.
This cleanups the internal driver operations for 3D.

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/16272>
 2022-05-03 08:10:10 +02:00
Timothy Arceri 180398f785 nir: fix sorting before assigning varying driver locations 
We need to make sure we also properly sort varyings sharing a single
slot otherwise we can end up assigning earlier components to the next
slot if we have already processed later components.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6392

Fixes: 1e93b0caa1 ("mesa/st: add support for NIR as possible driver IR")

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16208>
 2022-05-03 00:04:30 +00:00
Emma Anholt 067023dce2 util: Add some unit tests of the half-float conversions. 
Make sure that we're handling nans/infs correctly, in particular only
generating quiet nans when given a quiet nan.  This catches the previous
qNaN fix.

Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16233>
 2022-05-02 23:36:50 +00:00
Chia-I Wu 27e33d5c96 util: Keep quiet NaNs quiet when converting to half float. 
We don't want to be throwing exceptions and changing float values later by
emitting a signaling binary16 nan.

If we don't do this, then when we convert back to f32 in NIR constant
expression evaluation, the signaling NaN can end up giving NaN for
fmax(NaN, 0.0), instead of 0.0.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5933
Cc: mesa-stable
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16233>
 2022-05-02 23:36:50 +00:00
Lionel Landwerlin acf6bf88c0 iris: use new kernel uAPI to compute video memory 
v2: Use os_get_available_system_memory() when kernel memory region
    uAPI is not available (Lionel)

Cc: 22.1 <mesa-stable>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16210>
 2022-05-02 22:57:06 +00:00
Jason Ekstrand 763744aa8b panvk: Add address/range helpers for panvk_buffer 
The range helper is taken from ANV; the gpu_ptr one is original.  This
also fixes a few more bugs where we weren't adding offsets in properly.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16216>
 2022-05-02 17:37:40 -05:00
Alyssa Rosenzweig 631f47e83a panvk: Only write depth when depth test enabled 
Fixes dEQP-VK.pipeline.depth.format.*_test_disabled.depth_write_enabled

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16283>
 2022-05-02 20:28:48 +00:00
Alyssa Rosenzweig 46778f2b2a panvk: Lower image deref atomics 
Fixes dEQP-VK.image.atomic_operations.*

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16283>
 2022-05-02 20:28:48 +00:00
Alyssa Rosenzweig 54412afadc pan/bi: Handle texture offset + index 
Fixes dEQP-VK.glsl.opaque_type_indexing.sampler.uniform.vertex.sampler1d

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16283>
 2022-05-02 20:28:48 +00:00
Alyssa Rosenzweig 86d32eed38 panvk: Allocate stack for each thread 
Rather than a single stack for all threads to share -- that can't work! :-) We
use the same helper that the GLES driver does. Fixes anything using scratch or
spilling, including:

dEQP-VK.glsl.indexing.varying_array.vec3_static_write_dynamic_read

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16283>
 2022-05-02 20:28:48 +00:00
Alyssa Rosenzweig 7a7403f6aa panvk: Lower global vars to local 
Fixes dEQP-VK.glsl.indexing.varying_array.float_static_write_static_loop_read

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16283>
 2022-05-02 20:28:48 +00:00
Yiwei Zhang 3f44bad1ce venus: log performance envs upon instance creation with debug init 
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Change-Id: I8db93d6de28ac6a727c10fbaca94cffac092f8ea
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16256>
 2022-05-02 20:20:43 +00:00
Yiwei Zhang 84a53de76c venus: add VN_RELAX_BASE_SLEEP_US env var for tuning cpu utils 
Cap the minimum to the current base sleep us 10.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16256>
 2022-05-02 20:20:43 +00:00
Rob Clark f3fb1d6632 drm-shim: Cleanup on device file close 
Avoid leaking memory when the device file is closed, dropping bo
references and freeing the shim_fd.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16250>
 2022-05-02 19:50:33 +00:00
Rob Clark 409b76511c freedreno/drm-shim: Better iova handling 
We actually want to use util_vma to handle this.  But fortunately core
drm-shim alredy does this for mem offset, we can just delete a bunch of
code and re-use that.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16250>
 2022-05-02 19:50:33 +00:00
Rob Clark 6c465ad6c7 drm-shim: Add error handling for bo_init() 
Don't assert, return an error.  If we are fuzzing something using
drm-shim, we want to be more like the kernel and return an error
for OoM situation, rather than falling over.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16250>
 2022-05-02 19:50:33 +00:00
Rob Clark c4b5ebe1fc drm-shim: Better mmap offsets 
Using the bo pointer address as the offset doesn't go over well when
someone is fuzzing you.  But we already have the mem_addr, we can simply
use that instead.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16250>
 2022-05-02 19:50:33 +00:00
Rob Clark 97f4e48717 freedreno/drm-shim: Robustify error handling 
We can't be so sloppy if we are using drm-shim for fuzzing.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16250>
 2022-05-02 19:50:33 +00:00
Rob Clark d06fc7bb4f freedreno/drm-shim: Update to latest uapi version 
Needed for fuzzing virgl drm native context.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16250>
 2022-05-02 19:50:33 +00:00
Rob Clark a922997ffd drm-shim: Add GET_UNIQUE support 
Needed by drmOpenWithType(), which is used by virgl drm native context.
(We want to use drm-shim for fuzzing.)

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16250>
 2022-05-02 19:50:33 +00:00
Iván Briano bf04be17f7 vulkan/wsi/wayland: Fix double free on error condition 
If wsi_configure_native_image() fails, it will call
wsi_destroy_image_info() itself, so let's try to not call it again from
wsi_wl_swapchain_destroy().

Fixes the CTS tests:
dEQP-VK.wsi.wayland.swapchain.simulate_oom.*

Fixes: b626a5be43 ("vulkan/wsi/wayland: Split image creation")

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16257>
 2022-05-02 19:26:27 +00:00
Lionel Landwerlin 1e7ea18026 vulkan/runtime: 0-out pipeline cache object 
I'm running into crashes because cache->cache_object is uninitialized.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 591da98779 ("vulkan: Add a common VkPipelineCache implementation")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16275>
 2022-05-02 19:02:42 +00:00
Caio Oliveira 7cd9adeb41 intel/compiler: In XeHP prefer <1;1,0> regions before compacting 
Ken performed some tests with shader-db to evaluate the effects

```
Across all 145,848 shaders generated, the results were:

Total bytes compacted before: 3,326,224
Total bytes compacted after: 60,963,280
```

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15399>
 2022-05-02 18:03:01 +00:00
Mike Blumenkrantz 562fd98998 llvmpipe: add a ci flake 
not able to repro locally

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16280>
 2022-05-02 17:46:08 +00:00
Jesse Natalie 2505afc430 d3d12: Destroy empty residency bo set during early-return 
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 1e940f2ccf d3d12: Fix a couple over-releases from incorrect take-ownership flag 
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 32c1b5fd05 d3d12: Fix leaks in map with do-not-wait 
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 5b46ef9857 d3d12: Use a pipe_reference in d3d12_bo 
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 8de1e310a7 u_primconvert: Handle take_index_buffer_ownership 
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 12d88d500d u_primconvert: Refactor to remove recursion from util_primconvert_draw_vbo 
This prevents having unused copies of pipe_draw_info and
pipe_draw_start_count_bias on the stack, and makes it easier to do
things once for a multi-draw, which will matter in the next patch.

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 0474bbcfb9 u_debug_stack: Lock around stack dumps to prevent interleaving 
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 303c2754a3 u_debug_symbol: Allow re-init of DbgHelp to work around already-initialized issues 
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie f51840d87c u_debug_symbol: Use correct sizeof for DbgHelp 
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie d8024b2504 u_debug_refcnt: Don't loop for initial refcounts if the initial value is huge 
Some components use very large refcounts to bypass atomics in single-threaded
cases, and this produces near-infinite loops when these resources are created.

Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 0bcb5f919e bin: Add a script for filtering/sorting refcount logs 
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16182>
 2022-05-02 16:58:20 +00:00
Jesse Natalie 53a94fbdd5 mesa: Ensure ARB programs end in a newline 
If the last line of an ARB program has a comment, the program will
fail to parse, because the lexer only considers a comment valid if
it ends in a newline, not EOF. The parser then fails on the '#'.

Reviewed-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16230>
 2022-05-02 16:02:54 +00:00
Jesse Natalie 9f37579e1b d3d12: Correctly key off of polygon stipple enable cap 
Apparently we were only keying off the presence of a real stipple pattern
being set, and completely ignoring when the app does glDisable().

Add in the actual enable bit as an additional discriminator to determine
if we should be doing polygon stippling.

Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16254>
 2022-05-02 15:56:05 +00:00
Alyssa Rosenzweig ce6d5094c8 pan/va: Add non-IDVS VAR_TEX instructions 
Used in blit shaders.

Icecream95 supplied the test cases.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15588>
 2022-05-02 11:11:25 -04:00
Alyssa Rosenzweig 030e00dea2 pan/va: Add VAR_TEX_BUF_GRADIENT instruction 
Probably useful for ... something.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15588>
 2022-05-02 11:11:25 -04:00
Icecream95 0b17ea7536 pan/va: Improve texture instructions 
[Alyssa: Related improvements.]

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15588>
 2022-05-02 11:11:08 -04:00
Icecream95 2bfcfa005d pan/va: Add absneg modifier to V2F32_TO_V2F16 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15588>
 2022-05-02 11:11:08 -04:00
Icecream95 35c3073fc2 pan/va: Add SEG_ADD/SEG_SUB operation 
[Alyssa: Fixes to the instruction.]

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15588>
 2022-05-02 11:11:08 -04:00
Icecream95 822dfb76de pan/va: Add more transcendental operations 
TODO: Make the secondary opcode field wider so that FATAN_ASSIST can
be split into two instructions

[Alyssa: Fixes to the hardware behaviour.]

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15588>
 2022-05-02 11:11:08 -04:00