AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
190,642 Commits 1 Branch 0 Tags
08984e68fbfab0c7c968345f6700b390b8a5ca4f
Commit Graph

190642 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Alyssa Rosenzweig 08984e68fb gallium: remove ability to probe asahi 
The asahi uapi is unstable. Do not attempt to probe on it.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29620>
 2024-06-14 15:44:30 +00:00
Corentin Noël 8c5c93acba wsi: Make sure to return a valid wayland id string 
The result of asprintf was previously ignored, handle the case of failure by
not returning undefined content.

Signed-off-by: Corentin Noël <corentin.noel@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29724>
 2024-06-14 15:18:56 +00:00
Daniel Schürmann b7982152ff aco: use aco::monotonic_allocator for IDSet 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29713>
 2024-06-14 14:32:35 +00:00
Daniel Schürmann 97fd5d3f33 aco: make aco::monotonic_buffer_resource declaration visible for aco::IDSet 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29713>
 2024-06-14 14:32:35 +00:00
Daniel Schürmann 95967c2ca0 aco/reindex_ssa: replace live_var parameter with boolean 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29713>
 2024-06-14 14:32:35 +00:00
Daniel Schürmann a497d105e3 aco: move live var information into struct Program 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29713>
 2024-06-14 14:32:35 +00:00
Daniel Schürmann 2322ab427e aco/scheduler: remove unused register_demand parameter 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29713>
 2024-06-14 14:32:35 +00:00
Karol Herbst 169dc86e74 gallium: properly type fields of pipe_resource.usage 
This gets rid of most of the pointless bitfields and also moves away from
unsigned, which we really should stop using :)

Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29720>
 2024-06-14 13:56:34 +00:00
Karol Herbst 40785d9a52 gallium: properly type pipe_resource.usage with the enum 
This allows for a more strongly typed field on the Rust side.

Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29720>
 2024-06-14 13:56:34 +00:00
Karol Herbst f58247882a gallium: reduce pipe_resource.usage to 4 bits 
This works around a bindgen bug, but also moves the value into a single
byte as per the addition of compression_rate it spanned over two bytes.

Fixes: 5db7398672 ("gallium: add interface for fixed-rate surface/texture compression")
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29720>
 2024-06-14 13:56:34 +00:00
Danylo Piliaiev 59937f62a6 ir3/a7xx: Fix FS consts corruption when other FS has zero constlen 
Having zero consts in one FS may corrupt consts in follow up FSs,
on such GPUs blob never has zero consts in FS. The mechanism of
corruption is unknown.

Fixes geometry flickering in a number of games, including:
 Baldur's Gate 3
 Assasin's Creed Rogue

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29357>
 2024-06-14 13:27:01 +00:00
Danylo Piliaiev 98e3b1bc5f freedreno/a7xx: Update TPL1_DBG_ECO_CNTL1 to fix UBWC corruption 
Copying UBWC image via BLIT_OP_SCALE may be corrupted if previously
someone copied image with bit 18 set in TPL1_DBG_ECO_CNTL1.

Found by replaying blob's cmdstream on a740, but somehow this issue
doesn't happen on Android.

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29358>
 2024-06-14 12:21:03 +00:00
Eric Engestrom 44e6850016 bin/ci: escape literal url in regex 
Fixes: b24dd1fa1c ("ci: Fix parse GitLab pipeline url")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29718>
 2024-06-14 11:46:52 +00:00
Pierre-Eric Pelloux-Prayer 5c50e028d1 ac/sqtt: make VA helpers static 
They're only used from ac_sqtt so don't expose them.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25397>
 2024-06-14 10:32:17 +02:00
Pierre-Eric Pelloux-Prayer c44b20e8d6 radeonsi/sqtt: add AMD_THREAD_TRACE_INSTRUCTION_TIMING 
To control if instruction timing should be enabled (on by default).

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25397>
 2024-06-14 10:32:10 +02:00
Pierre-Eric Pelloux-Prayer 365fda834e radeonsi: use the common SQTT implementation 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25397>
 2024-06-14 10:32:04 +02:00
Sergi Blanch Torne 0c092dc5c4 ci: run_n_monitor, collect and summarize 
The job duration is printed during the test progress. It can be collected and
summarized at the end of the process. It is interesting when doing a stress
test as one will have this information together at the end.

Signed-off-by: Sergi Blanch Torne <sergi.blanch.torne@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29419>
 2024-06-14 08:04:56 +00:00
Konstantin Seurer 5726ecae3e khronos-update: Add ANDROID guards to vk_android_native_buffer.h 
Avoids adding back the code after every header update.

Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Acked-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29621>
 2024-06-14 06:58:35 +00:00
Samuel Pitoiset 3f9fe2dbe1 radv: use BDA in the DGC prepare shader 
Only for buffers that are managed by the application (ie. preprocess,
stream and sequence buffers). For future work.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29605>
 2024-06-14 06:35:18 +00:00
Samuel Pitoiset 730ba8322f radv: fix incorrect buffer_list advance for multi-planar descriptors 
If we have an array of multi-planar descriptors, buffer_list was
incorrectly incremented and this could have overwritten some BO entries.

In practice, this situation should be very rare because most of the
applications enable the global BO list.

Cc: mesa-stable
Closes: #10559
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28816>
 2024-06-14 06:14:30 +00:00
David Rosca 39fdd2aec0 radeonsi: Make si_compute_clear_image work with 422 subsampled formats 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29598>
 2024-06-14 03:19:24 +00:00
Qiang Yu 4a18809a56 radeonsi: add missing nir_intrinsic_bindless_image_descriptor_amd 
Otherwise we get shader compilation error when imageSize().

Fixes: d4fdeaa820 ("radeonsi: replace llvm resource code with nir lower")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29690>
 2024-06-14 02:52:52 +00:00
Qiang Yu b1d0ecd00d glsl: respect GL_EXT_shader_image_load_formatted when image is embedded in a struct 
Fix compilation failure when image is embedded in struct when
GL_EXT_shader_image_load_formatted is enabled:

  struct GpuPointShadow {
      image2D RayTracedShadowMapImage;
  };

  layout(std140, binding = 2) uniform ShadowsUBO {
      GpuPointShadow PointShadows[1];
  } shadowsUBO;

Compile log:
  error: image not qualified with `writeonly' must have a format layout qualifier

Fixes: 082d180a22 ("mesa, glsl: add support for EXT_shader_image_load_formatted")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29693>
 2024-06-14 02:21:59 +00:00
Faith Ekstrand 8307fa95ec nvk: Refactor build_cbuf_map() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 636604ea5a nvk: Only write draw parameters to cb0 when they change 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 9f7081b921 nvk: Use inline constant buffer updates for CB0 
This is what the old GL driver did and appears to be what the blob does
as well.  They should pipeline much better than full buffer re-binds
which appear to be causing stalling issues inside the GPU.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand f716bab6b7 nvk: Pass the queue to draw/dispatch_state_init() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand b2d85ca36f nvk: Use helper macros for accessing root descriptors 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 2423b0b295 nvk: Pass the base workgroup and global size to flush_compute_state() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 7a0237bdcf nvk: s/draw_idx/draw_index/g 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 20f21b1917 nvk: Use cbuf loads for variable pointers dynamic SSBO descriptors 
We want everything that touches the root table to go through ldc path
and not to use global loads.  This means that we need to do some
juggling to handle dynamic SSBO descriptors in the variable pointers
case.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 091a945b57 nvk: Be much more conservative about rebinding cbufs 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 8b5835af31 nvk: Use bindless cbufs on Turing+ 
These are much faster than ld.global.constant.  This takes The Witness
from 103 FPS to 130 FPS on my 4060 laptop GPU when run with
NVK_DEBUG=no_cbuf

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 248b22d158 nvk/descriptor_set_layout: Record which dynamic buffers are UBOs 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 6e41f2a28d nvk: Allow the cbuf optimization for VK_DESCRIPTOR_TYPE_MUTABLE_EXT 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 723e5cae59 nvk: Move the zero offset optimization to load_descriptor_for_idx_intrin() 
Looking at binding_layout->desc_type is sketchy in the face of mutable
descriptors.  It's safer for load_descriptor() to just return the
descriptor.  load_descriptor_for_idx_intrin() knows about the
descriptor's actual shader usage and we can do the optimization there.

This isn't actually a bug fix.  The optimization just didn't happen in
the presence of mutable descriptors.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 05e213f03e nvk/lower_descriptors: Add a descriptor_type_is_ubo/ssbo() helper 
This is a small behavioral change as we are no longer double-counting
inline uniform data loads when determining cbuf priority.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 903fb6f74a nvk: Make nvk_min_cbuf_alignment() inline 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand cbe62813a1 nvk: Rename nvk_cmd_buffr_get_cbuf_descriptor() 
This makes it clear that it returns an address, not a descriptor.  That
distinction will matter soon.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 59303584e3 nvk: Align buffer descriptors 
In theory, the app is supposed to do this for us but this gives us a bit
of protection against potential GPU faults.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 5685de8795 nvk: Split write_[dynamic_]buffer_desc into UBO and SSBO variants 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 6387ae7dfb nvk: Split SSBO and UBO address formats 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand dc7b08c41a nak: Implement nir_intrinsic_ldcx_nv 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 851b3ddd05 nak: Lower non-uniform ldcx_nv to global loads 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand e05cb967e7 nir: Add nir_foreach_block_in_cf_node_safe() iterators 
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 7b5856ebe9 nak: Implement [un]pin_cx_handle_nv 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand 12b79f814b nak: Implement r2ur_nv 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand dc99d9b2df nvk,nak: Switch to nir_intrinsic_ldc_nv 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:45 +00:00
Faith Ekstrand b107240474 nir: Add some new _nv intrinsics 
The ldc_nv and ldcx_nv intrinsics correspond to the index and bindless
forms of NVIDIA's LDC instruction, respectively.  ldc_nv is pretty much
load_ubo without some of the unnecessary constant bits while ldcx_nv
takes a 64-bit bindless handle instead of an index.  The other two give
us a little control over register allocation at the NIR level to ensure
that LDCX handles are placed in uniform registers.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:45 +00:00
Faith Ekstrand ab84cf11c7 nak/copy_prop: Don't propagate bindless cbufs into non-uniform blocks 
We can propagate within a non-uniform block just fine but not across
them because that might change live registers in unpredictable ways.
The real boundary here is that we can't propagate across an OpPin but
that's a lot harder to express.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:45 +00:00