c8a8543af7
vulkan: Fix incorrect bpcs value for padded formats
...
Skip padding channels and only consider valid color channels.
Add and use a common helper for this.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Signed-off-by: Valentine Burley <valentine.burley@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30821 >
2024-10-16 14:30:15 +00:00
f33e18ab39
vulkan/queue: Check for _mem_signal_temp before we submit
...
vk_queue_push_submit() takes ownership of the vk_queue_submit object and
potentially passes it to another thread. This fixes a race where, if
the other thread processes and deletes the vk_queue_submit before we get
to checking _mem_signal_temp, we may have a use-after-free.
Fixes: c95b646e23https://gitlab.freedesktop.org/mesa/mesa/-/issues/11988
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31597 >
2024-10-14 13:13:15 +00:00
746ea7b0cd
vulkan: Add a vk_fragment_shading_rate_is_disabled() helper
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
5bd825a4b8
vulkan: fix merging bind sparse submits together
...
Looks like this was completely broken because what we want is to merge
first/second binds to the merged object.
This fixes new CTS dEQP-VK.sparse_resources.image_sparse_binding.*.
Fixes: bba6bf33a8samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31545 >
2024-10-07 15:59:42 +00:00
5fe3f57d3f
vulkan: export vk_robustness_disabled
...
HK will use this for a HK_PERFTEST=norobust flag.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31466 >
2024-10-04 17:52:10 +00:00
1a2827e401
vulkan: merge driver internal bits
...
no reason for hk and vk_meta to diverge here.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31466 >
2024-10-04 17:52:10 +00:00
bba6bf33a8
vulkan/queue: Merge submits when possible
...
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25576 >
2024-10-03 22:11:40 +00:00
c3bf1a67a1
vulkan/queue: Split vk_queue_submit into create and submit
...
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25576 >
2024-10-03 22:11:40 +00:00
f0392779d9
vulkan/queue: Use a builder pattern for vk_queue_submit
...
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25576 >
2024-10-03 22:11:39 +00:00
899c774b9a
vulkan: Add a vk_queue_submit_has_bind() helper
...
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25576 >
2024-10-03 22:11:39 +00:00
c95b646e23
vulkan/queue: Use _mem_signal_temp instead of signal_mem_sync
...
The two checks should be equivalent. This just lets us use data in
struct vk_queue_submit rather than a local boolean.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25576 >
2024-10-03 22:11:39 +00:00
267b7f1deb
vulkan/queue: Move has_binary_permanent_semaphore_wait into the sumbit struct
...
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25576 >
2024-10-03 22:11:39 +00:00
9b21dc06c4
vulkan/queue: Don't use vk_semaphore in threaded payload stealing
...
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25576 >
2024-10-03 22:11:39 +00:00
3d7bf07089
vk/meta: Make some helpers public
...
vk_image_view_type_to_sampler_dim() and vk_image_view_type_is_array()
can be useful to driver-specific meta shaders.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31441 >
2024-10-03 09:53:34 +00:00
cd38fd37f7
vk/meta: Allow tracking of driver-specific objects in the meta list
...
Add VK_META_OBJECT_KEY_DRIVER_OFFSET to define an offset for
driver-specific key types.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31441 >
2024-10-03 09:53:34 +00:00
7fe4f64c3b
vk/meta: Support VkShaderExt objects to allow tracking internal shaders
...
PanVK has a few internal shaders that don't fit in the vk_meta
compute/graphics pipeline model. Teaching vk_meta about VkShaderEXT
allows us to keep track of those internal shaders without using yet
another hash table.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31441 >
2024-10-03 09:53:34 +00:00
790759dfaf
vk/image: Fix the extent adjustment of non-compressed views
...
When creating a non-compressed view of a compressed image, we need to
divide the extent by the image block size not the view block size.
Fixes: 8ddc527ba4boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31439 >
2024-10-02 21:54:53 +00:00
767a7d785b
vk/meta: Fix buffer -> image copy using a graphics pipeline
...
Unlike 3D image views, 2D array image views created by meta_copy only
contain the layers needed for rendering, so we shouldn't consider the
base layer an image offset.
Fixes: 07c6459cd8boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31439 >
2024-10-02 21:54:53 +00:00
bf013cf8de
vulkan: Add a comment explainint SEVERITY_WARNING_BIT for vk_error*()
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31449 >
2024-10-01 00:48:36 +00:00
b9af5564bb
vulkan: Add support for VK_EXT_depth_clamp_control
...
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31411 >
2024-09-30 22:18:27 +00:00
05ece3e219
vulkan: Use SEVERITY_WARNING_BIT for vk_error*()
...
Our usage of vk_error to expound on error messages returned by drivers
probably falls more into the warning category.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Reviewed-by: Eric Engestrom <eric@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31292 >
2024-09-27 21:28:17 +00:00
8097a8e81e
vk/runtime: add EXT DGC layout handling
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31386 >
2024-09-27 18:41:41 +00:00
5ebcb71e94
vulkan/pipeline: Patch through INDIRECT_BINDABLE_BIT
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31394 >
2024-09-26 17:18:25 +00:00
4ae2882bc1
vulkan: Add a vk_pipeline_ops::get_shader method
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31394 >
2024-09-26 17:18:25 +00:00
4a9a4721fe
vulkan: Add a stages field to vk_pipeline
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31394 >
2024-09-26 17:18:25 +00:00
2cf5ef37ba
vulkan/video: set HEVC nuh_temporal_id_plus1 and nal_unit_type correctly
...
This is only relevant for hierarchical coding using sub-layers.
Fixes: 72f52329cdzzoon@igalia.com >
Reviewed-by: Dave Airlie <airlied@redhat.com >
Signed-off-by: Colin Marc <hi@colinmarc.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31303 >
2024-09-26 04:18:11 +00:00
101a803858
vulkan: use standard sample locations if there's no VkPipelineSampleLocationsStateCreateInfoEXT
...
If the pipeline is created with no
VkPipelineSampleLocationsStateCreateInfoEXT, but
VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_ENABLE_EXT is set, we end up
dereferencing a NULL pointer.
Fixes future dEQP-VK.pipeline.*.extended_dynamic_state.*.sample_locations_enable_no_create_info
Fixes: 1deb83fb86faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31377 >
2024-09-26 03:43:57 +00:00
c4d6d9254a
vk/image: fix view creation for planar video aspects
...
drivers do implement this
cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Reviewed-by: Dave Airlie <airlied@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31318 >
2024-09-25 06:21:58 +00:00
e171579f51
vk/meta: Make sure texel is 32-bit in build_buffer_to_image_cs()
...
Just like fragment stores, image stores expect 32-bit values (at least
that's the case of the Bifrost backend), so make sure the value passed
to write_img() is always 32-bit, even when convert_texel() doesn't touch
the texel because the image view format matches the buffer format.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Tested-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31227 >
2024-09-20 18:21:50 +00:00
279b5ca10c
vulkan: skip the disk cache when disableInternalCache is true
...
The Vulkan spec says:
"disableInternalCache can be used to disable the driver’s internal
cache, allowing an application to take full control of both memory
and disk usage."
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30812 >
2024-09-10 08:19:52 +00:00
ce9e2dbc62
vk/runtime: Allow "require full subgroups" for Mesh/Task
...
Per description of VkPipelineShaderStageCreateFlags
```
VK_PIPELINE_SHADER_STAGE_CREATE_REQUIRE_FULL_SUBGROUPS_BIT specifies
that the subgroup sizes must be launched with all invocations active in
the task, mesh, or compute stage.
```
Future CTS tests will use that.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31023 >
2024-09-06 05:49:07 +00:00
1e84588887
vulkan/android: change vk_enum_defines.h path
...
To match everything else in vulkan/runtime.
Reviewed-by: Roman Stratiienko <r.stratiienko@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31030 >
2024-09-05 22:59:58 +00:00
9ba70bb1ed
vulkan: Expose a vk_cmd_set_rendering_attachment_locations() helper
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31033 >
2024-09-05 17:55:21 +00:00
ea783a96b8
vulkan: Allow pColorAttachmentLocations == NULL in CmdSetRenderingAttachmentLocationsKHR()
...
Fixes: fe19405c46alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31033 >
2024-09-05 17:55:21 +00:00
c626be82f4
vulkan: Add a helper for getting VkRenderingAttachmentLocaiontInfoKHR
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31033 >
2024-09-05 17:55:21 +00:00
af425a63f7
vulkan/video: fix vui encoding
...
This is a single bit field.
Fixes: d46162981azzoon@igalia.com >
Tested-by: Bernhard C. Schrenk <clemy@clemy.org >
Reviewed-by: Bernhard C. Schrenk <clemy@clemy.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30968 >
2024-09-02 21:15:13 +00:00
42114aa723
vulkan: Handle VIEW_INDEX_FROM_DEVICE_INDEX_BIT in the runtime
...
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30876 >
2024-08-29 03:30:31 +00:00
8c60f1461b
vulkan: Take a VkPipelineCreateFlags2KHR in vk_pipeline_*shader_stage*()
...
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Reviewed-by: Jesse Natalie <jenatali@microsoft.com >
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com >
Reviewed-by: Connor Abbott <cwabbott0@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30876 >
2024-08-29 03:30:31 +00:00
faf27a5989
dzn: Use vk_pipeline_shader_stage_to_nir()
...
Dozen is the last user so we can delete it now.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30876 >
2024-08-29 03:30:31 +00:00
c0191b20de
vulkan/pipeline: Handle VIEW_INDEX_FROM_DEVICE_INDEX_BIT
...
The rehash we're doing here is a bit of a hack but it's a back-portable
hack. We'll fix it properly in following commits.
Fixes: 9308e8d90divan.briano@intel.com >
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30876 >
2024-08-29 03:30:31 +00:00
03655dfda1
compiler, vk: Support subgroup size of 4
...
Relax the assert and assign it an enum value
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30876 >
2024-08-29 03:30:31 +00:00
e57d3551da
vk/render_pass: add Mesa-specific flag for dynamic rendering
...
VK_KHR_dynamic_rendering_local_read allows the application to sample
from a subpass input attachment where this attachment is also the color
attachment (aka. feedback loop). With legacy RPs, it's easy to detect
that at RP creation time by looking at the input<->color indices but
with DRLR this needs to be determined dynamically.
This flag would help when legacy RPs are converted to dynamic rendering
because it's not possible to know if a subpass used input attachments.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30124 >
2024-08-23 12:20:02 +00:00
07c6459cd8
vk/meta: Add copy/fill/update helpers
...
Add buffer copy/fill/update helpers using compute shaders. The driver
can select the optimal per-workgroup copy/fill/update size by specifying
a non-zero vk_meta_device::buffer_access::optimal_size_per_wg size.
If zero, the core will assume a 64-byte size (the usual cache-line size).
Buffer accesses will be done through SSBOs unless
vk_meta_device::buffer_access::use_global_address is true, in which
case the core will the buffer address using GetBufferDeviceAddress()
and pass that address as a push constant to the compute shader.
Image to buffer copies are always done through a compute shader. The
optimal workgroup size will be chosen based on
vk_meta_copy_image_properties::tile_size: the copy logic picks a
workgroup size matching the tile size, and aligns accesses on a tile.
The view format is selected by the driver. To optimize things on the
shader side, pick UINT formats (usually less work to do to pack data).
Buffer to image copies can be done done through the graphics pipeline
if needed (use_gfx_pipeline passed to vk_meta_copy_buffer_to_image()),
which is useful for vendor-specific compressed formats that can't be
written outside of the graphics pipeline. Drivers should normally prefer
compute-based copies when that's an option. Just like for image to buffer
copies, the workgroup size of compute shaders is picked based on the
image tile size, and the view format must be selected by the driver.
Image to image copies is just a mix of the above, with the driver being
able to select the pipeline type, as well as define the tile size and
view format to use. When using a compute pipeline, the workgroup size
will be MAX2(src_tile_sz, dst_tile_sz), and accesses will be aligned
on the selected reference image.
For compressed formats, the caller should pick an RGBA format matching
the compressed block size.
Co-developed-by: Boris Brezillon <boris.brezillon@collabora.com >
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29333 >
2024-08-20 15:11:14 +00:00
dad5c1184f
vk/meta: Let meta users pass the color attachment write mask explicitly
...
Will be needed for partial interleaved depth/stencil copies where the
image is treated as a color image with some components assigned to the
depth and others assigned to the stencil. If only one aspect is copies
using a graphics pipeline, we need to preserve components assigned to
the other aspect, and an easy way to do that is to tweak the color
attachment write mask.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Suggested-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29333 >
2024-08-20 15:11:14 +00:00
8ddc527ba4
vk/image: Fix the view extent of uncompressed views of compressed images
...
The view extent needs to be divided by the block width/height/depth in
that case.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29333 >
2024-08-20 15:11:14 +00:00
6ff9b8c36b
vk/image: Add a vk_image_buffer_range() helper
...
vk_image_buffer_range() returns the buffer range needed for an
image <-> buffer copy, which will help us initialize storage buffer
descriptors.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29333 >
2024-08-20 15:11:14 +00:00
2b2b66f497
vk/sync: Use the proper type in vk_filter_{src,dst}_access_flags2()
...
VkAccessFlags2 and VkPipelineStageFlags2 being both 64-bit bitmasks
the mistake is harmless, but let's fix that anyway to avoid any
confusion.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com >
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30735 >
2024-08-20 14:33:52 +00:00
eacb8f85a2
vulkan/runtime: fix GetBufferMemoryRequirements2 for maintenance4
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: 2649ee0724tapani.palli@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30714 >
2024-08-19 10:13:09 +00:00
1b4ae3d7af
vulkan/video: handle KHR_video_maintenance1 lack of profile list.
...
VK_KHR_video_maintenance1 allows no profile in which cases drivers
should just be pessimisitic.
Reviewed-by: Hyunjun Ko <zzoon@igalia.com >
Reviewed-by: Lynne <dev@lynne.ee >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30671 >
2024-08-16 13:31:41 +10:00
6ae401aa86
vulkan: Add null descriptor bits to vk_pipeline_robustness_state
...
Fixes: c9eac89da8alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30663 >
2024-08-15 22:25:13 +00:00
Valentine Burley