Italo Nicola [Fri, 3 Feb 2023 21:20:31 +0000 (21:20 +0000)]
panfrost/genxml: add Multiplanar Surface descriptor
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>
Italo Nicola [Thu, 22 Jun 2023 10:43:39 +0000 (10:43 +0000)]
panfrost/genxml: fix Surface With Stride descriptor alignment
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>
Luigi Santivetti [Wed, 2 Aug 2023 13:32:12 +0000 (14:32 +0100)]
pvr: do not claim support for ASTC texture compression
We don't support any ASTC formats yet, but the textureCompressionASTC_LDR
feature was incorrectly set to true. Fix this by setting it to false and
don't advertise ASTC support for texture compression.
Fixes dEQP-VK.api.info.format_properties.compressed_formats
Fixes:
8991e646 ("pvr: Add a Vulkan driver for Imagination Technologies PowerVR Rogue GPUs")
Signed-off-by: Luigi Santivetti <luigi.santivetti@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24448>
antonino [Mon, 7 Aug 2023 12:53:50 +0000 (14:53 +0200)]
freedreno: add ci flake
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24539>
antonino [Mon, 7 Aug 2023 12:04:14 +0000 (14:04 +0200)]
virgl: add ci flake
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24539>
Daniel Stone [Fri, 4 Aug 2023 11:39:40 +0000 (12:39 +0100)]
panfrost/vk: Use correct sampler dimensions for MSAA
2D-MS samplers are SAMPLER_DIM_MS, not SAMPLER_DIM_2D. There is no MSAA
for non-2D textures.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Closes: mesa/mesa#9474
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24494>
Lionel Landwerlin [Sat, 5 Aug 2023 09:31:58 +0000 (12:31 +0300)]
anv: enable INTEL_DEBUG=nofc
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24510>
Iago Toral Quiroga [Thu, 3 Aug 2023 06:37:58 +0000 (08:37 +0200)]
v3dv: drop cpu path for buffer to image copies
We really want to avoid cpu jobs as much as possible. Also,
the texel buffer path we have for this should be able to
handle most cases.
Reviewed-by: Alejandro PiƱeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24466>
Chris Spencer [Sun, 6 Aug 2023 20:54:38 +0000 (21:54 +0100)]
radv: initialize result when pipeline cache creation fails
Signed-off-by: Chris Spencer <spencercw@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24521>
Dave Airlie [Mon, 7 Aug 2023 01:01:27 +0000 (11:01 +1000)]
zink: use fprintf instead of printf to align the requirements warnings
This just seems to have been overlooked and nvk triggers it right now.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24527>
Mike Blumenkrantz [Thu, 3 Aug 2023 17:11:57 +0000 (13:11 -0400)]
zink: force image barriers after dmabuf import
if the image will be used, then it has to be on the right queue
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24492>
Mike Blumenkrantz [Fri, 4 Aug 2023 16:17:20 +0000 (12:17 -0400)]
zink: better handle separate shader dsl creation when no bindings exist
this otherwise underflows the array and provides a (probably huge) garbage
value for the binding id, which then causes the driver to massively overallocate
both the layout and set/pool/buffer
the main result of this is that on radv any simple test that should be near-instant
takes 2-3 seconds to execute, which somehow nobody noticed
Fixes:
e3b746e3a31 ("zink: use GPL to handle (simple) separate shader objects")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24501>
Dave Airlie [Sun, 6 Aug 2023 21:53:50 +0000 (07:53 +1000)]
docs: add two nvk exts to features.txt
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24523>
Jordan Justen [Wed, 28 Dec 2022 21:48:53 +0000 (13:48 -0800)]
intel/decoder: Make intel_spec_load_filename() have separate dir and name strings
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20889>
Jordan Justen [Wed, 28 Dec 2022 21:26:34 +0000 (13:26 -0800)]
intel/decoder: Add intel_spec_load_common()
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20889>
Jordan Justen [Wed, 28 Dec 2022 18:20:20 +0000 (10:20 -0800)]
intel/decoder: Fix xml filename when verx10 % 10 is not 0
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20889>
Gert Wollny [Sun, 6 Aug 2023 18:42:07 +0000 (20:42 +0200)]
r600/sfn: Take source uses into account when switching channels
When we switch the channels by re-creating vec4 values we have to
take into account that the source values may be used in an ALU op,
and with that we have to take read-port limitations into account.
Fixes:
18a8d148d862d4a0f5b83f4385f00cf8a4a71b6d
r600/sfn: Cleanup copy-prop into vec4 source values
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24519>
Faith Ekstrand [Sun, 6 Aug 2023 16:35:26 +0000 (11:35 -0500)]
nvk: Call nir_lower_int64
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24518>
Faith Ekstrand [Sun, 6 Aug 2023 16:55:43 +0000 (11:55 -0500)]
nvk: Use more consistent device variable names
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24518>
David Heidelberg [Sat, 5 Aug 2023 07:54:43 +0000 (10:54 +0300)]
ci/freedreno: document another flakes on Adreno 530
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24511>
Karol Herbst [Wed, 2 Aug 2023 13:34:29 +0000 (15:34 +0200)]
nv50/ir: convert system values to gl_system_value
Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24447>
Karol Herbst [Wed, 2 Aug 2023 13:32:37 +0000 (15:32 +0200)]
nv50/ir: use own info struct for sys vals
This makes it more obvious what's actually needed by the driver and allows
us to drop some code in the nir to nvir step.
Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24447>
Faith Ekstrand [Thu, 3 Aug 2023 23:32:45 +0000 (18:32 -0500)]
nvk: Enable the new UAPI
We also add a meson option to disable it and re-enable the legacy API.
This code probably won't last long but it gives developers the option.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 17:03:16 +0000 (12:03 -0500)]
nvk: Print an error message if VM_BIND support is missing
This should prevent a lot of bug reports.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 16:59:21 +0000 (11:59 -0500)]
nouveau/winsys: Allow nouveau_ws_device_new() without VM_BIND
Silently fall back to non-VM_BIND. The client can check for the
has_vm_bind flag in the device to detect whether VM_BIND is actually
available or not.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Sun, 23 Jul 2023 23:38:44 +0000 (09:38 +1000)]
nvk: add new internal bo allocation flag.
This is to allow the kernel to use a single resv object, this might
need more work.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 21 Jul 2023 20:03:25 +0000 (15:03 -0500)]
nvk: Use an empty EXEC for the empty submit case
The new UAPI allows for an EXEC with zero pushes just fine so we no
longer need the no-op push just for synchronization. This lets us drop
the whole empty push we allocate per-queue as well.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Thu, 20 Jul 2023 04:52:00 +0000 (14:52 +1000)]
nvk: enable sparse residency buffer on maxwell+
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Wed, 19 Jul 2023 01:27:33 +0000 (11:27 +1000)]
nvk: enable a bunch of external fence/semaphore bits
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Tue, 10 Jan 2023 22:20:10 +0000 (08:20 +1000)]
nvk: enable sparse features
There is a bug on my turing with
dEQP-VK.sparse_resources.buffer.ssbo.sparse_residency.buffer_size_2_24
but none of the others and we aren't seeing it on some other devices.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Wed, 22 Mar 2023 03:14:37 +0000 (13:14 +1000)]
nvk: Handle pre-turing indirect buffers with sparse
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Tue, 10 Jan 2023 22:19:41 +0000 (08:19 +1000)]
nvk/queue: add support for syncobjs and sparse binds
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Tue, 10 Jan 2023 22:18:44 +0000 (08:18 +1000)]
nvk: Add support for sparse images
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Wed, 19 Jul 2023 21:45:27 +0000 (16:45 -0500)]
nvk: Add support for sparse buffers
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Mon, 24 Jul 2023 19:26:09 +0000 (14:26 -0500)]
nouveau/mme: Support the new UAPI
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Tue, 10 Jan 2023 22:12:22 +0000 (08:12 +1000)]
nouveau/winsys: add support for the vma bind interfaces
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Wed, 2 Nov 2022 06:54:33 +0000 (16:54 +1000)]
nvk: always sync internal cmd bufs for vma lifetimes.
We always sync on the internal push buf submissions for zero vram
and ctx setup, these shouldn't be bottlenecks.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Mon, 31 Jul 2023 13:42:46 +0000 (08:42 -0500)]
nouveau/shim: Use the imported nouveau_drm.h headers
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Mon, 31 Jul 2023 13:42:14 +0000 (08:42 -0500)]
nvk: Use the imported nouveau_drm.h headers
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Mon, 31 Jul 2023 13:30:57 +0000 (08:30 -0500)]
nouveau/winsys: Use the imported nouveau_drm.h headers
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Mon, 31 Jul 2023 13:22:50 +0000 (08:22 -0500)]
drm-uapi: Import nouveau_drm.h
From https://cgit.freedesktop.org/drm-misc/
commit
e02238990b1ab2dfc8abb4c28369f1da6e863f81
Author: Danilo Krummrich <dakr@redhat.com>
Date: Fri Aug 4 20:23:43 2023 +0200
drm/nouveau: new VM_BIND uAPI interfaces
Signed-off-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 16:17:14 +0000 (11:17 -0500)]
nouveau: Move shader topology info to nv_device_info
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 15:46:05 +0000 (10:46 -0500)]
nvk: Decorate CmdBegin/EndRendering entrypoints
These decorations are necessary for proper building on Android and
Windows. You don't notice them in regular Linux builds, though.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 15:45:25 +0000 (10:45 -0500)]
nvk: Implement GetRenderingAreaGranularityKHR
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 15:43:15 +0000 (10:43 -0500)]
nvk: Implement CmdBindIndexBuffer2KHR
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 15:41:23 +0000 (10:41 -0500)]
nvk: Use VkBufferUsageFlags2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 15:35:24 +0000 (10:35 -0500)]
nvk: Implement the maintenance5 image layout queries
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 16:36:41 +0000 (11:36 -0500)]
nvk: Rework memory requirements to handle aspects correctly
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Fri, 4 Aug 2023 00:54:35 +0000 (19:54 -0500)]
nvk: Misc. style nits
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Wed, 26 Jul 2023 07:37:37 +0000 (17:37 +1000)]
nvk: enable EXT_depth_clip_enable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Wed, 26 Jul 2023 07:27:43 +0000 (17:27 +1000)]
nvk: enable EXT_depth_clip_control
This is already handled in the dynamic state.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 16:48:22 +0000 (11:48 -0500)]
nvk: Implement CopyQueryPoolResults with a compute shader
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 16:47:26 +0000 (11:47 -0500)]
nvk: Separate the MME query copy code out a bit
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 15:48:23 +0000 (10:48 -0500)]
nvk: Enable the SPIR-V DeviceGroup capability
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 12:26:10 +0000 (07:26 -0500)]
nvk: Add support for variable pointers
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 11:27:44 +0000 (06:27 -0500)]
nvk: Add a root_desc_addr to the root descriptor table
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 14:54:47 +0000 (09:54 -0500)]
nvk: Use nvk_root_descriptor_offset for drawInfoBase
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 09:34:10 +0000 (04:34 -0500)]
nvk: Rework multi-plane format features a bit
We need to avoid returning certain feature flags such as rendering or
storage from vkGetPhysicalDeviceFormatFeatures*() but still allow those
for single-plane views of the image which means allowing them in
vkGetPhysicalDeviceImageFormatProperties*().
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 2 Aug 2023 04:49:24 +0000 (23:49 -0500)]
nvk: Improve image format properties and limits
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Mohamed Ahmed [Tue, 4 Jul 2023 22:46:06 +0000 (01:46 +0300)]
nouveau/nvk: Enable VK_KHR_sampler_ycbcr
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Mohamed Ahmed [Tue, 25 Jul 2023 11:51:46 +0000 (14:51 +0300)]
nouveau/nvk: Add multiple sampler planes for CONVERSION_SEPARATE_RECONSTRUCTION_FILTER_BIT
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Mohamed Ahmed [Sat, 29 Jul 2023 20:55:51 +0000 (23:55 +0300)]
nouveau/nvk: Create helper function for sampler creation
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Mohamed Ahmed [Thu, 27 Jul 2023 09:46:22 +0000 (12:46 +0300)]
nouveau/nvk: Support multi-plane descriptors in nvk_nir_lower_descriptors.c
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Mohamed Ahmed [Tue, 25 Jul 2023 22:17:45 +0000 (01:17 +0300)]
nouveau/nvk: Add YCbCr sampler NIR lowering pass
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Tue, 1 Aug 2023 23:26:43 +0000 (18:26 -0500)]
nvk: Use the new NIL helper for image level extents for copies
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Tue, 1 Aug 2023 23:25:55 +0000 (18:25 -0500)]
nil: Add a nil_image_level_extent_px() helper
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Thu, 20 Jul 2023 05:35:12 +0000 (15:35 +1000)]
nvk: enable rgb32 texel buffer support
This is needed for zink.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Emma Anholt [Tue, 18 Jul 2023 18:37:14 +0000 (11:37 -0700)]
nvk: Disable shaderStorageImageReadWithoutFormat pre-Maxwell.
On Kepler and earlier, image loads must get lowered based on the declared
format, since we don't have a SULDP op.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Mon, 31 Jul 2023 22:27:40 +0000 (17:27 -0500)]
nouveau: Drop GART size from nv_device_info
The kernel always returns something rediculously high, like 40 bits or
something. There's really no point in us tracking it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
George Ouzounoudis [Sun, 30 Jul 2023 12:31:56 +0000 (15:31 +0300)]
nouveau/codegen: Do not keep redundant info for tessellation domain
Just add an assert instead.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
George Ouzounoudis [Sun, 30 Jul 2023 12:29:34 +0000 (15:29 +0300)]
nvk: Do not keep redundant info for tessellation domain
This is not needed after merge_tess_info that or's the info of both the
tesc and tese shaders.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Mon, 31 Jul 2023 07:37:15 +0000 (17:37 +1000)]
nvk: fix independent color write masks.
This fixes:
dEQP-VK.renderpass.suballocation.attachment_write_mask.attachment_count_*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 27 Jul 2023 21:03:38 +0000 (16:03 -0500)]
docs/relnotes: Stick something about NVK in new_features.txt
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 27 Jul 2023 21:01:31 +0000 (16:01 -0500)]
docs: Add NVK to features.txt
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 27 Jul 2023 21:17:43 +0000 (16:17 -0500)]
docs: Add a docs page for NVK
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 27 Jul 2023 22:06:57 +0000 (17:06 -0500)]
nvk: Don't skip zero-size bindings in GetDescriptorSetLayoutSupport
We still need to handle them in the case of variable descriptor counts.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Wed, 26 Jul 2023 23:38:04 +0000 (09:38 +1000)]
nvk: align geometry clip setting with nvc0
This fixes
dEQP-VK.clipping.clip_volume.depth_clamp.point_list,Fail
dEQP-VK.clipping.clip_volume.depth_clip.point_list,Fail
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 27 Jul 2023 19:12:05 +0000 (14:12 -0500)]
nvk: Flush more state on VI_BINDINGS_VALID dirty
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 27 Jul 2023 04:06:57 +0000 (23:06 -0500)]
nvk: Bind 3D depth/stencil images as 2D arrays
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Dave Airlie [Thu, 27 Jul 2023 02:58:28 +0000 (12:58 +1000)]
nvk/tess: don't emit patch control points in pipeline
Let the dynamic state thing emit these.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 26 Jul 2023 21:22:20 +0000 (16:22 -0500)]
vulkan/meta: Use vk_meta_get_pipeline_layout in blit/resolve
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 26 Jul 2023 21:20:50 +0000 (16:20 -0500)]
vulkan/meta: Add a get_pipeline_layout helper
This helper handles the cache lookup and constructs a pipeline layout
and a descriptor set layout, as needed, all in one go. This saves a bit
of boilerplate in the various meta functions.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 26 Jul 2023 19:22:58 +0000 (14:22 -0500)]
Revert "vulkan: Allow scissors or viewports to be set without counts"
This reverts commit
b551f26a0caad52de620222e2df8416e80332943.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 26 Jul 2023 16:15:19 +0000 (11:15 -0500)]
nvk: Use the new core vk_sampler struct
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Karol Herbst [Wed, 26 Jul 2023 00:30:32 +0000 (02:30 +0200)]
nvk: enable fp helper invocations loads on more gens
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Wed, 26 Jul 2023 14:31:34 +0000 (09:31 -0500)]
nvk: Require an environment variable for poorly tested hardware
We really only know that Turing and Ampere work right now. Maxwell has
recieved some testing but still can't survive a CTS run. Other hardware
is even less well tested.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Karol Herbst [Sat, 22 Jul 2023 17:33:45 +0000 (19:33 +0200)]
nouveau: fix max_warps_per_mp_for_sm for builds with asserts disabled
Fixes:
1b756be2bca ("nvk: properly calculate SLM region by taking per arch limits into account")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Karol Herbst [Thu, 20 Jul 2023 20:50:16 +0000 (22:50 +0200)]
nvk: properly calculate SLM region by taking per arch limits into account
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Karol Herbst [Thu, 20 Jul 2023 20:48:58 +0000 (22:48 +0200)]
nvk: replace mp with tpc
The tls space is calculate per TPC, but nouveau also doesn't report the MP
count to us, but the TPC count instead.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Karol Herbst [Thu, 20 Jul 2023 13:38:13 +0000 (15:38 +0200)]
nvk: fix num_gprs for Volta+
Fixes OOR_REG errors being thrown
Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 19:59:54 +0000 (14:59 -0500)]
nvk: Implement VK_EXT_physical_device_drm
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 20:02:12 +0000 (15:02 -0500)]
nvk: Don't enumerate pre-Kepler GPUs
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 17:49:34 +0000 (12:49 -0500)]
nvk: Move the winsys device to nvk_device
Vulkan requires that different VkDevices be unable to affect each other.
To correctly provide this invariant, each VkDevice really needs its on
virtual address space. This means opening a new nouveau_device per
nvk_device. This may make device creation a tiny bit less efficient but
we can fix that with some sort of caching if we absolutely need to.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 17:49:34 +0000 (12:49 -0500)]
nvk: Add a ws_dev to nvk_device and use it
At the moment, this is just a convenient short-hand but we're about to
permanently move the nouveau_ws_device to nvk_device so it will be the
only way to access it soon.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 17:47:57 +0000 (12:47 -0500)]
nvk: Rename nvk_device::ctx to ws_ctx
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 17:41:36 +0000 (12:41 -0500)]
nvk: Use nv_device_info for class checks
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 17:16:08 +0000 (12:16 -0500)]
nouveau: Move gart_size to nv_device_info
This is more complicated than the others since it is a calculated value.
For nv_device_info struct, we want the raw version that raw query from
the chip. We move the calculation into nvk_create_drm_physical_device().
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Tue, 31 Jan 2023 02:53:17 +0000 (20:53 -0600)]
nouveau: Move more stuff into nv_device_info
The device_id and is_integrated fields in nouveau_device can be dropped
entirely since they're redundant with nv_device_info::device_id and
nv_device_info::type.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Thu, 20 Jul 2023 16:45:40 +0000 (11:45 -0500)]
nvk: Set deviceType based on nv_device_info::type
The is_integrated bool is based on VRAM size which is a bit clunky and
not really what the Vulkan query means.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>
Faith Ekstrand [Tue, 31 Jan 2023 02:53:17 +0000 (20:53 -0600)]
nouveau: Stop using hex for SM numbers
There's no indication that NVIDIA considers this to be anything other
than a decimal number. Much to the contrary, the number of 5s and 9s in
the SM number table implies they very much are base 10. Given that
Lovelace is SM90, we'll likely be hitting 100 fairly soon at which point
this will stop fitting in a uint8_t. Switch to decimal to compress it
down a bit.
Also, rename the field to SM which beter matches the NVIDIA docs and
ensures that there are no users of this which are likely to be confused.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>