AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
166,422 Commits 1 Branch 0 Tags
a2fbd2fbd02cc989501496c5fd4d98cb6ee9f33e
Commit Graph

166422 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Emma Anholt a2fbd2fbd0 zink: Add tracing of blit operations. 
I found this useful in lining up some perfetto traces between zink+anv and
iris, and understanding what was going on in them.  Also it's a demo of
being able to insert annotations for work in the command stream, which I
suspect we'll want more of.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20657>
 2023-02-09 03:27:34 +00:00
Emma Anholt 4dc6390445 u_trace: Add an interface for checking trace enablement outside a context. 
For zink, we want to know if we should pass command stream markers down to
the underlying driver, but we don't have our own trace context we're
recording trace events with.  We definitely want those markers if the
underlying driver is going to be doing perfetto tracing, or is requesting
marker tracing.  So, create an interface for querying those flags before
they get copied down to an actual u_trace_context.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20657>
 2023-02-09 03:27:34 +00:00
José Roberto de Souza fb6e65c72c hasvk: Remove remaining bits of anv_i915_query() 
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20948>
 2023-02-09 02:57:32 +00:00
José Roberto de Souza 5ecb8966b2 anv: Remove remaining bits of anv_i915_query() 
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20948>
 2023-02-09 02:57:32 +00:00
José Roberto de Souza c8626a20bb anv: Start to move anv_gem_stubs.c to kmd backend 
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20948>
 2023-02-09 02:57:32 +00:00
José Roberto de Souza de79cf0512 anv: Add basic KMD backend infrastructure 
Functions that are in hot paths will have a different treatment to
support i915 and Xe KMD.

Each KMD will have an anv_kmd_backend that will have the hot path
functions set, this way we can avoid branch prediction misses.

Other functions will gradually be moved to anv_kmd_backend.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20948>
 2023-02-09 02:57:32 +00:00
José Roberto de Souza eaeda2107e anv: Use DRM_IOCTL_I915_GEM_CREATE_EXT in all supported kernels 
As we continue to refactor the code base to support Xe KMD here I'm
dropping anv_gem_create() and unifying all graphics memory allocation
calls to anv_gem_create_regions().

anv_gem_create_regions() will call DRM_IOCTL_I915_GEM_CREATE_EXT
for integrated platforms too only leaving DRM_IOCTL_I915_GEM_CREATE
calls to kernel versions that do not support
DRM_IOCTL_I915_GEM_CREATE_EXT.
This can be detected by devinfo->mem.use_class_instance as
DRM_I915_QUERY_MEMORY_REGIONS uAPI landed in the same kernel version
as DRM_IOCTL_I915_GEM_CREATE_EXT.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20948>
 2023-02-09 02:57:32 +00:00
José Roberto de Souza 099b9e5058 anv: Convert drm_i915_gem_memory_class_instance to intel_memory_class_instance 
Also using pointers to intel_device_info struct instead of replicate
the same information.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20948>
 2023-02-09 02:57:32 +00:00
José Roberto de Souza 9fc224631c intel: Add intel_memory_class_instance 
This is a KMD independent struct to hold memory class and instance
values.

drm_i915_gem_memory_class_instance usage will be gradually replaced.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20948>
 2023-02-09 02:57:32 +00:00
Emma Anholt cc0b749890 ci/freedreno: Don't forget to report flakes on a618, too. 
I just noticed that our flake reports are only coming from cheza, which is
a bummer because almost all the VK coverage is on a618 unless you're
kicking off a full run.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21092>
 2023-02-09 02:13:39 +00:00
Amber 20d4c592c9 freedreno: make sure depth/stencil layouts are always tiled 
Small depth/stencil textures were using linear tiling, but depth/stencil
attachments cannot use linear tiling for sysmem rendering.

Fixes:
 KHR-GL45.geometry_shader.layered_framebuffer.stencil_support
 KHR-GL45.geometry_shader.layered_framebuffer.depth_support

Signed-off-by: Amber Amber <amber@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21046>
 2023-02-08 22:55:52 +00:00
Eric Engestrom 885c315d10 docs: update calendar for 22.3.5 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21207>
 2023-02-08 22:26:56 +00:00
Eric Engestrom a0e1707b65 docs: add release notes for 22.3.5 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21207>
 2023-02-08 22:26:56 +00:00
SoroushIMG f7b2dbb2bd zink: relax bresenhamLines requirement for non-strictLine drivers 
non-strictLine Vulkan drivers use either parallelogram or bresenham
rasterization for default line modes.
This method of rasterisation produces close enough results that it
in practice is GL/GLES spec compliant (at least cts wise).
Don't emit a feature missing warning for this case.

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20985>
 2023-02-08 22:00:43 +00:00
Rhys Perry b4383821e7 aco: don't modify exec in p_interp_gfx11 
The RDNA3 ISA docs say that lds_param_load write the entire quad
regardless of exec, so this isn't needed.

fossil-db (gfx1100):
Totals from 5291 (3.93% of 134574) affected shaders:
Instrs: 4891396 -> 4789628 (-2.08%)
CodeSize: 25519032 -> 25111960 (-1.60%)
Latency: 36122982 -> 36074300 (-0.13%); split: -0.14%, +0.00%
InvThroughput: 4162436 -> 4161424 (-0.02%); split: -0.02%, +0.00%
Copies: 263862 -> 263838 (-0.01%)
PreSGPRs: 225012 -> 224179 (-0.37%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21171>
 2023-02-08 19:35:54 +00:00
Georg Lehmann 6e4598f7b9 aco: support omod/imod for v_fmac_f16 
Only matters for post-RA DPP16.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21174>
 2023-02-08 18:52:28 +00:00
Georg Lehmann 2deda5c0be aco: don't list imod/omod support v_fmaak_f32/v_fmamk_f32 
We can never use them anyway because these opcodes don't support VOP3/DPP16/SDWA

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21174>
 2023-02-08 18:52:28 +00:00
Georg Lehmann 4c9ac73064 aco: allow output modifiers for ldexp_f16 
It also supports imod for the first operand, but we cannot express that at
moment.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21174>
 2023-02-08 18:52:28 +00:00
Georg Lehmann b63aa2bb8e aco: don't allow output modifiers for v_cvt_pkrtz_f16_f32 
Cc: mesa-stable
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21174>
 2023-02-08 18:52:28 +00:00
Rob Clark dc93033234 freedreno: Indent fixes 
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21195>
 2023-02-08 18:27:55 +00:00
Rob Clark e9eee8781a freedreno/a2xx: Move pack_rgba() 
It is only used in fd2_draw.c

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21195>
 2023-02-08 18:27:55 +00:00
Rob Clark e7f88c41c4 freedreno: Quiet unused variable warnings 
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21195>
 2023-02-08 18:27:55 +00:00
Rob Clark eaf272aa93 ir3: Quiet unused variable warning 
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21195>
 2023-02-08 18:27:55 +00:00
Sil Vilerino 727afda21f d3d12: Video Encode - Fix ID3D12CommandAllocator leak 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21196>
 2023-02-08 18:00:41 +00:00
Sil Vilerino 997b91fead d3d12: Video processing - Fix out of bounds array access 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21196>
 2023-02-08 18:00:41 +00:00
Rob Clark 2c8c151591 freedreno/drm/virtio: Limit guest handles passed to virtgpu 
We only need to have virtgpu in the loop for shared buffers when
implicit sync is in play.  The rest of the time we can rely on userspace
fence waits.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21143>
 2023-02-08 17:36:23 +00:00
Rob Clark 53ba7d8867 freedreno/drm: Move sa_cpu_prep() to core 
Now that we can do a blocking wait on an fd_fence (which the suballoc
heap already depended on) we can just move the fence wait into core
leaving the backend cpu_prep() implementation only needing to care
about implicit sync on shared buffers.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21143>
 2023-02-08 17:36:23 +00:00
Rob Clark dec49ec50a freedreno/a6xx: Move CS state to PROG state group 
It is pretty easy to just cache the stateobj with the hwcso (since
unlike 3d, there is only a single shader state) and re-emit it by
pointer when it changes, now that the CS state doesn't depend on the
grid info.

This also moves immed consts into the PROG state, so they are only
updated when the PROG state is dirty.  And splits user consts and
driver param consts, so they are only re-emit when needed.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21127>
 2023-02-08 16:59:40 +00:00
Matt Coster 7ea98091c1 pvr: Add firmware stream support for transfer submit 
Move transfer jobs over to the new firmware stream interface. This had already
been done for render and compute jobs in 947e183ff5 ("pvr: Implement new
firmware stream interface").

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21163>
 2023-02-08 16:33:35 +00:00
Sarah Walker 5a2618c73c pvr: Update FWIF transfer queue register structures 
This matches changes made in FW 1.17.OS@6293467.

Signed-off-by: Sarah Walker <sarah.walker@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21163>
 2023-02-08 16:33:35 +00:00
Michel Dänzer 0dd89dacf2 mesa/st: Handle all 10 bpc types in st_choose_format 
Only GL_UNSIGNED_INT_2_10_10_10_REV was handled, add
GL_UNSIGNED_INT_10_10_10_2 & GL_UNSIGNED_INT_10_10_10_2_OES.

This makes sure that if the Gallium driver doesn't support the exact
corresponding format, another 10 bpc format is tried before an 8 bpc one
as a fallback.

Fixes the mutter test cogl-test-offscreen-texture-formats with iris.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21021>
 2023-02-08 16:06:38 +00:00
Michel Dänzer 98526b7d48 mesa/st: Fix GL_EXT_texture_type_2_10_10_10_REV name in comment 
Suggested by Adam Jackson in
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21021#note_1747761

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21021>
 2023-02-08 16:06:37 +00:00
Lionel Landwerlin 6390953fa5 anv: put the video extension behind a variable 
As initialized acked.

The WSI code is currently trying to blit on the video queue.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 633efa7af7 ("anv: enable video decode extensions.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21194>
 2023-02-08 15:38:27 +00:00
Simon Perretta 81af999a75 pvr: Clarify unreachable text 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta c13bed8bef pvr: Add block printing support during validation 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta 1bd84f4753 pvr: Validate instruction repeat and src/dst sizes 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta 8ac16c97cf pvr: Adjust instruction repeat offset 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta d431adaecd pvr: Keep NIR SSA defs instead of registers 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta d47d0ea101 pvr: Drop the ENUM_PACKED macro 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta 52bcb0e292 pvr: Commonise some instruction member defs 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta e74aaa347b pvr: Support dual-destination ALU instructions 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta 6187f1fd8c pvr: Add support for optional instruction params 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Simon Perretta d187418f63 pvr: Add new Rogue compiler framework 
Signed-off-by: Simon Perretta <simon.perretta@imgtec.com>
Acked-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20430>
 2023-02-08 15:15:17 +00:00
Georg Lehmann 5038a049f1 aco: add mov/cndmask opcodes to does_fp_op_flush_denorms 
For completeness sake also add v_mov_b32, even if we don't use imod for it
because it's only supported since gfx10.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21170>
 2023-02-08 13:07:46 +00:00
Georg Lehmann c8adf16278 aco: fix imod/omod for gfx11 VOP3 opcodes 
Fixes: d8d99c3c4f ("aco: add GFX11 opcode numbers")
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21170>
 2023-02-08 13:07:46 +00:00
Mike Blumenkrantz 8f5a63ebfe util/vbuf: fix multidraw unrolling 
* drawid needs to be incremented at end of loop
* rescope draw structs to initialize at top of loop

fix #8191

Fixes: 6b07893b31 ("util/vbuf: handle multidraws")

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Mark Janes <markjanes@swizzler.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21139>
 2023-02-08 12:41:32 +00:00
Tatsuyuki Ishi c6ebd2de71 radv: Fix noop FS not getting constructed for GPL pipelines. 
The condition was inverted, causing compilation to be actually skipped when
a noop FS is used and straight emitting the pipeline from the default
initialized struct.

Fixes: 3eb97b9d33 ("radv: skip compilation when possible with GPL fast-linking")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21166>
 2023-02-08 10:52:54 +00:00
Daniel Schürmann e5600d5257 radv: CSE ray_launch_{size|id} 
Stats for Q2RTX
Totals from 7 (0.01% of 134913) affected shaders: (GFX10.3)

VGPRs: 736 -> 704 (-4.35%)
CodeSize: 204424 -> 204400 (-0.01%); split: -0.03%, +0.02%
MaxWaves: 67 -> 69 (+2.99%)
Instrs: 37540 -> 37549 (+0.02%); split: -0.02%, +0.04%
Latency: 973556 -> 973267 (-0.03%)
InvThroughput: 209068 -> 200902 (-3.91%)
VClause: 920 -> 921 (+0.11%); split: -0.11%, +0.22%
SClause: 1045 -> 1043 (-0.19%)
Copies: 4853 -> 4865 (+0.25%); split: -0.08%, +0.33%
Branches: 1571 -> 1578 (+0.45%)
PreSGPRs: 421 -> 407 (-3.33%)
PreVGPRs: 638 -> 632 (-0.94%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21156>
 2023-02-08 10:10:27 +00:00
Timur Kristóf 8ebb34ee57 radv: Refactor radv_emit_ngg_culling_state so it's based on dirty flags. 
This significantly lowers the CPU overhead of this function.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
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/20980>
 2023-02-08 09:28:20 +00:00
Timur Kristóf 22947ef0cc radv: Remove NGG culling skip from command buffer. 
This code used to runtime-disable NGG culling for small draw calls.
However, this had too much CPU overhead, let's remove it.
It will be solved by the shaders instead.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
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/20980>
 2023-02-08 09:28:19 +00:00