Qiang Yu
d9f7902afb
radeonsi: init aco shader info for merged LS/HS
...
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/24204 >
2023-07-24 01:49:21 +00:00
Qiang Yu
7daa0857c0
radeonsi: extract si_get_prev_stage_nir_shader to be shared with aco
...
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/24204 >
2023-07-24 01:49:21 +00:00
Qiang Yu
ec17cc345f
radeonsi: aco does not pass LS outputs to HS by arg
...
aco has global input/output variables for this.
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/24204 >
2023-07-24 01:49:21 +00:00
Qiang Yu
599b50b448
aco,radv: replace tess_input_vertices shader info param
...
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com >
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/24204 >
2023-07-24 01:49:20 +00:00
David Heidelberg
826c570ab3
ci/freedreno: cover all texture gather flakes
...
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24300 >
2023-07-24 01:37:01 +02:00
Konstantin Seurer
01266f8119
llvmpipe: Fix compiling with LP_USE_TEXTURE_CACHE
...
Fixes: 36eb75d ("llvmpipe: move to common sampler/image binding code")
Closes : #9359
Reviewed-by: Roland Scheidegger <sroland@vmware.com >
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24243 >
2023-07-23 19:11:40 +00:00
Bas Nieuwenhuizen
c2e3986326
nir: Fix 16-component nir_replicate.
...
Fixes: f534c2c539 ("nir/builder: Add nir_replicate helper")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24286 >
2023-07-22 22:11:15 +00:00
Bas Nieuwenhuizen
e536d31a46
aco: Fix some constant patterns in 16-bit vec4 construction with s_pack.
...
Fixes: 04e3d7ad93 ("aco: improve nir_op_vec with constant operands")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24286 >
2023-07-22 22:11:15 +00:00
Bas Nieuwenhuizen
2fcf7c7014
aco: fix nir_op_vec8/16 with 16-bit elements.
...
Fixes: 5718347c2b ("aco: implement vec2/3/4 with subdword operands")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24286 >
2023-07-22 22:11:15 +00:00
Alyssa Rosenzweig
e890bb0e75
asahi: Don't depend on glibc to decode
...
fopencookie is a glibc feature, so we can't use it on macOS (and
probably other libc's?). It's only used for the hypervisor interface,
though, so we can just make the hypervisor piece glibc-only while
otherwise fixing the wrap.dylib build.
Fixes: ee83453f69 ("asahi: Add a shared library interface for decode")
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24293 >
2023-07-22 12:42:58 -04:00
Eric Engestrom
f997d32f9f
asahi: drop unused include paths
...
Signed-off-by: Eric Engestrom <eric@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24287 >
2023-07-22 10:10:03 +00:00
Christian Gmeiner
2572a96162
ci/etnaviv: update ci expectations
...
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24291 >
2023-07-22 04:16:32 +00:00
Chia-I Wu
5cca1124d1
amd/ci: update radv-stoney-aco-fails.txt for depth/stencil resolve
...
image_2d_16_64_6 ones have been fixed by the previous commit. The
others are outdated.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23959 >
2023-07-22 02:32:31 +00:00
Chia-I Wu
e7c4ebc0cd
radv: disable tc-compat htile for layered images on gfx8
...
sliceInterleaved may be true for layered images on gfx8. Such a htile
cannot be cleared with radv_clear_htile.
Fixes 24 failures in
dEQP-VK.renderpass2.depth_stencil_resolve.image_2d_16_64_6.* on GFX8.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23959 >
2023-07-22 02:32:31 +00:00
Thomas H.P. Andersen
d84d5ff0ce
tgsi: drop two unused functions
...
Removes:
* tgsi_util_get_src_from_ind
* tgsi_full_src_register_from_dst
The last usage of these got removed in !24175
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24283 >
2023-07-22 02:04:57 +00:00
Yiwei Zhang
2ed4f04869
venus: use in_render_pass to skip present_src counting
...
It's an early return also benefiting dynamic rendering. We then no
longer need to track the legacy pass from inheritance info.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
e47da97be6
venus: refactor more cmd states into cmd builder
...
This change:
- adds helpers for cmd begin/end rendering
- simplifies cmd reset
- updates ordering to align with cmd builder
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
10c791619c
venus: avoid redundant tracking of render pass
...
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
540242f9ff
venus: add helpers to track subpass view mask
...
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
311a0eeb21
venus: cleanup vn_cmd_begin_render_pass usage
...
For secondary command buffers, vn_cmd_begin_render_pass was only used to
track inherited render pass previously. So we clean it up.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
81b69f8e8b
venus: use tracked queue_family_index from the cmd pool
...
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
72728f83ed
venus: remove redundant fb tracking from cmd builder
...
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
f0b5a6335d
venus: move transient storage from cmd to pool
...
The storage is for command scope usage, so it fits better for the pool.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
566df7821b
venus: log and doc the broken query feedback in suspended render pass
...
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:43 +00:00
Yiwei Zhang
d13bf8f6b1
venus: fix cmd state leak across implicit reset
...
Reset cmd states during vkBeginCommandBuffer regardless of the
VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT for simplicity.
Cc: mesa-stable
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:42 +00:00
Yiwei Zhang
a9f058d614
venus: fix a cmd builder render_pass state leak across reset
...
Cc: mesa-stable
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24103 >
2023-07-22 01:49:42 +00:00
Dave Airlie
2b03938f27
gallivm: fix atomic global temporary storage.
...
Fixes regression on llvm15 with
piglit tests/cl/program/execute/builtin/atomic/atomic_xchg-global.cl
Fixes: f281290005 ("gallivm: Fix atomic_global types")
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com >
Tested-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24267 >
2023-07-21 20:48:16 +00:00
Charmaine Lee
9c17d9df43
svga: set clear_texture to NULL for vgpu9
...
With PIPE_CAP_CLEAR_TEXTURE removed, we need to set clear_texture to NULL
on svga vgpu9 device so it can use the fallback path.
Fixes: a1eabeff66 ("gallium: remove PIPE_CAP_CLEAR_TEXTURE")
Reviewed-by: Neha Bhende <bhenden@vmware.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24264 >
2023-07-21 17:36:31 +00:00
Yiwei Zhang
fd9a0ecc86
ci/venus: update venus-lavapipe expectations
...
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24263 >
2023-07-21 17:11:26 +00:00
Yiwei Zhang
8cb7bab341
lvp: avoid reading immutable sampler from desc write info
...
Lavapipe has switched to layer push descriptor support atop descriptor
updates internally since 12a7fc51c7 , so
it must skip retrieving immutable samplers from the write info even if
the update call itself is blessed by the spec to not hit that case.
Fixes: 12a7fc51c7 ("lavapipe: Rework descriptor handling")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24263 >
2023-07-21 17:11:26 +00:00
Lionel Landwerlin
1f3e320d15
vulkan: bump header register to 1.3.258
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24279 >
2023-07-21 16:36:26 +00:00
Connor Abbott
c75bf084d0
tu, freedreno/a6xx: Remove has_ccu_flush_bug
...
Based on the previous commit, this isn't actually a bug and is expected
behavior. Turnip should already be handling it correctly for user
flushes, we just have to make sure to handle it for flushes we insert
ourselves in turnip and freedreno.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24162 >
2023-07-21 16:02:50 +00:00
Connor Abbott
847834461e
tu: Fix and simplify execution dependency handling
...
When I wrote this code, I was under the impression that at most one
context from each cluster could be executing at a time. This would mean
that we could treat clusters as pipeline stages and only insert a WFI if
there was a bubble where an earlier stage depends on the result of a
later stage.
This mental model was wrong, though. Experiments on a6xx show it's
possible for two contexts to be executing simultaneously, even though
there are only two contexts - register writing is just stalled until the
earliest-launched context finishes.
This means that the mental model is now much simpler. Any draw can, in
theory, execute in parallel with any previous draw, blit, flush, etc,
although it seems that flushes do wait for any earlier work to finish.
Clusters are mostly just an implementation detail that only matter in
some corner cases, like setting a non-context register (written in the
last cluster) that is used by an earlier cluster that can race ahead of
the write.
An example where this makes a difference is a fragment shader that
writes an image via stib followed by a blit from that same image.
Because both operations happen in the same cluster and use the same
cache, we wouldn't emit anything in the barrier, however actually we
still need to WFI.
This was getting worse on a7xx because later clusters now have 4
contexts, making it easier for draws to be executed in parallel. However
AFAICT it was already a problem on a6xx.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24162 >
2023-07-21 16:02:50 +00:00
Connor Abbott
ed02094235
tu: Fix vk2tu_*_stage flag type
...
New flags were silently getting dropped.
Fixes: 59259a0167 ("tu: Convert to sync2 entrypoints")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24162 >
2023-07-21 16:02:50 +00:00
Eric Engestrom
e1abf85bf5
docs: update calendar for 23.1.4
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24281 >
2023-07-21 15:53:50 +00:00
Eric Engestrom
0720b0a613
docs: add sha256sum for 23.1.4
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24281 >
2023-07-21 15:53:50 +00:00
Eric Engestrom
f6de634e6b
docs: add release notes for 23.1.4
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24281 >
2023-07-21 15:53:50 +00:00
David Rosca
0bd9ab232b
gallium/auxiliary/vl: Fix RGB->YCbCr full range matrix
...
Also rename it to bt_709_rev_full as there already
is bt_709 which is used for YCbCr->RGB.
Fixes: 8a21efce3a ("frontends/va: Add postproc support for converting to full range")
Reviewed-by: Leo Liu <leo.liu@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24238 >
2023-07-21 15:16:26 +00:00
Sathishkumar S
0195e1204b
radeonsi/vcn: num bs_bufs must be proportional to num jpeg engines
...
using limited number of bs buffers constraints the simultaneous
use of all available jpeg engines especially when count is lesser than
that of the available engines. make sure the number of buffers
available are more than or equal to the number of jpeg engines on the asic.
Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com >
Reviewed-by: Leo Liu <leo.liu@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24240 >
2023-07-21 14:42:09 +00:00
Sathishkumar S
47a9d01641
radeonsi/vcn: support variable number of bs_bufs
...
add support to use variable number of bitstream buffers for decode
v2: remove the always true if condition (CI report)
Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com >
Reviewed-by: Leo Liu <leo.liu@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24240 >
2023-07-21 14:42:09 +00:00
Samuel Pitoiset
53d35c641d
radv: bind the pre-compiled PS epilog to the cmdbuf state
...
For PS epilogs we have two paths, the first one is to pre-compile PS
epilogs at pipeline creation time, while the second one is to compile
PS epilogs on-demand when some dynamic states are used.
Binding the pre-compiled PS epilog to the cmdbuf state allows us to
remove one more pipeline dependency when recording cmdbufs (for shader
objects).
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24254 >
2023-07-21 12:55:30 +00:00
Samuel Pitoiset
55f05309ca
radv: pass a shaders array for computing ia_multi_vgt_param
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24251 >
2023-07-21 12:19:28 +00:00
Samuel Pitoiset
b0f53c3a48
radv: remove unused param in radv_pipeline_emit_vgt_gs_out()
...
Also rename the function.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24251 >
2023-07-21 12:19:28 +00:00
Samuel Pitoiset
3348cd3a80
radv: stop using a pipeline for emitting VGT_VERTEX_REUSE_BLOCK_CNTL
...
Passing a TES is enough.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24251 >
2023-07-21 12:19:28 +00:00
Alyssa Rosenzweig
03b2c34793
nir: Remove register arrays
...
Nothing produces them any more, so remove them from NIR. This massively reduces
the size of nir_src, which should improve performance all over.
nir_src size reduced from 56 bytes -> 40 bytes (pahole results on arm64, x86_64
should be similar.)
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24253 >
2023-07-21 11:25:49 +00:00
Alyssa Rosenzweig
1466014184
nir: Rename lower_locals_to_reg_intrinsics back
...
The short name is freed up.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24253 >
2023-07-21 11:25:49 +00:00
Alyssa Rosenzweig
d2c94f9e71
nir: Remove nir_lower_locals_to_regs
...
No more users, all switched to the intrinsic version.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24253 >
2023-07-21 11:25:49 +00:00
Alyssa Rosenzweig
a8c0b6695f
panfrost: Remove unused helpers
...
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24253 >
2023-07-21 11:25:48 +00:00
Alyssa Rosenzweig
a08286f993
intel/fs: Don't read reg.base_offset
...
It's not set in the new intrinsics path.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24253 >
2023-07-21 11:25:48 +00:00
Rohan Garg
01965a2fe9
anv: drop CFE state validation checks
...
anv no longer needs to track if the CFE state is valid since we ensure
that the state is valid at pipeline creation time.
Signed-off-by: Rohan Garg <rohan.garg@intel.com >
Reviewed-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/23934 >
2023-07-21 10:46:08 +00:00