Chia-I Wu [Fri, 7 May 2021 18:23:38 +0000 (11:23 -0700)]
venus: remember render pass PRESENT_SRC barriers
We need them for wsi queue ownership transfer.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
Chia-I Wu [Wed, 5 May 2021 17:42:50 +0000 (10:42 -0700)]
venus: remember render pass PRESENT_SRC attachments
We need them for wsi queue ownership transfer.
v2: make VN_PRESENT_SRC_INTERNAL_LAYOUT public
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org> (v1)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
Chia-I Wu [Mon, 10 May 2021 20:04:15 +0000 (13:04 -0700)]
venus: handle VN_COMMAND_BUFFER_STATE_INVALID
If the state was set to VN_COMMAND_BUFFER_STATE_INVALID, return
VK_ERROR_OUT_OF_HOST_MEMORY in vn_EndCommandBuffer.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
Chia-I Wu [Thu, 13 May 2021 22:23:48 +0000 (15:23 -0700)]
venus: remember image wsi states
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
Chia-I Wu [Wed, 12 May 2021 21:15:33 +0000 (14:15 -0700)]
venus: move wsi_image_create_info parsing
Move it from vn_CreateImage to vn_wsi_create_image.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
Chia-I Wu [Thu, 3 Jun 2021 01:56:58 +0000 (18:56 -0700)]
venus: fix opaque fd re-import
dEQP-VK.api.external.memory.opaque_fd.* re-imports and fails because
external_handles was VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT on
allocation and is VK_EXTERNAL_MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT on
re-import.
Fixes:
ccefcb0baf9 ("venus: fix misaligned bo_flags between import and query")
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11154>
Chia-I Wu [Thu, 3 Jun 2021 03:17:22 +0000 (20:17 -0700)]
venus: fix asserts on mem bo
They failed to check the result before asserting.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11154>
Chia-I Wu [Thu, 3 Jun 2021 03:16:24 +0000 (20:16 -0700)]
venus: add vn_device_memory_alloc as a helper
It is used for the most common case (non-ahb, non-import, and
non-suballocate).
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11154>
Jason Ekstrand [Wed, 2 Jun 2021 19:01:24 +0000 (14:01 -0500)]
intel/nir,i965: Move HW generation check for UBO pushing to i965
Iris only runs on BDW+ and ANV already handles this by not even trying
on anything older than HSW. The only driver benefiting from this common
check is i965. Moving it out makes the pass more generic and if some
driver comes along which can push UBOs on IVB, it should work for that.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11145>
Sergii Melikhov [Thu, 27 May 2021 11:46:04 +0000 (14:46 +0300)]
util/format: Change the pointer offset.
Changed the pointer offset to 2 to account for the second structure variable.
Fixes:
90f98b56f85 ("mesa: Deduplicate _mesa_pack_uint_z_row().")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4685
Signed-off-by: Sergii Melikhov <sergii.v.melikhov@globallogic.com>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11060>
Chia-I Wu [Wed, 12 May 2021 20:11:52 +0000 (13:11 -0700)]
vulkan/wsi: provide more info in wsi_image_create_info
Always chain wsi_image_create_info to VkImageCreateInfo, which indicates
that the image is a wsi image and can be transitioned to/from
VK_IMAGE_LAYOUT_PRESENT_SRC_KHR.
Add prime_blit_buffer to the struct as well. When set, it indicates the
prime blit destination and implies that the image is a prime blit
source.
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10789>
Dave Airlie [Mon, 24 May 2021 00:11:25 +0000 (10:11 +1000)]
intel/gfx6: move xfb_setup outside the gs compiler into the driver.
This remove the use of a GL thing from the backend compiler
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11097>
Rhys Perry [Wed, 2 Jun 2021 15:30:35 +0000 (16:30 +0100)]
aco: don't create 4 and 5 dword NSA instructions on GFX10
"stability issues", apparently: https://reviews.llvm.org/D103348
fossil-db (Navi10):
Totals from 4512 (3.01% of 149839) affected shaders:
VGPRs: 221516 -> 223308 (+0.81%); split: -0.07%, +0.88%
CodeSize:
23000080 ->
23070672 (+0.31%); split: -0.08%, +0.39%
MaxWaves: 107718 -> 107496 (-0.21%); split: +0.11%, -0.32%
Instrs: 4321890 -> 4362822 (+0.95%); split: -0.00%, +0.95%
Latency:
71495710 ->
71581476 (+0.12%); split: -0.07%, +0.19%
InvThroughput:
11858568 ->
11938960 (+0.68%); split: -0.00%, +0.68%
VClause: 76575 -> 76585 (+0.01%); split: -0.05%, +0.07%
SClause: 168771 -> 168709 (-0.04%); split: -0.06%, +0.02%
Copies: 182305 -> 221948 (+21.75%); split: -0.00%, +21.75%
PreVGPRs: 194657 -> 195635 (+0.50%); split: -0.00%, +0.50%
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes:
c353895c922 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
Rhys Perry [Thu, 20 May 2021 09:47:47 +0000 (10:47 +0100)]
aco/tests: improve reporting of failed code checks
Instead of just reporting the failed statements, print where they
originated. This is useful for tests which have a number of similar
checks.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
Rhys Perry [Thu, 20 May 2021 09:41:37 +0000 (10:41 +0100)]
aco/tests: add tests for form_hard_clauses()
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
Rhys Perry [Wed, 19 May 2021 13:36:43 +0000 (14:36 +0100)]
aco: do not clause NSA instructions
According to LLVM, this has "unpredictable results on GFX10.1".
https://reviews.llvm.org/D102211
fossil-db (Navi10):
Totals from 26690 (17.81% of 149839) affected shaders:
CodeSize:
167935160 ->
167706280 (-0.14%); split: -0.14%, +0.00%
Instrs:
31801427 ->
31744142 (-0.18%); split: -0.18%, +0.00%
Latency:
732672435 ->
732622463 (-0.01%)
InvThroughput:
163361435 ->
163357838 (-0.00%); split: -0.00%, +0.00%
VClause: 546131 -> 546903 (+0.14%); split: -0.00%, +0.14%
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes:
c353895c922 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
Rob Clark [Wed, 2 Jun 2021 16:41:13 +0000 (09:41 -0700)]
freedreno/a6xx: Fix mh31 intermittent faults
It appears that CP can over-fetch push constants slightly. While it
otherwise has no problem fetching from an alignment of 32 bytes, if that
32 bytes is at the end of a mapped bo, this can trigger fetching up to
32 bytes beyond the patch, triggering an iova fault. While otherwise
"harmless", it is probably better to not have random intermittent
faults.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11142>
Emma Anholt [Wed, 2 Jun 2021 20:12:41 +0000 (13:12 -0700)]
docs/freedreno: Rewrite the section on array access.
We don't use collect/split for array access these days, instead use
ir3_array structs that the ir3_register can point to.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11147>
Emma Anholt [Wed, 2 Jun 2021 19:28:05 +0000 (12:28 -0700)]
docs/freedreno: Update for the fanin/fanout -> collect/split rename.
See
611258d5782c ("freedreno/ir3: rename fanin/fanout to collect/split")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11147>
Emma Anholt [Wed, 2 Jun 2021 17:30:54 +0000 (10:30 -0700)]
ci/freedreno: Add some more known flakes from recent marge runs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11144>
Eric Engestrom [Wed, 2 Jun 2021 20:32:14 +0000 (22:32 +0200)]
docs: update calendar and link releases notes for 21.1.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11148>
Eric Engestrom [Wed, 2 Jun 2021 19:00:20 +0000 (21:00 +0200)]
docs: add release notes for 21.1.2
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11148>
Jason Ekstrand [Fri, 24 Apr 2020 22:45:08 +0000 (17:45 -0500)]
intel/fs: Handle non-perspective-correct interpolation on gen4-5
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11125>
Dave Airlie [Tue, 18 May 2021 03:31:14 +0000 (13:31 +1000)]
st/nir: always revectorise if scalarising happens.
This fixes arb_gpu_shader_fp64-vs-non-uniform-control-flow-ssbo
on crocus.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11098>
Georg Lehmann [Tue, 1 Jun 2021 15:45:42 +0000 (17:45 +0200)]
zink: Add a missing VKAPI_ATTR.
Closes #4868
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reported-by: Roman Stratiienko <r.stratiienko@gmail.com>
Tested-by: Roman Stratiienko <r.stratiienko@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11115>
Dave Airlie [Sun, 23 May 2021 19:19:43 +0000 (05:19 +1000)]
llvmpipe: add the interesting bit of cpu detection to the cache.
This should detect if someone changes CPU configuration that matters like in a VM
Reviewed-by: Emma Anholt <emma@anholt.net>
Fixes:
6c0c61cb48e8 ("llvmpipe: add infrastructure for disk cache support")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10946>
Eric Anholt [Wed, 28 Apr 2021 21:18:09 +0000 (14:18 -0700)]
u_format: Use the computed BE channels/swizzles for bitmask formats.
No more error-prone encoding of swizzles in the .csv for non-planar
formats!
No change to generated u_format_table.c
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Wed, 28 Apr 2021 21:10:08 +0000 (14:10 -0700)]
u_format: Sanity check that BE swizzles are appropriately mapped from LE.
Once you read enough of them, there's an obvious pattern that we can just
write a little code for instead of making every dev write it out each time.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Tue, 27 Apr 2021 22:17:39 +0000 (15:17 -0700)]
u_format: Sanity check the BE channels for all bitmask formats.
Just check against the CSV (which has its codegen now tested with
u_format_test in CI) for now, so we know that our computed channels are
correct.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Tue, 27 Apr 2021 23:50:06 +0000 (16:50 -0700)]
u_format: Fix the BE channel ordering for R5G5B5A1_UINT.
It notably didn't fit the pattern of RGB5_A1_UNORM, and violated the
general pattern for bitmask format BE channels (channels are ordered
right-to-left in the BE columns in the CSV due to the parser walking them
in that order for historical reasons).
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Tue, 27 Apr 2021 23:48:20 +0000 (16:48 -0700)]
u_format: Define tests for r3g3b2 formats and fix BE swizzles for them.
These tests passed for LE, and the BE channel ordering specified obviously
didn't fit the pattern of the other BE formats (channels are listed
right-to-left in the BE columns for historical reasons).
Note that we can't write pure-integer format tests in u_format_tests.c
currently.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Tue, 27 Apr 2021 22:47:31 +0000 (15:47 -0700)]
u_format: Assert that array formats don't include BE swizzles.
Z32_FLOAT_S8X24_UINT and X32_S8X24_UINT are in fact the only non-bitmask
formats that have BE swizzles specified, but sorting out those two is
harder.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Emma Anholt [Tue, 1 Jun 2021 20:37:56 +0000 (13:37 -0700)]
u_format: Use the nice helper for reversing an array.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Tue, 27 Apr 2021 22:39:16 +0000 (15:39 -0700)]
u_format: Move the BE swizzle computation into Format init.
I wanted to do the next set BE changes here where I have Format's helper
functions available.
No changes in generated u_format_table.c.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Tue, 27 Apr 2021 22:56:23 +0000 (15:56 -0700)]
u_format: Drop redundant .name init.
It's the first member that's set.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Eric Anholt [Tue, 27 Apr 2021 22:40:21 +0000 (15:40 -0700)]
u_format: Fix some pep8 in u_format_parse.py.
My editor likes to enforce pep8, here's some low hanging fruit so I don't
have to do too much add -p.
Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10505>
Danylo Piliaiev [Tue, 1 Jun 2021 13:27:27 +0000 (16:27 +0300)]
turnip: fix register_index calculations of xfb outputs
nir_assign_io_var_locations() does not use outputs_written when
assigning driver locations. Use driver_location to avoid incorrectly
guessing what locations it assigned.
Copied from lavapipe
8731a1beb72da8f4165e5596519a56d16299c3f0
Will fix provoking vertex tf tests when VK_EXT_provoking_vertex
would be enabled:
dEQP-VK.rasterization.provoking_vertex.transform_feedback.*
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11111>
Danylo Piliaiev [Wed, 2 Jun 2021 11:15:46 +0000 (14:15 +0300)]
turnip: emit vb stride dynamic state when it is dirty
Due to incorrect condition we never emitted vb stride
if state was dynamically set.
Fixes vertex explosion with Zink.
See https://gitlab.freedesktop.org/mesa/mesa/-/issues/4738
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11133>
Kenneth Graunke [Thu, 27 May 2021 03:13:26 +0000 (20:13 -0700)]
iris: Use bo->mmap_mode in transfer map read check
The scenario we want to avoid is reading from WC or UC mappings,
so this is an easier to follow check.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 09:02:51 +0000 (02:02 -0700)]
iris: Pick a single mmap mode (WB/WC) at BO allocation time
Previously, iris_bufmgr had the ability to maintain multiple
simultaneous memory mappings for a BO, one in WB mode (with CPU caches),
and another in WC (streaming) mode. Depending on the flags passed to
iris_bo_map(), we would select one mode or the other.
The rules for deciding which to use were:
- Systems with LLC always use WB mode because it's basically free
- Non-LLC systems used...
- WB maps for all BOs where snooping is enabled (which translates to
when BO_ALLOC_COHERENT is set at allocation time)
- WB maps for reads unless persistent, coherent, async, or raw.
- WC maps for everything else.
This patch simplifies the system by selecting a single mmap mode at
BO allocation time, and always using that. Each BO now has at most one
map at a time, rather than up to two (or three before we deleted GTT
map support in recent patches).
In practical terms, this eliminates the capability to use WB maps for
reads of non-snooped BOs on non-LLC systems. Such reads would now be
slow, uncached reads. However, iris_transfer_map recently began using
staging blits for such reads - so the GPU copies the data to a snooped
buffer which will be mapped WB. So, rather than incurring slow UC
reads, we really just take the hit of a blit, and retain fast reads.
The rest of the rules remain the same.
There are a few reasons for this:
1. TTM doesn't support mapping an object as both WB and WC. The
cacheability is treated as a property of the object, not the map.
The kernel is moving to use TTM as part of adding discrete local
memory support. So it makes sense to centralize on that model.
2. Mapping the same BO as both WB and WC is impossible to support on
some CPUs. It works on x86/x86_64, which was fine for integrated
GPUs, but it may become an issue for discrete graphics paired with
other CPUs (should that ever be a thing we need to support).
3. It's overall much simpler. We only have one bo->map field, and
manage to drop a significant amount of boilerplate.
One issue that arises is the interaction with the BO cache: BOs with
WB maps and WC maps will be lumped together into the same cache. This
means that a cached BO may have the wrong mmap mode. We check that,
and if it doesn't match, we unmap it, waiting until iris_bo_map is
called to restore one with the desired mode. This may underutilize
cache mappings slightly on non-LLC systems, but I don't expect it to
have a large impact.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4747
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 09:22:51 +0000 (02:22 -0700)]
iris: Delete GTT mapping support
In the bad old days, i965 used GTT mapping for detiling maps. iris
never has, however. The only reason it used GTT maps was in weird
fallback cases for dealing with BO imports from foreign memory. We
now do staging blits for those, and never mmap them.
There are no more users of GTT mapping, so we can delete it.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 08:41:37 +0000 (01:41 -0700)]
iris: Drop fallback GEM_MMAP_GTT if GEM_MMAP with I915_MMAP_WC fails
XXX: This is actually wrong. The dmabuf imported case can be mapped via
GEM_MMAP_GTT if the iommu is working, according to Joonas, but GEM_MMAP
would fall over and fail. So we would need this fallback.
ALTERNATIVELY...we would need to flag such imported dmabufs as
unmappable, and then make iris_transfer_map/unmap always do blits
instead of direct mappings. That seems like the saner approach
We never want to use GEM_MMAP_GTT, as it does detiling maps, and iris
always wants direct maps. There were originally two cases that this
fallback path was attempting to handle:
1. The BO was allocated from stolen memory that we can't GEM_MMAP.
At one point, kernel patches were being proposed to use stolen
memory for userspace buffers, but these never landed. The kernel
has never given us stolen memory, so we cannot hit this case.
2. Imported objects may be from memory we can't GEM_MMAP.
For example, a DMABUF from a discrete AMD/NVIDIA GPU in a PRIME
setup would be backed by memory that we can't GEM_MMAP. We could
try and mmap these directly with GEM_MMAP_GTT, but that relies on
the IOMMU working. We could mmap the DMABUF fd directly (but have
never tried to do so), but there are complex rules there. Instead,
we now flag those imports, however, and rely on the iris_transfer_map
code to perform staging blits on the GPU, so we never even try to
map them directly. So this case won't reach us here any longer.
With both of those out of the way, there is no need for a fallback.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 07:46:30 +0000 (00:46 -0700)]
iris: Assert on mapping a tiled buffer without MAP_RAW
iris has never relied on detiled maps using hardware fences.
This code is a remnant of i965, where that was actually used.
We can just assert that callers don't do such a thing.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 18:22:22 +0000 (11:22 -0700)]
iris: Use staging blits for transfers involving imported BOs
Direct mappings of imported DMABUFs can be tricky. If they're allocated
from our own device, then we can probably mmap them and it'd be fine.
But they may come from a different device (such as a discrete GPU), in
which case I915_GEM_MMAP wouldn't work, I915_GEM_MMAP_GTT would require
a working IOMMU, and directly mmap'ing the DMABUF fd would come with a
bunch of rules and restrictions which are hard to get right.
CPU mapping an imported DMABUF image for writes seems very uncommon,
solidly in the "what are you even doing?" realm. Mapping an imported
DMABUF for reading might be a thing, in case someone wanted to do
glReadPixels on it. But in that case, the cost of doing a staging
blit is probably acceptable.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 07:14:53 +0000 (00:14 -0700)]
iris: Use staging blits for reads from uncached buffers.
If we're doing CPU reads of a resource that doesn't have CPU caches
enabled for the mapping (say, in device local memory, or WC mapped),
then blit it to a temporary that does have those caches enabled.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 17:55:53 +0000 (10:55 -0700)]
iris: Track imported vs. exported status separately
Not all external objects are the same. Imported buffers may be from
other devices (say a dmabuf from an AMD or NVIDIA discrete card) which
are backed by memory that we can't use with I915_GEM_MMAP. However,
exported buffers are ones that we know we allocated ourselves from our
own device. We may not know what other clients are doing with them,
but we can assume a bit more about where they came from.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 17:53:39 +0000 (10:53 -0700)]
iris: Make an iris_bo_is_external() helper and use it in a few places
I'd like to start tracking "imported" vs. "exported" for objects,
rather than a blanket "external" flag. Instead of directly checking
bo->external, use a new helper that will eventually be "imported or
exported".
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 18:25:35 +0000 (11:25 -0700)]
iris: Delete a comment suggesting we use tiled staging buffers
We basically tried this, and it performed worse, so delete the
suggestion in the comments that we may want to do it someday.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 20 May 2021 18:15:30 +0000 (11:15 -0700)]
iris: Promote to MAP_DIRECTLY when required before NULL return
In some cases, we have to map directly (e.g. coherent/persistent maps).
In other cases (e.g. tiled), we /cannot/ map directly. We should put
the code which adds the PIPE_MAP_DIRECTLY flag in mandatory cases before
the "bail and return NULL" check for cases where we can't do that.
We leave the "we would prefer to direct map this" cases after the error
check, since we -can- use blits for those, we'd just rather not. ASTC
also stays because even though it's tiled, our tiled memcpy paths work.
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Kenneth Graunke [Thu, 13 May 2021 05:19:42 +0000 (22:19 -0700)]
iris: Replace no_gpu flag with PIPE_MAP_DIRECTLY
Here, we're deciding when to map the buffer directly, rather than using
the GPU to blit to/from a temporary. There is already a flag for that,
PIPE_MAP_DIRECTLY, which has the added benefit of not being a negative
(such as "no_gpu").
Currently, we intend to map directly if:
1. Direct mappings were requested explicitly
2. Persistent or coherent mappings were requested (and so we must)
3. ASTC textures (we currently can't blit those correctly)
4. There is no need for a temporary (there's no image compression that
the CPU wouldn't understand, and we don't need to avoid stalls due
to the buffer being busy on the GPU)
Expressing "please memory map this directly" is easier to follow than
"please don't use the GPU as part of mapping this".
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10941>
Mike Blumenkrantz [Wed, 2 Jun 2021 16:39:14 +0000 (12:39 -0400)]
zink: ci updates
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11122>
Mike Blumenkrantz [Thu, 25 Mar 2021 19:55:47 +0000 (15:55 -0400)]
zink: handle null bufferview/imageview descriptors when robustness2 is missing
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11122>
Mike Blumenkrantz [Thu, 25 Mar 2021 19:55:33 +0000 (15:55 -0400)]
zink: create dummy surface/bufferview for null descriptor use
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11122>
Mike Blumenkrantz [Thu, 25 Mar 2021 19:44:18 +0000 (15:44 -0400)]
zink: replace context-based null framebuffer surfaces with internal ones
this is a bit cleaner and avoids accessing the context
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11122>
Mike Blumenkrantz [Thu, 25 Mar 2021 19:42:53 +0000 (15:42 -0400)]
zink: add a util function to create a null surface
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11122>
Andres Gomez [Mon, 31 May 2021 11:41:21 +0000 (14:41 +0300)]
ci: build the hang-detection tool into x86_test-vk
hang-detection is a vulkan-based lightweight wrapper from
parallel-deqp-runner that periodically submits empty command buffers
and waits for their completions. If the completion never happens, the
GPU is considered hung, the wrapped script is killed, and the job
should get aborted.
This should have no negative impact on the runtime of dEQP/traces/...,
but will allow saving time when the GPU gets hung as we can abort the
job immediately rather than waiting for the timeout.
In the case of B2C, we are using this tool's error message as a way to
trigger the reboot of the test machine and start again.
v2:
- Use hang-detection already with some jobs (Martin).
Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Martin Peres <martin.peres@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11087>
Andres Gomez [Mon, 31 May 2021 19:35:25 +0000 (22:35 +0300)]
ci: fix the vkd3d-proton runner
Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Martin Peres <martin.peres@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11087>
Danylo Piliaiev [Wed, 2 Jun 2021 13:07:13 +0000 (16:07 +0300)]
turnip: reset push descriptor set on command buffer reset
Otherwise it will store a pointer to already unmapped memory which
could lead to a crash in tu_CmdPushDescriptorSetWithTemplateKHR since
it tries to copy data from the old memory.
Fixes a crash with Zink's new lazy descriptor manager instroduced
in
bfdd1d8d
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11137>
Mike Blumenkrantz [Tue, 1 Jun 2021 15:11:00 +0000 (11:11 -0400)]
radv: explicitly load a desc set layout struct member during set allocate
accessing this variable repeatedly like this is a contended hotpath somehow,
so instead just create a const for reference
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11124>
Mike Blumenkrantz [Tue, 30 Mar 2021 20:22:27 +0000 (16:22 -0400)]
radv: declare index_va in a single call for indexed draw packet emit
this is an extreme hotpath, so having a single calculation in a const
variable is slightly better for compiler microoptimizing
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11124>
Erik Faye-Lund [Tue, 1 Jun 2021 10:22:00 +0000 (12:22 +0200)]
docs: add the doc-comment for fse-vars
This just copies the comment that's in the code where this state is
stored. It seems better than nothing.
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11106>
Erik Faye-Lund [Tue, 1 Jun 2021 09:30:08 +0000 (11:30 +0200)]
docs: use envvar role for envvars
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11106>
Erik Faye-Lund [Tue, 1 Jun 2021 09:36:46 +0000 (11:36 +0200)]
docs: use file-role for paths
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11106>
Erik Faye-Lund [Tue, 1 Jun 2021 09:44:50 +0000 (11:44 +0200)]
docs: fix header-levels in envvars.rst
It seems something went wrong during conversion of this article (or
maybe even before in the HTML version), where every header after the
"Gallium environment variables" header was nested below it.
That's clearly not what's meant here, so let's fix that.
This makes the toctree make a bit more sense for this article.
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11106>
Erik Faye-Lund [Tue, 1 Jun 2021 09:31:15 +0000 (11:31 +0200)]
docs: quote a few defines
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11106>
Mike Blumenkrantz [Fri, 28 May 2021 13:53:00 +0000 (09:53 -0400)]
aux/draw: fix aalines and aapoints for shaders with explicit FragData outputs
all color outputs need to be rewritten with coverage, not just FragColor
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11065>
Iago Toral Quiroga [Tue, 1 Jun 2021 07:41:48 +0000 (09:41 +0200)]
v3dv: implement external semaphore/fence extensions
This provides most of the implementation, but there are some
things we cannot enable until we improve of kernel submit
interface, namely:
We don't expose capacity to export SYNC_FD, although we do
have the implementation in place. This requires that we
improve our kernel interface and event wait implementation
first so we can cover the corner case where the application
submits a command buffer that includes a VkCmdWaitForEvents
and tries to export a SYNC_FD from its signal semaphores or
fence before it the event is signaled and the command buffer
is sent to the kernel for execution in full.
Likewise, we can't currently import semaphores. This is because
our current kernel submit interface can only take one syncobj.
We have been working around this so far by waiting on the last
syncobj produced from the device whenever we had to wait on any
semaphores (which is obviously suboptimal already), but this
won't work as soon as we allow importing external semaphores,
as those could (and would typically) be produced from a
different device.
Once we address the kernel bits, we should come back and enable
SYNC_FD exports as well as semaphore imports.
Relevant CTS tests:
dEQP-VK.api.external.fence.*
dEQP-VK.api.external.semaphore.*
dEQP-VK.synchronization.cross_instance.*
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11105>
Iago Toral Quiroga [Tue, 1 Jun 2021 07:33:36 +0000 (09:33 +0200)]
v3dv: don't keep an open file descriptor for imported fences/semaphores
We can (and should) close the descriptor immediately after the import.
Gets the following CTS test to pass without requiring to increase limits
for open file descriptors:
dEQP-VK.synchronization.basic.binary_semaphore.chain
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11105>
Mauro Rossi [Sat, 8 May 2021 08:30:52 +0000 (10:30 +0200)]
android: i965: remove brw_ff_gs_emit.c from Makefile.sources
Fixes the following building error:
clang: error: no such file or directory: 'external/mesa/src/mesa/drivers/dri/i965/brw_ff_gs_emit.c'
clang: error: no input files
Fixes:
897bcc1e6b42 ("i965: drop old brw ff gs code.")
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10718>
Mauro Rossi [Sat, 8 May 2021 09:18:38 +0000 (11:18 +0200)]
android: intel/compiler: add brw_compile_ff_gs.c to Makefile.sources
Fixes the following building error:
FAILED: out/target/product/x86_64/obj_x86/SHARED_LIBRARIES/i965_dri_intermediates/LINKED/i965_dri.so
...
ld.lld: error: undefined symbol: brw_compile_ff_gs_prog
>>> referenced by brw_ff_gs.c:56 (external/mesa/src/mesa/drivers/dri/i965/brw_ff_gs.c:56)
Fixes:
52e426fd8b57 ("intel/compiler: add support for compiling fixed function gs")
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10718>
Daniel Stone [Mon, 31 May 2021 20:01:42 +0000 (21:01 +0100)]
CI: Fix path confusion in OpenCL Piglit execution
When we remove the contents of the results directory, we `cd` into it.
The script expects that $PWD is /piglit, and $OLDPWD is the Mesa build
directory, however the cd into the results directory will make $OLDPWD
be $BUILDDIR/results.
This means that Piglit emits into results/results/ which looks weird,
but more importantly also fails OpenCL Piglit execution, because we
can't find our baseline result expectations.
Fix it by using an explicit variable rather than relying on history.
Fixes:
683ddf19dc85 ("ci: remove results directory content only with piglit runners")
Ref: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10856
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Martin Peres <martin.peres@mupuf.org>
Reviewed-by: Andres Gomez <agomez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11126>
Daniel Stone [Mon, 31 May 2021 08:50:40 +0000 (09:50 +0100)]
Revert "CI: Disable rk3399-gru-kevin jobs for now"
We've got 7/10 devices back now, that'll do.
This reverts commit
79a7f33710b84218a040a2bd540d7c39d1c0d2d4.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11127>
Matt Turner [Tue, 1 Jun 2021 21:38:19 +0000 (14:38 -0700)]
sparc: Avoid some redefinition warnings
These definitions are provided by m_vector_asm.h now.
Fixes:
67ffb853f05 ("sparc: Reuse m_vector_asm.h.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11119>
Tomeu Vizoso [Mon, 31 May 2021 07:03:04 +0000 (09:03 +0200)]
ci/lava: Switch LAVA jobs to x86 runners
So we don't need to provision aarch64 servers, which are these days
rarer than x8_64.
In the switch to the new runner tags, switch to one which contains the
device type, so we can dimension the runner jobs taking into account the
number of DUTs available.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11108>
Matt Turner [Tue, 1 Jun 2021 21:00:30 +0000 (14:00 -0700)]
freedreno/afuc: Print uintptr_t with PRIxPTR
Fixes a compilation error on 32-bit.
Fixes:
bba61cef38b ("freedreno/afuc: Add emulator mode to afuc-disasm")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11118>
Eric Anholt [Wed, 21 Apr 2021 21:00:34 +0000 (14:00 -0700)]
ci/android: Update to building for SDK 29 by default.
This will get us build coverage of a bunch of Vulkan features, plus the
ELF TLS support.
Reviewed-by: Roman Stratiienko <r.stratiienko@gmail.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10389>
Eric Anholt [Wed, 21 Apr 2021 20:58:58 +0000 (13:58 -0700)]
android: Fix ELF TLS support.
Android 29 introduced general-dynamic TLS variable support ("quick
function call to look up the location of the dynamically allocated
storage"), while Mesa on normal Linux has used initial-exec ("use some of
the startup-time fixed slots, hope for the best!"). Both would be better
options than falling all the way back to pthread_getspecific(), which is
the alternative we have pre-29.
Reviewed-by: Roman Stratiienko <r.stratiienko@gmail.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10389>
Emma Anholt [Fri, 21 May 2021 20:23:07 +0000 (13:23 -0700)]
util: Add a helper macro for defining initial-exec variables.
I'm going to add another case for Android shortly, and then we can keep
the logic all in one spot.
Reviewed-by: Roman Stratiienko <r.stratiienko@gmail.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10389>
Mike Blumenkrantz [Tue, 18 May 2021 15:10:02 +0000 (11:10 -0400)]
aux/draw: if pipe_draw_info::index_bias_varies is not set, ignore index_bias for N>1
this enables some optimizations in lavapipe
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10995>
Mike Blumenkrantz [Tue, 25 May 2021 12:00:21 +0000 (08:00 -0400)]
aux/trace: dump resource for samplerview and surface
this is the most important member
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10974>
Alyssa Rosenzweig [Tue, 25 May 2021 21:59:46 +0000 (17:59 -0400)]
docs/gallium: Document the index buffer format convention
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10990>
Alyssa Rosenzweig [Tue, 25 May 2021 21:57:37 +0000 (17:57 -0400)]
si: Use Rn_UINT instead of In_UINT for index buffers
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10990>
Alyssa Rosenzweig [Tue, 25 May 2021 21:57:31 +0000 (17:57 -0400)]
lima: Use Rn_UINT instead of In_UINT for index buffers
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10990>
Alyssa Rosenzweig [Tue, 25 May 2021 21:57:26 +0000 (17:57 -0400)]
freedreno: Use Rn_UINT instead of In_UINT for index buffers
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10990>
Alyssa Rosenzweig [Tue, 25 May 2021 21:57:21 +0000 (17:57 -0400)]
etnaviv: Use Rn_UINT instead of In_UINT for index buffers
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10990>
Alyssa Rosenzweig [Tue, 25 May 2021 21:57:41 +0000 (17:57 -0400)]
v3d: Use Rn_UINT instead of In_UINT for index buffers
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10990>
Alyssa Rosenzweig [Tue, 25 May 2021 21:57:09 +0000 (17:57 -0400)]
vc4: Use Rn_UINT instead of In_UINT for index buffers
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10990>
Mike Blumenkrantz [Tue, 1 Jun 2021 23:07:16 +0000 (19:07 -0400)]
zink: ci updates
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11093>
Mike Blumenkrantz [Wed, 10 Mar 2021 23:59:49 +0000 (18:59 -0500)]
zink: optimize buffer rebinds
by using the bind counts, the common cases of rebinds can be immediately
handled without unnecessary iteration, and following this each rebind can
be evaluated to ensure that every necessary descriptor was rebound in order
to catch any remaining corner cases that may not be handled in the optimized
rebind path
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11093>
Mike Blumenkrantz [Mon, 1 Feb 2021 18:36:46 +0000 (13:36 -0500)]
zink: unify more resource bind count tracking code
we want to avoid updating these values when possible in order to reduce
overhead, which means that if a descriptor is being replaced, it should
be updated only if the replacement is not the same resource
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11093>
Mike Blumenkrantz [Mon, 1 Feb 2021 18:06:26 +0000 (13:06 -0500)]
zink: unify code for updating res->bind_count values
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11093>
Mike Blumenkrantz [Mon, 4 Jan 2021 18:34:13 +0000 (13:34 -0500)]
zink: add a second descriptor manager
this is the "lazy" descriptor manager, which aims to perform the least
amount of work calculating updates while ignoring the overhead that an
update may incur: effectively the inverse of the caching manager
in this initial implementation, divergence exists between the descriptor
layouts of the cached manager and the lazy manager in order to avoid
incurring regressions in the existing descriptor architecture; this will
be reconciled in a followup MR that refactors and unifies descriptor layouts
during this interim period and until such reconciliation occurs,
the default descriptor manager is now the lazy manager for testing purposes as
there are no changes here which can affect the existing infrastructure
the caching descriptor manager can be selected with the ZINK_CACHE_DESCRIPTORS
env var and will be automatically used for vulkan drivers which don't support
the features required for lazy mode (templates)
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11093>
Alyssa Rosenzweig [Wed, 26 May 2021 23:00:00 +0000 (19:00 -0400)]
panfrost/ci: Report flakes on IRC
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11022>
Emma Anholt [Wed, 26 May 2021 21:13:47 +0000 (14:13 -0700)]
ci: Move the flakes channels to OFTC
Following the rest of our channels, move CI reporting over. Seems to
still work fine. This affects freedreno and iris.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11020>
Emma Anholt [Wed, 26 May 2021 17:45:33 +0000 (10:45 -0700)]
ci: Add known-flake handling for the IRC flake reports
Now, flakes that aren't in the *-flakes.txt get a "NEW" in their report so
I can watch for them.
The bash was unwieldy and made debugging hard, so I switched to python.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11020>
Mike Blumenkrantz [Tue, 1 Jun 2021 16:26:27 +0000 (12:26 -0400)]
anv: fix availability for copying timestamp query results
idx here is the index of the value being written, so if it isn't used/incremented
when the query result is written, the availability result will clobber it and
be written to the same buffer offset
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11116>
Alyssa Rosenzweig [Thu, 20 May 2021 21:09:14 +0000 (17:09 -0400)]
panfrost: Eliminate reserve_* functions
We always want to reserve _something_, so reserve what we need at batch
creation time and stop trying to re-reserve in a zillion places after.
This has a neglible (<128 bytes per batch) increase in memory usage for
compute-only workloads, but given the amount of simplication, that's a
fair tradeoff.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11074>
Alyssa Rosenzweig [Thu, 20 May 2021 23:25:00 +0000 (19:25 -0400)]
panfrost: Elucidate thread group split field
Last unknown field in the XML :-)
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11074>
Alyssa Rosenzweig [Thu, 20 May 2021 23:12:13 +0000 (19:12 -0400)]
panfrost: Bubble up errors
Instead of punting to PAN_MESA_DEBUG=msgs, which we can now remove.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11074>
Alyssa Rosenzweig [Thu, 20 May 2021 22:29:14 +0000 (18:29 -0400)]
panfrost: Minor changes to draw_vbo
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11074>