This is the first time we see an application running out of mmap().
We essentially allocate too many batches (+65k) and end up not being
able to mmap them, at which point we can't mmap anything anymore and
things go sideways.
This change allocates bigger batch BOs as we grow an existing command
buffer. This drastically reduces the number of BOs we need to allocate
(the benchmark that reported the issue now reaches a max of ~630 BOs,
instead of reaching 65k and failing previously).
v2: Track the total batch size of command buffers (Jason)
Just give 0 for batch_len to i915 (Jason)
v3: Fix indentation (Jason)
v4: Drop uncessary reshuffling of error labels (Jason)
v5: Remove empty lines (Marcin)
v6: Limit BO growing to chunks of 16Mb (Jason)
v7: Add assert on initial size (Jason)
v8: Add define for max size (Jason)
v9: Fixup v7 assert for non softpin platforms (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4956
Cc: mesa-stable
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11482>
(cherry picked from commit 40b67a2922)
Some infinite loop cases were already covered by other
restrictions (e.g. if the loop had a body), but the case with a single
block in the loop body wasn't yet.
This prevents an infinite loop when optimizing the shader in
dEQP-VK.reconvergence.subgroup_uniform_control_flow_ballot.compute.nesting2.3.2
and various others reconvergence tests.
Fixes: 0881e90c09 ("nir: Split ALU instructions in loops that read phis")
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> [v1]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11476>
(cherry picked from commit b951929795)
It turns out, some combinations of Meson and MSVC doesn't handle UTF-8
symbols in source-files as gracefully as we'd like. Luckily, there's a
work-around; forcing UTF-8 as the default-encoding.
Please see this Meson ticket for details:
https://github.com/mesonbuild/meson/issues/8263
Cc: mesa-stable
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11424>
(cherry picked from commit b437fb810a)
The second argument to debug_get_bool_option is a boolean, not a
pointer. Passing a NULL-pointer there triggers a warning, which we treat
as an error on MSVC:
---8<---
../src/gallium/auxiliary/driver_trace/tr_screen.c(1071): error C2220:
the following warning is treated as an error
../src/gallium/auxiliary/driver_trace/tr_screen.c(1071): warning C4047:
'function': 'bool' differs in levels of indirection from 'void *'
../src/gallium/auxiliary/driver_trace/tr_screen.c(1071): error C4024:
'debug_get_bool_option': different types for formal and actual parameter 2
---8<---
Fixes: c1270d4845 ("aux/trace: add screen deduplication for zink+lavapipe tracing")
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11424>
(cherry picked from commit bae4dca8e4)
Without this, I get the following error if I try to compile Zink without
any other drivers:
src/gallium/targets/libgl-gdi/libgl_gdi.c(210): error C2037: left of
'flush_frontbuffer' specifies undefined struct/union 'pipe_screen'
Fixes: fdfe4a4d30 ("libgl-gdi: add zink support")
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11536>
(cherry picked from commit 9aff548ff4)
These functions were added, but never used. Let's get rid of them.
Fixes: c2cb2dd3bc ("zink: Added support for MacOS MoltenVK APIs.")
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Acked-by: Hoe Hao Cheng <haochengho12907@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11554>
(cherry picked from commit 95cbb560cf)
Right now the accumulator-clearing move emitted by the generator for
Wa_14010017096 inherits the SWSB field from the previous instruction.
This can lead to redundant synchronization, or possibly more serious
issues if the previous instruction had a TGL_SBID_SET SWSB
synchronization mode. Take the SWSB synchronization information from
the IR.
Fixes: a27542c5dd ("intel/compiler: Clear accumulator register before EOT")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
(cherry picked from commit c19cfa9dc2)
This is unlikely to have had any negative side effect on the original
TGL, but will lead to issues on XeHP+ if the software scoreboard pass
isn't able to synchronize the accumulator writes.
Fixes: a27542c5dd ("intel/compiler: Clear accumulator register before EOT")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
(cherry picked from commit 63abc083ce)
This fixes an issue where one context only creates buffers while another
context only destroys buffers. Only the creating context can release its
buffers and the destroying context only turns them into zombie buffers.
This fix makes the creating context release its zombie buffers.
It's not a plot from an apocalyptic movie.
Fixes: e014e3b6be "mesa: don't count buffer references for the context that created them"
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4840
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11514>
(cherry picked from commit ec7e262349)
This message pops up in the Meson build logs, but has no context
and doesn't seem to be directed at end-users. Sounds like a leftover
from a debugging session.
Signed-off-by: Simon Ser <contact@emersion.fr>
Fixes: d0767fc045 ("amd/addrlib: use cpp.has_argument() to filter compiler arguments")
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11524>
(cherry picked from commit 17f0d7ef83)
Add previously missing line in HEVC SPS header generation on VCN 1.0
devices, for when an image does not need to be cropped.
Fixes: e62c7e7c6c ("radeon: Add cropping to encoded H.265 when padding is used")
Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11535>
(cherry picked from commit d977e7fbe1)
spirv_to_nir now requires NIR variables to be created for everything.
Fixes: 10b3eecd36 "spirv: Don't remove variables used by resource..."
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8637>
(cherry picked from commit 72437f6d54)
The Navi 1x NGG hardware can hang in certain conditions when
not every wave launched before s_sendmsg(GS_ALLOC_REQ).
As a workaround, to ensure this never happens, let's emit a
workgroup barrier at the beginning of NGG VS and TES.
Note that NGG GS already has a workgroup barrier so it doesn't
need this.
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10837>
(cherry picked from commit e6bf5cfe59)
Navi 1x GPUs have an issue: they can hang when the output vertex
and primitive counts are zero. The workaround is exporting a dummy
triangle.
This commit changes the dummy triangle's vertex so its positions
are all NaN. This should make sure the triangle is never rendered.
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10837>
(cherry picked from commit f9447abb36)
This is to fix a hang that can potentially happen when not every
wave had launched when a gs_alloc_req is executed.
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10837>
(cherry picked from commit 25314996d0)
mesa/mesa@2dcc9c7f54 from mesa/mesa!6639 added a resource_get_param
hook for llvmpipe, which was nice since it gave lavapipe more features.
One of those features was not exporting llvmpipe textures, so those
parts were stubbed out and landed in an assert(0).
This completely broke kms_swrast (llvmpipe+GBM) on non-release builds,
since that definitely does need to export llvmpipe textures.
The query codepath which caused this explosion does fall back to
resource_get_handle() - which is how it worked previously - but not all
callers do this, so just do what all other drivers implementing
resource_get_param() do and open-code the translation.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reported-by: Jonas Ådahl <jadahl@gmail.com>
Tested-by: Jonas Ådahl <jadahl@gmail.com>
Fixes: 2dcc9c7f54 ("llvmpipe: add resource get param support.")
Ref: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6639
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11504>
(cherry picked from commit 9102921109)
This avoids all locks for reads and using lock only while actually
writing.
This is enabled by doing two things:
1) Reading the index incrementally. This way we get new entries
written by other processes and do not write duplicate entries.
2) Taking the lock only during writes, and applying the incremental
read while holding the lock so we always append to the actual end of the file.
Fixes: eca6bb9540 ("util/fossilize_db: add basic fossilize db util to read/write shader caches")
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11485>
(cherry picked from commit 4f0f8133a3)
Output loads and stores are lowered to shared memory access,
so we have to update the barriers to also reflect this.
Closes: #4955
Fixes: bf966d1c1d
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11484>
(cherry picked from commit 72174a3eef)
If compressed rendering is only used for the depth aspect of a
depth/stencil image, stencil might also be compressed and it needs
to be decompressed. This only happens for non-TC compatible images.
As long as the driver needs to decompress the depth aspect, I don't
think that decompressing the stencil aspect introduces extra cost.
Fixes dEQP-VK.renderpass*late_fragment_tests*.d32_sfloat_s8_uint for
chips that don't support TC-compat HTILE.
Cc: 21.1 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11263>
(cherry picked from commit 213c4c5f44)
From the Vulkan spec 1.2.181:
"The difference of the size of memory and memoryOffset must be
greater than or equal to the size member of the
VkMemoryRequirements structure returned from a call to
vkGetImageMemoryRequirements with the same image"
This is invalid usage but adding a check in the driver is safe and
might avoid spurious failures.
This is a workaround for the inventory GPU hang with Cyberpunk 2077
which is actually a game bug. Luckily the game handles this error
gracefully.
Since the addrlib change from March, addrlib now selects a better
swizzle mode (4KB instead of 64KB) which reduces image size. Though,
the game assumes that an image with 2 mips is always smaller than the
same image but with 6 mips. This is not always true if the swizzle mode
is different. Then, it creates a D312 heap that is too small for the 2
mips image and the GPU hang with a memory violation, ugh...
Note that next vkd3d-proton release should also reject this but
fixing both sides is fine.
Cc: 21.1 mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4823
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4593
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11448>
(cherry picked from commit 50233d0daa)
When an OpBranchConditional that had two equal branches was parsed, we
were treating it as a regular OpBranch. However this doesn't work
well when there's an associated OpSelectionMerge. We ended up
skipping marking the merge block as such, and depending on what was
inside the construct we would end up trying to process the block
twice.
Fix this by keeping the vtn_if around, but when emitting NIR identify
the two equal branch case.
Fixes: 9c2a11430e ("spirv: Rewrite CFG construction")
Closes: #3786, #4580
Reviewed-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9297>
(cherry picked from commit 64cb143b92)
Read directly from the instruction getting spilt. Otherwise a fill
will be inserted before the spill writing the value, so the
instruction reading the spilt value gets garbage data.
Use the bundle_id to check if the instructions are in the same bundle.
Insert a move instruction, as the spill needs the value in a LD/ST
register such as AL0, while the ALU instruction reading the value
needs it in a work register such as R0.
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4857
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11212>
(cherry picked from commit 38e8d7afe3)
Commit 24342e499b changed how primitive
topology is handled on Gfx8+ but missed updating the Gfx7.x code.
As a result, tests which previously used topologies like PATCHLIST_3
instead started using bogus ones like LINESTRIP_ADJ. This caused a
GPU hangs in a bunch of Vulkan conformance tests involving tessellation.
This fixes those hangs.
Fixes: 24342e499b ("anv: fix dynamic primitive topology for tess")
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11434>
(cherry picked from commit 0510a947ba)
builtin_builder::~builtin_builder() and builtin_builder::release()
are running into race condition. This leads lightsmark to crash at
the end because both calls ralloc_free which mutates the arguments state
This patch fixes lightsmark2008 crash
Fixes: e4da8b9c33 ("mesa/compiler: rework tear down of builtin/types")
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Reviewed-by: Neha Bhende <bhenden@vmware.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Neha Bhende <bhenden@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11385>
(cherry picked from commit eb272f6571)
Buffers are created without a format in Vulkan, and we always pass in
R8_UNORM for them in Gallium. It's the view-formats we should have
checked, if anything.
But that's orthogonal to this. We shoudn't keep checking R8_UNORM
capabilities for buffers, all it's going to do is trigger asserts.
Fixes: 00dc0036b ("zink: flatten out buffer creation usage flags codepath")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11416>
(cherry picked from commit a8fc38b276)
We already set this bit unconditionally right before, no point in
repeating it.
Fixes: 00dc0036b ("zink: flatten out buffer creation usage flags codepath")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11416>
(cherry picked from commit 662dc70002)
Seems I missed that we already did an unconditional unmap here, and
forgot to remove it. Whoops.
Fixes: 5159f406d ("zink: use gallium api to copy to display-target")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11414>
(cherry picked from commit 0bb4703292)
With nir I encountered the case where the same value can be written to from
multiple surface operations. This caused some weird messups with the unions
as the def.rewrite operations caused unrelated instructions to get new their
value replaced as well.
In order to replace def.rewrite, we have to create a new temp value, write
to that one instead and move to the original value.
Fixes: 869e32593a ("gm107/ir: fix loading z offset for layered 3d image bindings")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11053>
(cherry picked from commit c1f938b647)
memset operates in bytes, and there are 8-bits in a byte. This is a
very easy to miss typo. :(
Fixes: 9758b1d416 ("util: add util_set_thread_affinity helpers including Windows support")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11228>
(cherry picked from commit a923e95b10)
