The users of exportable might have different expectations for what can
be exported, and some are more tight. So we need a new exportable_dmabuf
flag to track where dmabuf is actually needed.
If the underlying driver does not advertise dmabuf extension, requesting
dmabuf export violates the spec VU:
> VUID-VkMemoryGetFdInfoKHR-handleType-00671
>
> handleType must have been included in
> VkExportMemoryAllocateInfo::handleTypes when memory was created
Cc: mesa-stable
(cherry picked from commit 4f53828aa6)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
It is not actually available to all the platforms mesa can be compiled
to, so let's keep an opt-in list of supported platforms instead, and
compile it out on all other platforms.
Fixes: 48a0478126 ("zink: add renderdoc handling")
(cherry picked from commit 93390d4b73)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
fmin/fmax are not signed zero correct.
The name is taken from RDNA4's ISA, because the full IEEE name is a bit long.
Cc: mesa-stable
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
(cherry picked from commit f68b891a81)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
Otherwise we will have width of -1 later in null_fill_state_s.
Fixes: 2eb45daa9c ("gallium: de-pointerize pipe_surface")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
(cherry picked from commit 4ac7bbb1d1)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
In the case where we have a device that we want to choose after
probing, there is no point in asking the device select layer to do
any reordering at all.
This helps avoid a deadlock inside compositors where we don't need
device selection anyways.
Cc: mesa-stable
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
(cherry picked from commit 04071c5f9a)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
There are cases like zink where we have a file descriptors we are searching
for devices for, so we don't need device selecting reordering, we just want
the fastest path to get the devices so we can match them.
This also helps avoid some cases of deadlock inside compositors where
zink/vulkan initialises later and tries to connect to the compositor.
This uses a VK_EXT_layer_setting to add a bypass setting.
Cc: mesa-stable
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
(cherry picked from commit 833706e381)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
component_addr_off (in bytes) was used to offset a component index (in dwords).
Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
(cherry picked from commit 3552028e87)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
Currently the image creation failure handling codepath of
wsi_headless_surface_create_swapchain() just calls
wsi_headless_swapchain_destroy() , which will try to destroy all
`image_count` of images. However, some of these images might never be
successfully created because of the failure, which leads to double-free.
Set image_count to the number of successfully created images before
calling wsi_headless_swapchin_destroy() to prevent over-destroying.
Fixes dEQP-VK.wsi.headless.swapchain.simulate_oom.* on lavapipe and pvr,
although some of the tests got QualityWarning saying "Creating swapchain
did not succeed, callback limit exceeded" on lavapipe (Pass on pvr).
Cc: mesa-stable
Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
(cherry picked from commit e8c81652c9)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
The code originally disabled reuse whenever a pNext was passed to
`bo_create_internal`, however code has been added before it that messed
with the pointer disabling it more often that necessary.
Rearrange code to restore the original check.
cc: mesa-stable
(cherry picked from commit 6a68542167)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
Extracting from an SGPR could cause a wrong RegClass on
the operand which could later lead to selecting VOPD
instructions which falsely operate on the corresponding
VGPR.
Cc: mesa-stable
(cherry picked from commit d8481fd7cc)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
Fix compiler error:
../src/util/blob.c:344:8: error: assigning to 'uint8_t *' from
'const void *' discards qualifiers
[-Werror,-Wincompatible-pointer-types-discards-qualifiers]
glibc now provides C23-style type-generic string functions. memchr
returns const void * when passed a const void * argument. Update nul
declaration to const since it's only used to find the null terminator
position and calculate a size.
Fixes: 1c9877327e ("glsl: Add blob.c---a simple interface for serializing data")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
(cherry picked from commit b40850b24a)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
Only allow holes between the first and last used component.
Do not load unused components before the first used component.
This fixes test failures with a bunch of VK CTS tests
with allow_holes enabled on RADV:
dEQP-VK.tessellation.tess_io.max_in_out.with_f16.*
Fixes: 6286c1c66f
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
(cherry picked from commit 1981b9836b)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
Calling finalize_nir twice causes si_nir_lower_color_inputs_to_sysvals
to be called twice. The pass always sets color interpolation to FLAT
at the beginning, and then lowers color input loads and sets the correct
interpolation mode. However, when the pass is called for the second time,
it just sets color interpolation to flat and does nothing, which overrides
the original interpolation mode.
This fixes color interpolation for those cases. This only happens with
ATI_fragment_shader AFAIK.
Fixes: deda05e2b7 - nir: move nir_lower_color_inputs into radeonsi
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
(cherry picked from commit e5b1c568b9)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39462>
no, I don't know how this worked before.
fixes
KHR-GL45.conditional_render_inverted.functional
KHR-GL46.transform_feedback_overflow_query_ARB.*
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Cc: mesa-stable
Reviewed-by: Mary Guillemard <mary@mary.zone>
(cherry picked from commit 56b90a70b3)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
Per current OpenGL spec: When a cube map texture is sampled, the (s t r)
texture coordinates are treated as a direction vector (rx ry rz)
emanating from the center of a cube. The q coordinate is ignored.
Older spec had slightly different wording that allowed projecting the
coordinates, but to be consistent with GLSL/SPIRV and the expectations
of the real-world applications it's better to ignore the projector.
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14235
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
(cherry picked from commit d8e524cd6b)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
Applying the projector after the normalization breaks the coordinates, so
apply it early. Usually it's not even necessary for the cubemaps anyway,
but ARB_fragment_program and TGSI allow it.
Fixes: 52e71809 ("nir: Add a cubemap normalizing pass")
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
(cherry picked from commit f3eb98ec57)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
Subgroup ops make divergence information useless for our purpose,
we would need workgroup divergence.
The game affected here has control flow dependent on vote_any,
so it's possible that a wave only executes the code after culling/reordering
invocations.
That means we can't reuse the maybe undefined value from before culling.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14459
Cc: mesa-stable
(cherry picked from commit cbedced5e8)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
This broke a number of dEQP tests, for reasons that are unclear -- in
looking at cffdumps, the const setup looks appropriate, just changing from
e.g:
VS_CTRL_REG1(CONSTLEN=2)
FS_CTRL_REG1(CONSTLEN=1)
LOAD_STATE6(VERT, DIRECT, DST_OFF=2, NUM_UNIT=2)
LOAD_STATE6(FRAG, DIRECT, DST_OFF=0, NUM_UNIT=2)
to:
VS_CTRL_REG1(CONSTLEN=2)
FS_CTRL_REG1(CONSTLEN=1)
LOAD_STATE6(VERT, DIRECT, DST_OFF=2, NUM_UNIT=2)
LOAD_STATE6(FRAG, INDIRECT, DST_OFF=0, NUM_UNIT=2)
Fixes: 203ac73374 ("gallium: set prefer_real_buffer_in_constbuf0 for all drivers using tc")
(cherry picked from commit 974ca32643)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
The entire range of the allocated bo up to bo->size will be used, even if
alloc_size was way less, so to track the growth precisely for load factoring,
the allocated_batch_size should increase by bo->size.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
(cherry picked from commit 7b74321640)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
Although a specific size is requested, the entire range of the returned bo up
to bo->size may end up being used by anv_batch_chain, spamming memcheck errors.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
(cherry picked from commit 4de00e01dd)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
Since we use atomic mode setting now, the wsi->fd we use needs to have
the atomic client cap.
There are several different code paths where wsi can acquire a file
descriptor. For drm masters, the atomic client cap is set in
wsi_display_init_wsi. For leased drm fds, there are AcquireDrmDisplayEXT
and AcquireXlibDisplayEXT.
According to a comment we previously assumed wsi_display_get_connector
is common among all code paths, and that's why the atomic client cap was
set there. But that assumption can be broken based on the particular
order which the application invokes vulkan APIs in.
This commit simply push the drmSetClientCap to all entrypoints where a
drm fd comes through.
Fixes: 513ffea1d3 ("wsi/display: use atomic mode setting")
Signed-off-by: Yuxuan Shui <yshui@codeweavers.com>
(cherry picked from commit e0fbf9a908)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
Commit e626636e90 ("bin/symbols-check: fix fields length condition
before accessing fields") changed the condition from `or` to `and`
to prevent potential IndexError when accessing fields[1].
However, this broke macOS because `nm -gP` outputs different field
counts for undefined symbols:
- Linux: `_symbol U` (2 fields)
- macOS: `_symbol U 0 0` (4 fields)
The condition `len(fields) == 2 and fields[1] == 'U'` only matched
the Linux format, causing undefined symbols like
_mesa_glapi_tls_Dispatch to be incorrectly reported as
"unknown symbol exported" on macOS.
Fix by using `len(fields) >= 2` to handle both platforms while still
guarding against IndexError.
Fixes: e626636e90 ("bin/symbols-check: fix fields length condition before accessing fields")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/13451
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
(cherry picked from commit 191e4b8fe0)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39117>
