git.libre-soc.org Git - mesa.git/log

egl/wayland: Don't invalidate buffers on no-op resize

The Wayland platform's resize_callback is invoked from libwayland-egl
when wl_egl_window_resize() is called. The resize call is the only place
for the application to insert dx/dy arguments to wl_surface_attach().

When modifying the cursor hotspot (as in wayland/wayland#148), we want
to set dx/dy, but leave the surface size the same. If we get
wl_egl_window_resize() with the same width and height argument as we
already have, we do not need to invalidate our existing drawable.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4030>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4030>

Revert "glx: convert glx_config_create_list to one big calloc"

This reverts commit 35fc7bdf0e6ad6547e39099e7060a3d89539b56d.

Unfortunately mentioned commit introduced a memory leak because
`driwindowsMapConfigs` and `createDriMode` functions allocate
small memory portions for each element:
21,576 (232 direct, 21,344 indirect) bytes in 1 blocks are definitely lost in loss record 1,411 of 1,414
    at 0x483A7F3: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
    by 0x5D4AA09: createDriMode (dri_common.c:291)
    by 0x5D4ABF5: driConvertConfigs (dri_common.c:310)
    by 0x5D58414: dri3_create_screen (dri3_glx.c:945)
    by 0x5D39829: AllocAndFetchScreenConfigs (glxext.c:815)
    by 0x5D39C57: __glXInitialize (glxext.c:941)
    by 0x5D3290A: GetGLXPrivScreenConfig (glxcmds.c:174)
    by 0x5D34F38: glXQueryExtensionsString (glxcmds.c:1307)
    by 0x4F83038: glXQueryExtensionsString (in /usr/local/lib/libGL.so.1.7.0)
    by 0x4F2EA6B: ??? (in /usr/lib/x86_64-linux-gnu/libwaffle-1.so.0.6.0)
    by 0x4F2A0D7: waffle_display_connect (in /usr/lib/x86_64-linux-gnu/libwaffle-1.so.0.6.0)
    by 0x498F42A: wfl_checked_display_connect (piglit-util-waffle.h:74)

There is one more thing which disallow us to easily fix it are different element sizes
for instance: `glx_config_create_list` allocates memory just for `glx_config`,
`driwindowsMapConfigs` for `driwindows_config` and
`createDriMode` for `__GLXDRIconfigPrivate`.
Yes it is possible but size of such fix
will be more big and complex than original one.
So it make sense only if the malloc overhead
really is a big problem there.

Acked-by: Eric Engestrom <eric@engestrom.ch>
Signed-off-by: Andrii Simiklit <andrii.simiklit@globallogic.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3406>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3406>

zink. Don't set incorrect sType in VkImportMemoryFdInfoKHR struct

imfi.sType was being set to an invalid value, triggering a warning in Clang. The only valid value for imfi.sType is VK_STRUCTURE_TYPE_IMPORT_MEMORY_FD_INFO_KHR which is the value it is being given at initialisation time, a few lines earlier. The incorrect value, VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT, is supposed to be used in imfi.handleType instead - and indeed, handleType *is* being set to this value a few lines later.

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4034>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4034>

turnip: support indirect draw

Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3976>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3976>

android: gallium/auxiliary: fix "Unused source files" in tesselator

Avoids the following Android Build System error:

FAILED:
build/make/core/binary.mk:1245: error: external/mesa/src/gallium/auxiliary/Android.mk: libmesa_gallium: Unused source files: tessellator/tessellator.hpp
10:24:30 ckati failed with: exit status 1

Fixes: bd0188f ("gallium/auxiliary: add the microsoft tessellator and a pipe wrapper.")
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Acked-by: Dave Airlie <airlied@redhat.com>

ci: Flip db410c back to docker mode.

Turns out there's corporate policy to not deploy AGPL software, so I
have to take down the LAVA lab until we sort out how to do it without
a local server.

Reviewed-by: Daniel Stone <daniels@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4038>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4038>

intel/gen12+: Disable mid thread preemption.

Fixes a GPU hang in Car Chase.

Cc: mesa-stable@lists.freedesktop.org
v2: Add comment explaining why (Jason).

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4035>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4035>

Revert "gallium/swr: Fix min/max range index draw"

This reverts commit 5e9a2c603f38308f7027d6a5e4575e5fc24c1bd5

Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4032>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4032>

vc4: fix vc4_yuv_blit overwriting fragment constant buffer slot 0

vc4_yuv_blit calls util_blitter_restore_constant_buffer_state without
first calling util_blitter_save_fragment_constant_buffer_slot. This causes
subsequent crashes in vc4_write_uniforms when using fragment shaders that
reference YUV textures.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2581
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3997>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3997>

aco: handle v_add_co_u32_e64 in parse_base_offset()

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3902>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3902>

aco: fix carry-out size for wave32 v_add_co_u32_e64

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Fixes: e0bcefc3a0a ('aco/wave32: Use lane mask regclass for exec/vcc.')
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3902>

gallium/swr: fix corruptions in Unigine Heaven

Few changes to fix the last corruptions in Heaven:
- fix indirect TCS input when vertex/attribute
index is not the same for each patch
- use the correct functions to build loops in shader
- fix using vmask for writting TCS output

Reviewed-by: Krzysztof Raszkowski <krzysztof.raszkowski@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3980>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3980>

st/va: GetConfigAttributes: check profile and entrypoint combination

Added check if profile is supported or not for the entrypoint in
GetConfigAttributes.

Signed-off-by: Satyajit Sahu <satyajit.sahu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3889>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3889>

intel/isl: Implement D16_UNORM workarounds.

GEN:BUG:14010455700 (lineage 1808121037):

   "To avoid sporadic corruptions “Set 0x7010[9] when Depth Buffer
   Surface Format is D16_UNORM , surface type is not NULL & 1X_MSAA"

Required for fixing ttps://gitlab.freedesktop.org/mesa/mesa/issues/2501.

GEN:BUG:1806527549:

   "Set HIZ_CHICKEN (7018h) bit 13 = 1 when depth buffer is D16_UNORM."

This one could fix a GPU hang in some workloads.

v2: Implement WA in isl and add another similar WA (Jason).
v3: Add flushes before changing chicken registers (Jason)
v4: Depth flush and stall + end of pipe sync when changing registers
(Jason).

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3801>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3801>

aco: keep track of which events are used in a barrier

And properly handle unordered events so that they always wait for 0.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: 93c8ebfa780 ('aco: Initial commit of independent AMD compiler')
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3774>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3774>

st/va/postproc: reallocate interlaced destination buffer

When the source buffer is progressive source, re-allocate the
destination buffer as progressive if it isn't already - otherwise
transcoding will fail.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/1418
Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4001>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4001>

panfrost: fix transform feedback

Fix different use cases for transform feedback by setting:

- PIPE_CAP_PACKED_STREAM_OUTPUT=0
- PIPE_CAP_VIEWPORT_TRANSFORM_LOWERED=1
- PIPE_CAP_PSIZ_CLAMPED=1

This is enough for all dEQP xfb-related test cases to run
successfully.

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> (Update dEQP expectations)
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>

gallium: add PIPE_CAP_PSIZ_CLAMPED

This new capability indicates that the point size has been clamped.
This also means that the gl_PointSize has been modified and that
its value should be lowered for transform feedback, if needed.

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>

gallium: add PIPE_CAP_VIEWPORT_TRANSFORM_LOWERED

This new capability indicates that the nir_lower_viewport_transform
pass is enabled. This also means that the gl_Position value is
modified and should be lowered for transform feedback, if needed.

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>

gallium: add PIPE_CAP_PACKED_STREAM_OUTPUT

Setting this cap to 0 (default is 1) should disable packing
optimization for stream output (e.g. GL transform feedback captured
variables).

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>

glsl/linker: add xfb workaround for modified built-in variables

Some lowering passes modify the value of built-in variables in
order for drivers to work properly. However, modifying such values
will also break transform feedback as the captured value won't
match what's expected.

For example, on some hardware, the vertex shaders are expected to
output gl_Position in screen space. However, the transform
feedback captured value is still supposed to be the world-space
coordinates (see nir_lower_viewport_transform).

To fix that, we create a new variable that contains the
pre-transformation value and use it for transform feedback instead
of the built-in one.

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>

glsl/linker: handle array/struct members for DisableXfbPacking

When varying packing is disabled for transform feedback and a xfb
declaration points to an array element or structure member, the
element/member should be aligned to the start of a slot as well.
If that's not the case, a new varying is created and the
element/member value is copied.

There might a way to further optimize the number of slots allocated
or the number of copies necessary if the performance cost is
problematic. For example, in cases where simply padding the top
level variable might correctly align all the captured values.

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>

glsl/linker: add DisableTransformFeedbackPacking workaround

Some drivers (e.g. Panfrost) don't support packing of varyings when
used for transform feedback. This new constant ensures that any
varying used for xfb is aligned at the start of a slot and won't be
packed with other varyings.

Scenarios where transform feedback declarations are related to an
array element or a struct member will be handled in a subsequent
patch.

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> (Fix order of arguments to varying_matches())
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2433>

spirv: fix memory_barrier_tcs_patch emission

Shouldn't affect any driver, since all currently implement
memory_barrier_tcs_patch as a no-op. It also looks like optimizations are
fine

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4003>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4003>

spirv: improve creation of memory_barrier

It shouldn't check for atomic counters or return in case we also need to
create a TCS output barrier.

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4003>

lima: don't disable tiling if there's linear modifier in list

Instead we should disable it if tiling modifier is not here and we
already do that.

Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4029>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4029>

ac: rename min_vgpr_alloc to min_wave64_vgpr_alloc

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3975>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3975>

ac: rename vgpr_alloc_granularity to wave64_vgpr_alloc_granularity

And update the value.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3975>

ac: rename lds_size_per_cu to lds_size_per_workgroup

It's more accurate.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3975>

turnip: Execute main cs for secondary command buffers

Previously, we only added the secondary command buffer's draw and
draw epilogue command streams to the primary command buffer on
vkCmdExecuteCommands. However, we also need to merge the primary cs
for non-draw operations like vkCmdCopyBuffer and vkCmdBeginQuery.

Fixes dEQP-VK.memory.pipeline_barrier.host_write_transfer_src.*
and various other tests in dEQP-VK.api.command_buffers.*.

Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3988>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3988>

turnip: Promote tu_cs_get_size/is_empty to header

These will be used in tu_cmd_buffer.c.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3988>

nvc0: enable EXT_texture_shadow_lod

This passes all the CTS tests for this extension.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4014>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4014>

st/mesa: allow TXB2/TXL2 to work with cube array shadow textures

It's a bit asymmetric, but it's such a contrived use-case, and not a lot
of drivers will support it.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4014>

nv50,nvc0: add newly added PIPE_CAP's to list

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4014>

anv: multiply the scratch space by 4 on gen9-10 like iris and i965

My understanding is that there's no reason for the scratch space
allocation to be different between iris, i965 and anv. Let's make all
the functions behave the same.

I don't know if this fixes any specific gen9 bugs, it it might since
it increases the scratch space.

v2: Rebase.
v3: Rebase.
v4: Remove redundant gen 11 check (Jason).

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4006>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4006>

intel/device: bdw_gt1 actually has 6 eus per subslice

Found by inspection, I'm not aware of any bugs caused by this typo.

According to Lionel, it seems we only use this to generate masks
of available EUs for perfromance queries, and it's only used when we
can't query the fused parts of the GPU through DRM_IOCTL_I915_QUERY.
So this patch should help for the corner case where the Kernel is too
old to support the query ioctl.

v2: improve commit message, cc stable (Lionel).

Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4006>

intel: fix the gen 12 compute shader scratch IDs

This is the same idea as "intel: fix the gen 11 compute shader scratch
IDs".

The number of EUs on TGL is not the same as ICL, but the
MEDIA_VFE_STATE restrictions stay the same, so adapt the code to it.
Also, consider the base configuration instead of what we read from the
Kernel.

According to Mark, this fixes the following piglit tests on TGL:

    piglit.spec.arb_compute_shader.execution.shared-atomicmax-uint.tglm64
    piglit.spec.arb_compute_shader.execution.shared-atomicmax-int.tglm64
    piglit.spec.intel_shader_atomic_float_minmax.execution.shared-atomicmax-float.tglm64

v2: s/ICL+/Gen11+/ (Jason).

Cc: mesa-stable@lists.freedesktop.org
Tested-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4006>

intel: fix the gen 11 compute shader scratch IDs

Scratch space allocation is based on the number of threads in the base
configuration, and we only have one base configuration for ICL, with 8
subslices.

This fixes an issue with Aztec on Vulkan in a machine with a
configuration that's not the base. The issue looks like a regression
from b9e93db20896, but it seems things are broken since forever, just
not easily reproducible.

v2: Reimplement it using the subslices variable. Don't touch TGL.

Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4006>

pan/bi: Move some definitions from disasm to bifrost.h

These are generally useful outside the disassmbler.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Structify FMA_FADD

Just to make it easier to work with.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Squash LD_ATTR ops together

*whistles*

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Combine LOAD_VARYING_ADDRESS instructions by type

It's all a single opcode in fact.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Decode ADD_SHIFT properly

Just like FMA_SHIFT, but with some bits shuffled around.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Identify extended FMA opcodes

When the top 3 bits of the opcode are 111, it leads to a special
extended opcode mode instead.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Add v4i8 mode to FMA_SHIFT

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Decode FMA_SHIFT properly

The shift-bitwise ops are fairly configurable, let's decode this the
right way. Choo choo.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Move notes on ADD ops to notes file

Again, we'd like to see just the opcode table more clearly.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Introduce CSEL4 class

All of these "ops" are just variants on the same.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Move notes on FMA opcodes from disassembler

We're going to be shuffling around the opcode table, so let's get this
moved out first.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Add ICMP.GL.NEQ op

A fused not useful to feed into `discard`.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/bi: Add discard ops

These run on the ADD unit and evidently need to be their own clause
(probably treated as a high-latency instruction). Like csel, they can
either do a float comparison directly or ingest a 0/1 value.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/decode: Skip analysis for Bifrost tiler structures

We don't understand the Bifrost at all yet, so let's just print and move
on.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/decode: Fix tiler weights printing

Theoretical - still always zero.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/decode: Restore bifrost sample_locations

Code by Connor Abbott, reverting a part of
254f40fd535ef57dee2bcc4afd97840749ce5918 where it was removed during a
Midgard refactor.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

pan/decode: Calm an assert to a pandecode error

We'd like to see what the problem actually was...

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4025>

iris: Wait for the GPU to be idle before invalidating the aux table.

An end of pipe sync seems to satisfy this restriction. It takes care of
GPU hangs seen in dEQP-GLES31.functional.copy_image.* tests.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4005>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4005>

iris: Split aux map initialization from invalidation.

We can write the aux map address only once during the batch
initialization, and then only invalidate it once we modify it.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4005>

anv: Wait for the GPU to be idle before invalidating the aux table.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4005>

anv: Do end-of-pipe sync around MCS/CCS ops instead of CS stall

v2: Do end-of-pipe sync after clear depth stencil too (Jason).
v3: Also do end-of-pipe sync before clear depth stencil too (Jason).

Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4005>

anv: Use a proper end-of-pipe sync instead of just CS stall

Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4005>

anv: Use the PIPE_CONTROL instead of bits for the CS stall W/A

Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4005>

gallivm/tessellator: use private functions for min/max to avoid namespace issues

Different builds are failing because of namespace collisions here.

Just fix the MS code to avoid it.

Fixes: bd0188f9eab ("gallium/auxiliary: add the microsoft tessellator and a pipe wrapper.")
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2586
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4016>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4016>

egl: allow INVALID format for linux_dmabuf

As per
https://gitlab.freedesktop.org/wayland/wayland-protocols/commit/fb9b2a87317c77e26283da5f6c9559d709f6fdcd,
the compositor may advertise DRM_FORMAT_MOD_INVALID as a supported
modifier. This patch makes mesa recognize this fact and allow
linux_dmabuf usage with the INVALID modifier in this case.

In case the driver doesn't support modifiers, we can still use
linux-dmabuf protocol instead of the legacy wl_drm interface to create
wl_buffers. This will help compositors to handle these buffers better.

In this commit, the INVALID modifier is allowed to be added to the list
of supported modifiers, and create_wl_buffer will be able to use
linux_dmabuf with an INVALID modifier if the compositor advertised it as
supported.

Signed-off-by: Ivan Molodetskikh <yalterz@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2147>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2147>

lima: add RGBA5551 and RGBA4444 formats

We also need to set channel_layout in pp_frame reg (previously known as
foureight) depending on cbuf format.

Reviewed-by: Andreas Baierl <ichgeh@imkreisrum.de>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3972>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3972>

ci: Add a disabled-by-default job for GLES3 testing on db410c.

Now that we have 7 (soon 8) boards available, there's capacity to be
testing GLES 3.0.  However, due to (it looks like) buffer overflows in the
driver, we end up with flaky test results: 1/60 jobs spuriously failed,
and another 6/60 jobs reported flakes.  At 6 jobs per pipeline, that's way
too high of a failure rate to enable for non-freedreno developers.  Leave
the job present but disabled so that we can do manual test runs for
regressions.

Reviewed-by: Daniel Stone <daniels@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3661>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3661>

ci: Switch testing on db410c over to LAVA.

This should get us better stability of the db410c boards by having a
smaller per-board software stack, with no disks involved (just initramfs).
Additionally, the new cluster is 7 (soon 8) db410cs, while currently the
docker cluster only has 1/4 of its db410cs still running.

Unfortunately, we have to prepare the fastboot boot image during the ARM
drivers build stage, because LAVA relies on publicly available URLs for
the images to load into the bootloaders of the boards, and the only thing
we have for that is gitlab's artifacts.

Note that this testing relies on the boards being freshly flashed with the
linaro v136 firmware to pick up the initramfs size fixes and to stop the
boot at fastboot.

Reviewed-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3661>

r600/sfn: Don't try to catch exceptions, the driver doesn't throw any

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3974>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3974>

r600/sfn: Use static_cast when type is already known

In all these cases the type was tested before based, so don't use
dynamic_casts.

Closes #2566

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Tested-by: Mauro Rossi <issor.oruam@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3974>

r600/sfn: Avoid using dynamic_cast to identify type

v2: Fix typo (maurossi)

Related: #2566

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Tested-by: Mauro Rossi <issor.oruam@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3974>

docs/features: add v3d driver

Now that we bumped the GLES version to 3.1, it makes even more sense
to include the driver here.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2507
Reviewed-by: Jose Maria Casanova <jmcasanova@igalia.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3810>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3810>

aco: pass vars by const &

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3935>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3935>

Revert "gitlab-ci: disable panfrost runners"

The infrastructure issues, caused by building electrical works gone
wrong, have been fixed, and the Panfrost LAVA runners are available
again.

This reverts commit a86662c44d7cb2541c3f613805533064219ad11f.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4019>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4019>

aco: Minor optimization in spill_ctx constructor

'register_demand' is passed by value and only copied once; consider moving it to avoid unnecessary copies

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3968>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3968>

radv: update entrypoints generation from ANV

It's a massive rework loosely based on ANV. This introduces separate
dispatch tables for the instance, physical device and device objects.

This will help for implementing internal driver layers for SQTT.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3930>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3930>

radv/sqtt: add support for GFX10

All SQTT registers were moved to privileged space on GFX10, to emit
them we need a workaround with COPY_DATA.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4018>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4018>

ac/registers: add definitions for thread trace on GFX10

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/4018>

radv/sqtt: update SPI_CONFIG_CNTL.EXP_PRIORITY_ORDER value

It should be 3.

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/4018>

radv/sqtt: do not assume that the number of shader engines is 4

It's not always 4, for example on RAVEN there is only one.

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/4018>

radv/rgp: adjust trace memory/shader clocks to fix frame duration

To report microseconds instead of clocks.

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/4018>

mesa/st: fix formats required for EXT_texture_norm16

Earlier commit did not take in to account that lists required for
rendering and texturing are parsed separately. This commit simply
removes formats added to the other list.

Fixes: de4eb9a3bb9 ("mesa/st: toggle EXT_texture_norm16 based on format support")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3961>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3961>

lima: Add etc1 support

Layer stride has to be divided by 4. We also have to take care of
the array_size when returning the bo_size.

Drop the affected tests from the fails list.

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Andreas Baierl <ichgeh@imkreisrum.de>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3946>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3946>

doc: Update features.txt for r600 with misc supported features

Update features.txt with misc supported features for r600,
as reported by glxinfo for Cypress XT [Radeon HD 5870].

Reviewed-By: Gert Wollny <gert.wollny@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4010>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4010>

intel/tools/dump_gpu: fix getparam values

Don't return the pci_id for all params

Fixes: 76bf38eaf0b6 ("intel/tools/aub_dump: move aub file initialization to maybe_init()")
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3994>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3994>

meson: Enable -Wno-deprecated only for bison > 2.3.

Older versions of bison do not support the -W option.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2571
Fixes: 11a1cb2fa8da ("meson: Disable bison's -Wdeprecated since we still support old bison.")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3993>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3993>

mesa: fix _mesa_draw_nonzero_divisor_bits to return nonzero divisors

The bitmask is _EffEnabledNonZeroDivisor, so no need to invert it before
returning.

Fixes: fd6636ebc06d (st/mesa: simplify determination whether a draw needs min/max index)
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4009>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4009>

gitlab-ci: disable panfrost runners

They seem to be timing out.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4011>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4011>

radv: fix size of sqtt_file_chunk_asic_info on 32-bit system

The struct is actually 716 bytes, but on 64-bit systems the compiler
aligns it to 720. Add padding to make sure it's always 720.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2580
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2578
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3996>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3996>

radv: fix 32-bit build failure in radv_queue_internal_submit()

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2580
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2578
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/3996>

glsl: add some error checks to the nir uniform linker

These are optional for spirv but it shouldnt hurt to enable them.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>

glsl: fix sampler index calculation in nir linker

Here we reset the counter to 0 for each shader stage not each program.
We also make add a flag to stop iterating over indices that have
already been processed.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>

glsl: reset next_image_index count for each shader stage

This fixes the image index calculation in the nir linker. We need
to reset the counter to 0 for each shader stage not each program.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>

glsl: fix resizing of the uniform remap table

In the NIR linker we were not resizing the remap table correctly
for explicit locations when it was needed.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>

glsl: set the correct number of images in a shader

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>

glsl: set the correct number of samplers in a shader

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>

glsl: fix possible memory leak in nir uniform linker

Use UniformDataSlots for the context of UniformDataDefaults rather
than UniformStorage as in some cause UniformStorage may be NULL.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3992>

intel/compiler: Restrict cs_threads to 64

Our current GPGPU_WALKER code only supports up to 64 threads.

On HSW we could use up to 70 and TGL up to 112, but only if the walker
is adjusted so the width does not exceed 64. Work to support this is
in progress.

Previous to this change, we might try to downgrade to SIMD8 if the
SIMD16 shader spilled. Since HSW and TGL have the max number of
threads above 64, we would then try to emit an invalid GPGPU walker
command.

Fixes: 932045061b5 ("i965/cs: Emit compute shader code and upload programs")
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>

st/va: remove unneeded code

No need to explicitly set the 10-bit buffer format as the correct
buffer format will be allocated later

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3998>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3998>