AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
141,262 Commits 1 Branch 0 Tags
82c076d2c3147d632e951bc143a65d79ca183113
Commit Graph

130327 Commits

Author SHA1 Message Date
Gert Wollny 82c076d2c3 r600/sfn: Don't read return values of atomic ops that are not used 
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11348>
 2021-06-17 10:53:52 +00:00
Gert Wollny 800bfefde5 r600/sfn: Drop method for emit_atomic_add, it is handled in generic code 
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11348>
 2021-06-17 10:53:52 +00:00
Gert Wollny 2e8901a072 r600/sfn: don't read back unused image atomic result values 
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11348>
 2021-06-17 10:53:52 +00:00
Gert Wollny 5e1a703934 r600/sfn: don't designates initializers, since they are c++20 
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11348>
 2021-06-17 10:53:52 +00:00
Kai-Heng Feng f9d8d9acbb iris: Avoid abort() if kernel can't allocate memory 
When the system doesn't have enough memory, GNOME Shell may be crashed
by iris:
gnome-shell[1161]: iris: Failed to submit batchbuffer: Cannot allocate memory
gnome-shell[1161]: GNOME Shell crashed with signal 6

So don't abort() when kernel can't allocate memory to avoid crashing the
entire desktop.

Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11178>
 2021-06-17 09:04:17 +00:00
Iago Toral Quiroga d873e30d3b v3dv: expose VK_KHR_shader_non_semantic_info 
This is entirely implemented in the SPIR-V frontend.

Relevant CTS tests:
dEQP-VK.spirv_assembly.instruction.compute.non_semantic_info.*

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11440>
 2021-06-17 08:49:16 +00:00
Pierre-Eric Pelloux-Prayer 0fd152dab3 disk_cache: use UTIL_QUEUE_INIT_SCALE_THREADS 
Instead of spawning 4 threads when the cache is created,
spawn 1 and let u_queue grow the number of threads if
needed.

I wrote this patch because when running piglit's quick_shader
profile I had lots of samples in disk cache threads - mostly
in native_queued_spin_lock_slowpath kernel function.

Since these tests shouldn't really stress the cache, I assumed
it was caused only by thread creations.

After writing the patch and redoing the measurement, I got an
improvement but I still more hits in the same function for
shader_runner:$disk0 thread so something was wrong.

After digging more, I found out that my shader cache index was
corrupted: the on-disk size was 29MB but the index reported it
was way more than 1GB. So each disk cache thread was spending
a lot of time trying to evict files. Given that my cache had
a really low count of files, the LRU method based on randomly
generating subfolder names failed, so evicting was very slow.

Now that my cache index is fixed, the disk cache threads are
mostly idle but I still think it makes sense to grow the
number of threads instead of spawning 4 at the program start.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11296>
 2021-06-17 09:11:59 +02:00
Pierre-Eric Pelloux-Prayer 3713dc6b2a util/u_queue: add UTIL_QUEUE_INIT_SCALE_THREADS flag 
This flag allow to create a single thread initially, but set
max_thread to the request thread count.

If the queue is full and num_threads is lower than max_threads,
we spawn a new thread to help process the queue faster.

This avoid creating N threads at queue creation time.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11296>
 2021-06-17 09:11:59 +02:00
Pierre-Eric Pelloux-Prayer 0c88df1f6a util/u_queue: move function definition up 
Will be used by the next commit.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11296>
 2021-06-17 09:11:58 +02:00
Pierre-Eric Pelloux-Prayer edb77ec3c7 radeonsi: delay sample_pos_buffer creation until first use 
And use pipe_buffer_create_with_data instead of doing it
manually.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11296>
 2021-06-17 09:10:29 +02:00
Pierre-Eric Pelloux-Prayer 84dca32c51 vbo: delay vbo_exec_vtx_map call 
Instead of doing vbo_exec_vtx_map during initialization,
defer it until the first actual user.

v2: move init to vbo_exec_wrap_upgrade_vertex (Emma Anholt)

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11296>
 2021-06-17 09:10:29 +02:00
Vinson Lee 3333a96b6f nvc0/ir: Initialize CodeEmitterNVC0 member progType in constructor. 
Fix defect reported by Coverity Scan.

Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member progType is not initialized
in this constructor nor in any functions that it calls.

Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11350>
 2021-06-16 22:23:52 -07:00
Marek Olšák 849ab4ea0c mesa: execute glFlush asynchronously if no image has been imported/exported 
This improves viewperf performance and it shouldn't break synchronization
with external clients when it's indirectly implied by glFlush.

This should not break the cases described in:
    https://gitlab.freedesktop.org/mesa/mesa/-/issues/4903

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11341>
 2021-06-17 02:24:22 +00:00
Marek Olšák 0336b13e1e mesa: move _mesa_notifySwapBuffers into the x11 swrast driver 
It has no other use and no relevance to DRI drivers despite the name.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11341>
 2021-06-17 02:24:22 +00:00
Marek Olšák b7dc72380a mesa: add gallium flush_flags param into ctx->Driver.Flush 
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11341>
 2021-06-17 02:24:22 +00:00
Marek Olšák ee6929a779 st/mesa: move the st_flush_bitmap_cache call into st_flush 
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11341>
 2021-06-17 02:24:22 +00:00
Marek Olšák 37bf4184db st/mesa: fix an incorrect comment in st_context_flush 
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11341>
 2021-06-17 02:24:22 +00:00
Marek Olšák 3da170faae glthread: change when glFlush flushes asynchronously 
This fixes the flushing with external textures.

We don't know if we need to flush synchronously with multiple contexts,
so I removed that.

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11341>
 2021-06-17 02:24:22 +00:00
Kenneth Graunke 0510a947ba anv: Fix dynamic primitive topology for tess on Gfx7.x too 
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>
 2021-06-17 02:00:32 +00:00
Mike Blumenkrantz 0ea8becff0 zink: always defer image descriptor barriers 
this is simpler and guaranteed to be accurate

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11393>
 2021-06-17 01:18:39 +00:00
Mike Blumenkrantz 5341b985e4 zink: use fake buffer barriers for descriptors 
GL requires explicit glMemoryBarrier calls for shader synchronization
and only calls that map/copy buffers get implicit sync, so we don't actually
need barriers for any of these cases, only the state needs to be updated

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11393>
 2021-06-17 01:18:39 +00:00
Mike Blumenkrantz 3c2f343618 zink: check actual mem props to determine if resource object is coherent 
this was correct for what it has been used for until now, but it will no
longer be correct going forward

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11397>
 2021-06-17 01:07:35 +00:00
Mike Blumenkrantz b1b6aecf9d zink: key alloc cache on heap index, not heap flags 
this is a bit more sane

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11397>
 2021-06-17 01:07:35 +00:00
Mike Blumenkrantz c56a100494 zink: avoid caching visible vram allocations 
the visible vram heap is potentially going to be limited in size, so avoid
caching these allocations since that locks them to a given allocation size

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11397>
 2021-06-17 01:07:35 +00:00
Mike Blumenkrantz 18b71c0da3 zink: change a bunch of sparse buffer resource checks to host-visible checks 
(sparse buffer) is a subset of !host-visible, and !host-visible is actually the
more correct check to be using

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11397>
 2021-06-17 01:07:35 +00:00
Mike Blumenkrantz 5fee58bf59 zink: collapse host_visible and non-coherent alignment alloc cases 
* buffers can use normal mem prop checking to determine host_visible setting
* sparse buffers are never coherent, so this case can be dropped from the conditional

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11397>
 2021-06-17 01:07:35 +00:00
Martin Krastev eb272f6571 compiler/glsl: Use mutex lock while freeing up mem_ctx 
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>
 2021-06-16 23:14:23 +00:00
Mike Blumenkrantz 4a407e0ad8 zink: mark some functions inline 
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11429>
 2021-06-16 23:02:55 +00:00
Connor Abbott e19f112435 ir3/ra: Fix array parallelcopy confusion 
With array registers, there are two num's we care about:

1. The base num that the whole array starts at (->array.base)
2. The num that the instruction uses, plus possibly an indirect offset
   (->num or ->array.offset)

For parallel copies we always copy the whole array, so (2) is irrelevant
here. For phis and parallel copies inserted for phis, we used
assign_reg() which assigned ->array.base, but we forgot about this when
constructing our own parallel copies for live range splitting, just
setting ->num instead. The parallel copy lowering was also inconsistent
here, using ra_reg_get_num() (which looks at ->array.base for arrays)
for sources but looking at ->num directly for destinations. This makes
everything use ->array.base consistently.

While we're here, make sure to remove IR3_REG_SSA from liveout copies to
make sure printing works correctly.

Fixes: 0ffcb19 ("ir3: Rewrite register allocation")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11422>
 2021-06-16 22:45:13 +00:00
Connor Abbott 2c21dab36e ir3: Improve printing of array parallelcopies/phis 
Normally something with IR3_REG_ARRAY doesn't have a register assigned,
but we keep IR3_REG_ARRAY for parallel copies after RA because we need
to know the appropriate size. We want to see the register assigned for
these when printing the RA result before parallel copies are lowered.
The register is in ->array.base in this case, so initialize it to
INVALID_REG and print ->array.base if it's been assigned to something,
similar to ->num in the normal case.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11422>
 2021-06-16 22:45:13 +00:00
Marek Olšák d0d2108425 shader_enums: change VERT_BIT back to the 32-bit shift 
This reverts 0e2566a8. The warning is fixed differently.

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10993>
 2021-06-16 21:45:57 +00:00
Marek Olšák 45cbb08848 gallium/pb: change alignment to 32 bits 
This partially reverts 4a3f0444. The warning is fixed differently.

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10993>
 2021-06-16 21:45:57 +00:00
Marek Olšák 97a5753812 radeonsi: remove -Wstrict-overflow=0 since it doesn't seem to be needed 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11384>
 2021-06-16 21:29:13 +00:00
Marek Olšák 72a395b6de radeonsi: remove the chip_class dimension from the draw_vbo array 
We don't use/initialize draw_vbo callbacks for other generations anymore.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11384>
 2021-06-16 21:29:13 +00:00
Marek Olšák 1e4d91355f radeonsi: compile si_state_draw.cpp for each gfx generation separately 
It makes compilating faster.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11384>
 2021-06-16 21:29:13 +00:00
Marek Olšák 24895f020a radeonsi: move a few functions from si_state_draw.cpp into si_gfx_cs.c 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11384>
 2021-06-16 21:29:13 +00:00
Yiwei Zhang b8ce8530a0 venus: remove workarounds for multi-planar format interop 
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11282>
 2021-06-16 20:31:32 +00:00
Yiwei Zhang 357e7ae30b anv: fix build errors after commit 8b7ff78 
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Roman Stratiienko <r.stratiienko@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11373>
 2021-06-16 19:55:48 +00:00
Yiwei Zhang ec1968dcc9 radv: fix build errors after commit 8b7ff784 
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Roman Stratiienko <r.stratiienko@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11373>
 2021-06-16 19:55:48 +00:00
Mike Blumenkrantz 25972df4b7 zink: ci updates 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11391>
 2021-06-16 19:45:25 +00:00
Mike Blumenkrantz d0833f611a zink: support more RGBX formats 
the base formats are supported, so these should fall into place naturally

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11391>
 2021-06-16 19:45:25 +00:00
Mike Blumenkrantz 1e86cdd8b7 zink: also nope out of any dst alpha blends for rgbx formats 
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11391>
 2021-06-16 19:45:25 +00:00
Mike Blumenkrantz f091392c0d zink: smash dstAlphaBlendFactor to ZERO for RGBX attachments 
this is a no-op

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11391>
 2021-06-16 19:45:25 +00:00
Mike Blumenkrantz 2f2976e9e1 zink: add a more direct check for rgbx formats in create_sampler_view hook 
really the point of this is to clamp void channels for any permutation of rgbx
where all channels are the same (e.g., both rgbx8 and rgbx16), so the previous
helper isn't inclusive enough

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11391>
 2021-06-16 19:45:25 +00:00
Mike Blumenkrantz a3a6611e96 util/queue: add a global data pointer for the queue object 
this better enables object-specific (e.g., context) queues where the owner
of the queue will always be needed and various pointers will be passed in
for tasks

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11312>
 2021-06-16 15:10:09 -04:00
Marek Olšák d305a8fc5f radeonsi: set desc[3] of all buffer descriptors at context creation 
instead of at bind time. When we unbind, we shouldn't memset the last
element.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11340>
 2021-06-16 17:14:49 +00:00
Marek Olšák 42801731b8 radeonsi: remove no-op unref in si_set_constant_buffer 
buffer is already NULL here

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11340>
 2021-06-16 17:14:49 +00:00
Marek Olšák a29ff4c67e radeonsi: restructure si_set_sampler_views for faster unbinding trailing slots 
si_set_sampler_view updates only one slot. This rewrites it to update
multiple slots.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11340>
 2021-06-16 17:14:49 +00:00
Marek Olšák 10cf7b3031 radeonsi: don't clear register fields in si_set_mutable_tex_desc_fields 
They are always initialized to 0.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11340>
 2021-06-16 17:14:49 +00:00
Marek Olšák 810e84387b radeonsi: use the restrict keyword to set sampler view descriptors faster 
All places must set restrict for the same pointer.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11340>
 2021-06-16 17:14:49 +00:00