GitLab

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Logic is lifted from bi_layout.c, adapted to work on instructions (not
clauses) and for Valhall's off-by-one semantic which is annoyingly
different than Bifrost. (But the same as Midgard -- Bifrost was
annoyingly different than Midgard!)
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Implicit on Bifrost.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

We should really extend the IR.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Valhall removes certain instructions from Bifrost, requiring a canonical
lowering.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Valhall has a lookup table for common constants. Add a pass to take
advantage of it, lowering away immediate indices.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

These are pre-conditions required for packing.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Yo, I heard you like tests.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

This handles enough for basic compute shaders to compile.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Used to determine the instruction's action.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Just NOP everything out for now.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Annoyingly different from Bifrost.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Filled out the new structures from XML.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Specify gpu id with --gpu-id or marketing name with --gpu. Still have
compile/disasm as commands, but allow -v for verbose printing.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

I want to test SSBOs with the standalone compiler while retaining full
support for mediump qualifiers.

v2: Actually test SSBOs on compute shaders.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>

If we have 2 command buffers back to back, one with a query pool, one
without, we don't want to retain the second query pool value (NULL).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 0a7224f3

 ("anv: group as many command buffers into a single execbuf")
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <mesa/mesa!12107>

This avoids a deadlock condition when registered atexit handlers attempt to
acquire a mutex, but the program was going to exit with an error anyway
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Part-of: <mesa/mesa!11813>

This helps ensure we're not doing out of bounds access later.
Signed-off-by: Rohan Garg <rohan.garg@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <mesa/mesa!12075>

Signed-off-by: Rohan Garg <rohan.garg@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <mesa/mesa!12085>

These were fixed previously, but due to the CI not really running all
tests any more, I didn't notice these fixes. Let's bring the expected
results up to date.

Fixes: 2e29857b

 ("llvmpipe: only report supported shader-image formats")
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <mesa/mesa!12077>

No need to build the simulator with NEON; and also v3dv3 simulator
is not for VC4, so don't inherit v3dv3 requirement when building vc4
driver.

Fixes: #5126
Fixes: d198e26a

 ("broadcom/common: move v3d_tiling to common")
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <!12078>

This allows the variables decorated with RelaxedPrecision to have the
proper precision. It is worth to note that the decorator can be
applied on other cases, but those would be handled on the future.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <mesa/mesa!7614>

All vulkan drivers have been copying anv's code to convert
VkSpecializationInfo into nir_spirv_specialization.

Recently there was a Vulkan spec change on allowed values for
VkSpecializationInfo, and all drivers got affected.

This commits creates a new helper, and uses it on all Vulkan Mesa
drivers.

v2: use (uint8_t*) castings, instead of void*, to avoid C2036 with
    MSVC (detected by the CI, inspired on what radv was doing)
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <!12047>

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <mesa/mesa!12119>

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <mesa/mesa!12119>

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <mesa/mesa!12119>

Fragment isn't last for GLSL stages, compute is.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <!12082>

Part-of: <mesa/mesa!12071>

this uses a singly-linked list of timeline ids to compare against the device
queue's cmdbuf counter and determine which timeline id maps to which cmdbuf and
thus which fence can be waited on
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <mesa/mesa!12071>