pan/bi: Add CSEL.8 opcode

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4789>

pan/bi: Add FCMP.GL.v2f16 on ADD opcode

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4789>

pan/bi: Add 64-bit int compares

Likewise.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4789>

pan/bi: Add some 8-bit compares

Not all but enough to see the pattern.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4789>

pan/bi: Add CSEL.64 opcode

Chain twice for full 64-bit CSEL.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4789>

pan/bi: Add bool->float opcodes

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4789>

radv: enable FMASK for color attachments only

The reason behind this is that FMASK requires CMASK and also that
FMASK for non color attachments looks unnecessary. It's currently
much easier to add this simple check because the driver tries to
always enable DCC first and if we enable FMASK only if CMASK, we
might loose some FMASK compressions.

This helps fixing some new robustness2 tests which fails because
only FMASK is enabled.

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/4783>

anv: Expose CS workgroup sizes based on a maximum of 64 threads

Otherwise, we'll hit asserts in brw_compile_cs.

Fixes: cf12faef614ab "intel/compiler: Restrict cs_threads to 64"
Closes: #2835
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4746>

intel/devinfo: Compute the correct L3$ size for Gen12

Fixes: 8125d7960b6 "intel/dev: Add preliminary device info for Tigerlake"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Clayton Craft <clayton.a.craft@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4782>

radv: Determine memory type for import based on fd.

This would be necessary for an application to figure out if the
memory was allocated using a memory type with VK_MEMORY_PROPERTY_PROTECTED_BIT.

It also allows one to determine VRAM vs. GTT etc.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4751>

radv/winsys: Add function to get domains/flags from fd.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4751>

radv: Stop using memory type indices.

Lots of extra coding was involved in managing them.

And for protected memory I was thinking of making a function that
goes from domain+flags to memory types, which can reuse this array.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4751>

radv: Use actual memory type count for setting app-visible bitset.

Otherwise we might make a bitset that is too large.

Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4751>

radeonsi: Count planes for imported textures.

For the DRI2 lowered YUV import separate pipe_resources get created
but in the end the first resource just gets asked for NPLANES.

Since

1) (Almost) everything uses the first resource + a plane index in the
  Gallium interface.
2) This mirrors non-imported textures.

lets fix this in the driver.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4779>

r600: Enable tesselation for NIR

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Add tesselation shaders

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Add lowering passes for Tesselation IO

Lower the input and output intrinsics to r600 specific LDS intrinsics

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Move removing of unused variables

It doesn't make sense to do this in the optimization loop

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Handle LDS output in VS

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: derive the GS from the vertex stage for a common interface

The GS can also provide the primid

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: extract class to handle the VS export to different stages

This code can be shared with the TESS_EVAL shader

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Move some shader base methods to the public interface

This will be needed for handling the VS stage export better.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Add methods to valuepool to get a vector of values

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Move emission of barrier from compute shader to shader base

Tess shaders also use these barriers.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Emit some LDS instructions

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Handle umul24 and umad24

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Add IR instruction to fetch the TESS parameters

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Add TF write instruction

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Add LDS instruction to assembly conversion

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Add LDS IO instructions to r600 IR

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: Don't emit inline constants in the r600 IR

This can be handled when lowering to assembly, and it makes testing
for indirect buffer and sampler access easier.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600/sfn: simplify UBO lowering pass

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

r600: Handle texcoord semantics in LDS index evaluation

With NIR the texcoord semantic is enabled, and hence we have to handle
index evaluation differently here.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4714>

ci: bare-metal: power down device after tests

Helps to save electricity.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4754>

panfrost: Fix GL_EXT_vertex_array_bgra

Previously, attributes would always use an RGBA swizzle, even if the
format was BGRA.

Fixes piglit tests bgra-sec-color-pointer and bgra-vert-attrib-pointer.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4752>

ci: add llvmpipe paths to virgl rules

since llvmpipe changes will affect virgl

Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4777>

radv: do not expose GTT as device local memory mostly for APUs

On APUs, the memory is unified (all heaps are equally fast) and
apps should count all memory heaps together. But some games like
Id Tech games (Youngblood and such) don't manage memory correctly
on APUs and they spill everything when one VRAM heap is full.

Instead of spilling buffers, they should just allocate new buffers
in the second heap but it seems like these games are confused if
two memory heaps have the DEVICE_LOCAL_BIT set.

This is probably a first step towards better memory management on
APUs but there is still some work to do if we want to run most apps
with a small dedicated VRAM (256MB or so).

This gives a huge boost for Id Tech games on APUs, and doesn't
seem to reduce Feral games performance.

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/4771>

gallium/swr: Fix LLVM 11 compilation issues

Changes needed to adapt to LLVM API changes in vector
and pointer types.

Reviewed-by: Krzysztof Raszkowski <krzysztof.raszkowski@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4769>

ci/freedreno: Add a test run of a few driver options.

This lets us get coverage of corner cases of the driver that are tricky to
force a testcase to hit.  We don't want to do a full run of the CTS with
each option because that's a lot of runner time, so stack a bunch of
fractional runs in one test job to amortize the test run setup overhead.

Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4621>

ci: Clean up some excessive use of pipes in dEQP results processing.

Given that we use set -x in the script, this actually makes the user
experience of viewing logs nicer.

Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4621>

ci: Allow namespacing of dEQP run results files.

I want to do multiple runs of some bits of the CTS in one test job to test
some driver options, but I want to be able to see the results from any of
them.

Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4621>

freedreno: Fix calculation of the const buffer cmdstream size.

The HW packet requires padding the number of pointers you emit, and we
would assertion fail about running out of buffer space if the number of
UBOs to be uploaded was odd.

Fixes: b4df115d3f3c ("freedreno/a6xx: pre-calculate userconst stateobj size")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4621>

ci: Add sanity checking that dEQP gets the expected GL_RENDERER.

It's easy to get something wrong in the driver build or container or
something that results in falling back to swrast, and then your only clue
was runtime and how your failure cases suspiciously match a swrast
driver's.

Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4621>

ci: Enable --compact-display false on all dEQP runs.

We always want to see status updates happening in the logs, otherwise it
can like maybe your machine hung until the run actually completes.

Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4621>

zink: explicitly unref old fb object when setting new one

this object has a ref from being created, and its lifetime is expected to
be a single frame, so remove that initial ref when we expect to stop
using it

Closes: #2648
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4768>

zink: remove framebuffer cache

this can only match when re-rendering identical frames, which is not a
typical case. the lack of cache eviction also leads to memory ballooning.

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4768>

st/dri: Set next in template instead of after creation. (v2)

This should prevent horrors like Iris has with the delayed calls
to iris_resource_finish_aux_import just because info is not
available at allocation time.

AFAICT all drivers just copy the template except radeonsi/r600
which reset the next pointer.

AFAICT there is also no other place we get a state tracker setting
next ptrs on a resource.

v2: Updated Gallium docs.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3792>

mesa/st: call _mesa_initialize() early

This allows drivers to reliably do things like using the GLSL
type-system during initialization.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4765>

mesa/main: one_time_init() -> _mesa_initialize()

This exposes the logic inside one_time_init() as _mesa_initialize(), so
drivers who needs to use functionality initialized in one_time_init
earlier if they need.

This means we can reliably use the GLSL type-system when compiling
driver built-in shaders.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4765>

mesa/main: Do not pass context to one_time_init

There's no longer any reason to pass the context down to one_time_init,
because we always do the same thing regardless of the context, and we
don't change the context.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4765>

mesa/main: do not init remap-table per api

This hasn't really been nessecary since 8386088e3dc ("dispatch: stop
using _mesa_create_exec_table_es1() for GLES1."), when we stopped
diverging the logic here based on the context-API. So let's simplify the
code a bit.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4765>

mesa/main: do not pass context to one-time extension init

_mesa_problem doesn't use the ctx argument for anything, so there's no
reason to pass it. This saves us from needing a context passed down this
code-path in the first place.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4765>

mesa/main: do not store unrecognized extensions in context

We process extension overrides only when we initialize the first
context, which means that unrecognized extensions only appear in the
first context created.

Let's instead store them in a global array, so we can apply them to all
contexts. This has the added benefit of making the initialization of the
first context less special, which allows us to clean up code a bit more.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4765>

vulkan: add initial device selection layer. (v6.1)

This is code Bas has out of tree but I think mesa should be shipping it, and I've improved it.

Initially-written-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
v2: add infinite recursion fix (Bas)
v3: Fix wayland/xcb barrier, whitespace
v4: use a macro for getting apis, shorten some lines, use outarray
v5: rewrite in C, use hash_table/mutex.
v6: use once_init to init the mutex, fix freeing ht

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1766>

freedreno/ir3: Add support for disasm of cat2 float32 immediates.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4736>

freedreno/ir3: Refactor out print_reg_src().

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4736>

freedreno/ir3: Convert remaining disasm src prints to reginfo.

More lines of code, but they're much more intelligible.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4736>

freedreno/ir3: Add a unit test for our disassembler.

Makes sure that we can maintain consistent output from our disassembly as
we refactor.  I've only included stuff that matches qcom's disasm so far.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4736>

freedreno/ir3: Print a space after nop counts, like qcom's disasm.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4736>

freedreno/ir3: Fix the disasm of half-float STG dests.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4736>

ci: Enable GLES 3.1 testing on db820c (a530).

The driver exposes GLES3.1, so let's make sure we're not regressing its
featureset.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4685>

freedreno: Fix derivatives without texturing on a3xx-a5xx.

The shader variant tells us if we should set the PIXLODENABLE flag.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4685>

ci: Enable GLES3 testing on db410c/db820c (freedreno a306 and a530).

We haven't had it enabled due tointermittent failures.  Those failures
are, as far as I can tell, due to GPU faults from buffer overflows where a
failing test in a thread stomps an otherwise passing thread's buffers.  By
running deqp single-threaded, we can get more consistent failures, at the
cost of needing to do a tiny subset of the tests to keep runtime down.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4685>

ci: Drop redundant freedreno stage specification.

The source rules give us the stage.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4685>

freedreno/ir3: run nir_lower_pack

This lowers pack_32_2x16/unpack_32_2x16 into the scalar versions of those
instructions.

Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4738>

nir: add pack_32_2x16_split/unpack_32_2x16_split lowering

The new option replaces the two other _split lowering options, since
there's no need for separate options.

Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4738>

radv: Add WSI buffers to BO list only if they can be used.

Also reverse the BO list removal loop. This way typical WSI usage
should find the entry in O(active swapchains) iterations, which
should not be a performance issues. Tested with Doom(2106) which
found the entry in 1 iteration every time.

Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4306>

vulkan/wsi: Add callback to set ownership of buffer.

For radv BO list pruning.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4306>

ac,radeonsi: fix compilations issues with LLVM 11

Latest LLVM replaced LLVMVectorTypeKind.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2826
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4755>

gallium/gallivm: remove unused header include for newer LLVM

In the top of the trunk LLVM (11) llvm/IR/CallSite.h header
has been removed. The file compiles without this include also
for LLVM 8, but I'm not sure about 9, 10, and older versions
so I disable it only for the latest LLVM

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4748>

gallium/gallivm: fix compilation issues with llvm 11

Top of the trunk LLVM removes old vector type
and introduces two new ones - one for fixed-width
and one for scalable vectors. This commit fixes
compilation issues by switching from LLVMVectorTypeKind
to LLVMFixedVectorTypeKind for new LLVM.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4748>

ir3: Use shared mediump output lowering

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4716>

nir: Move nir_lower_mediump_outputs from ir3

(Original code from ir3)

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4716>

nir/algebraic: Detect some kinds of malformed variable names

I spent over an hour trying to debug a problem if a condition on a
variable not being applied.  The problem turned out to be
"a(is_not_negative" instead of "a(is_not_negative)".  This commit would
have detected that problem and failed to build.

v2: Just add $ to the end of the existing regex, and it will fail to
match a malformed string.  Suggested by Jason.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> [v1]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4720>

pan/bi: Implement 16-bit COMBINE lowering

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Fix RA wrt 16-bit swizzles

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bit: Add SELECT tests

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bit: Interpret BI_SELECT

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Force BI_SELECT arguments scalar

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Pack ADD SEL16

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Pack FMA SEL8

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Pack FMA SEL16

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Rename BI_SWIZZLE to BI_SELECT

The select version is more general.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Eliminate writemasks in the IR

Since the hardware doesn't support them, they're a burden to deal with,
so let's ensure we never get to a place where we would need to at all.

Disables COMBINE lowering for now.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Fix ADD.v4i8 opcode

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Add missing BI_VECTOR

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Assign blend descriptor for BLEND op

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Passthrough blend types

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Passthrough type for ATEST

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

pan/bi: Pack fp16 ATEST

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4766>

mesa: Skip 3-byte array formats in _mesa_array_format_flip_channels

Byte swapping makes no sense for 3-byte formats: Swapping the order of 2
or 4 bytes at a time would inevitably result in bytes getting mixed up
between neighbouring pixels.

Fixes crash with a debugging build on a big endian machine due hitting
the unreachable() at the end of the function.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2665
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4735>

mesa: replace GLenum target with gl_shader_stage in NewProgram

So that the GLSL compiler doesn't have to use the GLenum conversion
functions.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4756>

drm-uapi,radv,radeonsi: Add amdgpu_drm.h header.

Use it instead of the libdrm provided amdgpu_drm.h header. I used
the kernel revision from the README to get the header so the
header versions should be consistent.

Tested by removing /usr/include/libdrm/amdgpu_drm.h from my dev-machine.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4749>

mesa: extend _mesa_bind_vertex_buffer to take ownership of the buffer reference

This reduces overhead of _mesa_reference_buffer_object_ from 6% to 4%
with glthread when profiling the game "torcs" with non-VBO data uploaded
by glthread.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4758>

mesa: add offset_is_int32 param into _mesa_bind_vertex_buffer for glthread

glthread will pass signed integer offsets, so don't reset negative offsets
to 0 there.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4758>

mesa: add Const.BufferCreateMapUnsynchronizedThreadSafe & MESA_MAP_THREAD_SAFE

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4758>

gallium: add PIPE_CAP_MAP_UNSYNCHRONIZED_THREAD_SAFE for glthread

and add radeonsi support.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4758>

glthread: sort variables in marshal structures to pack them optimally

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4758>

glthread: use GLenum16 in batch buffers to save space

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4758>

glthread: reduce dereferences of the next batch

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4758>