panfrost: Factor out general varying case

To avoid hard-to-follow control flow now that we return void.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Pass varying descriptors by reference

Instead of returning by value, as a stepping stone to packing directly.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Reword comment

Avoid referencing data structures that no longer exist.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Use packs for vertex built-ins

These probably need a bigger cleanup at some point but let's handle one
thing at a time.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Use packs for vertex attribute buffers

Eliminates messy staging, I think.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Inline panfrost_vertex_instanced

We'd like to have attribute code in one place together so we can
refactor effectively; this routine as-is wouldn't fit perfectly outside
Gallium anyway so let's do the simple thing.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Hoist instance_shift/instance_odd fetch

We'll want to pass it as an argument shortly.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Use packs for attributes

Now that state is grouped, this is natural.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Move attr_meta emission to the draw routine

It's unfortunate that we can't do more at CSO time, but actually all we
really need is the format after all, and this lets us group the state.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Make attribute-buffer map explicit

This helps decouple the attribute and attribute buffer emission.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Simplify offset fixup proof

No need to put strides and such into it at all.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Use better packs for blits

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Add XML for attribute buffers

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Merge attribute packing routines

In preparation for streamlining the packing, we need related code in one
place.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: XMLify attributes

To keep review manageable, this only ports over pandecode and blitting.
The main uses for vertex data and varying linkage will come next, as
they touch considerably more code.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: XMLify mali_channel

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Squash 22-bit format field in attr_meta

This is an intermediate step to support v7 style formats. Now we don't
see the mali_format alone, instead together with the swizzle (and sRGB
flag, unused for attributes of course).

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Drop NXR format

It's not clear this was ever exposed under any GL/DX/CL/VK/... extension
whatsoever...

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

panfrost: Set STRIDE_4BYTE_ALIGNED_ONLY

It isn't clear to me why, but this breaks in some cases (such as some
dEQP cases where the blob does a translate too).

This reverts commit 9f2997dad061ee38e06f654f524dffbffc5b2a4b.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Cc: mesa-stable
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>

pipebuffer: Remove unused buffer event in slab bufmgr

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6225>

nir: handle load_input_vertex in nir_get_io_offset_src

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6328>

nir: remove nir_strip stub declaration

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6328>

compiler: add glsl_print_type

Move it from the glsl compiler. For debugging.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6328>

ac/llvm: handle static/shared llvm init separately

Having a single init function works as expected for shared llvm, but
when using a static llvm only one llvm will get initialized.

This commit introduces 2 separate init function:
- shared llvm = single public init function
- static llvm = one init function for each module using llvm

Fixes: 50d20dc055d ("ac/llvm: export ac_init_llvm_once in targets")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3376
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6253>

st/mesa: Enable clip planes lowering for geometry shaders

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6218>

st/mesa: factor ucp-lowering logic into helper

This makes the next step easier to follow.

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6218>

compiler/glsl: Initialize local variable to zero to fix MSVC RTC error

Without this, we get a run-time error about using an uninitialized
variable.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6221>

vallium: initial import of the vulkan frontend

This is the initial import of the vallium frontend for gallium.
This is only good enough to run the triangle and the gears demo
(wrongly) from Sascha demos.

Improvements are mostly on the llvmpipe side after this.

It contains an implementation of the Vulkan API which is mapped
onto the gallium API, and is suitable only for SOFTWARE drivers.

Command buffers are recordred into malloced memory, then later
they are played back against the gallium API. The command buffers
are mostly just Vulkan API marshalling but in some places the information is
processed before being put into the command buffer (renderpass stuff).

Execution happens on a separate "graphics" thread, againt the gallium API.

There is only a single queue which wraps a single gallium context.

Resources are allocated via the new resource/memory APIs.
Shaders are created via the context and bound/unbound in the
second thread.

(No HW for reasons - memory management, sw paths for lots of paths,
pointless CPU side queue)

v2: drop mesa_icd, drop cpp_args, drop extra flags, change meson config (Eric)
v2.1: use meson-gallium job

meson pieces:
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
overall:

Acked-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6082>

vulkan/wsi: add sw support. (v2)

This adds an option to the WSI support for a software path to be
used with the vulkan sw drivers. There is probably some changes
that could be made to improve this and use present, for now
just use put image.

v2: roll out flag across all drivers (Eric)

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6082>

util/format: add some ZS helpers for vallium

The vallium layer has a requirement to insert and extra the 24-bit
unorm value as a unorm value (not as a float etc). Add helpers
to facilitate that.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6082>

gallivm/nir: add load push constant support

This convets the push constant load to a ubo load against the 0 constant buffer,
which is where the vallium layer will bind things.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6082>

llvmpipe: add support for memory allocation APIs

This adds llvmpipe driver support for the vulkan split memory/resource API
and the the don't overallocate flag.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6082>

gallium: add a resource flag to say no over allocation.

llvmpipe overallocates buffers for buffers used as render targets,
however this breaks some vulkan apps (UE4), so add a workaround
flag to force llvmpipe to not overallocate certain buffers.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6082>

gallium: add an interface for memory allocations.

In order to support vulkan over gallium for the sw renderers,
there needs to be a vulkan-like memory allocation API.

It doesn't need to be overly complicated for the needs of the sw
renderers.

The vallium layer will allocate resources and memory separately
and bind them via this API.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6082>

radv: Use getter instead of setter to extract value.

Oops ...

Fixes: 64768111c30 "radv: Handle slightly different image dimensions."
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3417
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6337>

gallium: Use unpack_rgba() instead of fetch_rgba in translate_generic

This is the only user of fetch_rgba outside of llvmpipe, and it's in the
fallback path of this fallback path.  Looking at an example of these two
functions, b8g8r8a8's unpack_rgba is 2.7x as long as fetch_rgba.  It feels
reasonable to sacrifice some perf in this already slow (VBO readback, and
a function pointer call per attribute per vertex) path to reduce our
binary size.  And, if I ever finish getting unpack codegen to switch to
rows instead of rects, that factor will go back down.

Saves 40kb of binary on non-llvmpipe gallium drivers.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6305>

util: Move fetch_rgba to a separate function table.

Only llvmpipe and translate_generic use it, and only in fallbacks, so if
you're not building that then let's not bloat our binaries with it.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6305>

util: Mark the format description getter functions as const.

This lets the compiler CSE calls to them on the same format.  This is
particularly relevant for the description table lookup calls, which other
inlines might do internally.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6305>

util: Make all 3 fetch_rgba functions occupy the same function slot.

A single format either had the float, the sint, or the uint version.
Making the dst be void * lets us store them in the same slot and not have
logic in the callers to call the right one.

-6kb on gallium drivers

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6305>

uitl: Add R1_UNORM to the list of noaccess (no pack/unpack) formats.

The functions were just stubs, if you called them you would be terribly
disappointed.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6305>

egl: drop unused egldriver.h header

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop now empty egldriver.c

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: inline _eglInitializeDisplay() into eglInitialize()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: inline _eglGetDriverProc() into eglGetProcAddress()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused ${drv}_driver()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: replace `&_eglDriver`/`NULL` tested against `NULL` with simple `true`/`false`

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from MesaGLInteropEGL{QueryDeviceInfo,ExportObject}()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: replace replace _EGLDriver with _EGLDisplay->Driver in eglapi.c

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: replace _EGLDriver with _EGLDisplay->Driver in _eglGetSyncAttrib()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from _eglSurfaceAttrib()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver & _EGLDisplay from _eglQueryContext()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDisplay from _eglSetDamageRegionKHRClampRects()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from _eglGetConfigs()/_eglChooseConfig()/_eglGetConfigAttrib()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from SetBlobCacheFuncsANDROID()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from QueryDmaBuf{Formats,Modifiers}EXT()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from ExportDMABUFImage{,Query}MESA()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from QueryBufferAge()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from PostSubBufferNV()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from CreateWaylandBufferFromImageWL()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from {Bind,Unbind,Query}WaylandDisplayWL()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from {Create,Export}DRMImageMESA()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from DupNativeFenceFDANDROID()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from {Create,Destroy,ClientWait,Wait,Signal}SyncKHR()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from {Create,Destroy}ImageKHR()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from GetProcAddress()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver & _EGLDisplay from WaitNative()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from WaitClient()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from SetDamageRegion()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from CopyBuffers()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from SwapBuffers{,WithDamageEXT,RegionNOK}()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from SwapInterval()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from {Bind,Release}TexImage()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from QuerySurface()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from MakeCurrent()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from Create{Window,Pixmap,Pbuffer}Surface() & DestroySurface()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from {Create,Destroy}Context()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from Terminate()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: drop unused _EGLDriver from Initialize()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: replace _EGLDriver with _EGLDisplay->Driver in _eglQuerySurface()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: replace _EGLDriver param with _EGLDisplay->Driver in dri{2_x11,3}_create_window_surface()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

egl: replace _EGLDriver param with _EGLDisplay->Driver in _eglReleaseDisplayResources()

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6241>

spirv/vtn_cfg.c: Include util/debug.h for env_var_as_boolean.

Fixes https://gitlab.freedesktop.org/mesa/mesa/-/issues/3414

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6331>

llvmpipe: only read 0 for channels being read

KHR-GL45.robust_buffer_access_behavior.vertex_buffer_objects
on the CTS 4.6.0 branch and this fixes it.

Roland identified that if the vertex format doesn't contain channels
then we shouldn't be overriding them to 0, so RGB fetch out of bounds
should return 0 for RGB, but the A channel should still be getting back
1.

Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6287>

CI: temp disable t720/t760 jobs.

These are hanging somewhere.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6335>

anv: don't fail userspace relocation with perf queries

This is a left over from the earlier version of
VK_KHR_performance_query where we used kernel relocs to implement
multi passe queries.

We're using self modifying batches now so we shouldn't need any
relocation.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 2001a80d4a81f2 ("anv: Implement VK_KHR_performance_query")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6291>

nv50/ir/nir: fix smem size

Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6325>

clover/nir: fix mem_shared by using address_format_32bit_offset

Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6325>

android: util/format: fix generated sources rules

Changes are necessary after commit 84ed2d098
("util/format: expose generated format packing functions through a header")
because script for format/u_format_pack.h has different commandline
compared to existing pattern

Generated sources rules are ported from meson

Fixes: 84ed2d098 ("util/format: expose generated format packing functions through a header")
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6319>

spirv: Add a MESA_SPIRV_FORCE_UNSTRUCTURED environment variable

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

spirv: parse unstructured CFG

v2 (Boris Brezillon): handle functions with return values
v3: call structurizer
v4: entire rewrite
v5: fix handling of already visited default branches
v2 (Jason Ekstrand): Stop walking hash tables

Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

spirv: extract switch parsing into its own function

v2 (Jason Ekstrand):
 - Construct a list of vtn_case objects

Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

spirv: Move the emit a 'return value' store logic into own function

Right now, only the structured CF path emits 'return value' stores when
an SpvOpReturnValue opcode is found. Move the emit 'return value' logic
in a separate function so we can use it from the unstructured path as well.

v2 (Karol): rephrased and removed unstructured changes

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Add some debug prints

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Add a route::outside set

Acked-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Replace a tripple loop with a double loop

If there's some reason why this needs to be a tripple loop, I'm not
seeing it.  As far as I can tell, all the inner-most loop does is look
for the next remaining block not already in cur_level->blocks.  There's
no reason to re-walk the whole set every time just to do that.

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Add a block_for_singular_set helper

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Sort blocks in select_fork

Hash set ordering is non-deterministic so any time we make a decision
that may affect the final structure or order of instructions, we want to
use a sorted list of blocks.

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Rework some set union logic

I find the ternary a bit hard to read.  The optimization is fairly
obvious but the way it's coded makes things more dense than they
probably need to be.

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Rework handling of skip targets

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>

nir/lower_goto_if: Use util/list instead of exec_list

I'm trying to reduce exec_list usage in NIR.  Also, util_list has some
better helpers for a bunch of the operations this pass needs.

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2401>