radv: regroup fragment shader user SGPRs emission

In preparation to use a packed user SGPR.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23333>

radv: introduce SHIFT/MASK for unpacking shader input args

Loosely based on RadeonSI, looks cleaner and safer to me. I will also
use that to pack fragment shader arguments.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23333>

radv/nir: use ac_nir_unpack_arg() for packed shader input user SGPRS

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23333>

radv/ci: removed expected failures that are skipped now

These tests were failing because CTS wasn't checking a feature bit.
Now that CTS is fixed, these tests are skipped and can't fail at all.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23355>

freedreno/a6xx: Fix memory leak on error path.

Fix defect reported by Coverity Scan.

Resource leak (RESOURCE_LEAK)
leaked_storage: Variable fd6_ctx going out of scope leaks the storage it points to.

Fixes: de3b34df973 ("freedreno: Add a6xx backend")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23289>

panfrost: Use util_pipe_image_to_sampler_view

The common helper is based on what we had locally, with some more cases handled.
Let's use it instead of vendoring.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23273>

gallium: Add util_image_to_sampler_view helper

This helper is useful to implement images. This implementation is based on the
one in Panfrost and extended to handle all pipe_image_views (notably including
tex2d_from_buf which did not exist when the panfrost version was written). It
will be used in both Panfrost and Asahi.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23273>

drm-shim: Avoid assertion fail if someone does close(-1).

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23116>

ci: uprev VVL to 1.3.251

Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23343>

gitlab: prefill MR template with first multiline commit message

Resolve situation for one commit MR, with description inside.

Acked-By: Mike Blumenkrantz <michae
l.blumenkrantz@gmail.com>

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23346>

nir_opt_algebraic: Don't shrink 64-bit bitwise ops if pack_split is going to be lowered

Otherwise this can cause optimizations to fight resulting in infinite
optimization loops with opt_algebraic, constant_folding, and copy_prop.

Fixes: 368be872 ("nir/algebraic: shrink 64-bit bitwise operations with 0/-1 constant half")
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23192>

mesa/st: fix buffer overflow related to set_program_string()

For instance, this is triggered with "piglit/bin/ext_direct_state_access-named-program -auto -fbo":
==5695==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x606000050031 at pc 0x7f78dfca8d46 bp 0x7ffd9043b4a0 sp 0x7ffd9043ac50
READ of size 50 at 0x606000050031 thread T0
    #0 0x7f78dfca8d45  (/usr/lib64/libasan.so.6+0x3fd45)
    #1 0x7f78d450b18f in set_program_string ../src/mesa/main/arbprogram.c:385
    #2 0x7f78d3fdbd3e in execute_list ../src/mesa/main/dlist.c:13025
    #3 0x7f78d40c2564 in _mesa_CallList ../src/mesa/main/dlist.c:13451
    #4 0x7f78d42f380a in _mesa_unmarshal_CallList ../src/mesa/main/glthread_list.c:43
    #5 0x7f78d38e85c5 in glthread_unmarshal_batch ../src/mesa/main/glthread.c:122
    #6 0x7f78d38ea20d in _mesa_glthread_finish ../src/mesa/main/glthread.c:382
    #7 0x7f78d38ea20d in _mesa_glthread_finish ../src/mesa/main/glthread.c:347
    #8 0x7f78d3d73f69 in _mesa_marshal_IsProgramARB src/mapi/glapi/gen/marshal_generated2.c:4256

Fixes: 0b196b40a3ae ("mesa: don't compute the same SHA1 twice in glShaderSource")
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23295>

docs: increase contrast in dark-theme

The contrast-ratio between the foreground and background colors aren't
quite ideal in dark mode. This improves the situation a bit, by making
the body-text a bit brighter.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9043
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23236>

v3d/ci: update traces

Add new traces, remove old ones, and add more information for the
unsupported/crashes.

Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23319>

spirv: Assert sampler_dim is valid when building nir_tex_instr

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14244>

spirv: Validate Dim of OpTypeSampledImage and OpSampledImage

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14244>

intel: Fix support of kernel versions without DRM_I915_QUERY_ENGINE_INFO

As Matt Turner pointed out, the commit here fixed breaks in Iris and
ANV in kernel versions without support for DRM_I915_QUERY_ENGINE_INFO.
As compute engines are only present in gfx12 and newer, and support
for DRM_I915_QUERY_ENGINE_INFO was added before any gfx12 platform,
we can check for gfx version before trying to get engine info.

For ANV, this is done by checking if engine_info is not NULL, like in
other places in the ANV source code.

Fixes: a364f23a6cfa ("intel: Make gen12 URB space reservation dependent on compute engine presence")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9099
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Tested-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23257>

driconf: set vk_x11_strict_image_count for Wolfenstein II

Prevents crash with vsync turned off on xwayland.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9070

Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23326>

hasvk: enable single texel alignment

Signed-off-by: Rohan Garg <rohan.garg@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23321>

anv: fix incorrect asserts when combining CPS and per sample interpolation

CPS is dynamically turned off when per sample interpolation is active.
Update the asserts to reflect this.

Signed-off-by: Rohan Garg <rohan.garg@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 5644011f06 ("intel/compiler: Convert wm_prog_key::persample_interp to a tri-state")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23103>

mesa/st: fix refcnt imbalance related to st_feedback_draw_vbo()

This issue is happening on radeonsi. The reference was allocated
via _mesa_get_bufferobj_reference() with setup_arrays().
The same reference was never freed.

For instance, this issue is triggered on radeonsi with
"piglit/bin/gl-1.0-rendermode-feedback -auto -fbo"
while setting GALLIUM_REFCNT_LOG=refcnt.log.

Fixes: ff8c2a1748b9 ("mesa/bufferobj: rename bufferobj functions to be more consistent.")
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22921>

ci/windows: Update WARP to 1.0.7.1

Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23330>

c11: Remove _MTX_INITIALIZER_NP as it's not used anymore

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

vdpau: Replace usage of mtx_t with simple_mtx_t in htab.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

omx: Replace usage of mtx_t with simple_mtx_t in vid_omx_common.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

nine: Replace usage of mtx_t with simple_mtx_t in nine_lock.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

rtasm: Replace usage of mtx_t with simple_mtx_t in rtasm_execmem.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

rtasm: Trim trailing spaces and replace tab with 3 space

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

xlib: Replace usage of mtx_t with simple_mtx_t in xm_api.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

hud: Replace usage of mtx_t with simple_mtx_t in hud_sensors_temp.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

hud: Replace usage of mtx_t with simple_mtx_t in hud_nic.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

hud: Replace usage of mtx_t with simple_mtx_t in hud_diskstat.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

hud: Replace usage of mtx_t with simple_mtx_t in hud_cpufreq.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

drm: Replace usage of mtx_t with simple_mtx_t in nouveau_drm_winsys.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

drm: Replace usage of mtx_t with simple_mtx_t in drm/radeon_drm_winsys.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

drm: Replace usage of mtx_t with simple_mtx_t in virgl/drm/virgl_drm_winsys.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

drm-shim: Replace usage of mtx_t with simple_mtx_t in drm_shim.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

vc4: Replace usage of mtx_t with simple_mtx_t in vc4/vc4_simulator.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

v3d: Replace usage of mtx_t with simple_mtx_t in v3d_simulator.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

loader: Replace usage of mtx_t with simple_mtx_t in loader/loader_dri3_helper.c

This is a prepare for removing _MTX_INITIALIZER_NP.

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21284>

radv: fix msaa feedback loop without tc-compat cmask

When in an msaa feedback loop and when the image does not have tc-compat
cmask, we have to decompress and expand fmask.  This can happen on gfx9
when sample count > 2 or when RADV_DEBUG=notccompatcmask is specified.

Fixes: a38de4c011d ("radv: disable tc_compatible_cmask on GFX9 in some cases")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23331>

ci: Disable "opencl-external-clang-headers" when "microsoft-clc" is enabled

Signed-off-by: Dor Askayo <dor.askayo@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23255>

meson: add feature option for use of system Clang headers at runtime

Enabling or disabling the "microsoft-clc" option previously changed
shared logic for all compiler/clc users, which was surprising.

In addition, the option to avoid the use of system Clang headers at
runtime is useful outside the scope of Windows.

Separating the two concepts by making this a neutral feature option
addresses both matters.

Signed-off-by: Dor Askayo <dor.askayo@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23255>

aco: cleanup v_cmp_class usage

It's not the best way to check for NaN.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23310>

ci/zink+radv: mark flakes as such

They failed again when I retried, but apparently they are actually
flakes, so mark them as such.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23332>

pvr: Use common vkGetPhysicalDeviceFeatures2() implementation

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23311>

pvr: Use correct pbe format for VK_FORMAT_A8B8G8R8_UNORM_PACK32

Fixes:
  - dEQP-VK.api.image_clearing.core.clear_color_image.1d.linear
      .single_layer.a8b8g8r8_unorm_pack32_71x1
  - dEQP-VK.api.image_clearing.core.clear_color_image.2d.linear
      .single_layer.a8b8g8r8_unorm_pack32_33x128

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reported-by: Soroush Kashani <soroush.kashani@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23312>

pvr: Fix rect splitting logic in pvr_unwind_rects()

Fixes:
  dEQP-VK.api.copy_and_blit.core.image_to_buffer.buffer_offset_relaxed
  dEQP-VK.api.image_clearing.core.clear_color_image.3d.optimal
    .single_layer.r8_unorm_200x180x16

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reported-by: Soroush Kashani <soroush.kashani@imgtec.com>
Reported-by: Oskar Rundgren <oskar.rundgren@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23309>

etnaviv: don't flush implicit flush resources when forced

Resources only need to become visible when the application requested
the context flush, so we can safely skip the flushing when the flush
is forced internally by running out of commandstream space.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23329>

etnaviv: update derived state after forced commandstream flush

If we run out of space in the commandstream while emitting the
current state the drived states won't be recomputed for the
now fully dirty context as the state derivation is called before
any state emitted. Fix this by explicitly updating the derived
state after a forced flush.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8916
CC: mesa-stable
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23329>

etnaviv: nir: improve uniform usage for ALU opc

The current code in lower_alu(..) counts how many const values
are used by one ALU opc. If there are used more then one the
compiler tries to fix this issues by e.g. resolve with a single
combined const src.

We are doing this as some GPUs only allow one const src per
ISA instruction. But it is allowed to use the same const for
multiple srcs.

Lets have a closer look at a real world shader:

impl main {
        /* preds: */
        vec1 32 ssa_0 = load_const (0x3f800000 = 1.000000)
        vec1 32 ssa_1 = load_const (0x00000000 = 0.000000)
        vec4 32 ssa_2 = intrinsic load_uniform (ssa_1) (base=0, range=1, dest_type=bool32 /*38*/)       /* u_var */
        vec1 32 ssa_4 = fmul ssa_2.x, ssa_2.y
        vec1 32 ssa_11 = load_const (0x00000000 = 0.000000)
        vec1 32 ssa_13 = seq ssa_2.w, ssa_11
        vec1 32 ssa_6 = fmul ssa_2.z, ssa_13
        vec1 32 ssa_7 = fmul ssa_4, ssa_6
        vec1 32 ssa_9 = deref_var &gl_FragColor (shader_out vec4)
        vec4 32 ssa_10 = vec4 ssa_7, ssa_7, ssa_7, ssa_0
        intrinsic store_deref (ssa_9, ssa_10) (wrmask=xyzw /*15*/, access=0)
        /* succs: block_1 */
        block block_1:
}

The current compiler transforms it to:

impl main {
        block block_0:
        /* preds: */
        vec1 32 ssa_0 = load_const (0x3f800000 = 1.000000)
        vec4 32 ssa_14 = load_const (0x00000000, 0x00000001, 0x00000002, 0x00000003) = (0.000000, 0.000000, 0.000000, 0.000000)
        vec2 32 ssa_15 = load_const (0x00000000, 0x00000001) = (0.000000, 0.000000)
        vec1 32 ssa_4 = fmul ssa_15.x, ssa_15.y
        vec2 32 ssa_16 = load_const (0x00000003, 0x00000000) = (0.000000, 0.000000)
        vec1 32 ssa_13 = seq ssa_16.x, ssa_16.y
        vec1 32 ssa_6 = fmul ssa_14.z, ssa_13
        vec1 32 ssa_7 = fmul ssa_4, ssa_6
        vec1 32 ssa_9 = deref_var &gl_FragColor (shader_out vec4)
        vec1 32 ssa_17 = mov ssa_0
        vec4 32 ssa_10 = vec4 ssa_7, ssa_7, ssa_7, ssa_17
        intrinsic store_deref (ssa_9, ssa_10) (wrmask=xyzw /*15*/, access=0)
        /* succs: block_1 */
        block block_1:
}

There is no need to create ssa_15 as we can use ssa_14 for the first fmul.

With this change the compiler creates the following shader:

impl main {
        block block_0:
        /* preds: */
        vec1 32 ssa_0 = load_const (0x3f800000 = 1.000000)
        vec4 32 ssa_14 = load_const (0x00000000, 0x00000001, 0x00000002, 0x00000003) = (0.000000, 0.000000, 0.000000, 0.000000)
        vec1 32 ssa_4 = fmul ssa_14.x, ssa_14.y
        vec2 32 ssa_15 = load_const (0x00000003, 0x00000000) = (0.000000, 0.000000)
        vec1 32 ssa_13 = seq ssa_15.x, ssa_15.y
        vec1 32 ssa_6 = fmul ssa_14.z, ssa_13
        vec1 32 ssa_7 = fmul ssa_4, ssa_6
        vec1 32 ssa_9 = deref_var &gl_FragColor (shader_out vec4)
        vec1 32 ssa_16 = mov ssa_0
        vec4 32 ssa_10 = vec4 ssa_7, ssa_7, ssa_7, ssa_16
        intrinsic store_deref (ssa_9, ssa_10) (wrmask=xyzw /*15*/, access=0)
        /* succs: block_1 */
        block block_1:
}

This change reduces immediate pressure and reduces spend CPU cycles.

No piglit or deqp regression seen.

shader-db results for GC2000:

total instructions in shared programs: 955128 -> 955128 (0.00%)
instructions in affected programs: 0 -> 0
helped: 0
HURT: 0

total temps in shared programs: 85689 -> 85689 (0.00%)
temps in affected programs: 0 -> 0
helped: 0
HURT: 0

total immediates in shared programs: 155428 -> 155240 (-0.12%)
immediates in affected programs: 1840 -> 1652 (-10.22%)
helped: 34
HURT: 1
helped stats (abs) min: 4 max: 16 x̄: 5.65 x̃: 4
helped stats (rel) min: 2.94% max: 33.33% x̄: 16.92% x̃: 16.67%
HURT stats (abs)   min: 4 max: 4 x̄: 4.00 x̃: 4
HURT stats (rel)   min: 14.29% max: 14.29% x̄: 14.29% x̃: 14.29%
95% mean confidence interval for immediates value: -6.57 -4.17
95% mean confidence interval for immediates %-change: -19.83% -12.23%
Immediates are helped.

total loops in shared programs: 0 -> 0
loops in affected programs: 0 -> 0
helped: 0
HURT: 0

LOST:   0
GAINED: 0

Total CPU time (seconds): 102.55 -> 96.35 (-6.05%)

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23323>

Fix DGC bug where indirect count > maxSequencesCount.

Need to explicitly clamp the indirect count against maxSequencesCount,
or we risk writing bogus commands into spill region.

Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23252>

aco: remove nir_intrinsic_load_barycentric_at_sample occurences

This is lowered earlier and shouldn't get there.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23307>

radv/meta: fix uninitialised stack memory usage.

==10199== Conditional jump or move depends on uninitialised value(s)
==10199==    at 0xA107B13: radv_resume_queries (radv_meta.c:93)
==10199==    by 0xA108097: radv_meta_restore (radv_meta.c:225)
==10199==  Uninitialised value was created by a stack allocation
==10199==    at 0xA1145B2: fill_buffer_shader (radv_meta_buffer.c:171)

saved_state is never memset, so the value should be inited.

Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23327>

r600/sfn: Initialize BlockScheduler member m_chip_family.

Fix defect reported by Coverity Scan.

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

Fixes: e57643cf548 ("r600/sfn: Add handling for R600 indirect access alias handling")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23290>

v3dv: Update texture padding logic to match v3d changes

Piglit tests for v3d highlighted issues with the padding
computation when allocating memory for slices. This change
moves the fixes from v3d to v3dv.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23260>

v3d: Add support for ASTC texture compression

Add proper support for ASTC texture compression in the v3d
gallium driver, instead of relying on the fallback software
conversion from gallium, as the hardware has native support
for ASTC compressed textures.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23260>

vulkan: use cmd size array for queued cmd allocations

minor simplification for consistency

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23322>

vulkan/cmd_queue: expose cmd sizes

now that cmds are more precisely allocated, it's necessary for drivers
to have some way to determine what the allocation size is

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23322>

vulkan/cmd_queue: allocate cmds based on the size of the cmd

the base size of a vk_cmd_queue_entry is massive since there are a couple
union entries that have a trillion params. by allocating conditionally using
the union member size, memory can be reduced, which will affect some user-facing
api properties

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23322>

vulkan: reorder vk_cmd_queue_entry

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23322>

zink/ci: update expected results

Remove the now passing arb_compute_variable_group_size tests

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23244>

zink: fix layout(local_size_variable) for vk1.3+

Use the correct exec-mode op for LocalSizeId
Corrected typo `gl_LocalGroupSize` -> `gl_LocalGroupSizeARB`

Fixes: 99bd1eaf ("zink: use spir-v 1.6 local-size when needed")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23244>

docs/venus: advertise VK_EXT_image_2d_view_of_3d

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23090>

venus: enable VK_EXT_image_2d_view_of_3d

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23090>

venus: sync protocol for VK_EXT_image_2d_view_of_3d

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23090>

venus: sync to latest protocol from header v1.3.248

This is to make later protocol update CL to be easily backported to
older branches.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23090>

venus: silence -Wuninitialized

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23090>

intel/dev: switch defect identifiers to use lineage numbers

Update existing workarounds when necessary to match changed
identifiers.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23226>

freedreno: Reallocate on unshared export

If we need to export a handle on a resource which was not originally
allocated with PIPE_BIND_SHARED, then re-allocate with shared flag and
try again.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9110
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23324>

freedreno: Add aux-context support

A global aux-context can be created on-demand for cases where we need to
(for example) blit a resource when we only have a screen ptr.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23324>

freedreno: Handle export error handling

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23324>

freedreno/drm: Don't try to export suballoc bo

Suballoc BOs don't have a real handle, so attempting to dmabuf export
won't work.  Likewise for handle or flink-name export.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23324>

nir/print: Print locations for geometry shader inputs

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23259>

intel/blorp: Use nir_trim_vector

With Coccinelle patch:

@@
expression b, V;
@@

-nir_vec2(b, nir_channel(b, V, 0), nir_channel(b, V, 1))
+nir_trim_vector(b, V, 2)

@@
expression b, V;
@@

-nir_vec3(b, nir_channel(b, V, 0), nir_channel(b, V, 1), nir_channel(b, V, 2))
+nir_trim_vector(b, V, 3)

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23259>

radv/query: Use nir_trim_vector

With Coccinelle patch:

@@
expression b, V;
@@

-nir_vec2(b, nir_channel(b, V, 0), nir_channel(b, V, 1))
+nir_trim_vector(b, V, 2)

@@
expression b, V;
@@

-nir_vec3(b, nir_channel(b, V, 0), nir_channel(b, V, 1), nir_channel(b, V, 2))
+nir_trim_vector(b, V, 3)

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23259>

pan/lower_framebuffer: Use nir_replicate

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Italo Nicola <italonicola@collabora.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23259>

treewide: Use nir_replicate

Via coccinelle.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23259>

nir/builder: Add nir_replicate helper

Splat a scalar to all components of a vector.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23259>

radeonsi: return kernel queried video capability for HEVC and JPEG

Query and return the values obtained from kernel for VCN_1 and above.
Earlier the HEVC and JPEG capabilities were returned based on
pre-defined values.

Signed-off-by: Veerabadhran Gopalakrishnan <veerabadhran.gopalakrishnan@amd.com>
Reviewed-by: Thong Thai <thong.thai@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23067>

radeonsi: return kernel queried video capability for HEVC and JPEG

Query and return the values obtained from kernel for VCN_1 and above.
Earlier the HEVC and JPEG capabilities were returned based on
pre-defined values.

Signed-off-by: Veerabadhran Gopalakrishnan <veerabadhran.gopalakrishnan@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23067>

gallivm: Fix gather/scatter types for newer llvm

The types changed with opaque pointer support.

Fixes a bunch of lavapipe regressions.
Cc: mesa-stable
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23293>

docs/calendar: add 23.2 branchpoint and release candidates

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23205>

docs: update calendar for 23.0.4

At this point I'm calling 23.0 done. Please use 23.1 for future updates

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23325>

docs: Add sha256 sum for 23.0.4

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23325>

docs: add release notes for 23.0.4

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23325>

microsoft/compiler: Use image formats to determine texture types

Fixes some tests when bindless is disabled, where the image format is
R32, we do atomics on it, but we didn't set the "typed UAV load with
additional formats" feature bit because when we loaded from it, we
only loaded one component. Since the image format on the DXIL side
was declared as U32x4, the DXIL validator said that we should have.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23266>

spirv2dxil: Assign formats to image vars before lowering to bindless

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23266>

microsoft/compiler: Add a pass to assign image formats based on number of components

For loads/stores without formats, let's guess one based on how it's used.
The actual format doesn't matter, we just want to use it for the number
of components it has.

Also copy image formats from variables to intrinsics, to ensure that
deref-based intrinsics have formats assigned and lowered intrinsics
are up to date.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23266>

microsoft/compiler: Enable emitting type info for textures with <4 comps

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23266>

radv/amdgpu: Report 48-bit VAs in bo logs.

More ergonomic when correlating with page fault addresses.

Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23256>

radv: fix gl_SampleMaskIn for sample shading

When sample shading, we need

  gl_SampleMaskIn = SampleCoverage & (PsIterMask << gl_SampleID);

Add a new shader arg, ps_iter_mask, to pass PsIterMask to ps.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23265>

ac, radeonsi: add and use ac_get_ps_iter_mask

It is more natural for ac_get_ps_iter_mask to take sample count.
Replace samplemask_log_ps_iter by ps_iter_smples in
ac_nir_lower_ps_options accordingly.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23265>

aco: fix alignment check in emit_load

align_offset already takes const_offset into consideration.  Remove the
adjustment.

Fixes: 542733dbbf7 ("aco: add emit_load helper")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9097
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23242>

anv: remove unused functions

I thought I removed those, it seems my rebase got screwed up :(

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 64f20cec28 ("anv: prepare image/buffer views for non indirect descriptors")
Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23317>

anv: add support for VK_EXT_dynamic_rendering_unused_attachments

Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23294>

zink: do not lower line-smooth for non-lines

We used to do this correctly, but it seems this accidentally got
dropped.

Fixes: d80a35a7f76 ("zink: unified `zink_set_primitive_emulation_keys` and `zink_create_primitive_emulation_gs`")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23316>

zink: compute correct location for line-smooth gs

The GS and the FS needs to agree on the driver_location. But we just
used the num_outputs variable for the GS instead of matching the logic
from lower_aaline_instr in nir_draw_helpers.c.

This does that, but cleans up our copy slightly to avoid computing the
needless location, as well as using unsigned values.

This used to *mostly* work before, but only because we were lucky and
not too much crazy stuff went on with the inputs / outputs in
smooth-line cases.

Fixes: edecb66b018 ("nir: avoid generating conflicting output variables")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23316>

zink: keep gl46_optimal extensions/features sorted

These are mostly sorted, which makes it easy to know where to insert new
stuff. Let's make them completely sorted.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23316>

zink: explicitly avoid ci errors due to unrecognized extensions in VVL

lavapipe still had this from shader object, but this should always be in
place for all drivers so that old VVL doesn't block merges

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23315>