swr/rast: Make SIMDLib templated types easier to use

"typename SIMD_T::TypeName" --> "TypeName<SIMD_T>"

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Be more explicit when fetching next component

Use a new function to denote that we want to get offset to next component
and hide the fact that GEP is used underneath.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Fix bug related to passing AR handle

We were passing a garbage handle. Let's not do that.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Fix primitive replication issue in tesselation PA.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Use llvm intrinsic masked gather

Use llvm intrinsic masked.gather instead of manual unroll for the cases
where we have vector of pointers.  Improves llvm IR debug experience by
reducing a ton of IR to a single intrinsic call. Also seems to reduce
overall stack use considerably.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Misc cleanup

Together with correct detection of clipDistance NaNs when no cullDistance is set

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Renamed variable in vertexbufferstate

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Fix GATHERPS to avoid assertions.

With the pBase type change, LLVM was asserting because of wrong types.
Cast appropriately.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: More precise user clip distance interpolation

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Cull prims when all verts have negative clip distances

Performance optimization, and fixes some clipping issues.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: whitespace and comment cleanup

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Fix invalid number of attributes

Fix invalid number of attributes passed into tesselation PA.
Needs to take into account any offsets from the shader.
Innocuous issue, but removes an assert firing in debug.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Add clipper stats.

Clipper event is now:

event ClipperEvent
{
    uint32_t drawId;
    uint32_t trivialRejectCount;
    uint32_t trivialAcceptCount;
    uint32_t mustClipCount;
};

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Separate event types to public and private

Split into two proto files and modify appropriate build rules for
configure / scons / meson builds.

There are private internal events (proxy) that communicate information
from rasterizer to ArchRast. ArchRast can use these events to calculate
a final answer and then emit other public events which will be saved to
file. Users will use the public proto file and not the private one.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Clean up event types and remove BE events

Begin/End events not needed anymore.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Removed unused variable

Gets rid of zillions of unused variable warnings, made worse by templates.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Separate RDTSC code from archrast

Renamed rdstc defines more appropriately

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Cleanup of mpPrivateContext in Builder

Provide access functions for mpPrivateContext in Builder.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Remove some JIT debug code

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Don't include private context in gather args

Move mpPrivateContext to compensate

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Cleanup knob definitions

Rename some of the categories and move some options around.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Add missing parameter to a few gather functions

We now pass pDrawContext as a default parameter

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

etnaviv: add useful information to BO import errors

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

egl/wayland: Always use in-tree wayland-egl-backend.h

A recent patchset to Wayland[0] migrated Mesa's libwayland-egl backend
into Wayland itself, so implementations could provide backends. Mesa
still uses its own, and the two have already diverged[1].

The include from egl_dri2.h could pick up either the installed Wayland
wayland-egl-backend.h (with a 'driver_private' member), or the Mesa
internal wayland-egl-backend.h (with a 'private' member), failing the
build in the first instance.

Add an explicit directory prefix to the include, so we always get our
in-tree version.

[0]: https://patchwork.freedesktop.org/series/31663/
[1]: https://cgit.freedesktop.org/wayland/wayland/commit/?id=9fa60983b579

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105103
Fixes: 198af27c679c ("wayland-egl: rename wayland-egl-{priv,backend}.h")

meson: Move Wayland dmabuf to wayland-drm

As the comment notes: linux-dmabuf has nothing to do with wayland-drm,
but we need a single place to build these files we can use from both EGL
and Vulkan, which is guaranteed to be included before both EGL and
Vulkan WSI.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.co.uk>

egl/wayland: check for invalid format index

v2: just tell the compiler to assume the format will always be found, as
it comes from the table itself to begin with. (DanielS)

CID: 1429516
Fixes: d32b23f3830099a328b91 "egl/wayland: Add bpp to visual map"
Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

glsl: fix sizeof(pointer) bug

Doesn't really change anything to the test though ¯\_(ツ)_/¯

CID: 1429511
Fixes: e8495646afb06a9dd7786 "glsl/tests: changes to test_disk_cache_create test"
Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>

radeonsi/nir: set TGSI_PROPERTY_FS_EARLY_DEPTH_STENCIL correctly

We set this for post_depth_coverage in addition to early_fragment_tests.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

virgl: remap query types to hw support.

The gallium query types changed, so we need to remap from the
gallium ones to the virgl ones.

Fixes:
dEQP-GLES3.functional.transform_feedback.basic_types*

"This also fixes:

dEQP-GLES3.functional.transform_feedback.array.separate*
dEQP-GLES3.functional.transform_feedback.array_element*
dEQP-GLES3.functional.transform_feedback.interpolation.*

Gallium's p_defines.h and virglrenderer's p_defines.h have diverged
quite a bit, so not including
PIPE_QUERY_OCCLUSION_PREDICATE_CONSERVATIVE there makes sense for now."
 - Gurchetan Singh

Fixes: 3f6b3d9db (gallium: add PIPE_QUERY_OCCLUSION_PREDICATE_CONSERVATIVE)
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>

i965/icl: Add render target flush after uploading binding table

From PIPE_CONTROL command description in gfxspecs:

"Whenever a Binding Table Index (BTI) used by a Render Taget Message
 points to a different RENDER_SURFACE_STATE, SW must issue a Render
 Target Cache Flush by enabling this bit. When render target flush
 is set due to new association of BTI, PS Scoreboard Stall bit must
 be set in this packet."

V2: Move the PIPE_CONTROL to update_renderbuffer_surfaces() in
    brw_wm_surface_state.c (Ken).

Fixes a fulsim error and a GPU hang described in below JIRA.
JIRA: MD5-322
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/icl: Enable float blend optimization and Wa3DStateMode

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

intel/common/icl: Add has_sample_with_hiz flag in gen_device_info

Sampling from hiz is enabled in i965 for GEN9+ but this feature has
been removed from gen11. So, this new flag will be useful to turn
the feature on/off for different gen h/w. It will be used later
in a patch adding device info for gen11.

Suggested-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/icl: Add assertions to check dispatch mode is SIMD8

SIMD4x2 dispatch mode has been removed in GEN11. We're not using
it anyways in Mesa. Adding few asserts to make it explicit.

Use GEN_GEN macro in place of devinfo->gen (Ken)

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/icl: Update switch statements

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/icl: Update the assert in brw_memory_barrier()

Nothing is changed here from gen10 to gen11. So, just update
the assert.

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/icl: Define and use icl mocs settings

Gen11 MOCS settings are duplicate of Gen10 MOCS settings.

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/icl: Update the comment for maximum number of threads per PSD

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/icl: Build and use gen11 functions for genxml state-upload and blorp

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>

i965/icl: Don't set ResetGatewayTimer

This field is removed in gen11+

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

intel/icl: Do StateCacheInvalidation for indirect clear color

StateCacheInvalidation is required on all gen7+ platforms. We
don't need to update this check for every new gen h/w unless
this requirement is changed. So, dropping the check for latest
gen h/w.

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

intel/isl/icl: Build and use gen11 surface state emit functions

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>

intel/isl/icl: Add the maximum surface size limit

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

intel/genxml/icl: Update genx_bits header

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

intel/genxml/icl: Generate packing headers

Move build system changes in to one patch (Ken, Emil)

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>

intel/genxml/icl: Add gen11.xml

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

i965: Drop EXEC_OBJECT_CAPTURE defines.

These only existed to avoid making people update libdrm for new uABI
headers.  A while ago we imported those headers into the Mesa repo,
so the dependency is gone and these are no longer useful.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

clover: Fix build after llvm r325155 and r325160

r325155 ("Pass a reference to a module to the bitcode writer.")
and
r325160 ("Pass module reference to CloneModule")

change function interface from pointer to reference.

v2: Fix indentation (tab instead of spaces)

Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>

radv: Always lower indirect derefs after nir_lower_global_vars_to_local.

Otherwise new local variables can cause hangs on vega.

CC: <mesa-stable@lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105098
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

meson: fix xvmc target linkage

This needs to link the state tracker with --whole-archive to expose the
right symbols.

v4: - Always add libswdri and libswkmsdri to the link_with list

Fixes: 22a817af8a89eb3c7 ("meson: build gallium xvmc state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: Fix xa target linkage

This needs to use --whole-archive (link_whole in meson) to properly
expose symbols.

v4: - Always add libswdri and libswkmsdri to link_with list

Fixes: 0ba909f0f111824 ("meson: build gallium xa state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: Fix omx-bellagio target linkage

This needs to use --whole-archive (link_whole in meson) to properly
expose symbols.

v4: - Always add libswdri and libswkmsdri to link_with

Fixes: 1d36dc674d528b93b ("meson: build gallium omx state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: fix va target linkage

The state tracker needs to be linked with whole-archive (like
autotools). As a result there are symbols from libswdri and libswkmsdri
that are needed, so link those as well.

v4: - Always add libswdri and libswkmsdri to link_with list

Fixes: 5a785d51a6d6 ("meson: build gallium va state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: fix vdpau target linkage

The VDPAU state tracker needs to be linked with whole-archive (autotools
does this). Because we are linking the whole archive we alos need to
link with libswdri and libswkmsdri if those have been enabled.

v4: - Always add libswdri and libswkmsdri to link_with list

Fixes: 68076b87474e7959 ("meson: build gallium vdpau state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: Actually link xvmc target with libxvmc

Unlike vdpau this is required.

Fixes: 22a817af8a89eb3c7 ("meson: build gallium xvmc state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: actually link with libomxil-bellagio

This state tracker actually needs to link, unlike vdpau.

Fixes: 1d36dc674d528b93b ("meson: build gallium omx state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: link dri3 xcb libs into vlwinsys instead of into each target

This makes the dependencies easier to manage, since each media target
doesn't need to worry about linking to half a dozen libraries.

Fixes: b1b65397d0c4978e3 ("meson: Build gallium auxiliary")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: use va-api version reported by pkg-config

Fixes: 5a785d51a6d6 ("meson: build gallium va state tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: add libswdri and libswkmsdri to dri link_with

Fixes: b154b44ae342 ("meson: build radeonsi gallium driver")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: add libswdri and libswkmsdri to d3dadaptor link_with

v5: - Fix libswdi -> libswdri typo

Fixes: 6b4c7047d571 ("meson: build gallium nine state_tracker")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: define empty variables for libswdri and libswkmsdri

This allows these variables to unconditionally included in `link_with`
lists, even if they're not used. This allows deleting duplicated logic
in nearly every gallium target implemented in meson today. This also
removes the now useless `build_by_default` flag from swdri and swkmsdri.

v4: - add this patch

Fixes: 66c94b9313a697ce8f2b222f4ba353035e4b8726
       ("meson: build gallium winsys for dri, null, and wrapper")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

meson: add convenience variable for anv_extensions.py depdendency

Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>

meson: use depend_files for adding extra file dependencies

cc: Jason Ekstrand <jason.ekstrand@intel.com>
Fixes: dd088d4bec74f37ffe4 ("anv/extensions: Generate a header file with extension tables")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>

meson: use depend_files to track extra file dependencies

cc: Jason Ekstrand <jason.ekstrand@intel.com>
Fixes: f93994080993bda ("anv: Split anv_extensions.py into two files")
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>

Revert "anv/meson: Make anv_entrypoints_gen.py depend on anv_extensions.py"

This reverts commit 10d1b0be8e9c463dbc35cd66968299f33c76672c.

This is unnecessary, the depend_files argument is for adding
dependencies on files that are not part of the input, which is already
done.

cc: Jason Ekstrand <jason.ekstrand@intel.com>
Fixes: 10d1b0be8e9c463dbc35cd66968299f33c76672c
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>

svga: replace gotos with else clauses

Simple clean-up.

Reviewed-by: Charmaine Lee <charmainel@vmware.com>

svga: s/unsigned/enum pipe_shader_type/

Reviewed-by: Neha Bhende <bhenden@vmware.com>

svga: move duplicated code for setting fillmode/flatshade state

Move the calls to svga_hwtnl_set_fillmode() and svga_hwtnl_set_flatshade()
out of the two retry_draw_*() functions to the svga_draw_vbo() function.

Reviewed-by: Charmaine Lee <charmainel@vmware.com>

svga: move svga_update_state() call in draw code

This fixes a few Piglit transform feedback regressions caused by
commit 7a1401938b351.

In that change I moved the moved svga_update_state() into the loops,
after the calls to svga_hwtnl_set_flatshade().  But
svga_hwtnl_set_flatshade() actually depends on some derived shader
state.  This patch moves the svga_update_state() call into
svga_draw_vbo() so it's not duplicated in two places.

Fixes: 7a1401938b351 ("svga: clean up retry_draw_range_elements(),
retry_draw_arrays()")

Reviewed-by: Charmaine Lee <charmainel@vmware.com>

svga: call tgsi_scan_shader() for dummy shaders

If we fail to compile the normal VS or FS we fall back to a simple/
dummy shader.  We need to rescan the the shader to update the shader
info.  Otherwise, this can lead to further translations failures
because the shader info doesn't match the actual shader.

Found by adding some extra debug assertions in the state-update code
while debugging something else.

v2: also update shader generic_inputs/outputs, etc. per Charmaine

Reviewed-by: Charmaine Lee <charmainel@vmware.com>

ac/nir: do not reserve user SGPRs for unused descriptor sets

In theory this might lead to corruption if we bind a descriptor
set which is unused, because LLVM is smart and it can re-use
unused user SGPRs. In practice, this doesn't seem to fix
anything.

As a side effect, this will reduce the number of emitted
SH_REG packets.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

ac/shader: fix gathering of desc_set_used_mask

This was quite wrong.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

ac/shader: be a little smarter when scanning vertex buffers

Although meta shaders don't use any vertex buffers, there is no
behaviour change but I think it's better to do this. Though,
this saves two user SGPRs for push constants inlining or
something else.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

dri: fromPlanar() can return NULL as a valid result

It was assumed that fromPlanar() could return NULL to mean
that the planar image is the same as the parent DRI image.
That assumption wasn't made everywhere though.

Let's fix things and make sure that all callers understand
a NULL result

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Daniel Stone <daniels@collabora.com>

docs: correct link to the 17.3.3 release notes

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

docs: update calendar, add news and link release notes to 17.3.4

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

docs: add sha256 checksums for 17.3.4

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
(cherry picked from commit 26c84b8af9ebdedd2df75596c25173639ed5e4b2)

docs: add release notes for 17.3.4

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
(cherry picked from commit 2f9820c553c943f776513ddc2887fb6b1cf0f29a)

nvc0: disable MS Images for sample_count == 1 on Maxwell

fixes KHR-GL45.multi_bind.dispatch_bind_textures on Maxwell

Suggested-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>

mesa: don't clamp just based on ARB_viewport_array extension

The ARB_viewport_array spec says:

"Dependencies
    OpenGL 1.0 is required.

    OpenGL 3.2 or the EXT_geometry_shader4 or ARB_geometry_shader4 extensions
    are required.

    This extension is written against the OpenGL 3.2 (Compatibility)
    Specification."

As such, we should ignore it for GLES2 contexts.

Fixes:
dEQP-GLES2.functional.state_query.integers.viewport_getinteger
dEQP-GLES2.functional.state_query.integers.viewport_getfloat

on llvmpipe and virgl.

v2: Use _mesa_has_* (Ilia)

Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Cc: 17.3 18.0 <mesa-stable@lists.freedesktop.org>

meson: use a custom target instead of a generator for i965 oa

Generators really are never the thing you want. The problem in this case
is that a generator must create a file that contains any file that the
generated target depends on. Since brw_oa.py doesn't generate such a
file the generated sources are not regenerated even if the xml files
they should depend on changes.

While we could change brw_oa.py to write such a file, that's silly, it
depends on itself and the xml file. So we'll just use a custom target
instead, which will have the correct dependency behavior and doesn't
really add that much code.

Fixes: 3218056e0eb3 ("meson: Build i965 and dri stack")
CC: Ian Romanick <idr@freedesktop.org>
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

isl: Don't use surface format R32_FLOAT for typed atomic integer operations

From Skylake PRM Surface Formats section:

   "The surface format for the typed atomic integer operations must
    be R32_UINT or R32_SINT."

Fixes an error and a piglit GPU hang in simulation environment.
Piglit test: gl45-imageAtomicExchange-float.shader_test

Suggested-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.co
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "18.0 17.3" <mesa-stable@lists.freedesktop.org>

radeonsi/nir: fix si_nir_load_tcs_varyings() for outputs

We were incorrectly using the input info for outputs.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

ac: implement nir_intrinsic_image_samples

Fixes cts test:
KHR-GL45.shader_texture_image_samples_tests.image_functional_test

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

st: add NIR GL_ARB_get_program_binary support

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

st/shader_cache: add st_{de}serialise_nir_program() helpers

These will be used for NIR GL_ARB_get_program_binary support.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

ac/nir_to_llvm: fix image size for arrays of arrays

Fixes cts test:
KHR-GL44.shader_image_size.advanced-changeSize

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

radeonsi/nir: fix shader ballot return value bitsize

Fixes cts test:
KHR-GL46.shader_ballot_tests.ShaderBallotFunctionBallot

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

intel/aubinator: Correctly decode INTERFACE_DESCRIPTOR_DATA

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965: Add gl_state_index casts for PATCH_VERTICES_IN

This fixes the build in clang

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105088
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/miptree: Initialize mcs with a linear map

When initializing mcs, map with MAP_RAW and fill in the linear
map. Removes a place where gtt mapping is used.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/tiled_memcpy: change linear pointer from (0, 0) to (xt1, yt1)

In all current uses, the linear surface is only allocated starting
at (xt1, yt1) anyway, so this improves the calling ergonomics.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/tiled_memcpy: linear_to_ytiled a cache line at a time

TileY's low 6 address bits are: v1 v0 u3 u2 u1 u0
Thus a cache line in the tiled surface is composed of a 2d area of
16x4 bytes of the linear surface.

Add a special case where the area being copied is 4-line aligned
and a multiple of 4-lines so that entire cache lines will be
written at a time.

On Apollolake, this increases tiling throughput to wc maps by
84.0103% +/- 0.862818%

v2: Split [y0, y1) and [y2, y3) loops apart for clarity (Jason Ekstrand)
v3: Don't reset src var (Jason), Ensure y0 <= y1 <= y2 <= y3

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

docs: Add Cannonlake support to 18.0 release notes.

17.4 is actually 18.0.

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Cc: "18.0" mesa-stable@lists.freedesktop.org
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

anv/gen10: Remove warning message.

Gen10 seems pretty stable so far, remove "alpha support" message.

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Cc: "18.0" mesa-stable@lists.freedesktop.org
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

i965/gen10: Remove warning message.

Gen10 seems pretty stable so far, so there's no reason to keep this
message.

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Cc: "18.0" mesa-stable@lists.freedesktop.org
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

egl/x11: Fix leak in dri3_create_image_khr_pixmap

bp_reply wasn't properly free'd

Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

i965/compiler: clean up nir_intrinsic_load_input for vertex shaders

This code to re-set the type of the source and destination is not
necessary since we never manipulate the types. Looks like a
left over from a time where we had to retype to float temporarily
to handle 64-bit inputs.

Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>

intel/compiler: fix first_component for 64-bit types on vertex inputs

Divide it by two as we do for other stages. This is because the
component layout qualifier is always in 32-bit units.

Fixes issues in a new CTS test (still WIP):
KHR-GL45.enhanced_layouts.varying_double_components

Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>

ac/nir: rename nir_to_llvm_context to radv_shader_context

There is still more to do in that area, but it's a good start.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

ac: remove nir_to_llvm_context from ac_nir_translate()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>