pan/bi: Add support for if-else blocks

Branch lowering code lifted from Midgard as usual.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

pan/bi: Call nir_lower_io_to_temporaries in cmdline

Normally mesa/st would do this for us, but we're using the standalone
compiler (in advance of having the hardware) and need this pass
particularly for fragment writeout.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

pan/bi: Add instruction emit/remove helpers

As we start descending into code generation these will be of interest.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

pan/bi: Print branch target

...if it's present, anyway.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

pan/bi: Don't print types for unconditional branches

There's nothing to type!

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

pan/bi: Improve block printing

Skip predecessor printing if there are none and match a missing brace,
also fixup the spacing.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

pan/bi: Walk through the NIR control flow graph

Copypaste from Midgard with some cleanups. That seems to be a trend
these days. Hopefully boilerplate will come to a close soon.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

pan/bi: Lower and optimize NIR

Pretty much a copypaste from Midgard except where architectural
decisions diverge around vectorization. On that note, we will need our
own ALU scalarization pass at some point (or rather we'll need to extend
nir_lower_alu_scalar) to allow partial lowering for 8/16-bit ops. I.e.
we'll approximately need to lower

   vec4 16 ssa_2 = fadd ssa_0, ssa_1

to

   vec2 16 ssa_2 = fadd ssa_0.xy, ssa_1.xy
   vec2 16 ssa_3 = fadd ssa_0.zw, ssa_1.zw
   vec4 16 ssa_4 = vec4 ssa_2.x, ssa_2.y, ssa_3.x, ssa_4

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4097>

anv: Flatten the logic add_aux_surface_if_supported (v3)

Reduces the function's max indentation level from 5 to 3 inside the big
'if' tree. And enables more comments to be attached to the condition
they describe.

v2:
  - Add missing DEBUG_NO_RBC check.
v3:
  - Return early on DISABLE_AUX_BIT.
  - Restore original order of gen7 hiz check.

Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4096>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4096>

anv: Refactor creation of aux surfaces (v2)

make_surface() contained a giant if-tree for creation of aux surfaces.
Move the if-tree into its own function, add_aux_surface_if_supported().
This will simplify future changes for VK_EXT_image_drm_format_modifier.

This patch merely moves the code verbatim, then extracts duplicate
assertions to the top.

v2: Rename func to add_aux_surface_if_supported [for jekstrand].

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4096>

anv: Add anv_image_plane_needs_shadow_surface() (v2)

The function returns true if hardware limitations require the image
plane to use a shadow surface. It replaces equivalent code in
make_surface().

Refactor only. No intended change in behavior.

Why extract this code out of vkCreateImage? If an image requires
a shadow surface, then that may impact its support for DRM format
modifiers, which must be evaluated during
vkGetPhysicalDeviceImageFormatProperties2.

v2:
  - Use early return. [for jekstrand]
  - Unexport function.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4096>

glsl: add subroutine support to nir linker

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: dont try to assign uniform storage for uniform blocks

Fixes a crash in some shaders.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: add support for builtins to the nir uniform linker

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: set ShaderStorageBlocksWriteAccess in the nir linker

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: nir linker fix setting of ssbo top level array

This helps correcly set it for each top level member and correctly
handle unsized arrays.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: find the base offset for block members from unnamed blocks

These block member have been split into individual vars so we need
to set the correct offsets for each member in the new glsl nir
linker. We also take this opportunity to set the correct location
for the variable.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: correctly set explicit offsets for struct members

This correctly sets offsets set in glsl when using the nir linker.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: add std140 and std430 layouts to nir uniform linker

The current ARB_gl_spirv linking only supports explicit layouts so
we need to update it to support std140 and std430 layouts before
we can use the linker for glsl.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

nir: add glsl_get_std430_size() helper

This will be used by the nir glsl linker for linking uniforms.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

nir: add glsl_get_std430_base_alignment() helper

This will be used by the nir glsl linker for linking uniforms.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

nir: add glsl_get_std140_size() helper

This will be used by the nir glsl linker for linking uniforms.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

nir: add glsl_get_std140_base_alignment() helper

This will be used by the nir glsl linker for linking uniforms.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

nir: add glsl_get_internal_ifc_packing() helper

This will be used by the nir glsl linker for linking uniforms.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: correctly find block index when linking glsl with nir linker

The existing code for spirv expects all vars to have explicit
bindings set which is not true for glsl.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: add name support to nir uniform linker

Name support is optional for spirv support but is required for glsl
support.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: move get_next_index() earlier in nir link uniforms

We will use get_next_index() in more of the helper functions in
the following patches.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: move add_parameter() earlier in nir link uniforms

We will use add_parameter() in more of the helper functions in
the following patches.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

glsl: move nir link uniforms struct defs earlier

We will need to use the state in more of the helper functions in
the following patches.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4050>

lima: gpir: enforce instruction limit earlier

Enforce instruction limit of 512 instructions earlier. This is a
workaround for infinite loops in gpir compiler and allows us to
pin point the tests that are affected.

Reviewed-by: Andreas Baierl <ichgeh@imkreisrum.de>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4055>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4055>

intel/compiler: Calculate num_instructions in O(1) during register pressure calculation

And mark the variable declaration as const.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Move register pressure calculation into IR analysis object

This defines a new BRW_ANALYSIS object which wraps the register
pressure computation code along with its result.  For the rationale
see the previous commits converting the liveness and dominance
analysis passes to the IR analysis framework.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

entel/compiler: Simplify new_idom reduction in dominance tree calculation

Trivial, just use a few less tokens to do the same thing.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Move dominance tree data structure into idom_tree object

It makes sense to keep the result of analysis passes independent from
the IR itself.  Instead of representing the idom tree as a pointer in
each basic block pointing to its immediate dominator, the whole
dominator tree is represented separately from the IR as an array of
pointers inside the idom_tree object.  This has the advantage that
it's no longer possible to use stale dominance results by accident
without having called require() beforehand, which makes sure that the
idom tree is recalculated if necessary.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Move idom tree calculation and related logic into analysis object

This only does half of the work.  The actual representation of the
idom tree is left untouched, but the computation algorithm is moved
into a separate analysis result class wrapped in a BRW_ANALYSIS
object, along with the intersect() and dump_domtree() auxiliary
functions in order to keep things tidy.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Drop invalidate_live_intervals()

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler/vec4: Switch liveness analysis to IR analysis framework

This involves wrapping vec4_live_variables in a BRW_ANALYSIS object
and hooking it up to invalidate_analysis() so it's properly
invalidated.  Seems like a lot of churn but it's fairly
straightforward.  The vec4_visitor invalidate_ and
calculate_live_intervals() methods are no longer necessary after this
change.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler/fs: Switch liveness analysis to IR analysis framework

This involves wrapping fs_live_variables in a BRW_ANALYSIS object and
hooking it up to invalidate_analysis() so it's properly invalidated.
Seems like a lot of churn but it's fairly straightforward.  The
fs_visitor invalidate_ and calculate_live_intervals() methods are no
longer necessary after this change.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler/vec4: Add live interval validation pass

This could be improved somewhat with additional validation of the
calculated live in/out sets and by checking that the calculated live
intervals are minimal (which isn't strictly necessary to guarantee the
correctness of the program).  This should be good enough though to
catch accidental use of stale liveness results due to missing or
incorrect analysis invalidation.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler/fs: Add live interval validation pass

This could be improved somewhat with additional validation of the
calculated live in/out sets and by checking that the calculated live
intervals are minimal (which isn't strictly necessary to guarantee the
correctness of the program).  This should be good enough though to
catch accidental use of stale liveness results due to missing or
incorrect analysis invalidation.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Pass single backend_shader argument to the vec4_live_variables constructor

The IR analysis framework requires the analysis result to be
constructible with a single argument.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Pass single backend_shader argument to the fs_live_variables constructor

This removes the dependency of fs_live_variables on fs_visitor.  The
IR analysis framework requires the analysis result to be constructible
with a single argument -- The second argument was redundant anyway.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Restructure live intervals computation code

This makes the structure of the vec4 live intervals calculation more
similar to the FS back-end liveness analysis code.  The non-CF-aware
start/end computation is moved into the same pass that calculates the
block-local def/use sets, which saves quite a bit of code, while the
CF-aware start/end computation is moved into a separate
compute_start_end() function as is done in the FS back-end.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Move all live interval analysis results into vec4_live_variables

This moves the following methods that are currently defined in
vec4_visitor (even though they are side products of the liveness
analysis computation) and are already implemented in
brw_vec4_live_variables.cpp:

> int var_range_start(unsigned v, unsigned n) const;
> int var_range_end(unsigned v, unsigned n) const;
> bool virtual_grf_interferes(int a, int b) const;
> int *virtual_grf_start;
> int *virtual_grf_end;

It makes sense for them to be part of the vec4_live_variables object,
because they have the same lifetime as other liveness analysis results
and because this will allow some extra validation to happen wherever
they are accessed in order to make sure that we only ever use
up-to-date liveness analysis results.

The naming of the virtual_grf_start/end arrays was rather misleading,
they were indexed by variable rather than by vgrf, this renames them
start/end to match the FS liveness analysis pass.  The churn in the
definition of var_range_start/end is just in order to avoid a
collision between the start/end arrays and local variables declared
with the same name.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Move all live interval analysis results into fs_live_variables

This moves the following methods that are currently defined in
fs_visitor (even though they are side products of the liveness
analysis computation) and are already implemented in
brw_fs_live_variables.cpp:

> bool virtual_grf_interferes(int a, int b) const;
> int *virtual_grf_start;
> int *virtual_grf_end;

It makes sense for them to be part of the fs_live_variables object,
because they have the same lifetime as other liveness analysis results
and because this will allow some extra validation to happen wherever
they are accessed in order to make sure that we only ever use
up-to-date liveness analysis results.

This shortens the virtual_grf prefix in order to compensate for the
slightly increased lexical overhead from the live_intervals pointer
dereference.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Mark virtual_grf_interferes and vars_interfere as const

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Pass detailed dependency classes to invalidate_analysis()

Have fun reading through the whole back-end optimizer to verify
whether I've missed any dependency flags -- Or alternatively, just
trust that any mistake here will trigger an assertion failure during
analysis pass validation if it ever poses a problem for the
consistency of any of the analysis passes managed by the framework.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Define more detailed analysis dependency classes

I've deliberately separated this from the general analysis pass
infrastructure in order to discuss it independently.  The dependency
classes defined here refer to state changes of several objects of the
program IR, and are fully orthogonal and expected to change less often
than the set of analysis passes present in the compiler back-end.

The objective is to avoid unnecessary coupling between optimization
and analysis passes in the back-end.  By doing things in this way the
set of flags to be passed to invalidate_analysis() can be determined
from knowledge of a single optimization pass and a small set of well
specified dependency classes alone -- IOW there is no need to audit
all analysis passes to find out which ones might be affected by
certain kind of program transformation performed by an optimization
pass, as well as the converse, there is no need to audit all
optimization passes when writing a new analysis pass to find out which
ones can potentially invalidate the result of the analysis.

The set of dependency classes defined here is rather conservative and
mainly based on the requirements of the few analysis passes already
part of the back-end.  I've also used them without difficulty with a
few additional analysis passes I've written but haven't yet sent for
review.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Introduce backend_shader method to propagate IR changes to analysis passes

The invalidate_analysis() method knows what analysis passes there are
in the back-end and calls their invalidate() method to report changes
in the IR.  For the moment it just calls invalidate_live_intervals()
(which will eventually be fully replaced by this function) if anything
changed.

This makes all optimization passes invalidate DEPENDENCY_EVERYTHING,
which is clearly far from ideal -- The dependency classes passed to
invalidate_analysis() will be refined in a future commit.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Introduce simple IR analysis pass framework

Motivated in detail in the source code.  The only piece missing here
from the analysis pass infrastructure is some sort of mechanism to
broadcast changes in the IR to all existing analysis passes, which
will be addressed by a future commit.  The analysis_dependency_class
enum might seem a bit silly at this point, more interesting dependency
categories will be defined later on.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Reverse inclusion dependency between brw_vec4_live_variables.h and brw_vec4.h

brw_vec4.h (in particular vec4_visitor) is logically a user of the
live variables analysis pass, not the other way around.
brw_vec4_live_variables.h requires the definition of some VEC4 IR data
structures to compile, but those can be obtained directly from
brw_ir_vec4.h without including brw_vec4.h.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Reverse inclusion dependency between brw_fs_live_variables.h and brw_fs.h

brw_fs.h (in particular fs_visitor) is logically a user of the live
variables analysis pass, not the other way around.
brw_fs_live_variables.h requires the definition of some FS IR data
structures to compile, but those can be obtained directly from
brw_ir_fs.h without including brw_fs.h.  The dependency of
fs_live_variables on fs_visitor is rather accidental and will be
removed in a future commit, a forward declaration is enough for the
moment.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Nest definition of live variables block_data structures

When this commit was originally written, these two structures had the
exact same name. Subsequently in commit 12a8f2616a2f (intel/compiler:
Fix C++ one definition rule violations) they were renamed.

Original commit message:

> These two structures have exactly the same name which prevents the two
> files from being included at the same time and could cause serious
> trouble in the future if it ever leads to a (silent) violation of the
> C++ one definition rule.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Reverse inclusion dependency between brw_cfg.h and brw_shader.h

This reflects the natural dependency relationship between brw_cfg.h
and brw_shader.h.  brw_cfg.h only requires the base IR definitions
which are now part of a separate header.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

intel/compiler: Move base IR definitions into a separate header file

This pulls out the i965 IR definitions into a separate file and leaves
the top-level backend_shader structure and back-end compiler entry
points in brw_shader.h.  The purpose is to keep things tidy and
prevent a nasty circular dependency between brw_cfg.h and
brw_shader.h.  The logical dependency between these data structures
looks like:

 backend_shader (brw_shader.h) -> cfg_t (brw_cfg.h)
    -> bblock_t (brw_cfg.h) -> backend_instruction (brw_shader.h)

This circular header dependency is currently resolved by using forward
declarations of cfg_t/bblock_t in brw_shader.h and having brw_cfg.h
include brw_shader.h, which seems backwards and won't work at all when
the forward declarations of cfg_t/bblock_t are no longer sufficient in
a future commit.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4012>

etnaviv: add etna_constbuf_state object

With this new state object we keep track of enabled pipe_constant_buffer
and only mark them as read when needed.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4088>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4088>

st/va: add check for P010 and P016 encode/decode support

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4033>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4033>

radeon: add support for 10-bit HEVC encoding to VCN 2.0

Signed-off-by: Thong Thai <thong.thai@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4033>

radeonsi: add 10-bit HEVC encode support for VCN2.0 devices

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4033>

nir/linker: remove reference to just SPIR-V linking

Several files had a initial comment about the purpose of such files,
including a reference that the NIR linker was implemented with just
ARB_gl_spirv in mind.

Since the nice job Timothy is doing to use the NIR linker on GLSL,
that is not true anymore, so let's remove that reference and also
tweak some other comments.

Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4081>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4081>

bin/gen_release_notes.py: fix commit list command

Fixes: 86079447da1e00d49db0 ("scripts: Add a gen_release_notes.py script")
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4069>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4069>

docs: fix typos in the release docs

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Andres Gomez <agomez@igalia.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4067>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4067>

radeonsi: remove AMD_DEBUG=sisched option

sisched is not maintained anymore in LLVM.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4059>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4059>

nir/lower_input_attachments: remove bogus assert in try_lower_input_texop()

It can be a sampler too.

Fixes: 84b08971fbd ("nir/lower_input_attachments: lower nir_texop_fragment_{mask}_fetch")
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2558
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4043>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4043>

radv/rgp: report correct system ram size

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4023>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4023>

radv/rgp: report correct cu_mask info

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4023>

ac: add ac_gpu_info::cu_mask to store bitmask of compute units

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4023>

radv/sqtt: abort if SQTT is used on GFX6-GFX7

RGP only supports GFX8+.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4022>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4022>

radv/sqtt: add support for GFX8

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4022>

ac/registers: adjust some definitions for thread trace on GFX8

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4022>

radv/sqtt: add radv_copy_thread_trace_info_regs() helper

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4022>

radv/sqtt: tidy up radv_emit_thread_trace_{start,stop}

Check for GFX10 first.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4022>

radv/sqtt: fix wrong check in radv_is_thread_trace_complete()

Oops, should be equal actually.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4022>

radv/winsys: fix missing initializations of shader info in the null device

To avoid divide by zero when computing shader stats.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3999>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3999>

iris: Don't skip fast depth clears if the color changed

We depend on BLORP to convert the clear color and write it into the
clear color buffer for us.  However, we weren't bothering to call blorp
in the case where the state is ISL_AUX_STATE_CLEAR.  This leads to the
clear color not getting properly updated if we have back-to-back clears
with different clear colors.  Technically, we could go out of our way to
set the clear color directly from iris in this case but this is a case
we're unlikely to see in the wild so let's not bother.  This matches
what we already do for color surfaces.

Cc: mesa-stable@lists.freedesktop.org
Reported-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4073>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4073>

swr: Fix non-pod-varargs error.

../src/gallium/drivers/swr/rasterizer/jitter/functionpasses/lower_x86.cpp:391:24: error: cannot pass object of non-trivial type 'std::string' (aka 'basic_string<char>') through variadic function; call will abort at runtime [-Wnon-pod-varargs]
                       pFunc->getName().str());
                       ^

Fixes: ff8265b64ff1 ("gallium/swr: Fix llvm11 compilation issues")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Jan Zielinski <jan.zielinski@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4008>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4008>

glthread: fall back if a param size is non-zero and a pointer param is NULL

So that we don't crash. This is a GL error anyway.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: fix a crash with incorrect glShaderSource parameters

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add custom marshalling for glNamedBuffer(Sub)DataEXT

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: merge glBufferSubData and glNamedBufferSubData into 1 set of functions

This is a big cleanup.
GL_EXTERNAL_VIRTUAL_MEMORY_BUFFER_AMD also doesn't sync anymore.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: merge glBufferData and glNamedBufferData into 1 set of functions

This is a big cleanup.
GL_EXTERNAL_VIRTUAL_MEMORY_BUFFER_AMD also doesn't sync anymore.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: replace custom glBindBuffer marshalling with generated one

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: sync instead of disabling glthread for non-VBO pointers

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: track for each VAO whether the user has set a user pointer

This commit mainly adds basic infrastructure for tracking vertex array
state.

If glthread gets a non-VBO pointer, this commit delays disabling
glthread until glDraw is called. The next will change that to "sync"
instead of "disable".

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add marshal_call_after and remove custom glFlush and glEnable code

Instead of implementing marshalling manually, this XML property allows us
to insert additional code into code-generated functions.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: don't insert an empty line after (void) cmd;

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add support for glMemoryObjectParameteriv, glSemaphoreParameterui64v

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add support for glCallLists, glPatchParameterfv

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add support for glClearNamedFramebuffer, glMaterial, glPointParameter

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add support for glFog, glLight, glLightModel, glTexEnv, glTexGen

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add support for TexParameteri and SamplerParameteri functions

It's straightfoward except that I had to hack the python scripts to add
"marshal_count", which behaves just like "count" except that "variable_param"
is ignored. ("variable_param" changes the behavior of "count", which I don't
want)

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: replace custom ClearBuffer marshalling with generated one

If the count attribute contains "enum", the count is evaluated only once.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: check the size of all variable params and clean up the code

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: handle complex pointer parameters and support GL functions with strings

The python changes add a local variable that computes the parameter size
only once.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add/update count and marshal fields for many GL functions

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: add GL_DRAW_INDIRECT_BUFFER tracking and generator support

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: don't increment variable_data if it's the last variable-size param

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: don't insert _mesa_post_marshal_hook into every function

Let the developer decide that in the python script.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: simplify repeated function sequences in marshal_generated.c

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>

glthread: use int instead of size_t where it's OK

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3948>