GitLab

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>

Instead of gen_device_info instance pass anv_physical_device instance
which also has refernce to gen_device_info. This will allow for future
patches to use the driconf options.
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>

On Fallout4, enabling HIZ_CCS_WT compression for D16_UNORM format
regress the performance by 2%, in order to avoid that disable
compression via driconf option.
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>

Passes at least:
dEQP-VK.dynamic_state.vp_state.viewport_array
dEQP-VK.draw.shader_viewport_index.*
dEQP-VK.draw.shader_layer.*
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <mesa/mesa!5832>

Does the same thing as layer_zero, but for VARYING_SLOT_VIEWPORT.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <mesa/mesa!5832>

With VK_EXT_shader_viewport_index_layer, these stages can all output the
viewport or layer id, and not just GS anymore.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <mesa/mesa!5832>

Add an allowlist to make an exception when deriving images from
interlaced buffers. Normally, the function should fail if the surface
needs to be modified to derive the image. But some applications do not
follow the fall-back method of using vaCreateImage + vaPutImage as
mentioned in the VAAPI documentation, so we have to make an exception.
Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <mesa/mesa!5942>

Allow vaDriveImage to derive a vaImage from interlaced buffers
by creating a new progressive buffer.

v2: Keeps the surface used by DeriveImage untouched (Pierre-Eric)
v3: Fixed a segfault reported by Roman Elshin

Closes: mesa/mesa#1428

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <mesa/mesa!5942>

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <mesa/mesa!5942>

We don't use exceptions or RTTI at all, so pass this flag to the compiler
to allow it to create better code.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <mesa/mesa!6315>

The restriction found in removed comment is not found on new specs.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <mesa/mesa!6593>

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Anuj Phogat anuj.phogat@gmail.com
Part-of: <mesa/mesa!6678>

Fix defects reported by Coverity Scan.

Uninitialized pointer field (UNINIT_CTOR)
Non-static class member exit is not initialized in this
constructor nor in any functions that it calls
Non-static class member immInsertPos is not initialized in this
constructor nor in any functions that it calls.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <mesa/mesa!6695>

Fix defect reported by Coverity Scan.

Argument cannot be negative (NEGATIVE_RETURNS)
negative_returns: fd is passed to a parameter that cannot be
negative.

Fixes: 1ea4ef0d

 ("freedreno: slurp in decode tools")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <mesa/mesa!6696>

Fix defect reported by Coverity.

Sizeof not portable (SIZEOF_MISMATCH)
suspicious_sizeof: Passing argument vec_size * 8UL /* sizeof
(LLVMValueRef *) */ to function __builtin_alloca and then casting
the return value to LLVMValueRef * is suspicious. In this
particular case sizeof (LLVMValueRef *) happens to be equal to
sizeof (LLVMValueRef), but this is not a portable assumption.

Fixes: ca74603b

 ("ac/llvm: add better code for isign")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <mesa/mesa!6682>

Changes are necessary to properly build libmesa_aco static library for Android.
was removed in 47de5532 "aco/isel: Move context initialization code to a dedicated file"
so filter-out of aco_instruction_selection_setup.cpp in Android is removed.

Fixes the following building errors:

FAILED: out/target/product/x86_64/obj/SHARED_LIBRARIES/vulkan.radv_intermediates/LINKED/vulkan.radv.so
...
ld.lld: error: undefined symbol: aco::setup_isel_context(aco::Program*, unsigned int, nir_shader* const*, ac_shader_config*, radv_shader_args*, bool)
>>> referenced by aco_instruction_selection.cpp:10879 (external/mesa/src/amd/compiler/aco_instruction_selection.cpp:10879)
>>>               aco_instruction_selection.o:(aco::select_program(aco::Program*, unsigned int, nir_shader* const*, ac_shader_config*, radv_shader_args*)) in archive out/target/product/x86_64/obj/STATIC_LIBRARIES/libmesa_aco_intermediates/libmesa_aco.a
...
ld.lld: error: undefined symbol: aco::init_context(aco::isel_context*, nir_shader*)
>>> referenced by aco_instruction_selection.cpp:10885 (external/mesa/src/amd/compiler/aco_instruction_selection.cpp:10885)
>>>               aco_instruction_selection.o:(aco::select_program(aco::Program*, unsigned int, nir_shader* const*, ac_shader_config*, radv_shader_args*)) in archive out/target/product/x86_64/obj/STATIC_LIBRARIES/libmesa_aco_intermediates/libmesa_aco.a

Fixes: 47de5532

 ("aco/isel: Move context initialization code to a dedicated file")
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <mesa/mesa!6703>

total instructions in shared programs: 105044 -> 103312 (-1.65%)
instructions in affected programs: 27328 -> 25596 (-6.34%)
helped: 55
HURT: 0
helped stats (abs) min: 4 max: 96 x̄: 31.49 x̃: 24
helped stats (rel) min: 1.18% max: 23.08% x̄: 8.26% x̃: 6.88%
95% mean confidence interval for instructions value: -37.81 -25.18
95% mean confidence interval for instructions %-change: -9.66% -6.85%
Instructions are helped.

total temps in shared programs: 2960 -> 2917 (-1.45%)
temps in affected programs: 425 -> 382 (-10.12%)
helped: 41
HURT: 6
helped stats (abs) min: 1 max: 3 x̄: 1.20 x̃: 1
helped stats (rel) min: 7.14% max: 25.00% x̄: 13.79% x̃: 12.50%
HURT stats (abs)   min: 1 max: 1 x̄: 1.00 x̃: 1
HURT stats (rel)   min: 12.50% max: 25.00% x̄: 17.52% x̃: 16.67%
95% mean confidence interval for temps value: -1.17 -0.66
95% mean confidence interval for temps %-change: -13.30% -6.29%
Temps are helped.

total immediates in shared programs: 16772 -> 16468 (-1.81%)
immediates in affected programs: 4304 -> 4000 (-7.06%)
helped: 59
HURT: 0
helped stats (abs) min: 4 max: 16 x̄: 5.15 x̃: 4
helped stats (rel) min: 2.22% max: 33.33% x̄: 9.31% x̃: 7.14%
95% mean confidence interval for immediates value: -5.88 -4.43
95% mean confidence interval for immediates %-change: -11.14% -7.49%
Immediates are helped.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <mesa/mesa!6572>

Broken by the various GenXML refactors, oops!
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <mesa/mesa!6712>

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Fixes: ec37c712 ("bifrost: Emit "d3d" variant of comparison instructions")
Part-of: <mesa/mesa!6712>

Probably harmless but looks funny in the decode.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <mesa/mesa!6712>

Use the stack. (note: we already do for drm_msm_gem_submit_cmd array, and
using calloc() for heap allocations in a VK driver is wrong
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <mesa/mesa!6687>

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <mesa/mesa!6604>

49d35f3d moved gl_Layer/gl_ViewportIndex/gl_ViewportMask
as builtins but fields size wasn't increased.

Closes: mesa/mesa#3508
Fixes: 49d35f3d

 ("glsl: declare gl_Layer/gl_ViewportIndex/gl_ViewportMask as vs builtins")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <mesa/mesa!6688>

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <mesa/mesa!6699>

Make it less error-prone and more consistent with other helpers.
Pass the masks as a single argument rather than two.
In wave64 mode, split the argument into low and high halves in
emit_mbcnt rather than where it is called.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <mesa/mesa!6699>

Use the lane mask instead of s2 for the register class.

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <mesa/mesa!6699>

GLSL Desktop spec 1.30.x:
   "New built-ins: trunc(), round(), roundEven(), isnan(), isinf(), modf()"

For ES, 3.00.x is the first ES spec that mentions the builtins.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Part-of: <mesa/mesa!6455>

It seems that the mali400 pp is unable to load vec3 unaligned varyings.
This can happen in the current state with mesa if a varying float is put
into the first component of a vec4 and a vec3 is packed right after it.
This would be fine as by default nir would create a vec4 load followed
by a mov with swizzle to realign the components into a vec3.
In lima_nir_split_load_input, this becomes a separate vec3 load
expecting the unaligned load.
Since this can't happen, skip the load input splitting for this special
case.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <mesa/mesa!6507>

Avoid having to deal with BO tracking. However, the kernel still requires a
bo list, so keep a global one which can be re-used for every submit.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Part-of: <mesa/mesa!6686>

This has been shown to help performance on TGL and DG1. This could be
applied to gen9+, but we still need to show if it helps with those
platforms.

Rework:
 * Make change in src/intel/vulkan/genX_cmd_buffer.c too. (Ken)
 * Keep mask as 3 for gen < 12

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <mesa/mesa!6684>

Anv doesn't do multi-layer fast-clear tracking, but TGL may add
fast-clears to multiple layers. Disable CCS_E for image arrays on TGL+
until anv gets more clear color tracking abilities.

With this change, anv+TGL now passes:
* dEQP-VK.multiview.readback_implicit_clear.15_15_15_15
* dEQP-VK.multiview.readback_implicit_clear.8_1_1_8
* dEQP-VK.multiview.readback_implicit_clear.1_2_4_8_16_32
* dEQP-VK.multiview.renderpass2.readback_implicit_clear.15_15_15_15
* dEQP-VK.multiview.renderpass2.readback_implicit_clear.8_1_1_8
* dEQP-VK.multiview.renderpass2.readback_implicit_clear.1_2_4_8_16_32

v2. Mention HSD 14010672564. (Sagar)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <mesa/mesa!6528>

This fixes a bunch of dEQP GLES tests.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <mesa/mesa!6694>

sync_accumulate() does not take ownership.

Fixes: bf23ff83

 ("freedreno: fence_server_sync() fixes")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <mesa/mesa!6692>

syncobj wait takes int64_t timeout and won't clamp it
in kernel code, so we have to pass in INT64_MAX instead
of OS_TIMEOUT_INFINITE which is UINT64_MAX. Otherwise
syncobj wait with OS_TIMEOUT_INFINITE case just return
fail.

Fixes: c638301b

 "radeonsi: fix syncobj wait timeout"
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <mesa/mesa!6676>

Pre-V3D 4.3 hardware has a quirk where it expects XY coordinates in
.8 fixed-point format, but then it will internally round it to .6 fixed-point,
introducing a double rounding. The double rounding can cause very slight
differences in triangle raterization coverage that can actually be noticed by
some CTS tests.

The correct fix for this as recommended by Broadcom is to convert to
.8 fixed-point with ffloor().

Fixes:
dEQP-VK.renderpass.suballocation.subpass_dependencies.late_fragment_tests.*
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Part-of: <mesa/mesa!6677>

This should be helpful if someone chooses to implement cache support on
windows. Also providing this greater level of abstraction makes it easier
to implement alterative cache layouts in future.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <mesa/mesa!6197>

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <mesa/mesa!6197>

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <mesa/mesa!6197>