Harish Krupo [Mon, 12 Aug 2019 10:07:04 +0000 (12:07 +0200)]
egl/android: Delete set_damage_region from egl dri vtbl
The intension of the KHR_partial_update was not to send the damage back
to the platform but to send the damage to the driver to ensure that the
following rendering could be restricted to those regions.
This patch removes the set_damage_region from the egl_dri vtbl and all
the platfrom_*.c files.
Then upcomming patches add a new dri2 interface for the drivers to
implement
Signed-off-by: Harish Krupo <harishkrupo@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Tested-by: Qiang Yu <yuq825@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Jordan Justen [Fri, 2 Aug 2019 08:21:12 +0000 (01:21 -0700)]
iris: Implement pipe_screen::resource_get_param
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jordan Justen [Sat, 3 Aug 2019 08:16:48 +0000 (01:16 -0700)]
gallium/dri2: Use pipe_screen::resource_get_param in image queries
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Eric Anholt <eric@anholt.net>
Jordan Justen [Mon, 24 Jun 2019 17:39:03 +0000 (10:39 -0700)]
gallium/dri2: Support images with multiple planes for modifiers
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Eric Anholt <eric@anholt.net>
Jordan Justen [Sat, 3 Aug 2019 08:14:01 +0000 (01:14 -0700)]
gallium/dri2: Refactor image property queries
This refactor will let us more easily use
pipe_screen::resource_get_param as an alternative to
pipe_screen::resource_get_handle.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Eric Anholt <eric@anholt.net>
Jordan Justen [Fri, 2 Aug 2019 20:52:29 +0000 (13:52 -0700)]
state_tracker/winsys_handle: Add plane input field
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Eric Anholt <eric@anholt.net>
Jordan Justen [Fri, 2 Aug 2019 21:00:08 +0000 (14:00 -0700)]
gallium/dri2: Support creating multi-planar modifier images
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Eric Anholt <eric@anholt.net>
Jordan Justen [Mon, 8 Jul 2019 22:05:04 +0000 (15:05 -0700)]
gallium/dri2: Implement dri2ImageExtension.queryDmaBufFormatModifierAttribs
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Eric Anholt <eric@anholt.net>
Jordan Justen [Thu, 1 Aug 2019 20:30:21 +0000 (13:30 -0700)]
gallium/screen: Add pipe_screen::resource_get_param
This function retrieves individual parameters selected by enum
pipe_resource_param. It can be used as a more direct alternative to
pipe_screen::resource_get_handle.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Acked-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Wed, 31 Jul 2019 08:33:08 +0000 (10:33 +0200)]
vc4: clamp gl_PointSize to a minimum of 1.0
The OpenGL ES spec requires that the value of gl_PointSize is clamped
to an implementation-dependent range matching what is advertised by
GL_ALIASED_POINT_SIZE_RANGE. For VC4 this is [1.0, 512.0], but the
hardware won't clamp to the minimum side of the range and won't render
points with a size strictly smaller than 1.0 either, so we need to
clamp manually. For points larger than the maximum size of the range
the hardware clamps automatically.
Fixes piglit test:
spec/!opengl 2.0/vs-point_size-zero
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Wed, 31 Jul 2019 07:50:34 +0000 (09:50 +0200)]
v3d: clamp gl_PointSize to a minimum of 1.0
The OpenGL ES spec requires that the value of gl_PointSize is clamped
to an implementation-dependent range matching what is advertised by
GL_ALIASED_POINT_SIZE_RANGE. For V3D this is [1.0, 512.0], but the
hardware won't clamp to the minimum side of the range and won't render
points with a size strictly smaller than 1.0 either, so we need to
clamp manually. For points larger than the maximum size of the range
the hardware clamps automatically.
Fixes piglit test:
spec/!opengl 2.0/vs-point_size-zero
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Wed, 31 Jul 2019 07:47:08 +0000 (09:47 +0200)]
nir: add a pass to clamp gl_PointSize to a range
The OpenGL and OpenGL ES specs require that implementations clamp the
value of gl_PointSize to an implementation-depedent range. This pass
is useful for any GPU hardware that doesn't do this automatically
for either one or both sides of the range, such as V3D.
v2:
- Turn into a generic NIR pass (Eric).
- Make the pass work before lower I/O so we can use the deref variable
to inspect if we are writing to gl_PointSize (Eric).
- Make the pass take the range to clamp as parameter and allow it
to clamp to both sides of the range or just one side.
- Make the pass report progress.
v3:
- Fix copyright header (Eric)
- use fmin/fmax instead of bcsel to clamp (Eric)
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Wed, 7 Aug 2019 06:32:58 +0000 (08:32 +0200)]
v3d: line length style fixes
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Wed, 7 Aug 2019 06:20:35 +0000 (08:20 +0200)]
v3d: honor the write mask on store operations
v2:
- Fix incremental update of the const offset when we need to emit a sequence
with more than one write because of the writemask.
- Do not move the tmu write emission to a separate helper.
v3:
- Get the store writemask before the loop, use ffs to get the first component
to write and clear writemask bits as we process the components (Eric).
- Simplified the code that figured out the number of components for the TMU
config based on the number of tmu writes for stores and atomics.
v4:
- Code clean-ups (Eric).
Fixes:
KHR-GLES31.core.shader_image_load_store.advanced-cast-cs
KHR-GLES31.core.shader_image_load_store.advanced-cast-fs
KHR-GLES31.core.shader_storage_buffer_object.advanced-switchBuffers-cs
KHR-GLES31.core.shader_storage_buffer_object.advanced-switchPrograms-cs
KHR-GLES31.core.shader_storage_buffer_object.basic-operations-case1-cs
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Tue, 6 Aug 2019 10:18:17 +0000 (12:18 +0200)]
v3d: refactor ntq_emit_tmu_general() slightly
When we implement write masks on store operations we might need to
emit multiple write sequences for a given store intrinsic. To make
that easier, let's split the emission of the tmud instructions to
their own block after we are done with the code that only needs to
run once no matter how many write sequences we need to emit.
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Mon, 12 Aug 2019 06:36:37 +0000 (08:36 +0200)]
v3d: do not automatically flush current job for SSBOs and shader images
If the current job has a sequence of draw calls involving SSBOs and/or
shader images, we would flush the job in between each draw call.
With this change, we won't flush the current job and we rely on the
application inserting correct barriers by issuing glMemoryBarrier()
when needed.
v2 (Eric):
- When mapping a buffer for writing, we always need to flush.
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Thu, 8 Aug 2019 10:46:43 +0000 (12:46 +0200)]
v3d: only process glMemoryBarrier() for SSBOs and images
PIPE_BARRIER_UPDATE is defined as:
PIPE_BARRIER_UPDATE_BUFFER | PIPE_BARRIER_UPDATE_TEXTURE
Which means we were flushing for any flags other than these two, but
this was intended to only flush for ssbos and images.
Actually, the driver automatically flushes jobs as we need, including
writes/reads involving SSBOs and images, so we don't really need to
flush anything when the program emits a barrier. However, this may
lead to excessive flushing in some cases, so we will soon change this
to avoid atutomatic flushing of the current job for SSBOs and images,
meaning that we will rely on the application to emit correct memory
barriers for these that we should make sure to process here.
Reviewed-by: Eric Anholt <eric@anholt.net>
Iago Toral Quiroga [Fri, 9 Aug 2019 11:02:32 +0000 (13:02 +0200)]
v3d: fix flushing of SSBOs and shader images
If the current draw call includes SSBOs, then we must flush any jobs
that are writing to the same SSBOs (so that our SSBOs reads are correct),
as well as jobs reading from the same SSBO (so that our SSBO writes don't
stomp previous SSBO reads).
The exact same logic applies to shader images. In this case we were already
flushing previous writes, but we should also flush previous reads.
Note that We don't need to call v3d_flush_jobs_reading_resource() and
v3d_flush_jobs_writing_resource() separately though, since flushing
jobs that read a resource also flushes those writing to it.
Suggested-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Eric Anholt <eric@anholt.net>
Caio Marcelo de Oliveira Filho [Mon, 12 Aug 2019 19:55:37 +0000 (12:55 -0700)]
intel/tools: Fix aub_file initialization in intel_dump_gpu
The `device` can be set earlier either by a command line or a by
intercepting an ioctl call to get the I915_PARAM_CHIPSET_ID done by
the application early. In both cases `aub_file` and `devinfo` would
not be initialized.
Fix by splitting the conditions
- `device == 0`: use the FD to get both device and devinfo.
- Or `devinfo.gen == 0`: use `device` to initialize it.
And separatedly, initialize aub_file the first time it is needed.
Fixes: d594d2a0524 ("intel/tools: use device info initializer")
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Rafael Antognolli [Mon, 22 Jul 2019 22:49:24 +0000 (15:49 -0700)]
i965/gen11: Emit SLICE_HASH_TABLE when pipes are unbalanced.
If the pixel pipes have a different number of subslices, emit a slice
hashing table that will ensure proper workload distribution.
v2: Set Mask field to 0xffff for workaround (Ken).
Rafael Antognolli [Mon, 22 Jul 2019 17:56:53 +0000 (10:56 -0700)]
anv/gen11: Emit SLICE_HASH_TABLE when pipes are unbalanced.
If the pixel pipes have a different number of subslices, emit a slice
hashing table that will ensure proper workload distribution.
v2: Don't need to set the mask - it's mbo (Ken).
Rafael Antognolli [Fri, 19 Jul 2019 22:29:00 +0000 (15:29 -0700)]
iris/gen11: Emit SLICE_HASH_TABLE when pipes are unbalanced.
If the pixel pipes have a different number of subslices, emit a slice
hashing table that will ensure proper workload distribution.
v2: Don't need to set the mask - it's mbo (Ken).
v3: Don't keep a reference to the resource used for emitting the table
(Ken).
Rafael Antognolli [Mon, 1 Jul 2019 20:34:08 +0000 (13:34 -0700)]
intel: Get information about pixel pipes subslices.
v2: Use 1 instead of 1UL (Ken).
Rafael Antognolli [Fri, 19 Jul 2019 22:30:56 +0000 (15:30 -0700)]
intel/gen_decoder: Decode SLICE_HASH_TABLE.
Rafael Antognolli [Fri, 19 Jul 2019 20:37:25 +0000 (13:37 -0700)]
intel/genxml: Update 3D_MODE and add SLICE_HASH_TABLE.
Add these fields and the 3DSTATE_SLICE_TABLE_STATE_POINTERS instruction
so we can properly configure the slice and subslice hashing on ICL+
v2: Make 'Mask' field a mbo (Ken).
Jason Ekstrand [Wed, 24 Apr 2019 08:02:35 +0000 (03:02 -0500)]
anv: Implement VK_KHR_pipeline_executable_properties
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 24 Apr 2019 07:21:01 +0000 (02:21 -0500)]
anv: Add a ralloc context to anv_pipeline
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Thu, 25 Apr 2019 00:56:39 +0000 (19:56 -0500)]
anv: Force a full re-compile when CAPTURE_INTERNAL_REPRESENTATION_TEXT is set
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Thu, 25 Apr 2019 00:54:14 +0000 (19:54 -0500)]
anv/pipeline: Split setting up per-stage keys into its own loop
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 24 Apr 2019 07:00:25 +0000 (02:00 -0500)]
anv: Record shader compile stats in the pipeline cache
We're going to want these to be available regardless of caching.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 24 Apr 2019 08:19:25 +0000 (03:19 -0500)]
anv/pipeline: Stash generated code in the pipeline stage
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Thu, 25 Jul 2019 17:00:23 +0000 (12:00 -0500)]
intel/fs: Add SLM size to brw_cs_prog_data
We don't need it for state setup but it's a useful statistic we want to
pass on to developers.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 24 Apr 2019 04:19:56 +0000 (23:19 -0500)]
intel/compiler: Fill a compiler statistics struct
This commit is all annoying plumbing work which just adds support for a
new brw_compile_stats struct. This struct provides a binary driver
readable form of the same statistics we dump out to stderr when we
INTEL_DEBUG is set with a shader stage.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Khaled Emara [Sun, 4 Aug 2019 15:51:24 +0000 (17:51 +0200)]
freedreno: disable tiling for cubemaps
Tiling doesn't work quite well with cubemaps.
Revert to linear textures, until it's fixed.
Khaled Emara [Sat, 3 Aug 2019 12:59:18 +0000 (14:59 +0200)]
freedreno: add tiling parameters for 2D/2DArray/3D
Khaled Emara [Sat, 3 Aug 2019 12:30:53 +0000 (14:30 +0200)]
freedreno: simplified slices setup for a3xx
a3xx doesn't support ASTC and layout_first always returns false
Khaled Emara [Sat, 3 Aug 2019 12:29:43 +0000 (14:29 +0200)]
freedreno: enable tiled textures for debug builds
Paulo Zanoni [Thu, 11 Jul 2019 22:08:03 +0000 (15:08 -0700)]
intel/fs: add 64 bit integer multiplication lowering
While NIR's lower_imul64() solves the case of 64 bit integer multiplications
generated early, we don't have a way to lower such instructions when they are
generated by our own backend, such as the scan/reduce intrinsics. We'll need
this soon, so implement it now.
An easy way to test this is to simply disable nir_lower_imul64 to let
those operations reach the backend.
v2:
- Fix Q/UQ copy/paste errors (Caio).
- Transform an 'if' into 'else if' (Caio).
- Add an extra comment to clarify the need for 64b = 32b * 32b
(Caio).
- Make private functions private (Caio).
v3:
- Remove ambiguity with 'b' and 'd' variables (Caio).
- Allocate potentially less regs for the dwords (Caio).
Cc: Jason Ekstrand <jason.ekstrand@intel.com>
Cc: Matt Turner <matt.turner@intel.com>
Cc: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Thu, 11 Jul 2019 00:03:48 +0000 (17:03 -0700)]
intel/compiler: invert the logic of lower_integer_multiplication()
Invert the logic of how progress is handled: remove the continue
statements and mark progress inside the places where it actually
happens.
We're going to add a new lowering that also looks for BRW_OPCODE_MUL,
so inverting the logic here makes the resulting code much easier to
follow.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Wed, 10 Jul 2019 23:48:01 +0000 (16:48 -0700)]
intel/compiler: don't instantiate a builder for each instruction
Don't instantiate a builder for each instruction during
lower_integer_multiplication(). Instantiate one only when needed.
On the other hand, these unneeded builders don't seem to cost much to
init, so I don't expect any significant difference in performance:
this is mostly about code organization.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Thu, 11 Jul 2019 23:56:05 +0000 (16:56 -0700)]
intel/compiler: extract subfunctions of lower_integer_multiplication()
The lower_integer_multiplication() function is already a little too
big. I want to add more to it, so let's reorganize the existing code
first. Let's start with just extracting the current code to
subfunctions. Later we'll change them a little more.
v2: Make private functions private (Caio).
v3: Fix typo (Caio).
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Rhys Perry [Wed, 24 Jul 2019 18:23:21 +0000 (19:23 +0100)]
nir: merge and extend nir_opt_move_comparisons and nir_opt_move_load_ubo
v2: add to series
v3: update Makefile.sources
v4: don't remove a comment and break statement
v4: use nir_can_move_instr
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Rhys Perry [Wed, 22 May 2019 19:23:03 +0000 (20:23 +0100)]
nir: replace nir_move_load_const() with nir_opt_sink()
This is mostly the same as nir_move_load_const() but can also move
undef instructions, comparisons and some intrinsics (being careful with
loops).
v2: actually delete nir_move_load_const.c
v3: fix nir_opt_sink() usage in freedreno
v3: update Makefile.sources
v4: replace get_move_def with nir_can_move_instr and nir_instr_ssa_def
v4: handle if uses
v4: fix handling of nested loops
v5: re-write adjust_block_for_loops
v5: re-write setting of use_block for if uses
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Co-authored-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Eric Anholt <eric@anholt.net>
Francisco Jerez [Thu, 18 Jul 2019 01:41:38 +0000 (18:41 -0700)]
anv/gen9: Optimize slice and subslice load balancing behavior.
See "i965/gen9: Optimize slice and subslice load balancing behavior."
for the rationale. According to Jason, improves Aztec Ruins
performance by 2.7%.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (v1)
v2: Undo CPU performance micro-optimization done in i965 and iris due
to lack of data justifying it on anv. Use
cmd_buffer_apply_pipe_flushes wrapper instead of emitting pipe
control command directly. (Jason)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Andreas Baierl [Thu, 4 Jul 2019 15:35:09 +0000 (17:35 +0200)]
lima/ppir: Add fddx and fddy
Lower fddx and fddy and set the right bits in codegen.
Signed-off-by: Andreas Baierl <ichgeh@imkreisrum.de>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Bas Nieuwenhuizen [Sat, 1 Jun 2019 19:08:20 +0000 (21:08 +0200)]
radv: Enable VK_KHR_pipeline_executable_properties.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Sat, 1 Jun 2019 16:48:51 +0000 (18:48 +0200)]
radv: Implement radv_GetPipelineExecutableStatisticsKHR.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Fri, 31 May 2019 00:23:44 +0000 (02:23 +0200)]
radv: Implement radv_GetPipelineExecutableInternalRepresentationsKHR.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Thu, 30 May 2019 23:57:43 +0000 (01:57 +0200)]
radv: Implement radv_GetPipelineExecutablePropertiesKHR.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Thu, 30 May 2019 23:06:27 +0000 (01:06 +0200)]
radv: Keep shader info when needed.
This allows enabling the shader info keeping on a per shader basis.
Also disables the cache on a per shader basis.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Sat, 1 Jun 2019 19:45:39 +0000 (21:45 +0200)]
radv: Add VK_KHR_pipeline_executable_properties in disabled state.
So we can add the functions.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Sat, 1 Jun 2019 18:54:35 +0000 (20:54 +0200)]
radv: Use string for nir dumping.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Allows us to easily dump all nir shaders for combined variants in
vega and simplifies ownership.
Bas Nieuwenhuizen [Sat, 1 Jun 2019 18:25:47 +0000 (20:25 +0200)]
radv: Get max workgroup size without nir.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Sat, 1 Jun 2019 16:46:21 +0000 (18:46 +0200)]
radv: Add utility function to calculate max waves.
Not AC because a lot of it is data extraction out of radv structs.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Francisco Jerez [Sat, 10 Aug 2019 19:45:46 +0000 (12:45 -0700)]
iris/gen9: Optimize slice and subslice load balancing behavior.
See "i965/gen9: Optimize slice and subslice load balancing behavior."
for the rationale.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Francisco Jerez [Thu, 18 Jul 2019 01:30:45 +0000 (18:30 -0700)]
intel/genxml: Add GT_MODE hashing defs for Gen9.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Francisco Jerez [Wed, 17 Jul 2019 21:56:18 +0000 (14:56 -0700)]
i965/gen9: Optimize slice and subslice load balancing behavior.
The default pixel hashing mode settings used for slice and subslice
load balancing are far from optimal under certain conditions (see the
comments below for the gory details). The top-of-the-line GT4 parts
suffer from a particularly severe performance problem currently due to
a subslice load balancing issue. Fixing this seems to improve
graphics performance across the board for most of the benchmarks in my
test set, up to ~20% in some cases, e.g. from SKL GT4:
unigine/valley: 3.44% ±0.11%
gfxbench/gl_manhattan31: 3.99% ±0.13%
gputest/pixmark_piano: 7.95% ±0.33%
synmark/OglTexFilterAniso: 15.22% ±0.07%
synmark/OglTexMem128: 22.26% ±0.06%
Lower-end platforms are also affected by some subslice load imbalance
to a lesser degree, especially during CCS resolve and fast clear
operations, which are handled specially here due to rasterization
ocurring in reduced CCS coordinates, which changes the semantics of
the pixel hashing mode settings.
No regressions seen during my tests on some SKL, KBL and BXT
configurations. Additional benchmark reports welcome on any Gen9
platforms (that includes anything with Skylake, Broxton, Kabylake,
Geminilake, Coffeelake, Whiskey Lake, Comet Lake or Amber Lake in your
renderer string).
P.S.: A similar problem is likely to be present on other non-Gen9
platforms, especially for CCS resolve and fast clear operations.
Will follow-up with additional patches fixing the hashing mode
for those once I have enough performance data to justify it.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Alyssa Rosenzweig [Tue, 6 Aug 2019 00:40:55 +0000 (17:40 -0700)]
pan/midgard: Handle 64-bit address in mir_mask_of_read_components
This is a bit of a hack, but it'll hold us over until we have 64-bit
support wired through.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 9 Aug 2019 17:40:23 +0000 (10:40 -0700)]
pan/midgard: Allocate separate spill indices for lowered moves
This helps RA be slightly more reasonable.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 9 Aug 2019 17:40:01 +0000 (10:40 -0700)]
pan/midgard: Extend liveness analysis to trinary ops
Fixes RA fails with multiple indirect SSBO writes.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 9 Aug 2019 15:56:44 +0000 (08:56 -0700)]
pan/midgard: Fix load/store pairing
This used a delicate hack to try to find indirect inputs and skip them
as candidates for pairing. Let's use a better criterion -- no sources --
and pair based on that.
We could do better, but that would require more complex data flow
analysis than we're interested in doing here.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 21:07:10 +0000 (14:07 -0700)]
pan/midgard: Implement nir_intrinsic_load_num_work_groups
Just a sysval to route through.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 20:47:17 +0000 (13:47 -0700)]
pan/midgard: Implement some compute builtins
We implement gl_WorkGroupID and gl_LocalInvocationID, which map to
ld_compute_id with special sources.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 20:46:56 +0000 (13:46 -0700)]
pan/midgard: Rename ld_global_id -> ld_compute_id
It's used for more general loads within a compute shader.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 01:17:27 +0000 (18:17 -0700)]
pan/midgard: Handle partial writes in liveness analysis
This allows liveness analysis within a loop to be more fine grained,
fixing RA failures with partial spilled movs within a loop, as well as
enabling a slight reduction of register pressure more generally:
total registers in shared programs: 350 -> 347 (-0.86%)
registers in affected programs: 12 -> 9 (-25.00%)
helped: 3
HURT: 0
helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
helped stats (rel) min: 25.00% max: 25.00% x̄: 25.00% x̃: 25.00%
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 01:09:36 +0000 (18:09 -0700)]
pan/midgard: Dump "no spill"?
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 01:06:24 +0000 (18:06 -0700)]
pan/midgard: Absorb nonexistance sources
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 00:59:53 +0000 (17:59 -0700)]
pan/midgard: Pretty-print destinations
They're not "sources" but they follow the same conventions.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 00:58:52 +0000 (17:58 -0700)]
pan/midgard: Pretty-print units
Since we are seeing some use of MIR post-scheduling, let's get this
printed right.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 6 Aug 2019 00:52:56 +0000 (17:52 -0700)]
pan/midgard: Print mask in dumped MIR
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 5 Aug 2019 22:00:11 +0000 (15:00 -0700)]
pan/midgard: Add no_spill flag
Hint for the RA to avoid infinite spilling loops.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 5 Aug 2019 21:47:28 +0000 (14:47 -0700)]
pan/midgard: Generalize mir_mask_of_read_components
This now works for load/store and texture instructions as well as ALU.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 1 Aug 2019 17:03:02 +0000 (10:03 -0700)]
pan/midgard: Implement SSBO access
Just laying the groundwork. Reads and writes should be supported (both
direct and indirect, either int or float, vec1/2/3/4), but no bounds
checking is done at the moment.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 5 Aug 2019 16:19:39 +0000 (09:19 -0700)]
pan/midgard: Pipe uniform mask through when spilling
This is a corner case that happens a lot with SSBOs. Basically, if we
only read a few components of a uniform, we need to only spill a few
components or otherwise we try to spill what we spilled and RA hangs.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Sat, 3 Aug 2019 00:10:18 +0000 (17:10 -0700)]
pan/midgard: Clamp sysval component count
We don't want to load a 128-bit sysval when 64-bits will do. Fixes RA
failures with SSBO indirect writes.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Sat, 3 Aug 2019 00:09:54 +0000 (17:09 -0700)]
pan/midgard: Pass uploaded midgard_instruction through
We want to edit it after emission in some cases.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 2 Aug 2019 18:06:21 +0000 (11:06 -0700)]
pan/midgard: Allow sysval destination override
Sometimes a sysval is used to facilitate an instruction but is not the
instruction itself.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 9 Aug 2019 15:32:49 +0000 (08:32 -0700)]
panfrost: Force flush every compute job
This is of course suboptimal for performance, forcing each
glDispatchCompute call to be submitted separately to the kernel and
finish to completion. However, for the initial bring-up of compute jobs,
this simplifies quite a bit.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 1 Aug 2019 18:03:15 +0000 (11:03 -0700)]
panfrost: Add SSBO system value
For each SSBO index we get from Gallium/NIR, we need two pieces of
information in the shader:
1. The address of the SSBO in GPU memory. Within the shader, we'll be
accessing it with raw memory load/store, so we need the actual address,
not just an index.
2. The size of the SSBO. This is not strictly necessary, but at some
point, we may like to do bounds checking on SSBO accesses.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 8 Aug 2019 21:27:41 +0000 (14:27 -0700)]
gallium/util: Add u_stream_outputs_for_vertices helper
This u_prim.h helper determines the number of outputs for stream output,
given a particular primitive type and a vertex count. This is useful for
statically calculating sizes of stream output buffers (i.e. when there
is no geometry/tessellation shader in use).
This helper will be used in Panfrost's transform feedback
implementation, as you can probably guess since why else would I be
submitting it....
See also dEQP's getTransformFeedbackOutputCount routine.
v2: Simplify definition using new helpers, which also extends to non-ES2
primitive types (Eric).
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Marek Olšák [Thu, 1 Aug 2019 19:49:39 +0000 (15:49 -0400)]
radeonsi: remove the always_nir option
tgsi_to_nir is no longer optional if NIR is enabled.
Marek Olšák [Thu, 1 Aug 2019 18:47:03 +0000 (14:47 -0400)]
radeonsi/nir: implement default tess level system values
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 18:38:40 +0000 (14:38 -0400)]
compiler: add SYSTEM_VALUE_TESS_LEVEL_OUTER/INNER_DEFAULT
TCS system values for internal passthru TCS, needed by radeonsi NIR support
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 18:21:41 +0000 (14:21 -0400)]
gallium: add TGSI_SEMANTIC_DEFAULT_OUTER/INNER_LEVEL
for radeonsi NIR support.
Marek Olšák [Thu, 1 Aug 2019 18:16:01 +0000 (14:16 -0400)]
tgsi_to_nir: handle tess level inner/outer varyings
for internal radeonsi shaders
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 16:29:05 +0000 (12:29 -0400)]
tgsi_to_nir: add support for the stencil FS output
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 16:25:12 +0000 (12:25 -0400)]
tgsi_to_nir: add support for TEX_LZ
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Marek Olšák [Thu, 1 Aug 2019 01:52:48 +0000 (21:52 -0400)]
compiler: add SYSTEM_VALUE_USER_DATA_AMD
for internal radeonsi shaders
Marek Olšák [Thu, 1 Aug 2019 01:33:37 +0000 (21:33 -0400)]
compiler: add shader_info.cs.user_data_components_amd
Marek Olšák [Thu, 1 Aug 2019 01:30:22 +0000 (21:30 -0400)]
tgsi_to_nir: add basic compute shader support
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 01:17:47 +0000 (21:17 -0400)]
tgsi_to_nir: add support for LOAD & STORE with SSBOs and images
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 01:17:05 +0000 (21:17 -0400)]
tgsi_to_nir: make setup_texture_info reusable
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Marek Olšák [Wed, 31 Jul 2019 21:31:52 +0000 (17:31 -0400)]
tgsi_to_nir: add support for TXF_LZ
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Wed, 31 Jul 2019 20:55:33 +0000 (16:55 -0400)]
compiler: add shader_info.vs.blit_sgprs_amd
for internal radeonsi shaders
Marek Olšák [Wed, 31 Jul 2019 21:16:23 +0000 (17:16 -0400)]
tgsi_to_nir: be careful about not losing any TGSI properties silently (v2)
v2: squash with Timur Kristof's commit
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Wed, 31 Jul 2019 21:15:18 +0000 (17:15 -0400)]
tgsi/scan: don't set GS_INVOCATIONS for all shader stages
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 01:16:12 +0000 (21:16 -0400)]
compiler: add ACCESS_STREAM_CACHE_POLICY
radeonsi will use this.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Marek Olšák [Thu, 1 Aug 2019 01:28:40 +0000 (21:28 -0400)]
gallium: add AMD-specific compute TGSI enums
for tgsi_to_nir
Marek Olšák [Wed, 31 Jul 2019 20:45:21 +0000 (16:45 -0400)]
gallium: add TGSI_PROPERTY_VS_BLIT_SGPRS_AMD for tgsi_to_nir
needed by radeonsi NIR support
Marek Olšák [Fri, 2 Aug 2019 23:11:53 +0000 (19:11 -0400)]
st/mesa: don't allocate mipmapped texture for NEAREST_MIPMAP_LINEAR
Reviewed-by: Brian Paul <brianp@vmware.com>