Vasily Khoruzhick [Mon, 23 Sep 2019 05:03:22 +0000 (22:03 -0700)]
lima/ppir: add NIR pass to split varying loads
NIR may emit a single instrinsic to load several packed varyings,
but that's suboptimal for Utgard PP for several reasons:
- varyings that are used as sampler inputs can be passed using
pipeline register with increased precision
- we have small number of regs, so using a vec4 regs for storing
two vec2 varyings increases reg pressure.
Add NIR pass to split a single load into several loads and utilize
it in lima.
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Timur Kristóf [Thu, 26 Sep 2019 07:37:16 +0000 (09:37 +0200)]
radv: Fix L2 cache rinse programming.
According to radeonsi, GLM doesn't support WB alone, so
we have to set INV too when WB is set.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Jonathan Marek [Thu, 26 Sep 2019 04:37:26 +0000 (00:37 -0400)]
turnip: emit texture and uniform state
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 04:31:56 +0000 (00:31 -0400)]
turnip: add some shader information in pipeline state
This information is needed by texture/uniform descriptors.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 04:30:22 +0000 (00:30 -0400)]
turnip: use nir_opt_copy_prop_vars
Avoids getting a "load_output" in a case like this:
gl_Position = ubuf.MVP * ubuf.position[gl_VertexIndex];
frag_pos = gl_Position.xyz;
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 04:29:26 +0000 (00:29 -0400)]
turnip: lower samplers and uniform buffer indices
Lower these to something compatible with ir3, and save the descriptor set
and binding information.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 03:00:16 +0000 (23:00 -0400)]
turnip: basic descriptor sets (uniform buffer and samplers)
Mostly copy-paste from radv, with a few modifications.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 18:58:49 +0000 (14:58 -0400)]
turnip: enable linear filtering
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 18:56:17 +0000 (14:56 -0400)]
turnip: align layer_size
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 02:47:02 +0000 (22:47 -0400)]
turnip: use linear tiling for scanout image
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Thu, 26 Sep 2019 02:24:13 +0000 (22:24 -0400)]
turnip: implement image view descriptor
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Wed, 25 Sep 2019 16:55:14 +0000 (12:55 -0400)]
turnip: implement sampler state
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Wed, 25 Sep 2019 16:48:39 +0000 (12:48 -0400)]
turnip: fix vertex_id
ir3 uses non-zero based vertex id for a6xx
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jonathan Marek [Wed, 25 Sep 2019 16:46:04 +0000 (12:46 -0400)]
turnip: emit shader immediates
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Acked-by: Eric Anholt <eric@anholt.net>
Jason Ekstrand [Wed, 25 Sep 2019 15:02:15 +0000 (10:02 -0500)]
util/rb_tree: Stop relying on &iter->field != NULL
The old version of the iterators relies on a &iter->field != NULL check
which works fine on older GCC but newer GCC versions and clang have
optimizations that break if you do pointer math on a null pointer. The
correct solution to this is to do the null comparisons before we do any
sort of &iter->field or use rb_node_data to do the reverse operation.
Acked-by: Michel Dänzer <mdaenzer@redhat.com>
Tested-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 25 Sep 2019 15:01:27 +0000 (10:01 -0500)]
util/rb_tree: Also test _safe iterators
Acked-by: Michel Dänzer <mdaenzer@redhat.com>
Tested-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Eric Anholt [Wed, 28 Aug 2019 20:24:41 +0000 (13:24 -0700)]
freedreno/a3xx: Mostly fix min-vs-mag filtering decisions on non-mipmap tex.
This is based on the fix I used for the same problem on V3D. In this
case, it fixes all but the the
dEQP-GLES2.functional.texture.filtering.2d.*_npot cases of
dEQP-GLES2.functional.texture.filtering.2d.*'s failures.
Acked-by: Rob Clark <robdclark@chromium.org>
Maya Rashish [Thu, 26 Sep 2019 14:14:34 +0000 (17:14 +0300)]
intel/compiler: avoid truncating int64_t to int
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Maya Rashish <maya@netbsd.org>
Icenowy Zheng [Thu, 26 Sep 2019 03:10:18 +0000 (11:10 +0800)]
lima: support rectangle texture
As Vasily discovered, the bit 7 of the word 1 of the texture descriptor
is set when reloading the framebuffer, to use framebuffer-based offset
rather than normalized one. This bit also works for regular textures to
enable accessing with non-normalized offset.
Add support for rectangle texture by setting this bit for
PIPE_TEXTURE_RECT.
Suggested-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Icenowy Zheng <icenowy@aosc.io>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Michel Dänzer [Thu, 26 Sep 2019 09:02:46 +0000 (11:02 +0200)]
loader: Avoid use-after-free / use of uninitialized local variables
Per the valgrind output below, we were returning the pointer to freed
memory if none of the later conditional pointer assignments were
executed. This caused dEQP CI jobs to crash on certain runners,
presumably due to a double-free down the line.
Also, we were skipping to the out: label before the vendor_id & chip_id
variables used by it were initialized, resulting in broken
LIBGL_DEBUG=verbose output such as
libGL: pci id for fd 4:
51108f00:
51108f00, driver radeonsi
Fixes: 5a545e355b23 "loader: always map the "amdgpu" kernel driver name to radeonsi (v2)"
==403== Invalid read of size 1
==403== at 0x4AFD576: surfaceless_probe_device (platform_surfaceless.c:316)
==403== by 0x4AFD915: dri2_initialize_surfaceless (platform_surfaceless.c:391)
==403== by 0x4AF5EEA: dri2_initialize (egl_dri2.c:984)
==403== by 0x4AF5EEA: dri2_initialize (egl_dri2.c:958)
==403== by 0x4AF1EEC: _eglMatchAndInitialize (egldriver.c:75)
==403== by 0x4AF1F3B: _eglMatchDriver (egldriver.c:96)
==403== by 0x4AE9367: eglInitialize (eglapi.c:617)
==403== by 0x1D99C9: tcu::surfaceless::EglRenderContext::EglRenderContext(glu::RenderConfig const&, tcu::CommandLine const&) [clone .constprop.57] (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x1DABB0: tcu::surfaceless::ContextFactory::createContext(glu::RenderConfig const&, tcu::CommandLine const&, glu::RenderContext const*) const (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x53EBD1: glu::createRenderContext(tcu::Platform&, tcu::CommandLine const&, glu::RenderConfig const&, glu::RenderContext const*) (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x53EFE9: glu::createDefaultRenderContext(tcu::Platform&, tcu::CommandLine const&, glu::ApiType) (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x1DE07A: deqp::gles2::Context::Context(tcu::TestContext&) (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x1DB5EF: deqp::gles2::TestPackage::init() (in /deqp/modules/gles2/deqp-gles2)
==403== Address 0x56bd340 is 0 bytes inside a block of size 4 free'd
==403== at 0x48369AB: free (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==403== by 0x4B01767: loader_get_driver_for_fd (loader.c:464)
==403== by 0x4AFD553: surfaceless_probe_device (platform_surfaceless.c:308)
==403== by 0x4AFD915: dri2_initialize_surfaceless (platform_surfaceless.c:391)
==403== by 0x4AF5EEA: dri2_initialize (egl_dri2.c:984)
==403== by 0x4AF5EEA: dri2_initialize (egl_dri2.c:958)
==403== by 0x4AF1EEC: _eglMatchAndInitialize (egldriver.c:75)
==403== by 0x4AF1F3B: _eglMatchDriver (egldriver.c:96)
==403== by 0x4AE9367: eglInitialize (eglapi.c:617)
==403== by 0x1D99C9: tcu::surfaceless::EglRenderContext::EglRenderContext(glu::RenderConfig const&, tcu::CommandLine const&) [clone .constprop.57] (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x1DABB0: tcu::surfaceless::ContextFactory::createContext(glu::RenderConfig const&, tcu::CommandLine const&, glu::RenderContext const*) const (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x53EBD1: glu::createRenderContext(tcu::Platform&, tcu::CommandLine const&, glu::RenderConfig const&, glu::RenderContext const*) (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x53EFE9: glu::createDefaultRenderContext(tcu::Platform&, tcu::CommandLine const&, glu::ApiType) (in /deqp/modules/gles2/deqp-gles2)
==403== Block was alloc'd at
==403== at 0x483577F: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==403== by 0x4EE5E09: strndup (strndup.c:43)
==403== by 0x4B010B1: loader_get_kernel_driver_name (loader.c:101)
==403== by 0x4B016AF: loader_get_driver_for_fd (loader.c:462)
==403== by 0x4AFD553: surfaceless_probe_device (platform_surfaceless.c:308)
==403== by 0x4AFD915: dri2_initialize_surfaceless (platform_surfaceless.c:391)
==403== by 0x4AF5EEA: dri2_initialize (egl_dri2.c:984)
==403== by 0x4AF5EEA: dri2_initialize (egl_dri2.c:958)
==403== by 0x4AF1EEC: _eglMatchAndInitialize (egldriver.c:75)
==403== by 0x4AF1F3B: _eglMatchDriver (egldriver.c:96)
==403== by 0x4AE9367: eglInitialize (eglapi.c:617)
==403== by 0x1D99C9: tcu::surfaceless::EglRenderContext::EglRenderContext(glu::RenderConfig const&, tcu::CommandLine const&) [clone .constprop.57] (in /deqp/modules/gles2/deqp-gles2)
==403== by 0x1DABB0: tcu::surfaceless::ContextFactory::createContext(glu::RenderConfig const&, tcu::CommandLine const&, glu::RenderContext const*) const (in /deqp/modules/gles2/deqp-gles2)
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Adam Jackson [Thu, 26 Sep 2019 15:07:42 +0000 (11:07 -0400)]
Revert "glx: Lift sending the MakeCurrent request to top-level code"
Apparently this provokes crashes elsewhere in code unrelated to
MakeCurrent. I hate GLX so very very much.
This reverts commit
999c2aed8826f403b071f52b040ce25b56d35f9d.
Gitlab: https://gitlab.freedesktop.org/mesa/mesa/issues/1207
Adam Jackson [Thu, 26 Sep 2019 15:07:13 +0000 (11:07 -0400)]
Revert "glx: Implement GLX_EXT_no_config_context"
This reverts commit
0d635ccc912d7122f35f81eec27d8b2c0a2a7a28.
Gitlab: https://gitlab.freedesktop.org/mesa/mesa/issues/1207
Timur Kristóf [Wed, 18 Sep 2019 12:39:10 +0000 (14:39 +0200)]
radv: Add debug option to dump meta shaders.
This new option can help debug shader compiler problems when
there are issues with the meta shaders.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Timur Kristóf [Wed, 25 Sep 2019 14:40:07 +0000 (16:40 +0200)]
amd/common: Introduce ac_get_fs_input_vgpr_cnt.
Add a function called ac_get_fs_input_vgpr_cnt which will return
the number of input VGPRs used by an AMD shader. Previously,
radv and radeonsi had the same code duplicated, but this commit also
allows them to share this code.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Timur Kristóf [Fri, 13 Sep 2019 13:53:09 +0000 (15:53 +0200)]
radv: Set shared VGPR count in radv_postprocess_config.
This commit allows RADV to set the shared VGPR count according to
the shader config.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Timur Kristóf [Fri, 13 Sep 2019 13:38:50 +0000 (15:38 +0200)]
amd/common: Add num_shared_vgprs to ac_shader_config for GFX10.
In GFX10 wave64 mode, shared VGPRs allow the two wave halves to
share some data with each other.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Timur Kristóf [Wed, 25 Sep 2019 12:10:18 +0000 (14:10 +0200)]
amd/common: Extract some helper functions to ac_shader_util.
This commit moves ac_get_tbuffer_format, ac_get_sampler_dim and
ac_get_image_dim into ac_shader_util, thus enabling them to be used
by compilers other than LLVM.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Timur Kristóf [Wed, 25 Sep 2019 12:05:19 +0000 (14:05 +0200)]
amd/common: Move ac_export_mrt_z to ac_llvm_build.
The aim of this commit is to keep ac_shader_util LLVM-free,
since we would like to use it in ACO later.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Rhys Perry [Mon, 23 Sep 2019 13:31:24 +0000 (14:31 +0100)]
aco: CSE readlane/readfirstlane/permute/reduce with the same exec mask
v2: rename pass_temp to pass_flags
v2: also CSE reductions
v3: add ds_swizzle_b32 support
v3: check gds/offset0/offset1 fields
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Rhys Perry [Sat, 21 Sep 2019 15:00:45 +0000 (16:00 +0100)]
aco: don't CSE v_readlane_b32/v_readfirstlane_b32
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Rhys Perry [Wed, 25 Sep 2019 10:48:04 +0000 (11:48 +0100)]
aco,radv: rename record_llvm_ir/llvm_ir_string to record_ir/ir_string
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Tue, 24 Sep 2019 14:25:07 +0000 (15:25 +0100)]
radv/aco: return a correct name and description for the backend IR
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Tue, 24 Sep 2019 14:23:46 +0000 (15:23 +0100)]
aco: store printed backend IR in binary
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Tue, 24 Sep 2019 14:21:06 +0000 (15:21 +0100)]
aco,radv/aco: get dissassembly for release builds if requested
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Wed, 25 Sep 2019 11:04:51 +0000 (12:04 +0100)]
radv/aco: actually disable ACO when unsupported
We were setting this twice. The second time, we weren't later disabling
it if unsupported.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tapani Pälli [Tue, 24 Sep 2019 11:34:40 +0000 (14:34 +0300)]
mesa/st: calculate texture size based on EGLImage miplevel
Fixes issues with 'egl-gl_oes_egl_image' Piglit test.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Dylan Baker [Wed, 25 Sep 2019 23:25:27 +0000 (23:25 +0000)]
meson: fix logic for generating .pc files with old glvnd
We want to generate PC files for non-glvnd builds and for builds with
old glvnd, but the current logic doesn't do that, it builds them
unconditionally, and for GLES it builds the shared libraries, which is
also not what we want. This does not generate .pc files for gles1 or
gles2. Which it we weren't doing before either, making this not a
regression but a return to status-quo.o
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/1838
Fixes: 93df862b6affb6b8507e40601212a58012bfa873
("meson: re-add incorrect pkg-config files with GLVND for backward compatibility")
Reviewed-by: Matt Turner <mattst88@gmail.com>
Ian Romanick [Tue, 13 Aug 2019 00:28:35 +0000 (17:28 -0700)]
nir/range-analysis: Use types to provide better ranges from bcsel and mov
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
All Gen7+ platforms had similar results. (Ice Lake shown)
total instructions in shared programs:
16328255 ->
16315391 (-0.08%)
instructions in affected programs: 218318 -> 205454 (-5.89%)
helped: 988
HURT: 0
helped stats (abs) min: 1 max: 72 x̄: 13.02 x̃: 10
helped stats (rel) min: 0.33% max: 16.04% x̄: 6.27% x̃: 4.88%
95% mean confidence interval for instructions value: -13.69 -12.35
95% mean confidence interval for instructions %-change: -6.55% -5.99%
Instructions are helped.
total cycles in shared programs:
363683977 ->
363615417 (-0.02%)
cycles in affected programs:
1475193 ->
1406633 (-4.65%)
helped: 923
HURT: 36
helped stats (abs) min: 1 max: 624 x̄: 75.78 x̃: 48
helped stats (rel) min: 0.08% max: 13.89% x̄: 5.20% x̃: 5.08%
HURT stats (abs) min: 1 max: 179 x̄: 38.58 x̃: 4
HURT stats (rel) min: 0.06% max: 16.56% x̄: 3.33% x̃: 0.29%
95% mean confidence interval for cycles value: -75.88 -67.10
95% mean confidence interval for cycles %-change: -5.10% -4.66%
Cycles are helped.
Sandy Bridge
total instructions in shared programs:
10785779 ->
10785654 (<.01%)
instructions in affected programs: 13855 -> 13730 (-0.90%)
helped: 67
HURT: 0
helped stats (abs) min: 1 max: 15 x̄: 1.87 x̃: 1
helped stats (rel) min: 0.20% max: 3.45% x̄: 0.97% x̃: 0.78%
95% mean confidence interval for instructions value: -2.47 -1.26
95% mean confidence interval for instructions %-change: -1.13% -0.81%
Instructions are helped.
total cycles in shared programs:
153704799 ->
153704481 (<.01%)
cycles in affected programs: 101509 -> 101191 (-0.31%)
helped: 38
HURT: 13
helped stats (abs) min: 1 max: 38 x̄: 12.53 x̃: 16
helped stats (rel) min: 0.07% max: 2.69% x̄: 0.87% x̃: 0.53%
HURT stats (abs) min: 1 max: 36 x̄: 12.15 x̃: 7
HURT stats (rel) min: 0.06% max: 2.53% x̄: 0.73% x̃: 0.44%
95% mean confidence interval for cycles value: -10.24 -2.24
95% mean confidence interval for cycles %-change: -0.75% -0.17%
Cycles are helped.
LOST: 2
GAINED: 0
No shader-db change on Iron Lake or GM45.
Ian Romanick [Tue, 13 Aug 2019 00:28:35 +0000 (17:28 -0700)]
nir/range-analysis: Use types in the hash key
This allows the reslut of mov and bcsel to be separately interpreted as
float or int depending on the use.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Ian Romanick [Tue, 24 Sep 2019 22:55:49 +0000 (15:55 -0700)]
nir/range-analysis: Bail if the types don't match
Some shaders are hurt by this change because now a
load_const(0x00000000) is not recognized as eq_zero when loaded as a
float. This behavior is restored in a later patch (nir/range-analysis:
Use types to provide better ranges from bcsel and mov).
v2: Add a comment about reinterpretation of int/uint/bool. Suggested by
Caio. Rewrite condition the check for types being float versus checking
for types not being all the things that aren't float.
Fixes: 405de7ccb6c ("nir/range-analysis: Rudimentary value range analysis pass")
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
All Gen7+ platforms had similar results. (Ice Lake shown)
total instructions in shared programs:
16327543 ->
16328255 (<.01%)
instructions in affected programs: 55928 -> 56640 (1.27%)
helped: 0
HURT: 208
HURT stats (abs) min: 1 max: 16 x̄: 3.42 x̃: 3
HURT stats (rel) min: 0.33% max: 6.74% x̄: 1.31% x̃: 1.12%
95% mean confidence interval for instructions value: 3.06 3.79
95% mean confidence interval for instructions %-change: 1.17% 1.46%
Instructions are HURT.
total cycles in shared programs:
363682759 ->
363683977 (<.01%)
cycles in affected programs: 325758 -> 326976 (0.37%)
helped: 44
HURT: 133
helped stats (abs) min: 1 max: 179 x̄: 33.61 x̃: 5
helped stats (rel) min: 0.06% max: 14.21% x̄: 2.47% x̃: 0.29%
HURT stats (abs) min: 1 max: 157 x̄: 20.28 x̃: 14
HURT stats (rel) min: 0.07% max: 14.44% x̄: 1.42% x̃: 0.73%
95% mean confidence interval for cycles value: 0.38 13.39
95% mean confidence interval for cycles %-change: -0.06% 0.96%
Inconclusive result (%-change mean confidence interval includes 0).
Sandy Bridge
total instructions in shared programs:
10787433 ->
10787443 (<.01%)
instructions in affected programs: 1842 -> 1852 (0.54%)
helped: 0
HURT: 10
HURT stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
HURT stats (rel) min: 0.33% max: 1.85% x̄: 0.73% x̃: 0.49%
95% mean confidence interval for instructions value: 1.00 1.00
95% mean confidence interval for instructions %-change: 0.36% 1.10%
Instructions are HURT.
total cycles in shared programs:
153724543 ->
153724563 (<.01%)
cycles in affected programs: 8407 -> 8427 (0.24%)
helped: 1
HURT: 3
helped stats (abs) min: 18 max: 18 x̄: 18.00 x̃: 18
helped stats (rel) min: 0.98% max: 0.98% x̄: 0.98% x̃: 0.98%
HURT stats (abs) min: 4 max: 18 x̄: 12.67 x̃: 16
HURT stats (rel) min: 0.21% max: 0.75% x̄: 0.56% x̃: 0.72%
95% mean confidence interval for cycles value: -21.31 31.31
95% mean confidence interval for cycles %-change: -1.11% 1.46%
Inconclusive result (value mean confidence interval includes 0).
No shader-db changes on Iron Lake or GM45.
Lionel Landwerlin [Wed, 25 Sep 2019 14:43:07 +0000 (17:43 +0300)]
intel: Add new Comet Lake PCI-ids
Commit
bfc4c359b282 ("drm/i915/cml: Add Missing PCI IDs") in i915
added 3 new CML PCI ids.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Lionel Landwerlin [Wed, 25 Sep 2019 14:40:50 +0000 (17:40 +0300)]
intel: use proper label for Comet Lake skus
Fixes: 82f6a746e8 ("intel: Add support for Comet Lake")
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Kristian H. Kristensen [Fri, 20 Sep 2019 17:10:57 +0000 (10:10 -0700)]
freedreno/a6xx: Move instrlen and obj_start writes to fd6_emit_shader
Consolidate a few more generic shaders setup regs in fd6_emit_shader.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Kristian H. Kristensen [Thu, 19 Sep 2019 22:04:09 +0000 (15:04 -0700)]
freedreno/a6xx: Emit const and texture state for HS/DS/GS
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Kristian H. Kristensen [Thu, 19 Sep 2019 20:59:36 +0000 (13:59 -0700)]
freedreno/ir3: Add HS/DS/GS to shader key and cache
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Kristian H. Kristensen [Thu, 19 Sep 2019 20:55:35 +0000 (13:55 -0700)]
freedreno/a6xx: Add generic program stateobj support for HS/DS/GS
This add generic stage state setup for HS/DS/GS to the program state
object.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Kristian H. Kristensen [Thu, 19 Sep 2019 20:45:44 +0000 (13:45 -0700)]
freedreno: Move fs functions after geometry pipeline stages
Let's try to always order the stages in the pipeline order.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Kristian H. Kristensen [Thu, 19 Sep 2019 20:25:00 +0000 (13:25 -0700)]
freedreno: Add state binding functions for HS/DS/GS
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Kristian H. Kristensen [Thu, 19 Sep 2019 20:40:31 +0000 (13:40 -0700)]
freedreno: Rename vp and fp to vs and fs in fd_program_stateobj
We're using vs and fs now, and adding hs, ds and gs soon. It's
confusing enough that we have both DS/TCS and HS/TES. At least for VS
and FS there doesn't have to be multiple names.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Kristian H. Kristensen [Thu, 19 Sep 2019 20:19:34 +0000 (13:19 -0700)]
freedreno/a6xx: Factor out const state setup
We'll be sharing this logic for new shader stages soon.
Signed-off-by: Kristian H. Kristensen <hoegsberg@google.com>
Eric Engestrom [Tue, 24 Sep 2019 15:58:31 +0000 (16:58 +0100)]
glsl: turn runtime asserts of compile-time value into compile-time asserts
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Eric Engestrom [Wed, 25 Sep 2019 18:20:00 +0000 (19:20 +0100)]
docs/release-calendar: add missing <td> and </td>
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Wed, 25 Sep 2019 18:19:05 +0000 (19:19 +0100)]
docs/release-calendar: fix bugfix release numbers
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Lionel Landwerlin [Wed, 25 Sep 2019 13:26:52 +0000 (16:26 +0300)]
anv: gem-stubs: return a valid fd got anv_gem_userptr()
Fixes invalid close(-1) in the unit tests.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Danylo Piliaiev [Tue, 24 Sep 2019 11:12:39 +0000 (14:12 +0300)]
st/nine: Ignore D3DSIO_RET if it is the last instruction in a shader
RET as a last instruction could be safely ignored.
Remove it to prevent crashes/warnings in case underlying driver
doesn't implement arbitrary returns.
A better way would be to remove the RET after the whole shader
is parsed which will handle a possible case when the last RET is
followed by a comment.
CC: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Axel Davy <davyaxel0@gmail.com>
Dylan Baker [Mon, 23 Sep 2019 18:05:16 +0000 (11:05 -0700)]
bin/get-pick-list: use --oneline=pretty instead of --oneline
--oneline shortens hashes, while --oneline=pretty doesn't, otherwise
they are the same. Having full hashes is convenient as that is the
format that the bin/.cherry-ignore script requires to work correctly.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Dylan Baker [Fri, 20 Sep 2019 17:37:14 +0000 (10:37 -0700)]
release: Push 19.3 back two weeks
The main reason to do this is that 19.2 has slipped by two weeks, and
such the 19.3 branch is due to happen extremely close to the release of
19.2.0. I think it would be better to have a little more time between
releases for developers and for packagers.
This would still have the 19.3 release out before December, even if it
slips by 1 week.
Acked-By: Karol Herbst <kherbst@redhat.com>
Acked-by: Juan A. Suarez <jasuarez@igalia.com>
Dylan Baker [Wed, 25 Sep 2019 17:40:00 +0000 (10:40 -0700)]
docs: update calendar, add news item, and link release notes for 19.2.0
Dylan Baker [Wed, 25 Sep 2019 17:39:51 +0000 (10:39 -0700)]
docs: add SHA256 sum for 19.2.0
Dylan Baker [Wed, 25 Sep 2019 16:55:33 +0000 (09:55 -0700)]
docs: Add release notes for 19.2.0
Andreas Baierl [Thu, 19 Sep 2019 06:53:18 +0000 (08:53 +0200)]
lima/ppir: Add various varying fetch sources to disassembler
Signed-off-by: Andreas Baierl <ichgeh@imkreisrum.de>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Eric Engestrom [Thu, 19 Sep 2019 13:18:55 +0000 (14:18 +0100)]
meson: re-add incorrect pkg-config files with GLVND for backward compatibility
This is a bit counter-intuitive, but the issue is that GLVND is broken
in versions <= 1.1.1, so we need to keep wrongly providing these files
to cover up their mistake, otherwise the rest of the world ends up
broken.
Suggested-by: Dylan Baker <dylan@pnwbakers.com>
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Rhys Perry [Tue, 24 Sep 2019 14:46:37 +0000 (15:46 +0100)]
aco: check for duplicate opcode numbers
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Sat, 21 Sep 2019 13:10:38 +0000 (14:10 +0100)]
aco: fix opcode for s_mul_hi_i32
Fixes dEQP-VK.glsl.builtin.function.integer.imulextended.*_compute
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Tue, 24 Sep 2019 14:45:48 +0000 (15:45 +0100)]
aco: fix v_subrev_co_u32_e64 opcode
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Tue, 24 Sep 2019 13:34:28 +0000 (14:34 +0100)]
aco: fix GFX9 opcode for v_xad_u32
Fixes various dEQP-VK.image.store.* tests.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Wed, 25 Sep 2019 11:16:34 +0000 (12:16 +0100)]
aco: implement 64-bit ineg
We currently lower them, but nir_opt_algebraic() can add new ones because
lower_sub=true.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Rhys Perry [Tue, 24 Sep 2019 14:15:26 +0000 (15:15 +0100)]
aco: run nir_lower_int64() before nir_lower_idiv()
nir_lower_idiv() asserts on 64-bit integers.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Connor Abbott [Tue, 24 Sep 2019 15:29:53 +0000 (17:29 +0200)]
nir: Fix overlapping vars in nir_assign_io_var_locations()
When handling two variables with overlapping locations, we process the
one with lower location first, and then extend the location ->
driver_location map to guarantee that it's contiguous for the second
variable too. But the loop had the wrong bound, so we weren't extending
the map 100%, which could lead to problems later such as an incorrect
num_inputs. The loop index i is an index into the slots of the variable,
so we need to stop at the final slot of the variable (var_size) instead
of the number of unassigned slots.
This fixes
spec@arb_enhanced_layouts@execution@component-layout@vs-fs-array-interleave-range
on radeonsi NIR.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Karol Herbst [Fri, 20 Sep 2019 11:08:50 +0000 (13:08 +0200)]
clover: eliminate "ignoring attributes on template argument" warning
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Karol Herbst [Fri, 20 Sep 2019 10:45:11 +0000 (12:45 +0200)]
clover/codegen: remove unused get_symbol_offsets function
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Karol Herbst [Fri, 20 Sep 2019 10:43:10 +0000 (12:43 +0200)]
clover/llvm: remove harmful std::move call
both clang and gcc warn with:
"moving a local object in a return statement prevents copy elision"
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Tapani Pälli [Mon, 2 Sep 2019 10:02:33 +0000 (13:02 +0300)]
iris: disable aux on first get_param if not created with aux
This moves the fix from commit
361f3d19f1f to happen in get_param
(used now instead of get_handle by st/dri). This fixes artifacts
seen with Xorg and CCS_E.
Fixes: fc12fd05f56 "iris: Implement pipe_screen::resource_get_param"
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Erik Faye-Lund [Tue, 24 Sep 2019 14:57:03 +0000 (16:57 +0200)]
glsl: correct bitcast-helpers
Without this, we'll incorrectly round off huge values to the nearest
representable double instead of keeping it at the exact value as
we're supposed to.
Found by inspecting compiler-warnings.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Fixes: 85faf5082f ("glsl: Add 64-bit integer support for constant expressions")
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Vasily Khoruzhick [Sat, 14 Sep 2019 18:00:16 +0000 (11:00 -0700)]
lima/ppir: add support for indirect load of uniforms and varyings
Utgard PP supports indirect load of uniforms and varyings, so let's
enable it.
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Vasily Khoruzhick [Sat, 14 Sep 2019 18:01:03 +0000 (11:01 -0700)]
lima/ppir: add node dependency types
Currently we add dependecies in 3 cases:
1) One node consumes value produced by another node
2) Sequency dependencies
3) Write after read dependencies
2) and 3) only affect scheduler decisions since we still can use pipeline
register if we have only 1 dependency of type 1).
Add 3 dependency types and mark dependencies as we add them.
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Vasily Khoruzhick [Tue, 24 Sep 2019 04:20:07 +0000 (21:20 -0700)]
lima/ppir: don't attempt to clone tex coords if it's not varying
It makes no sense to clone texture coords if it's not varying, moreover
we don't support cloning ALU nodes.
Fixes: 1c1890fa7077 ("lima/ppir: clone uniforms and load_coords into each successor")
Reviewed-by: Andreas Baierl <ichgeh@imkreisrum.de>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Timothy Arceri [Fri, 20 Sep 2019 06:54:31 +0000 (16:54 +1000)]
radeonsi/nir: lower load constants to scalar
We call nir_lower_load_const_to_scalar in the state trackers linker
however some later passes can reintroduce constant vectors. Here
we lower these to scalar and perform optimisations. The Intel
drivers do a similar call in their backend..
shader-db results VEGA 64:
Totals from affected shaders:
SGPRS: 152168 -> 151976 (-0.13 %)
VGPRS: 135224 -> 135112 (-0.08 %)
Spilled SGPRs: 4027 -> 4163 (3.38 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size:
10670028 ->
10654776 (-0.14 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 13122 -> 13135 (0.10 %)
Wait states: 0 -> 0 (0.00 %)
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Jonathan Marek [Tue, 24 Sep 2019 18:39:55 +0000 (14:39 -0400)]
turnip: use image tile_mode for gmem configuration
Fixes at least this deqp test:
dEQP-VK.api.smoke.triangle
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Jonathan Marek [Tue, 24 Sep 2019 18:36:53 +0000 (14:36 -0400)]
turnip: fix binning shader compilation
ir3 segfaults if nonbinning is NULL for the bininng pass shader.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Rhys Perry [Mon, 23 Sep 2019 13:48:22 +0000 (14:48 +0100)]
nir/opt_remove_phis: handle phis with no sources
This can happen with loops with unreachable exits which are later
optimized away.
Fixes assertion in dEQP-VK.graphicsfuzz.unreachable-loops with RADV.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Michel Dänzer [Thu, 19 Sep 2019 00:18:39 +0000 (20:18 -0400)]
radeonsi: fix VAAPI segfault due to various bugs
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111236
Marek Olšák [Tue, 17 Sep 2019 22:22:08 +0000 (18:22 -0400)]
gallium/vl: don't set PIPE_HANDLE_USAGE_EXPLICIT_FLUSH
because vl doesn't call flush_resource and I wasn't able to find
all places where flush_resource needs to be called.
This fixes corrupted / unflushed surfaces with fullscreen videos on Raven.
Cc: 19.1 19.2 <mesa-stable@lists.freedesktop.org>
Marek Olšák [Tue, 17 Sep 2019 02:31:48 +0000 (22:31 -0400)]
radeonsi: initialize displayable DCC using the retile blit to prevent hangs
Cc 19.2 <mesa-stable@lists.freedesktop.org>
Connor Abbott [Tue, 24 Sep 2019 10:43:29 +0000 (12:43 +0200)]
nir/opt_large_constants: Handle store writemasks
This fixes some piglit tests on radeonsi NIR where a varying is
initialized to a constant array in the vertex shader. Varying packing
after nir_lower_io_to_temporaries creates writemasked stores which
persist after pulling the constant initialization down into the fragment
shader.
While we're here, rewrite handle_constant_store() to do the loop over
components outside the switch, so that we don't have to duplicate the
writemask checking for every bitsize.
Fixes: 1235850522c ("nir: Add a large constants optimization pass")
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Eric Engestrom [Mon, 23 Sep 2019 17:53:22 +0000 (18:53 +0100)]
meson: split more compiler options to their own line
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Mon, 23 Sep 2019 17:37:01 +0000 (18:37 +0100)]
meson: drop -Wno-foo bug workaround for Meson < 0.46
This was a workaround for a bug in Meson that was fixed in 0.46 [1].
[1] https://github.com/mesonbuild/meson/pull/2284
Fixes: f7b6a8d12fdc446e3251 ("meson: bump required version to 0.46")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Mon, 23 Sep 2019 16:48:55 +0000 (17:48 +0100)]
radv: fix s/load/store/ copy-paste typo
Fixes: cdc6efddf918bc07d30d ("radv: implement all depth/stencil resolve modes using graphics")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Stephen Barber [Tue, 24 Sep 2019 00:51:43 +0000 (17:51 -0700)]
nouveau: add idep_nir_headers as dep for libnouveau
Fixes a compilation error when building libnouveau:
In file included from ../src/gallium/drivers/nouveau/nv50/nv50_program.c:25:
../src/compiler/nir/nir.h:1115:10: fatal error: nir_intrinsics.h: No such file or directory
#include "nir_intrinsics.h"
^~~~~~~~~~~~~~~~~~
compilation terminated.
Fixes: f014ae3c7cce504afe5d ("nouveau: add support for nir")
Signed-off-by: Stephen Barber <smbarber@chromium.org>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 00:53:21 +0000 (02:53 +0200)]
radv: Add workaround for hang in The Surge 2.
Released today and hangs on RADV. We don't have the root cause yet,
but this should unblock people playing the game.
No drirc because the radv debugflags are not usable from drirc and
I want this backported.
CC: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Andres Gomez [Mon, 23 Sep 2019 22:37:57 +0000 (01:37 +0300)]
i965/fs: set rounding mode when emitting the flrp instruction
flrp was forgotten when already adding the rounding mode for other
instructions.
Fixes: ba1e25e1aa6 ("i965/fs: set rounding mode when emitting fadd, fmul and ffma instructions")
Suggested-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Andres Gomez [Mon, 23 Sep 2019 22:16:11 +0000 (01:16 +0300)]
i965/fs: add a comment about how the rounding mode in fmul is set
After
1711bf6cf2d ("intel/fs: Generate better code for fsign multiplied by a value"),
the conflicts resolution for setting the rounding mode after the
fused fmul and fsign optimization is non obvious.
Basically, the optimization doesn't really result in a MUL, or any
other operation which would need to have the rounding mode set. Hence,
we set it just before the actual MUL in the treatment of fmul.
Fixes: ba1e25e1aa6 ("i965/fs: set rounding mode when emitting fadd, fmul and ffma instructions")
Suggested-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Juan A. Suarez Romero [Tue, 10 Sep 2019 08:30:43 +0000 (10:30 +0200)]
bin/get-pick-list.sh: sha1 commits can be smaller than 8 chars
The script only handles commits with "Fixes: <sha1>" where <sha1> is
equal or great than 8 chars. But <sha1> can be smaller, like 7 chars.
This commit relax the restriction to handle <sha1> 4 or more chars.
Fixes: 533fead4236 ("bin/get-pick-list.sh: tweak the commit sha matching pattern")
Acked-by: Eric Engestrom <eric.engestrom@intel.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Connor Abbott [Wed, 18 Sep 2019 17:47:28 +0000 (00:47 +0700)]
lima/gpir: Fix 64-bit shift in scheduler spilling
There are 64 physical registers so the shift must be 64 bits.
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Connor Abbott [Wed, 18 Sep 2019 11:13:08 +0000 (18:13 +0700)]
lima/gpir: Don't emit movs when translating from NIR
The scheduler doesn't expect them. To do this, I had to refactor the
registration part of gpir_node_create_dest() to be separate from
creating and inserting the node, since the last two now aren't done when
handling moves. This adds more code but creates the possibility of
automatically inserting input dependencies when inserting nodes, similar
to what's done in NIR with the use-def lists (this isn't done yet).
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Connor Abbott [Wed, 18 Sep 2019 05:29:45 +0000 (12:29 +0700)]
lima/gpir: Fix postlog2 fixup handling
We guarantee that a complex1 op is always used by postlog2 directly by
rewriting the postlog2 op to be a move when there would be a move
inserted between them. But we weren't doing this in all circumstances
where there might be a move. Move the logic to place_move() so that it
always happens. Fixes a few log tests that happened to start failing due
to changes in the register allocator leading to a different scheduling
order.
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Connor Abbott [Fri, 13 Sep 2019 06:23:56 +0000 (13:23 +0700)]
lima/gpir: Use registers for values live in multiple blocks
This commit adds the framework for cross-basic-block register
allocation. Like ARM's compiler, we assume that the value registers
aren't usable across branches, which means we have to use physical
registers to store any value that crosses a basic block. There are three
parts to this:
1. When translating from NIR, we rely on the NIR out-of-ssa pass to
coalesce values into registers. We insert store_reg instructions for
values used in more than one basic block, and load_reg instructions for
values not defined in the same basic block (or defined after their use,
for loops). So by the time we've translated out of NIR we've already
split things into values (which are only used in the same basic block)
and registers (which are only used in different basic blocks than where
they're defined).
2. We allocate the registers at the same time that we allocate the
values, before the final scheduler. Unlike the values, where the
assigned color is fake, we assign the actual physical index & component
to physregs at this stage. load_reg and store_reg are treated as moves
in the allocator and when creating write-after-read dependencies.
3. Finally, in the main scheduler we have to avoid overwriting existing
live physregs when spilling. First, we have to tell the scheduler which
physical registers are live at the end of each block, to avoid
overwriting those. If a register is only live at the beginning, we can
reuse it for spilling after the last original use in the final program
happens, i.e. before any original use is scheduled, but we have to be
careful to add the proper dependencies so that the spill write is
scheduled before the original reads. To handle this we repurpose
reg_link for uses to be used by the scheduler.
A few register-related things copied over from NIR or from other
drivers can be dropped.
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Connor Abbott [Wed, 28 Aug 2019 08:57:35 +0000 (10:57 +0200)]
lima/gpir: Support branch instructions
Because branch conditions have to be in the pass slot, there is no
unconditional branch, and realistically the pass slot has to contain a
move when branching (there's nothing it does that would be useful for
operating on booleans, so we can't use it for anything when computing
the branch condition), we put the branch instruction in the pass slot
and at codegen time turn it into a move of the branch condition. This
means that it doesn't have to be special-cased like store instructions
are in the scheduler. Because of this decision we can remove the
half-implemented BRANCH codegen slot. Finally, we (ab)use the existing
schedule_first mechanism to make sure that branches are always last in
the basic block.
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Connor Abbott [Tue, 10 Sep 2019 14:11:42 +0000 (21:11 +0700)]
lima/gpir: Only try to place actual children
When picking a node to be scheduled, we try to schedule its children as
well. But we shouldn't try to schedule nodes which only have a fake
dependency on the original node, since this isn't the point of
scheduling children at the same time and can break some expectations of
the rest of the code.
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Connor Abbott [Fri, 13 Sep 2019 06:18:29 +0000 (13:18 +0700)]
lima/gpir: Fix compiler warning
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>