Christian Gmeiner [Thu, 3 Oct 2019 20:14:20 +0000 (22:14 +0200)]
etnaviv: nir: start to make use of compile_error(..)
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Michel Dänzer [Fri, 11 Oct 2019 09:10:01 +0000 (11:10 +0200)]
gitlab-ci: Disable meson-mingw32-x86_64 job again for now
The wrapdb.mesonbuild.com SSL certificate expired, causing the job to
fail: https://gitlab.freedesktop.org/mesa/mesa/-/jobs/731864
Switching to http:// doesn't avoid it:
https://gitlab.freedesktop.org/daenzer/mesa/-/jobs/732043
Michel Dänzer [Thu, 10 Oct 2019 13:27:17 +0000 (15:27 +0200)]
gitlab-ci: Add .use-debian-10 template
It simplifies the definitions of jobs using the Debian 10 image.
The needs: was previously missing from the llvmpipe/softpipe test jobs,
so they could spuriously run if the debian-10 job failed or was
cancelled.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Michel Dänzer [Tue, 8 Oct 2019 13:58:27 +0000 (15:58 +0200)]
gitlab-ci: Remove redundant .meson-cross template script
It was identical to the one inherited from the .meson-build template.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Dave Airlie [Thu, 10 Oct 2019 03:27:30 +0000 (13:27 +1000)]
gallivm: fix coroutines on aarch64 with llvm 8
The coroutine split pass is missing a dependency before LLVM 9.0,
and fails to initialise properly if the CallGraphWrapperPass hasn't
be initialised earlier (x86 does it due to some of it's passes
requiring it).
This is a workaround for llvm 8 (coroutines are only supported in 8
and higher). It adds another pass that has a dependency on the pass
the coroutines split requires. This pass shouldn't have any raal
effects.
Fixes: d32690b43c9 (gallivm: add coroutine pass manager support)
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Dave Airlie [Wed, 9 Oct 2019 03:12:31 +0000 (13:12 +1000)]
llvmpipe: add support for tg4 component selection.
This is needed as part of GLES3.1 and helps for ARB_gpu_shader5.
Fixes: KHR-GLES31.core.texture_gather.* cases
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Dave Airlie [Wed, 9 Oct 2019 03:09:00 +0000 (13:09 +1000)]
st/glsl: add support for alternate TG4 encoding.
This will encode the component selection value (0, 1, 2, 3) into
the X swizzle of the sampler, if the driver requests it.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Dave Airlie [Wed, 9 Oct 2019 03:07:04 +0000 (13:07 +1000)]
gallium: add a a new cap for changing the TGSI TG4 instruction encoding
Accessing the TG4 component via immediates in the llvmpipe backend is quite
messy (like really messy). Roland suggested we change the instruction encoding,
so introduce a cap to allow the component to be selected to be store in the
sampler swizzle, which should be otherwise unused.
I could probably switch all drivers over, but virgl would need some work that
I'd prefer not to rush it.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Dave Airlie [Fri, 30 Aug 2019 00:30:00 +0000 (10:30 +1000)]
gallivm/sample: add gather component selection to the key.
This allows for component selection to work as per ARB_gpu_shader5/GLES3.1
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Roland Scheidegger [Thu, 10 Oct 2019 18:10:34 +0000 (20:10 +0200)]
llvmpipe: increase max texture size to 2GB
The 1GB limit was arbitrary, increase this to 2GB (which is the max
possible without code changes).
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Dylan Baker [Thu, 19 Sep 2019 17:21:51 +0000 (10:21 -0700)]
gitlab-ci: Add a mingw x86_64 job
Acked-by: Eric Engestrom <eric.engestrom@intel.com> (v1)
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 23 Jan 2019 22:21:26 +0000 (14:21 -0800)]
appveyor: Add support for meson as well as scons on windows
This job uses the vs2017 backend of meson (msbuild) as opposed to the
ninja backend used on MacOS and Linux.
v7: - rebase on master
- remove llvm (we'll add that back later)
- remove cygwin (we'll add that back later too)
v6: - rebase on master, including the addition of cygwin
- consolidate 3 appveyor patches into this one patch
v5 - use the new b_vscrt option instead of manually specifying the crt
v4: - rebase on python3 generators
- cache meson wraps
- Build x86 instead of x86_64, since that's what the pre-built LLVM
is
- update to vs2017 from vs2015
- set the default-library to static
- use the new vscrt override
- add the /m switch to msbuild to make the build somewhat faster
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 11 Jun 2019 18:25:31 +0000 (11:25 -0700)]
docs: update meson docs for windows
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Mon, 7 Oct 2019 20:03:58 +0000 (13:03 -0700)]
glsl/tests: Handle no-exec errors
Currently meson doesn't correctly handle passing compiled binaries to
scripts in tests. This patch looks to the future (0.53) when meson will
have this functionality, but also immediately it fixes these tests in
cross compiles by causing them to return 77, which meson interprets as
skip.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 8 Oct 2019 20:38:01 +0000 (13:38 -0700)]
meson/util: Don't run string_buffer tests on mingw
They succeed with MSVC but not with MinGW. I don't understand why they
fail.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Fri, 20 Sep 2019 20:09:54 +0000 (13:09 -0700)]
meson: glcpp tests are expected to fail on windows
v2: - Exclude the tests rather than xfail them
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Fri, 20 Sep 2019 19:19:52 +0000 (12:19 -0700)]
meson: only build timspec test if timespec is available
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Thu, 27 Jun 2019 21:52:40 +0000 (14:52 -0700)]
meson: don't error on formaters with mingw
MSVC is generally happy, but mingw errors. I've spent as much time
(several days) trying to squash all of these warnings and I'm done with
it, just leave them as warnings with MinGW.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 29 Jan 2019 00:07:20 +0000 (16:07 -0800)]
meson: add msvc compat args to swr
This has always been present in the scons build, so it should be in
the meson build as well.
Acked-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 10 Jul 2018 18:09:27 +0000 (11:09 -0700)]
meson: maintain names of shared API libraries
Mesa uses the lib prefix, and doesn't use a version for it's dynamic
libraries, which meson defaults to.
v2: - this patch
Acked-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 23 May 2018 00:32:21 +0000 (17:32 -0700)]
meson: don't build or run mesa-sha1 test on windows
It crashes hard (pop-up window and all).
v2: - Change comment to FIXME
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 23 May 2018 00:22:45 +0000 (17:22 -0700)]
meson: disable graw tests on mingw
I can't figure out why symbols are being exposed that shouldn't.
v2: - change comment to FIXME
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 23 May 2018 00:06:35 +0000 (17:06 -0700)]
meson: don't build gallium trivial tests on windows
They require the pipe-loaders, which require xmlconfig, which doesn't
build with msvc.
Acked-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 23 May 2018 00:06:09 +0000 (17:06 -0700)]
meson: Set visibility and compat args for graw
Acked-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 22 May 2018 22:00:35 +0000 (15:00 -0700)]
meson: Add msvc compat args to util/tests
To keep this building with msvc
Acked-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 22 May 2018 21:57:14 +0000 (14:57 -0700)]
meson: Add idep_getopt for tests
There are quite a few tests that require getopt, when using MSVC we need
to use the bundled version of getopt since there isn't a system version.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 22 May 2018 21:08:36 +0000 (14:08 -0700)]
meson: don't define USE_ELF_TLS for windows
Because the macros for exporting dll symbols and using TLS are mutually
exclusive.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Tue, 22 May 2018 17:54:12 +0000 (10:54 -0700)]
meson: add switches for SWR with MSVC
This makes two changes for SWR,
The first is that it reorders the arguments to try to put the ICL ones
first. This is required to support older versions of meson that don't
add enough "error in this case" switches to ICL, which causes it to
happy accept -mavx (for example) even though it doesn't support them,
resulting in compilation failures.
The second is to fix the names of the libraries, setting the soversion
to '' will result in <lib>.dll, instead of <lib>-0.dll. Since these are
not versioned dll's, but implement an internal API we should communicate
that. It's also what scons does.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Thu, 17 May 2018 21:45:10 +0000 (14:45 -0700)]
meson: disable sse4.1 optimizations with msvc
There isn't an obvious command line switch here, /arch:AVX *might* be
the right thing, but meson doesn't know what to do here either and
leaves the -msse4.1 and -mstackrealign.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 23 May 2018 17:05:20 +0000 (10:05 -0700)]
meson: force inclusion of inttypes.h for glcpp with msvc
Because we provide a copy if MSVC doesn't, and we need it to make flex
do what we want.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Thu, 17 May 2018 20:42:50 +0000 (13:42 -0700)]
meson: Add support for using win_flex and win_bison on windows
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Thu, 17 May 2018 20:32:12 +0000 (13:32 -0700)]
meson: don't look for rt on windows
v6: - Minor refactor
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Thu, 17 May 2018 18:52:26 +0000 (11:52 -0700)]
meson: fix pipe-loader compilation for windows
v2: - Add missing D to pound define
- Simply define the variable rather than set it to 1 (mirrors
android.mk not scons)
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Thu, 17 May 2018 18:33:35 +0000 (11:33 -0700)]
util/xmlconfig: include strndup.h for windows
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:54:12 +0000 (13:54 -0700)]
meson: Don't check for posix_memalign on windows
There's a mingw bug for this, it exports __builtin_posix_memalign but
not posix_memalign, so the check will succeed, but compiling will fail.
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:29:26 +0000 (13:29 -0700)]
meson: fix gallium-osmesa to build for windows
v2: - set so_version to '' (only affects windows)
- always set lib prefix to 'lib', even on msvc
v5: - key NO_EXPORTS on shared glapi instead of gles.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:29:04 +0000 (13:29 -0700)]
meson: build graw-gdi target
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:28:38 +0000 (13:28 -0700)]
meson: build libgl-gdi target
v4: - Fix check for broken mingw (should be for x86 not x86_64)
- Add comment about why check is needed
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:28:04 +0000 (13:28 -0700)]
meson: build wgl state tracker
v4: - Handle enable gles properly
- Add comments about what various #defines do
v5: - key NO_EXPORTS on shared glapi instead of gles.
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:27:13 +0000 (13:27 -0700)]
meson: build gallium gdi winsys
v6: - use null_dep instead of []
Reviewed-by: Eric Anholt <eric@anholt.net> (v5)
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:23:48 +0000 (13:23 -0700)]
meson: Add necessary defines for mesa_gallium on windows
v4: - Retain scons comments for windows specific defines
v5: - key GLAPI_NO_EXPORTS off of shared-glapi instead of gles
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Wed, 18 Apr 2018 20:22:35 +0000 (13:22 -0700)]
meson: Add windows defines to glapi
These are needed to control the export or symbols due to differences
between the way windows and *nix handle symbol exports.
Reviewed-by: Eric Anholt <eric@anholt.net> (v2)
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
v5: - key NO_EXPORT off of shared-glapi instead of gles
Dylan Baker [Wed, 18 Apr 2018 20:19:54 +0000 (13:19 -0700)]
meson: add windows compiler checks and libraries
v4: - Fix typo in warning code (4246 -> 4267)
- Copy comments from scons for what MSVC warnings codes do
- Merge linker argument changes into this commit
v5: - Add /GR- on windows if LLVM is build without rtti (equivalent to
GCc's -fno-rtti')
- Add /wd4291, which is catching the same hting that
-Wno-non-virtual-dtor is on GCC/Clang
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Dylan Baker [Thu, 19 Sep 2019 18:04:21 +0000 (11:04 -0700)]
util: use _WIN32 instead of WIN32
MinGW defines only _WIN32, but doesn't have fcntl, so we need to use the
windows path.
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Acked-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Wed, 9 Oct 2019 16:47:13 +0000 (09:47 -0700)]
freedreno/ir3: handle multi component alu src when propagating shifts
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Wed, 9 Oct 2019 16:15:45 +0000 (09:15 -0700)]
freedreno/ir3: drop unused param
Signed-off-by: Rob Clark <robdclark@chromium.org>
Marek Olšák [Thu, 10 Oct 2019 22:43:47 +0000 (18:43 -0400)]
clover: fix the nir_serialize build failure
Fixes: dd4cc56ebd0 "nir: add a strip parameter to nir_serialize"
Dave Airlie [Wed, 9 Oct 2019 05:11:14 +0000 (15:11 +1000)]
llvmpipe/draw: handle UBOs that are < 16 bytes.
Not sure if this is a bug in the user or not, but some CTS
tests fail due to using an 8 byte constant buffer.
Fixes: KHR-GLES31.core.layout_binding.block_layout_binding_block_VertexShader
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Dave Airlie [Wed, 9 Oct 2019 04:13:19 +0000 (14:13 +1000)]
llvmpipe/draw: fix image sizes for vertex/geometry shaders.
since images are a single level, minify before passing the w/h
to draw.
Fixes: KHR-GLES31.core.shader_image_size.basic-nonMS-vs-*
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Dave Airlie [Sun, 1 Sep 2019 23:52:55 +0000 (09:52 +1000)]
llvmpipe: make texture buffer offset alignment == 16
Due to use vmovdqa instructions in the asm, which require 16-byte
aligned buffers.
This fixes a crash in
KHR-GLES31.core.texture_buffer.texture_buffer_texture_buffer_range
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Eric Engestrom [Mon, 30 Sep 2019 13:08:29 +0000 (14:08 +0100)]
meson: skip installation of GLVND-provided headers
Fixes: 93df862b6affb6b8507e ("meson: re-add incorrect pkg-config files with GLVND for backward compatibility")
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/1846
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Mon, 30 Sep 2019 13:06:07 +0000 (14:06 +0100)]
meson: split Mesa headers as a separate installation
Fixes: 93df862b6affb6b8507e ("meson: re-add incorrect pkg-config files with GLVND for backward compatibility")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Mon, 30 Sep 2019 13:00:51 +0000 (14:00 +0100)]
meson: split headers one per line
Fixes: 93df862b6affb6b8507e ("meson: re-add incorrect pkg-config files with GLVND for backward compatibility")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Mon, 30 Sep 2019 12:40:22 +0000 (13:40 +0100)]
meson: move a couple of include installs around
Preparation for a later commit.
Fixes: 93df862b6affb6b8507e ("meson: re-add incorrect pkg-config files with GLVND for backward compatibility")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Mon, 30 Sep 2019 09:26:43 +0000 (10:26 +0100)]
meson: rename `glvnd_missing_pc_files` to `not glvnd_has_headers_and_pc_files`
This reflects better what is provided by glvnd or not.
Fixes: 93df862b6affb6b8507e ("meson: re-add incorrect pkg-config files with GLVND for backward compatibility")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Thu, 10 Oct 2019 15:26:52 +0000 (16:26 +0100)]
GL: drop symbols mangling support
SCons and Meson have never supported that feature, and Autotools was
deleted over 6 months ago and no-one complained yet, so it's pretty
obvious nobody cares about it.
Fixes: 95aefc94a941701616fd ("Delete autotools")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Rhys Perry [Thu, 10 Oct 2019 11:17:04 +0000 (12:17 +0100)]
aco: update print_ir
Mostly adds GFX10 stuff.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Rhys Perry [Thu, 10 Oct 2019 11:15:49 +0000 (12:15 +0100)]
aco: rework scratch resource code
Fix compute, cleanup and add GFX10 support.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Rhys Perry [Thu, 10 Oct 2019 11:13:28 +0000 (12:13 +0100)]
aco/gfx10: disable GFX9 1D texture workarounds
Navi added back support for 1D textures.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Rhys Perry [Thu, 10 Oct 2019 11:12:29 +0000 (12:12 +0100)]
aco/gfx10: fix inline uniform blocks
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Rhys Perry [Thu, 10 Oct 2019 11:10:15 +0000 (12:10 +0100)]
radv/aco: disable NGG when ACO is used
Note that radv_device.c still has to be modified to use ACO with Navi.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-By: Timur Kristóf <timur.kristof@gmail.com>
Marek Olšák [Tue, 8 Oct 2019 04:08:19 +0000 (00:08 -0400)]
ac/nir: add back nir_op_fmod
radeonsi doesn't lower it for doubles.
This partially reverts commit
d861401554b52b2c2fc6721c69bdfe1697ee608f.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Tue, 8 Oct 2019 02:50:36 +0000 (22:50 -0400)]
gallium: remove PIPE_SHADER_CAP_SCALAR_ISA
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Tue, 8 Oct 2019 02:48:41 +0000 (22:48 -0400)]
tgsi_to_nir: use nir_shader_compiler_options::lower_to_scalar
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Tue, 8 Oct 2019 02:47:23 +0000 (22:47 -0400)]
st/mesa: use nir_shader_compiler_options::lower_to_scalar
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Tue, 8 Oct 2019 02:46:00 +0000 (22:46 -0400)]
nir: add nir_shader_compiler_options::lower_to_scalar
This will replace PIPE_SHADER_CAP_SCALAR_ISA.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Thu, 26 Sep 2019 23:34:31 +0000 (19:34 -0400)]
tgsi_to_nir: add #ifdef header guards
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Tue, 1 Oct 2019 02:16:55 +0000 (22:16 -0400)]
nir/drawpixels: fix what appears to be a copy-paste bug in get_texcoord_const
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Tue, 1 Oct 2019 01:29:11 +0000 (21:29 -0400)]
nir/drawpixels: handle load_color0, load_input, load_interpolated_input
for radeonsi
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Marek Olšák [Fri, 27 Sep 2019 20:43:31 +0000 (16:43 -0400)]
nir: move gl_nir_opt_access from glsl directory
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Marek Olšák [Wed, 9 Oct 2019 17:27:07 +0000 (13:27 -0400)]
nir: add a strip parameter to nir_serialize
so that drivers don't have to call nir_strip manually.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 17:44:01 +0000 (19:44 +0200)]
radv: Enable VK_ANDROID_external_memory_android_hardware_buffer.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 8 Oct 2019 21:24:02 +0000 (23:24 +0200)]
radv: Check the size of the imported buffer.
This is a security feature to disallow malicious apps from passing
a buffer that is too small.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 8 Oct 2019 20:15:43 +0000 (22:15 +0200)]
radv: Expose image handle compat types for Android handles.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 17:42:49 +0000 (19:42 +0200)]
radv: Allow Android image binding.
Using delayed layout of images.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Wed, 10 Jul 2019 12:45:01 +0000 (14:45 +0200)]
radv/android: Add android hardware buffer import/export.
Support does not include images yet.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 14:33:39 +0000 (16:33 +0200)]
radv: Deal with Android external formats.
To abstract things a bit, this adds a helper function in radv_android.c.
However, this means we have to link in radv_android.c on non-android as
well, which means some scaffolding changes.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Mon, 8 Jul 2019 13:33:06 +0000 (15:33 +0200)]
radv: Derive android usage from create flags.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 8 Oct 2019 20:14:44 +0000 (22:14 +0200)]
radv: Disallow sparse shared images.
Since we really cannot share them ever.
Also remove an unused switch.
Fixes: b70829708ac "radv: Implement VK_KHR_external_memory"
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 25 Jun 2019 09:42:03 +0000 (11:42 +0200)]
radv/android: Add android hardware buffer queries.
Derived from the Intel code.
For the internal format we just use the internal Vulkan format,
as we have Vulkan formats for all android formats we care about.
For the ycbcr properties we just do something. I do not have a real
clue what would be recommended.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Wed, 10 Jul 2019 12:27:37 +0000 (14:27 +0200)]
radv/android: Add android hardware buffer field to device memory.
You cannot go from BO to Android hardware buffer, so for export we
have to remember it.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 11:56:05 +0000 (13:56 +0200)]
radv: Add VK_ANDROID_external_memory_android_hardware_buffer.
Still disabled but now we can add entrypoints.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 11:42:31 +0000 (13:42 +0200)]
radv: Unset vk_info in radv_image_create_layout.
For better test coverage of this corner case.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 11:23:36 +0000 (13:23 +0200)]
radv: Handle slightly different image dimensions.
The minigbm comment really says it all. We should
fix minigbm as well, but for now this is the more
robust solution.
Note that this only changes width and height for
the surface creation, not for the image and hence
also not for the sampler, where it would wreak
havoc due to the normalized coords.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Mon, 23 Sep 2019 14:42:39 +0000 (16:42 +0200)]
radv: Delay patching for imported images until layout time.
We want this flexibility because in GFX10 we lose any stride fields,
so we have to make sure our width/height are in alignment with
the external image we import.
Furthermore, we need the ability to inject tiling modifiers on import
time which is strictly after create time for Android. So, with the
layout & patch functions being fully independent of pCreateInfo, we
can delay it until import/bind time.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Mon, 23 Sep 2019 14:14:50 +0000 (16:14 +0200)]
radv: Split out layout code from image creation.
So we can delay the layout until later in some import cases.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Wed, 10 Jul 2019 12:21:11 +0000 (14:21 +0200)]
radv: Handle device memory alloc failure with normal free.
Less duplication/complexity.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Bas Nieuwenhuizen [Tue, 24 Sep 2019 15:26:41 +0000 (17:26 +0200)]
radv: Cleanup buffer_from_fd.
Unused stride/offset args.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tomeu Vizoso [Sun, 6 Oct 2019 15:49:56 +0000 (08:49 -0700)]
gitlab-ci/lava: Test Lima driver with dEQP
Run dEQP on boards with Mali 400 and 450 in Baylibre's lab.
There's lots of skipped tests because of crashes and undetermined
behavior. May be a good idea to run the tests with valgrind and fix any
issues found.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Tomeu Vizoso [Sun, 6 Oct 2019 22:21:39 +0000 (15:21 -0700)]
gitlab-ci/lava: Use files to list tests to skip
As the non-LAVA runner script does, have per-GPU version files listing
the tests that are to be skipped, due to being very slow, unstable, etc.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Rafael Antognolli [Mon, 16 Sep 2019 16:07:26 +0000 (09:07 -0700)]
intel/tools: Support multiple contexts in intel_dump_gpu.
Create basic aub_context on GEM_CONTEXT_CREATE.
Set it up and submit a context + ring + pphwsp during execbuf
submission, if it has not been initialized yet.
v2: Write the HWSP only once per engine (Lionel).
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Rafael Antognolli [Fri, 13 Sep 2019 22:13:31 +0000 (15:13 -0700)]
intel/tools: Add basic aub_context code and helpers.
v2:
- Only dump context if there were no erros (Lionel).
- Store counter for context handles in aub_file (Lionel).
v3:
- Add a comment about aub_context -> GEM context (Lionel).
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Rafael Antognolli [Tue, 17 Sep 2019 01:12:02 +0000 (18:12 -0700)]
intel/tools: Use common code for GGTT address allocation.
We want to be able to create contexts on demand, and increase the GGTT
as needed for that. Use the aub_map_ggtt() function for that.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Rafael Antognolli [Mon, 9 Sep 2019 21:24:41 +0000 (14:24 -0700)]
intel/tools: Factor out GGTT allocation.
We want to reuse it in execlists_setup().
v2: Rename it to write_ggtt_ptes() (Lionel).
v3: Rename it to aub_map_ggtt() (Lionel).
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Bas Nieuwenhuizen [Thu, 10 Oct 2019 09:40:27 +0000 (11:40 +0200)]
radv: Implement & enable VK_EXT_texel_buffer_alignment.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Samuel Pitoiset [Wed, 9 Oct 2019 11:34:52 +0000 (13:34 +0200)]
radv: use a compute shader for copying timestamp query results
When the timestamp is not ready (ie. UINT64_MAX), the availabily bit
should be zero. The previous code used to copy the timestamp value
as the availabily bit and that's completely wrong.
Because it's not that simple to emit a conditional with the CP, the
driver now uses a compute shader for copying timestamp query results.
Fixes dEQP-VK.pipeline.timestamp.misc_tests.reset_query_before_copy.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Wed, 9 Oct 2019 12:30:49 +0000 (14:30 +0200)]
radv: sync before resetting query pools if timestamps have been written
Otherwise, the GPU might write timestamp queries after the reset
operation. This is similar to other query operations.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Timur Kristóf [Wed, 9 Oct 2019 08:40:24 +0000 (10:40 +0200)]
aco: Clean up usages of PhysReg::reg from aco_assembler.
These are not needed anymore, since PhyReg has an implicit
conversion operator that can convert it to unsigned int,
which is equivalent to accessing this field.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Timur Kristóf [Thu, 3 Oct 2019 17:32:48 +0000 (19:32 +0200)]
aco: Add extra assertion for number of FS input VGPRs.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Timur Kristóf [Tue, 17 Sep 2019 17:59:17 +0000 (19:59 +0200)]
aco: Fix s_dcache_wb on GFX10.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>