Nicolai Hähnle [Fri, 9 Sep 2016 08:08:11 +0000 (10:08 +0200)]
radeonsi: support ARB_compute_variable_group_size
Not sure if it's possible to avoid programming the block size twice (once for
the userdata and once for the dispatch).
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Lionel Landwerlin [Fri, 7 Oct 2016 12:53:04 +0000 (13:53 +0100)]
anv: turn on samplerAnisotropy in VkPhysicalDeviceFeatures
According to the Vulkan spec 5.63.4 :
samplerAnisotropy indicates whether anisotropic filtering is supported. If
this feature is not enabled, the maxAnisotropy member of the
VkSamplerCreateInfo structure must be 1.0.
Since we already set maxAnisotropy to 16 and program the hardware according
to the VkSamplerCreateInfo.maxAnisotropy, it seems we can turn this on.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Edward O'Callaghan [Fri, 7 Oct 2016 11:19:19 +0000 (22:19 +1100)]
radv: Use proper header guards over 'pragma once' directives
Signed-off-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Tapani Pälli [Fri, 7 Oct 2016 05:41:15 +0000 (08:41 +0300)]
mesa: throw error if bufSize negative in GetSynciv on OpenGL ES
Fixes following dEQP tests:
dEQP-GLES31.functional.debug.negative_coverage.callbacks.state.get_synciv
dEQP-GLES31.functional.debug.negative_coverage.get_error.state.get_synciv
dEQP-GLES31.functional.debug.negative_coverage.log.state.get_synciv
v2: drop _mesa_is_gles check (Kenneth)
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98133
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tapani Pälli [Fri, 7 Oct 2016 05:23:41 +0000 (08:23 +0300)]
glsl: prohibit lowp, mediump precision on atomic_uint
Fixes following dEQP tests:
dEQP-GLES31.functional.debug.negative_coverage.callbacks.atomic_counter.atomic_precision
dEQP-GLES31.functional.debug.negative_coverage.get_error.atomic_counter.atomic_precision
dEQP-GLES31.functional.debug.negative_coverage.log.atomic_counter.atomic_precision
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98131
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tapani Pälli [Fri, 30 Sep 2016 15:12:12 +0000 (08:12 -0700)]
glsl: optimize copy_propagation_elements pass
Changes make copy_propagation_elements pass faster, reducing link
time spent in test case of bug 94477. Does not fix the actual issue
but brings down the total time. No regressions seen in CI.
v2 (idr): Formatting / whitespace fixes. Embed the acp_ref in the
acp_entry.
v3 (idr): Delete unused copy constructor. Use while(pop_head) instead
of foreach() { remove }.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Dave Airlie [Mon, 10 Oct 2016 00:06:52 +0000 (10:06 +1000)]
radv: don't build without SHA1.
Just copy the section from anv above this.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98167
Signed-off-by: Dave Airlie <airlied@redhat.com>
Edward O'Callaghan [Fri, 7 Oct 2016 12:21:32 +0000 (23:21 +1100)]
docs/features.txt: Add GL_KHR_robustness supported on ES 3.2
Both radeonsi and nvc0 should also support ES so fixup doc.
Signed-off-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Lionel Landwerlin [Tue, 4 Oct 2016 14:15:50 +0000 (15:15 +0100)]
intel: aubinator: enable loading dumps from standard input
In conjuction with an intel_aubdump change, you can now look at your
application's output like this :
$ intel_aubdump -c '/path/to/aubinator --gen=hsw' my_gl_app
v2: Add print_help() comment about standard input handling (Eero)
Remove shrinked gtt space debug workaround (Eero)
v3: Use realloc rather than memcpy/free (Ben)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Sirisha Gandikota <Sirisha.Gandikota@intel.com>
Lionel Landwerlin [Wed, 5 Oct 2016 22:52:51 +0000 (23:52 +0100)]
intel: aubinator: enable loading xml files from a given directory
This might be useful for people who debug with out of tree descriptions.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Sirisha Gandikota <Sirisha.Gandikota@intel.com>
Lionel Landwerlin [Tue, 4 Oct 2016 15:19:43 +0000 (16:19 +0100)]
intel: aubinator: generate a standalone binary
Embed the xml files into the binary, so aubinator can be used from any
location.
v2: Split generation packing into another patch (Jason)
Check for xxd (Jason)
v3: Fix out of tree builds (Jason)
Generate custom variable name rather than names generated by xxd
(Lionel)
v4: Move generated _xml.h files to genxml/ (Sirisha)
v5: Remove newline from makefile (Jason)
v6: Add comment on gen*_xml.h creation (Jason)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Nanley Chery [Fri, 7 Oct 2016 19:07:33 +0000 (12:07 -0700)]
anv/TODO: Update the HiZ task
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Nanley Chery [Fri, 7 Oct 2016 19:07:31 +0000 (12:07 -0700)]
anv: Enable fast depth clears
Provides an FPS increase of ~30% on the Sascha triangle and multisampling
demos.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Chad Versace [Thu, 6 Oct 2016 22:21:53 +0000 (15:21 -0700)]
anv/cmd_buffer: Enable rendering to HiZ
Nanley Chery:
(rebase)
- Resolve conflicts with new anv_batch_emit macro
(amend)
- Handle a QPitch TODO
- Emit 3DSTATE_HIER_DEPTH_BUFFER on pre-BDW systems
- Only use HiZ for single-subpass renderpasses
- Emit the HiZ instruction before the stencil instruction to follow the
optimized clear sequence specified in the PRMs
- Don't modify clear params
- Enable resolves when a HiZ buffer is used to ensure depth buffer validity
Provides an FPS increase of ~15% on the Sascha triangle and multisampling
demos.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Nanley Chery [Fri, 7 Oct 2016 19:07:27 +0000 (12:07 -0700)]
anv/cmd_buffer: Add code for performing HZ operations
Create a function that performs one of three HiZ operations -
depth/stencil clears, HiZ resolve, and depth resolves.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Jason Ekstrand [Thu, 6 Oct 2016 22:21:51 +0000 (15:21 -0700)]
anv/image: Memset hiz surfaces to 0 when binding memory
Nanley Chery (amend):
- Change memset value from 0xff to 0 (a defined value for HiZ).
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jason Ekstrand [Thu, 6 Oct 2016 22:21:50 +0000 (15:21 -0700)]
anv: Move BindImageMemory to anv_image.c
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Chad Versace [Thu, 6 Oct 2016 22:21:49 +0000 (15:21 -0700)]
anv: Allocate hiz surface
Nanley Chery:
(rebase)
- Use isl_surf_get_hiz_surf()
(amend)
- Only add a HiZ surface onto a depth/stencil attachment
- Add comment above HiZ surface addition
- Hide HiZ behind INTEL_VK_HIZ prior to BDW
- Disable HiZ for untested cases
- Remove DISABLE_AUX_BIT instead of preventing it from being added
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Chad Versace [Thu, 6 Oct 2016 22:21:48 +0000 (15:21 -0700)]
anv: Add func anv_image_has_hiz()
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Chad Versace [Thu, 6 Oct 2016 22:21:47 +0000 (15:21 -0700)]
anv: Add anv_image::hiz_surface
Unused.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Nanley Chery [Fri, 7 Oct 2016 19:07:22 +0000 (12:07 -0700)]
isl: Correct a comment in the isl_format enum
HiZ is not a color surface, but an auxiliary depth surface.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Rob Clark [Fri, 7 Oct 2016 15:58:16 +0000 (11:58 -0400)]
gallium: add missing zero-init for resource templates
Mostly test code, plus one spot I noticed in r600.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Rob Clark [Fri, 7 Oct 2016 15:59:28 +0000 (11:59 -0400)]
freedreno: don't try to shadow layered textures
We will only hit this with multi-planar YUV external images, so we would
probably never hit this code path in the first place. But if we did, it
wouldn't do the right thing so just bail.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Mon, 3 Oct 2016 20:24:07 +0000 (16:24 -0400)]
freedreno/a3xx+a4xx: fix clip-plane lowering state
If enabled clip-planes have changed, we need to mark program state
dirty.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Ian Romanick [Wed, 5 Oct 2016 20:21:53 +0000 (13:21 -0700)]
glsl: Let cache_test build when the shader cache is not enabled
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Tested-by: Aaron Watry <awatry@gmail.com>
Lionel Landwerlin [Fri, 7 Oct 2016 16:16:55 +0000 (17:16 +0100)]
anv: pipeline cache: fix return value of vkGetPipelineCacheData
According to the spec - 9.6. Pipeline Cache :
If pDataSize is less than the maximum size that can be retrieved by the
pipeline cache, at most pDataSize bytes will be written to pData, and
vkGetPipelineCacheData will return VK_INCOMPLETE.
Fixes the following test from Vulkan CTS :
dEQP-VK.pipeline.cache.pipeline_from_incomplete_get_data.vertex_stage_fragment_stage
dEQP-VK.pipeline.cache.pipeline_from_incomplete_get_data.vertex_stage_geometry_stage_fragment_stage
dEQP-VK.pipeline.cache.misc_tests.invalid_size_test
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Timothy Arceri [Fri, 7 Oct 2016 10:10:37 +0000 (21:10 +1100)]
util: remove unused variable
Also initialise page at declaration.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Martin Peres [Thu, 6 Oct 2016 14:10:35 +0000 (17:10 +0300)]
loader/dri3: import prime buffers in the currently-bound screen
This tries to mirrors the codepath taken by DRI2 in IntelSetTexBuffer2()
and fixes many applications when using DRI3:
- Totem with libva on hw-accelerated decoding
- obs-studio, using Window Capture (Xcomposite) as a Source
- gstreamer with VAAPI
v2:
- introduce get_dri_screen() in the dri3 loader's vtable (krh)
Tested-by: Timo Aaltonen <tjaalton@ubuntu.com>
Tested-by: Ionut Biru <biru.ionut@gmail.com>
Cc: mesa-stable@lists.freedesktop.org
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71759
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
Martin Peres [Thu, 6 Oct 2016 14:07:22 +0000 (17:07 +0300)]
loader/dri3: add get_dri_screen() to the vtable
This allows querying the current active screen from the
loader's common code.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
Jason Ekstrand [Fri, 7 Oct 2016 00:16:51 +0000 (17:16 -0700)]
anv/entrypoints: Save off the entire devinfo rather than a pointer
Since the gen_device_info structs are no longer just constant memory, a
pointer to one is not a pointer to something in the .data section so we
shouldn't be storing it in a static variable. Instead, we should just
store the entire device_info structure.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Dave Airlie [Fri, 7 Oct 2016 02:08:26 +0000 (12:08 +1000)]
radv: drop all uint for unsigned.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Eric Anholt [Thu, 6 Oct 2016 21:50:04 +0000 (14:50 -0700)]
vc4: Don't worry about partial Z/S clear if the other is already cleared.
We have to be careful to not smash the value they're clearing to, but
other than that we're fine. Avoids quad clears in Processing, which likes
to do glClear(Z|S); glClear(Z).
Improves performance of Processing's QuadRendering demo at 5000 quads by
5.46507% +/- 1.35576% (n=15 before, 32 after)
Eric Anholt [Thu, 6 Oct 2016 21:45:23 +0000 (14:45 -0700)]
vc4: Try to fix the HW-2116 workaround.
We were incrementing the count at the end of vc4_start_draw(), except that
that function returns immediately if we've already started drawing on this
batch. It also failed to count the statechanges from the GFXH-515
workaround.
This incidentally allows repeated glClear() to be coalesced, because the
fast clears aren't counted in draw_calls_queued any more. Fixes most of
the extra flushes in Processing, which emits glClear(Z|S); glClear(Z);
glClear(C) during its frame setup.
Improves performance of Processing's QuadRendering demo at 5000 quads by
3.33538% +/- 2.05846% (n=21 before, 15 after)
Eric Anholt [Thu, 6 Oct 2016 21:35:58 +0000 (14:35 -0700)]
vc4: Drop dead argument from vc4_start_draw().
Eric Anholt [Wed, 5 Oct 2016 21:10:30 +0000 (14:10 -0700)]
vc4: Fix fallback to quad clears of depth in GLX.
The fix in the vc4-jobs series ended up triggering the fallback path on
GLX apps that use depth but not stencil.
Eric Anholt [Wed, 5 Oct 2016 21:22:09 +0000 (14:22 -0700)]
vc4: Add the format name in miptree_debug.
I was curious if my Z/S buffer was actually ZS or ZX, and the vc4 format
of "0" didn't tell me much.
Eric Anholt [Wed, 5 Oct 2016 21:05:51 +0000 (14:05 -0700)]
vc4: Fix perf debug formatting on partial Z/S clear.
Eric Anholt [Wed, 5 Oct 2016 16:21:37 +0000 (09:21 -0700)]
vc4: Drop destination register when it's unused.
This slightly reduces instructions on shader-db, but I think it's just
perturbing register allocation -- the allocator should have always
trivially colored these nodes, before. This commit is just to make QIR
code failing more intelligible when register allocation fails.
Eric Anholt [Wed, 5 Oct 2016 16:07:46 +0000 (09:07 -0700)]
vc4: Fix live intervals analysis for screening defs in if statements.
If a conditional assignment is only conditioned on the exec mask, that's
still screening off the value in the executed channels (and, since we're
not storing to the unexcuted channels, we don't care what's in there).
Fixes a bunch of extra register pressure on Processing's Ribbons demo,
which is failing to allocate.
Eric Anholt [Tue, 4 Oct 2016 23:29:26 +0000 (16:29 -0700)]
vc4: Fix simulator when more than one vc4_screen is opened.
We would assertion fail in setting up the simulator the second time
around. This at least postpones the assertion failure until we've closed
all of the first set of screens and started opening a new set.
Eric Anholt [Thu, 6 Oct 2016 23:46:35 +0000 (16:46 -0700)]
vc4: Fix assertion fails from trying to cast non-ALU instrs to ALU.
Fixes 100 piglit tests since the assertions were added to nir.h. What's
amazing is that these tests used to pass, even when casting garbage.
Jason Ekstrand [Wed, 5 Oct 2016 23:54:57 +0000 (16:54 -0700)]
anv/cmd_buffer: Move the clear_subpasses calls to set_subpass
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Jason Ekstrand [Wed, 5 Oct 2016 23:51:02 +0000 (16:51 -0700)]
anv/cmd_buffer: Don't call set_subpass in a secondary
Initially, we had intended set_subpass to be an interesting function that
did whatever (presumably a lot) setup we needed for a subpass. In reality,
it just sets a pointer and a dirty bit and then emits depth and stencil
state. When we call BeginCommandBuffer on a secondary, there's no point in
setting depth and stencil state since it will already be set by the
primary. Instead, the only thing we need to do at the start of a secondary
is set the subpass pointer and the dirty bit.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Jason Ekstrand [Thu, 6 Oct 2016 22:50:21 +0000 (15:50 -0700)]
anv/cmd_buffer: Rework descriptor dirtying in set_subpass
We have a DIRTY_RENDER_TARGETS flag and that makes a lot more sense than
just dirtying fragment descriptors. We're checking for it in some of the
gen7 code but unfortunately, nothing was setting it and it didn't do what
it was supposed to do in cmd_buffer_flush_state.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Jason Ekstrand [Wed, 5 Oct 2016 20:23:22 +0000 (13:23 -0700)]
anv/wsi: Advertise UNORM formats as well as sRGB
Because WSI images are created with VkImageCreateInfo::flags explicitly set
to 0, they don't ever have the VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT set.
This means that you can't create an image view of it with a different
format so applications can't render directly in sRGB (without automatic
encoding) unless we actually advertise UNORM formats. There are a lot of
applications that want to do their own sRGB conversion, so we should allow
for that. We do, however, make UNORM come after sRGB in the list so that
the default for dumb apps that just grab the first thing is to render in
linear and let the sRGB conversion happen automatically.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Dave Airlie [Thu, 6 Oct 2016 23:27:36 +0000 (09:27 +1000)]
radv: fix configure.ac check
This should be positive test.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Gustaw Smolarczyk [Wed, 5 Oct 2016 23:09:54 +0000 (01:09 +0200)]
radv: Skip already signalled fences.
If the user created a fence with VK_FENCE_CREATE_SIGNALED_BIT set, we
shouldn't fail to wait for a fence if it was not submitted since that is
not necessary.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Thu, 6 Oct 2016 23:16:09 +0000 (09:16 +1000)]
radv: add initial non-conformant radv vulkan driver
This squashes all the radv development up until now into
one for merging.
History can be found:
https://github.com/airlied/mesa/tree/semi-interesting
This requires llvm 3.9 and is in no way considered
a conformant vulkan implementation. It can run a number
of vulkan applications, and supports all GPUs using
the amdgpu kernel driver.
Thanks to Intel for providing anv and spirv->nir,
and Emil Velikov for reviewing build integration.
Parts of this are:
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Acked-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Authors: Bas Nieuwenhuizen and Dave Airlie
Signed-off-by: Dave Airlie <airlied@redhat.com>
Samuel Pitoiset [Thu, 6 Oct 2016 23:08:54 +0000 (01:08 +0200)]
nv50/ir: fix wrong check when optimizing MAD to SHLADD
Checking if MAD is supported is definitely wrong, and it's
more likely a typo I introduced few days ago which breaks
NV50 because SHLADD is not supported there.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Lionel Landwerlin [Tue, 4 Oct 2016 15:29:55 +0000 (16:29 +0100)]
intel: aubinator: use getopt to parse arguments
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Sirisha Gandikota <sirisha.gandikota@intel.com>
Samuel Pitoiset [Thu, 6 Oct 2016 22:43:51 +0000 (00:43 +0200)]
nvc0: dump program binary only when NV50_PROG_DEBUG is set
When the chipset is forced with NV50_PROG_CHIPSET, we actually
only want to output the binary if NV50_PROG_DEBUG is also
enabled. Otherwise, this pollutes the shader-db output.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Jason Ekstrand [Thu, 6 Oct 2016 22:46:22 +0000 (15:46 -0700)]
nir: Fix the control flow tests for nir_loop_first_block changes
Commit
2ed17d46de045404042f13c6591895a1cf31b167 changed
nir_loop_first_cf_node and friends to return a nir_block instead of a
nir_cf_node. This broke one of the NIR control flow tests.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98128
Samuel Pitoiset [Sun, 11 Sep 2016 15:46:24 +0000 (17:46 +0200)]
docs: mark ARB_compute_variable_group_size as done for nvc0
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Samuel Pitoiset [Sat, 10 Sep 2016 14:45:32 +0000 (16:45 +0200)]
nvc0: expose ARB_compute_variable_group_size
Only expose 512 threads/block on Fermi to not be limited by
32 GPRs/thread.
v4: - use 512 threads on Fermi, 1024 on Kepler+
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Samuel Pitoiset [Tue, 6 Sep 2016 22:12:51 +0000 (00:12 +0200)]
nv50/ir: set number of threads/block for variable local size
When a variable local size is defined as specified by
ARB_compute_variable_group_size, the fixed local size is set to 0
and a SIGFPE occurs when we compute the maximum number of regs.
This allows to use 64 GPRs/thread.
v4: - use 512 threads on Fermi, 1024 on Kepler+
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Samuel Pitoiset [Wed, 7 Sep 2016 16:00:16 +0000 (18:00 +0200)]
st/mesa: expose ARB_compute_variable_group_size
This extension is only exposed if the underlying driver supports
ARB_compute_shader and if PIPE_COMPUTE_MAX_VARIABLE_THREADS_PER_BLOCK
is set.
v3: - initialize max_variable_threads_per_block to 0
v2: - expose the ext based on that new cap
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Tue, 6 Sep 2016 18:15:00 +0000 (20:15 +0200)]
st/mesa: add support for dispatching a variable local size
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Thu, 8 Sep 2016 19:35:36 +0000 (21:35 +0200)]
st/mesa: add mapping for SYSTEM_VALUE_LOCAL_GROUP_SIZE
gl_LocalGroupSizeARB can be translated into TGSI_SEMANTIC_BLOCK_SIZE
which represents the block size in threads.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Sat, 10 Sep 2016 14:31:27 +0000 (16:31 +0200)]
gallium: add PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK
v3: - use a new case statement in r600_pipe_common.c
- fix compilation of softpipe...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Thu, 8 Sep 2016 19:28:11 +0000 (21:28 +0200)]
glsl: add gl_LocalGroupSizeARB as a system value
v2: - only add it if the ext is enabled (Ilia)
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Tue, 6 Sep 2016 20:46:42 +0000 (22:46 +0200)]
glsl/linker: handle errors when a variable local size is used
Compute shaders can now include a fixed local size as defined by
ARB_compute_shader or a variable size as defined by
ARB_compute_variable_group_size.
v2: - update formatting spec quotations (Ian)
- various cosmetic changes (Ian)
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Wed, 7 Sep 2016 19:37:53 +0000 (21:37 +0200)]
glsl: reject compute shaders with fixed and variable local size
The ARB_compute_variable_group_size specification explains that
when a compute shader includes both a fixed and a variable local
size, a compile-time error occurs.
v2: - update formatting spec quotations (Ian)
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Tue, 6 Sep 2016 19:48:42 +0000 (21:48 +0200)]
glsl: process local_size_variable input qualifier
This is the new layout qualifier introduced by
ARB_compute_variable_group_size which allows to use a variable work
group size.
v4: - add missing '%s' in the monster format string
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Wed, 7 Sep 2016 16:05:52 +0000 (18:05 +0200)]
glsl: add enable flags for ARB_compute_variable_group_size
This also initializes the default values for the standalone compiler.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Wed, 7 Sep 2016 19:52:08 +0000 (21:52 +0200)]
mesa/main: add support for ARB_compute_variable_groups_size
v5: - replace fixed_local_size by !LocalSizeVariable (Nicolai)
v4: - slightly indent spec quotes (Nicolai)
- drop useless _mesa_has_compute_shaders() check (Nicolai)
- move the fixed local size outside of the loop (Nicolai)
- add missing check for invalid use of work group count
v2: - update formatting spec quotations (Ian)
- move the total_invocations check outside of the loop (Ian)
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Tue, 6 Sep 2016 14:23:58 +0000 (16:23 +0200)]
glapi: add entry points for GL_ARB_compute_variable_group_size
v2: - correctly sort that new extension (Ian)
- fix up the comment (Ian)
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Karol Herbst [Wed, 5 Oct 2016 22:36:23 +0000 (00:36 +0200)]
nv50/ir: optimize sub(a, 0) to a
helped some ue4 demos and divinity OS shaders
total instructions in shared programs :
2818674 ->
2818606 (-0.00%)
total gprs used in shared programs : 379273 -> 379273 (0.00%)
total local used in shared programs : 9505 -> 9505 (0.00%)
total bytes used in shared programs :
25837792 ->
25837192 (-0.00%)
local gpr inst bytes
helped 0 0 33 33
hurt 0 0 0 0
Signed-off-by: Karol Herbst <karolherbst@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Pierre Moreau <pierre.morrow@free.fr>
Brian Paul [Mon, 3 Oct 2016 15:57:50 +0000 (09:57 -0600)]
st/mesa: move all sampler view code into new st_sampler_view.[ch] files
Previously, the sampler view code was scattered across several different
files.
Note, the previous REALLOC(), FREE() for st_texture_object::sampler_views
are replaced by realloc(), free() to avoid conflicting macros in Mesa vs.
Gallium.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Acked-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Brian Paul [Fri, 30 Sep 2016 22:41:46 +0000 (16:41 -0600)]
st/mesa: optimize pipe_sampler_view validation
Before, st_get_texture_sampler_view_from_stobj() did a lot of work to
check if the texture parameters matched the sampler view (format,
swizzle, min/max lod, first/last layer, etc). We did this every time
we validated the texture state.
Now, we use a ctx->Driver.TexParameter() callback and a couple other
checks to proactively release texture views when we know that
view-related parameters have changed. Then, the validation step is
simplified:
- Search the texture's list of sampler views (just match the context).
- If found, we're done.
- Else, create a new sampler view.
There will never be old, out-of-date sampler views attached to texture
objects that we have to test.
Most apps create textures and set the texture parameters once. This
make sampler view validation much cheaper for that case.
Note that the old texture/sampler comparison code has been converted
into a set of assertions to verify that the sampler view is in fact
consistent with the texture parameters. This should help to spot any
potential regressions.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Brian Paul [Mon, 3 Oct 2016 15:40:56 +0000 (09:40 -0600)]
mesa: call ctx->Driver.TexParameter() in texture_buffer_range()
To inform drivers of texture buffer offset/size changes, as we do for
other texture object parameters.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Brian Paul [Fri, 23 Sep 2016 19:36:22 +0000 (13:36 -0600)]
st/mesa: consolidate view format setup code
Before, we had code to compute the sampler view's format spread across two
different functions: in update_single_texture() and
st_get_texture_sampler_view_from_stobj(). Now it's all in one new function.
Also, use _mesa_texture_base_format() to simplify the code.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Brian Paul [Fri, 23 Sep 2016 18:24:03 +0000 (12:24 -0600)]
st/mesa: add some const qualifiers in st_atom_texture.c
And minor code reformatting.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Brian Paul [Fri, 23 Sep 2016 18:18:46 +0000 (12:18 -0600)]
st/mesa: simplify some code in get_texture_format_swizzle()
There's no need to cast to st_texture_image. Just use gl_texture_image.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Brian Paul [Fri, 23 Sep 2016 18:59:56 +0000 (12:59 -0600)]
mesa: make _mesa_texture_buffer_range() static
Not called from any other file. Also, add a comment.
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Brian Paul [Sat, 24 Sep 2016 13:33:59 +0000 (07:33 -0600)]
mesa: add const qualifier, comment on can_avoid_reallocation()
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Brian Paul [Fri, 23 Sep 2016 18:59:29 +0000 (12:59 -0600)]
mesa: add comment/assertion on get_tex_level_parameter_buffer()
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Jason Ekstrand [Thu, 6 Oct 2016 02:56:54 +0000 (19:56 -0700)]
nir: Remove some no longer needed asserts
Now that the NIR casting functions have type assertions, we have a bunch of
assertions that aren't needed anymore.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Thu, 6 Oct 2016 02:08:57 +0000 (19:08 -0700)]
nir: Make nir_foo_first/last_cf_node return a block instead
One of NIR's invariants is that control flow lists always start and end
with blocks. There's no good reason why we should return a cf_node from
these functions since we know that it's always a block. Making it a block
lets us remove a bunch of code.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Thu, 6 Oct 2016 01:09:25 +0000 (18:09 -0700)]
nir: Add asserts to the casting functions
This makes calling nir_foo_as_bar a bit safer because we're no longer 100%
trusting in the caller to ensure that it's safe. The caller still needs to
do the right thing but this ensures that we catch invalid casts with an
assert rather than by reading garbage data. The one downside is that we do
use the casts a bit in nir_validate and it's not a validate_assert.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Steven Toth [Thu, 6 Oct 2016 15:26:51 +0000 (11:26 -0400)]
gallium/hud: Remove superfluous debug
No longer required.
Signed-off-by: Steven Toth <stoth@kernellabs.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Thu, 6 Oct 2016 14:42:04 +0000 (15:42 +0100)]
amd: add amd_kernel_code_t.h to the sources list
Otherwise it won't be picked in the tarball and the build will fail.
Fixes: 91ec6e56649 ("radeonsi/compute: Use the HSA abi for non-TGSI
compute shaders v3")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Thu, 6 Oct 2016 14:21:58 +0000 (15:21 +0100)]
svga: add svga_mksstats.h to the sources list
Otherwise it won't be picked in the tarball and the build will fail.
Fixes: 0035f7f1365 ("svga: add guest statistic gathering interface")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Fri, 30 Sep 2016 10:01:28 +0000 (11:01 +0100)]
glx: rename choose_visual(), drop const argument
The function deals with fb (style) configs, thus using "visual"
in the name is misleading. Which in itself had led to the use of
fbconfig_style_tags argument.
Rename the function to reflect what it does and drop the unneeded
argument.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Emil Velikov [Fri, 30 Sep 2016 10:01:27 +0000 (11:01 +0100)]
glx: return GL_FALSE from glx_screen_init where applicable.
Return GL_FALSE if we fail to find any fb/visual configs, otherwise we
end up with all sorts of chaos further down the GLX stack.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Emil Velikov [Fri, 30 Sep 2016 10:01:26 +0000 (11:01 +0100)]
glx: correctly mask the drawableType for GLX_ARB_fbconfig_float
The comment/spec says - only for pbuffer drawables, while the code
clears the window/pixmap bit. Practise what you preach and apply the
trivial tweak. In practise this should not cause functional change.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Chuck Atkins [Tue, 4 Oct 2016 15:05:30 +0000 (11:05 -0400)]
autoconf: Make header install distinct for various APIs (v2)
This fixes a problem where GL headers would only get installed if
glx was enabled. So if osmesa was enabled but not glx, then the
GL headers required by osmesa would be missing from the install.
v2: Dropped unneeded mesa_glinterop.h redundant osmesa.h install
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Chuck Atkins <chuck.atkins@kitware.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Thu, 25 Aug 2016 15:59:01 +0000 (16:59 +0100)]
mesa: annotate AttribFuncsARB[] as const
It's read-only data, so annotate it accordingly.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Emil Velikov [Thu, 25 Aug 2016 15:59:00 +0000 (16:59 +0100)]
mapi/glapi: remove unused _glapi_check_table()
Similar to earlier commit - symbol was never part of the public API so
we're safe to remove it.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Emil Velikov [Thu, 25 Aug 2016 15:58:59 +0000 (16:58 +0100)]
glapi/hgl: remove the final user of _glapi_check_table()
The symbol is a no-op since, the EXTRA_DEBUG macro is not set in the
build. Unused by !Haiku people/platforms since 2010 (commit
a73c6540d9a7f6e26d8568ba2fc522cb865f0a6c) while the Haiku C++ wrapper
has no obvious users.
Cc: Alexander von Gluck IV <kallisti5@unixzen.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Emil Velikov [Thu, 25 Aug 2016 15:58:58 +0000 (16:58 +0100)]
mapi/glapi: remove unused _glapi_check_table_not_null
Function was never part of the API/ABI and the final user was removed
with commit
a73c6540d9a7f6e26d8568ba2fc522cb865f0a6c, back in 2010.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Emil Velikov [Mon, 19 Sep 2016 14:35:32 +0000 (15:35 +0100)]
st/xvmc/tests: force enable assertions
Similar to the other 'tests', enable assertions in xvmc_bench.
This silences the GCC warnings about unused-variable(s), makes the
program actually useful, as the XvMC API called. Atm the function
calls are omitted, since they're called within the assert.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Mon, 19 Sep 2016 12:14:39 +0000 (13:14 +0100)]
anv: automake: ship intel_icd.json.in in the tarball
Otherwise we'll fail to (re)generate intel_icd.json.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Mon, 19 Sep 2016 11:46:37 +0000 (12:46 +0100)]
intel: automake: reference the correct header
The header was renamed with earlier commit, so update the
Makefile.sources respectively.
{vulkan/genX_multisample.h => common/gen_sample_positions.h}
Fixes: c779ad3e661("intel: Move Vulkan sample positions to common code")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Lionel Landwerlin [Tue, 4 Oct 2016 15:28:43 +0000 (16:28 +0100)]
intel: aubinator: add missing return characters
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Kenneth Graunke [Wed, 5 Oct 2016 18:58:59 +0000 (11:58 -0700)]
nir: Delete open coded type printing.
glsl_print_type() prints arrays of arrays incorrectly. For example,
a type with name float[3][7] would be printed as float[7][3]. (This
is an array of length 3 containing arrays of 7 floats.) cdecl says
that the type name is correct.
glsl_print_type() doesn't really do anything above and beyond printing
type->name, and glsl_print_struct() wasn't used at all. So, drop them.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Philipp Zabel [Wed, 5 Oct 2016 23:48:04 +0000 (01:48 +0200)]
anv: fix GetPhysicalDeviceProperties to return timestampPeriod in ns
According to chapters 16.5. (Timestamp Queries) and 30.2 (Limits) of the
Vulkan Specification 1.0.29, the .limits.timestampPeriod field returned
by vkGetPhysicalDeviceProperties is measured in nanoseconds, not in
seconds.
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Timothy Arceri [Wed, 5 Oct 2016 05:46:39 +0000 (16:46 +1100)]
i965: remove remaining tabs in brw_draw.c
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Timothy Arceri [Wed, 5 Oct 2016 05:45:27 +0000 (16:45 +1100)]
i965: get inputs read from nir info
This is a step towards dropping the GLSL IR version of
do_set_program_inouts() in i965 and moving towards native nir support.
This is important because we want to eventually convert to nir and
use its optimisations passes before we can call this GLSL IR pass.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Timothy Arceri [Wed, 5 Oct 2016 05:24:56 +0000 (16:24 +1100)]
i965: get outputs written from nir info
This is a step towards dropping the GLSL IR version of
do_set_program_inouts() in i965 and moving towards native nir support.
This is important because we want to eventually convert to nir and
use its optimisations passes before we can call this GLSL IR pass.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Timothy Arceri [Wed, 5 Oct 2016 04:56:21 +0000 (15:56 +1100)]
i965: get outputs read from nir info
This is a step towards dropping the GLSL IR version of
do_set_program_inouts() in i965 and moving towards native nir support.
This is important because we want to eventually convert to nir and
use its optimisations passes before we can call this GLSL IR pass.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>