Mathias Fröhlich [Fri, 3 May 2019 04:53:21 +0000 (06:53 +0200)]
mesa: Rip out now unused gl_context::aelt_context.
Now this part of gl_context state is unused and can be removed.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Wed, 3 Apr 2019 11:06:33 +0000 (13:06 +0200)]
mesa: Implement _mesa_array_element by walking enabled arrays.
In glArrayElement, use the bitmask trick to just walk the enabled
vao arrays. This should be about equivalent in execution time to
walk the prepare aelt_context list. Finally this will allow us to
reduce the _mesa_update_state calls in a few patches.
v2: Add comments.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Wed, 3 Apr 2019 10:47:36 +0000 (12:47 +0200)]
mesa: Use glVertexAttrib*NV functions for fixed function attribs.
In the glArrayElement implementation, use glVertexAttrib*NV type
functions for fixed function attributes. We do the same in display
execution when the list is replayed using immediate mode attribute
functions. Using a single set of function pointers enables to
use a unified loop to walk the vertex array attributes.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Wed, 3 Apr 2019 10:33:37 +0000 (12:33 +0200)]
mesa: Factor out index function that will have multiple use.
For access to glArrayElement methods factor out a function to
get the table lookup index for normalized/integer/double access.
The function will be used in the next patch at least twice.
v2: Use vertex_format_to_index instead of NORM_IDX.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Jason Ekstrand [Sat, 5 Jan 2019 15:13:44 +0000 (09:13 -0600)]
nir: Add a SSA type gathering pass
This new pass (which isn't even compile-tested) attempts to determine
the ALU type of all the SSA values in a function impl. It takes a
greedy approach and assigns intness or floatness to everything it thinks
can possibly contain an int or a float. Some values will be labled as
both int and float and some will be labled as neither and it is up to
the caller to decide what to do with this information. However, for a
"nice" shader where the original source contained no bit-casts and no
implicit bit-casts were introduced by optimizations, there shouldn't be
any overlap in the two sets save for the odd CSEd zero constant.
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Kenneth Graunke [Fri, 26 Apr 2019 01:30:57 +0000 (18:30 -0700)]
iris: Delete bucketing allocators
These add a lot of complexity, and I currently can't measure any
performance benefit from having them. In the past, I seem to recall
seeing a benefit in drawoverhead scores, but currently it looks like
dropping them is either a wash or 1-2% faster.
Drop them to simplify allocations.
Kenneth Graunke [Sat, 27 Apr 2019 01:53:57 +0000 (18:53 -0700)]
iris: Force VMA alignment to be a multiple of the page size.
This should happen regardless, but let's be paranoid.
Kenneth Graunke [Fri, 3 May 2019 18:58:57 +0000 (11:58 -0700)]
iris: leave the top 4Gb of the high heap VMA unused
This ports commit
9e7b0988d6e98690eb8902e477b51713a6ef9cae from anv
to iris. Thanks to Lionel for noticing that it was missing!
Kenneth Graunke [Sat, 27 Apr 2019 00:12:24 +0000 (17:12 -0700)]
iris: Fix 4GB memory zone heap sizes.
The STATE_BASE_ADDRESS "Size" fields can only hold 0xfffff in pages,
and 0xfffff * 4096 =
4294963200, which is 1 page shy of 4GB.
So we can't use the top page.
Julien Isorce [Thu, 2 May 2019 22:36:04 +0000 (15:36 -0700)]
st/va: check resource_get_info nullity in vlVaDeriveImage
This pipe_screen function is not implemented by all backends.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110443
Signed-off-by: Julien Isorce <jisorce@oblong.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Jason Ekstrand [Fri, 3 May 2019 03:00:02 +0000 (22:00 -0500)]
anv,i965: Stop warning about incomplete gen11 support
Both drivers are feature-complete and should be running more-or-less at
perf at this point. Drop the warning.
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Connor Abbott [Thu, 2 May 2019 20:22:01 +0000 (22:22 +0200)]
nir/algebraic: Don't emit empty initializers for MSVC
Just don't emit the transform array at all if there are no transforms
v2:
- Don't use len(array) > 0 (Dylan)
- Keep using ARRAY_SIZE to make the generated C code easier to read
(Jason).
Kenneth Graunke [Thu, 2 May 2019 22:42:10 +0000 (15:42 -0700)]
iris: Resolve textures used by the program, not merely bound textures
st/mesa's PBO upload path binds a vertex shader that doesn't use any
textures, but leaves the existing sampler views bound in place. This
was tricking us into thinking the PBO destination might be bound for
texturing in some cases. In Civilization VI, this fixes a false self-
dependency issue that was preventing CCS_E compression on upload.
Fixing this slightly improves frame times.
Dylan Baker [Tue, 22 May 2018 22:24:16 +0000 (15:24 -0700)]
meson: Don't build glsl cache_test when shader cache is disabled
v2: - Use new with_shader_cache variable instead of
host_machine.system() == 'windows'
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Mon, 11 Jun 2018 22:11:07 +0000 (15:11 -0700)]
tests/vma: fix build with MSVC
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Tue, 22 May 2018 22:32:22 +0000 (15:32 -0700)]
glsl/tests: define ssize_t on windows
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Tue, 22 May 2018 22:00:35 +0000 (15:00 -0700)]
util/tests: Use define instead of VLA
To allow the this test to be built with MSVC, which doesn't support
VLAs.
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Thu, 17 May 2018 20:44:30 +0000 (13:44 -0700)]
meson: make nm binary optional
This makes nm not required, but used if found. In general I imagine that
this means that on windows nm wont be found, and on other platforms it
will.
v2: - fix gbm and egl symbols check tests to only be run if nm is found
- reword commit message to reflect the code change
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Wed, 18 Apr 2018 20:20:28 +0000 (13:20 -0700)]
meson: Make shader-cache a trillean instead of boolean
So that it can be implicitly disabled on windows, where it doesn't
compile.
v2: - Use an auto-option rather than automagic.
- fix shader_cache check (== -> !=)
v4: - Use new with_shader_cache instead of get_option('shader-cache')
elsewhere in the meson build
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Fri, 19 Oct 2018 16:47:41 +0000 (09:47 -0700)]
meson: switch gles1 and gles2 to auto options
This allows them to default to false on windows, but default to true
elsewhere. As a side effect turning off shared-glapi now automatically
turns off gles. Shared glapi remains a boolean defaulting to true.
v5: - new in this version
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Wed, 18 Apr 2018 17:44:44 +0000 (10:44 -0700)]
glsl: fix general_ir_test with mingw
Somewhere down in the depths of the mingw headers 'interface' is
defined, change it to iface like a similar patch did.
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Dylan Baker [Wed, 18 Apr 2018 20:09:07 +0000 (13:09 -0700)]
meson: always define libglapi
This allows the identifier to be used even if shared-glapi isn't build,
which simplifies a bunch of things.
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Chuck Atkins [Fri, 3 May 2019 16:06:22 +0000 (12:06 -0400)]
meson: Fix missing glproto dependency for gallium-glx
Signed-off-by: Chuck Atkins <chuck.atkins@kitware.com>
Cc: mesa-stable <mesa-stable@lists.freedesktop.org>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Samuel Pitoiset [Fri, 3 May 2019 09:45:34 +0000 (11:45 +0200)]
radv: apply the indexing workaround for atomic buffer operations on GFX9
Because the new raw/struct intrinsics are buggy with LLVM 8
(they weren't marked as source of divergence), we fallback to the
old instrinsics for atomic buffer operations only. This means we need
to apply the indexing workaround for GFX9. The load/store
operations still use the new LLVM 8 intrinsics.
The fact that we need another workaround is painful but we should
be able to clean up that a bit once LLVM 7 support will be dropped.
This fixes a GPU hang with AC Odyssey and some rendering problems
with Nioh.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110573
Fixes:
31164cf5f70 ("ac/nir: only use the new raw/struct image atomic intrinsics with LLVM 9+")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Alyssa Ross [Fri, 19 Apr 2019 18:14:37 +0000 (18:14 +0000)]
get_reviewer.pl: improve portability
Not all package managers / users will install perl into /usr/bin,
but /usr/bin/env /should/ always be present.
Using /usr/bin/env means that we can't give the -w argument to Perl,
so I added `use warnings' in the script.
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Lionel Landwerlin [Thu, 2 May 2019 14:08:42 +0000 (15:08 +0100)]
anv: fix crash when application does not provide push constants
Found while running Talos Principle.
As far as I can tell running a draw call with a pipeline having push
constants without the application having called vkCmdPushConstants
gives undefined push constant values.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: mesa-stable@lists.freedesktop.org
Samuel Pitoiset [Thu, 2 May 2019 16:07:11 +0000 (18:07 +0200)]
radv: fix radv_get_aspect_format() for D+S formats
This restores the previous behaviour before YCBCR landed. For D+S
formats, it returns the depth format.
This fixes an assertion with Thrones of Britannia.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110540
Fixes:
66507cc6563 ("radv: Add single plane image views & meta operations")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Caio Marcelo de Oliveira Filho [Wed, 1 May 2019 23:31:14 +0000 (16:31 -0700)]
intel/fs: Assert when brw_fs_nir sees a nir_deref_instr
Since
09f1de97a76 "anv,i965: Lower away image derefs in the driver"
the backend compiler is not expected to handle any derefs, so let's
assert on it.
This helps identifying problems when a deref is not lowered and
"leaks" into the backend compiler.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Julien Isorce [Thu, 2 May 2019 22:57:04 +0000 (15:57 -0700)]
r600: implement resource_get_info
Factoring code with resource_get_handle.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110443
Signed-off-by: Julien Isorce <jisorce@oblong.com>
Reviewed-by: Dave Airlie airlied@redhat.com
Dave Airlie [Mon, 29 Apr 2019 03:17:41 +0000 (13:17 +1000)]
util/bitset: fix bitset range mask calculations.
The MASK macro is used in the RANGE macro, and it should
return the pre-bitset word mask for the (b) value.
i.e.
BITSET_MASK(0) should be undefined since it's meaningless.
BITSET_MASK(31) should give 0x7fffffff
BITSET_MASK(32) should give 0xffffffff
BITSET_MASK(33) should give 0x00000001
BITSET_MASK(64) should give 0xffffffff
However then BITSET_RANGE ends up broken for cases where
it's (b) value is the 0,32,64 value as in that case the lower
mask would be 0 not 0xffffffff.
This fixes the unit tests that I've added, and my code that
uses bitsets.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Fixes:
bb38cadb1c5f2 "More GLSL code"
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Dave Airlie [Mon, 29 Apr 2019 02:24:58 +0000 (12:24 +1000)]
util/tests: add basic unit tests for bitset
The last test here currently fails as there is a bug in bitset.h
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Dave Airlie [Fri, 3 May 2019 03:23:45 +0000 (13:23 +1000)]
nir: fix lower vars to ssa for larger vector sizes.
This has a couple of hardcoded vec4 limits in it, change them
to the proper sizing to avoid future issues.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Dave Airlie [Fri, 3 May 2019 02:53:56 +0000 (12:53 +1000)]
spirv: fix SpvOpBitSize return value.
The spir-v spec says this returns a bool.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Kenneth Graunke [Fri, 3 May 2019 04:14:49 +0000 (21:14 -0700)]
iris: Disable dual source blending when shader doesn't handle it
This is a port of Danylo's
eca4a6548d07bbbb02a7768edb397bad7b72cfc2
which fixed the hang on i965. It fixes GPU hangs in his new Piglit
test, arb_blend_func_extended-dual-src-blending-discard-without-src1.
I avoided my own review feedback here, and decided to simply adjust
3DSTATE_PS_BLEND rather than BLEND_STATE_ENTRY[0]. It has never been
clear to me which the hardware uses in every case. However, whacking
the enable in 3DSTATE_PS_BLEND seems to be sufficient to fix the hang,
and that packet is already dynamic, so it's easy to handle. I'd rather
avoid making BLEND_STATE_ENTRY[0] dynamic unless I have to.
Jason Ekstrand [Mon, 29 Apr 2019 05:46:10 +0000 (00:46 -0500)]
anv: Stop including POS in FS input limits
It is an input but it comes in as part of the shader payload and doesn't
count towards the limits.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Rob Clark [Thu, 2 May 2019 21:48:41 +0000 (14:48 -0700)]
nir: fix nir tex print harder
Fixes:
691d5a825a6 nir: rework tex instruction printing
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Erik Faye-Lund [Thu, 2 May 2019 18:24:28 +0000 (20:24 +0200)]
docs: fixup mistake in contents
During a rebase, it seems I accidentally broke the contents-menu,
leading to a duplicate link to freedesktop.org. This was obviously not
intended. Let's fix this.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Fixes:
7eee13c4679 ("docs: use dl/dd instead of blockquote for
freedesktop link")
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Erico Nunes [Tue, 23 Apr 2019 17:36:34 +0000 (19:36 +0200)]
lima/ppir: support nir_op_ftrunc
Support nir_op_ftrunc by turning it into a mov with a round to integer
output modifier.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Eric Engestrom [Tue, 16 Apr 2019 16:06:53 +0000 (18:06 +0200)]
gitlab-ci: merge meson-glvnd into meson-swr
There's no need to have a whole build just for that flag, we can add it
to any build.
v2: Add a note about why we put glvnd where we did (by anholt).
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com> (v2)
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Tue, 16 Apr 2019 16:06:30 +0000 (18:06 +0200)]
gitlab-ci: simplify meson job names
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Tue, 16 Apr 2019 16:06:12 +0000 (18:06 +0200)]
gitlab-ci: meson-gallium-radeonsi was a subset of meson-gallium-clover-llvm
Let's just drop it.
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Eric Engestrom [Fri, 8 Mar 2019 16:33:07 +0000 (16:33 +0000)]
gitlab-ci: merge several meson jobs
Merge the following into `meson-main`/`meson-loader-classic-dri`/
`meson-gallium-swr`:
- meson-vulkan
- meson-gallium-drivers-other
- meson-gallium-st-other
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
[ Michel Dänzer ]
* Rebase and fix up commit log.
* Don't set VULKAN_DRIVERS in meson-loader-classic-dri.
* Remove extraneous whitespace.
* Squash in follow-up fixes.
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
[ anholt]
* Add a note why nine and swrast landed where they did.
* Switch from s/meson-vulkan/meson-main/ to
s/meson-loader-classic-dri/meson-main/ which I think was the original
intent
Reviewed-by: Eric Anholt <eric@anholt.net>
Acked-by: Eric Engestrom <eric.engestrom@intel.com> (anholt changes)
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Heinrich [Tue, 30 Apr 2019 12:21:04 +0000 (14:21 +0200)]
gbm: Improve documentation of BO import
- Add GBM_BO_IMPORT_FD_MODIFIER to documentation of supported foreign
object types
- Add newline before documentation block
- Improve language
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Samuel Pitoiset [Thu, 2 May 2019 15:44:39 +0000 (17:44 +0200)]
radv: only need to force emit the TCS regs on Vega10 and Raven1
Other GFX9 chips aren't affected.
Cc: "19.0" "19.1" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Marek Olšák [Wed, 24 Apr 2019 17:16:07 +0000 (13:16 -0400)]
glsl: fix and clean up NV_compute_shader_derivatives support
- make sure compute shader derivatives are exposed for all extensions
- unify duplicated code
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Marek Olšák [Mon, 22 Apr 2019 21:11:00 +0000 (17:11 -0400)]
st/dri: decrease input lag by syncing sooner in SwapBuffers
It's done by:
- decrease the number of frames in flight by 1
- flush before throttling in SwapBuffers
(instead of wait-then-flush, do flush-then-wait)
The improvement is apparent with Unigine Heaven.
Previously:
draw frame 2
wait frame 0
flush frame 2
present frame 2
The input lag is 2 frames.
Now:
draw frame 2
flush frame 2
wait frame 1
present frame 2
The input lag is 1 frame. Flushing is done before waiting, because
otherwise the device would be idle after waiting.
Nine is affected because it also uses the pipe cap.
Erik Faye-Lund [Fri, 12 Apr 2019 15:51:30 +0000 (17:51 +0200)]
meson: add build-summary
This roughly mirrors what we get from autotools. There's a few
differences, though:
1. The "exec_prefix" output has been dropped. Meson doesn't support
this, so it makes no sense here.
2. The "llvm-config" output has been dropped. Meson abstracts dependency
discovery a bit more than our autotools build-system does, so it's
not easy to get this information as-is.
3. HUD extra stats, SWR archs, Shared/Static libs and CFLAGS / CXXFLAGS /
LDFLAGS has been dropped. These can be inspected by "meson configure".
4. How we set defines works quite differently in our Meson build-system,
and the result isn't quite the same. In particular, the DEFINES output
has been dropped, to avoid having to refactor the code too much.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=109326
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Erik Faye-Lund [Fri, 12 Apr 2019 15:51:08 +0000 (17:51 +0200)]
meson: give dri- and gallium-drivers separate vars
Variables are cheap, and there's little reason for the dri and gallium
drivers to work on the same variable for the driver list. So let's split
these in two separate lists instead.
This makes it easier to inspect these after-the fact, for instance
for generating a summary of build-settings.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Erik Faye-Lund [Mon, 15 Apr 2019 08:11:43 +0000 (10:11 +0200)]
meson: lift driver-collection out into parent build-file
This way we can mark the dri_drivers and dri_link arrays as temporary,
as all knowledge about them are contained in a single build-file with
clearly visible limited life-span.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Rob Clark [Thu, 2 May 2019 17:04:17 +0000 (10:04 -0700)]
docs: mark KHR_blend_equation_advanced done on a6xx
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Mon, 29 Apr 2019 17:11:44 +0000 (10:11 -0700)]
freedreno/a6xx: smaller hammer for fb barrier
We just need to do a sequence of commands to flush the cache.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Tue, 30 Apr 2019 17:07:02 +0000 (10:07 -0700)]
freedreno/a6xx: KHR_blend_equation_advanced support
Wire up support to sample from the fb (and force GMEM rendering when we
have fb reads). The existing GLSL IR lowering for blend_equation_advanced
does the rest.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Tue, 30 Apr 2019 17:05:30 +0000 (10:05 -0700)]
freedreno/ir3: fb read support
Lower load_output to txf_ms_fb and add support for the new texture fetch
instruction.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Tue, 30 Apr 2019 17:03:54 +0000 (10:03 -0700)]
freedreno/drm: expose GMEM_BASE address
Needed for sampling from tile buffer (GMEM).
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Fri, 26 Apr 2019 17:05:08 +0000 (10:05 -0700)]
nir: add pass to lower fb reads
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Fri, 26 Apr 2019 18:39:42 +0000 (11:39 -0700)]
nir: fix lower_wpos_ytransform in load_frag_coord case
Apparently we never hit this path. Or at least haven't for a rather
long time. But in either case (load_deref or load_frag_coord), we can
just directly use the intrinsic's ssa dest. So stop passing the
nir_variable (which would be NULL in the load_frag_coord case) around
and instead just use &intr->dest.ssa.
(This ofc means we need to setup the cursor to insert *after* the
instruction, which seems to be another bug of the original
implementation.)
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Sun, 28 Apr 2019 15:06:12 +0000 (08:06 -0700)]
nir: rework tex instruction printing
The extra comma at the end was annoying me.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Rob Clark [Thu, 2 May 2019 16:37:21 +0000 (09:37 -0700)]
freedreno/ir3: add some ubo range related asserts
And a comment.. since we are mixing units of bytes/dwords/vec4,
hopefully this will avoid some unit confusion.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Thu, 2 May 2019 16:33:40 +0000 (09:33 -0700)]
freedreno/ir3: add IR3_SHADER_DEBUG flag to disable ubo lowering
It isn't quite as simple as not running the pass, since with packed
varyings we get load_ubo for block==0 (ie. the "real" uniforms). So
instead run the pass normally but decline to lower anything in
block > 0
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Thu, 2 May 2019 16:25:36 +0000 (09:25 -0700)]
freedreno/ir3: fix lowered ubo region alignment
Since we emit UBO regions INDIRECTly (ie. not copied into cmdstream but
emit by EXT_SRC_ADDR) we need to keep them 4*vec4 aligned. Which the
code already mostly did, except for aligning the first UBO region itself
(ie. the one after block==0 which is the "real" uniforms).
Fixes:
893425a607a freedreno/ir3: Push UBOs to constant file
Fixes:
3c8779af325 freedreno/ir3: Enable PIPE_CAP_PACKED_UNIFORMS
Signed-off-by: Rob Clark <robdclark@chromium.org>
Rob Clark [Wed, 1 May 2019 17:41:51 +0000 (10:41 -0700)]
freedreno/ir3: fix shader variants vs UBO analysis
Otherwise we zero out the state again, but all the UBO loads that we
could lower are already lowered. End result is that we didn't emit the
uniforms for lowered UBO access in any case where multiple shader
variants are used.
Fixes:
893425a607a freedreno/ir3: Push UBOs to constant file
Fixes:
3c8779af325 freedreno/ir3: Enable PIPE_CAP_PACKED_UNIFORMS
Signed-off-by: Rob Clark <robdclark@chromium.org>
Lionel Landwerlin [Thu, 28 Feb 2019 13:02:49 +0000 (13:02 +0000)]
vulkan/overlay: add TODO list
Keen on having other people contribute.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Lionel Landwerlin [Sun, 3 Mar 2019 01:51:43 +0000 (01:51 +0000)]
vulkan/overlay: make overriden functions static
And fix the unused CmdDrawIndirect.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Sun, 3 Mar 2019 00:09:04 +0000 (00:09 +0000)]
vulkan/overlay: make overlay size configurable
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Sun, 3 Mar 2019 18:08:36 +0000 (18:08 +0000)]
vulkan/overlay: add a frame counter option
This is useful to normalize the numbers written into the output file
as those number are accumulated over a period of time and number of
frames.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Sat, 2 Mar 2019 17:49:21 +0000 (17:49 +0000)]
vulkan/overlay: record all select metrics into output file
The output looks something like this (csv style) :
fps, frame, frame_timing(us), submit, draw_indexed, pipeline_graphics, acquire_timing(us), vert_invocations, frag_invocations, gpu_timing(ns)
480.55, 242, 501512, 247, 1444, 1204, 714, 5827272,
113043296,
121424174
467.80, 234, 500214, 234, 1412, 1176, 648, 5635680,
109436188,
117743760
424.37, 213, 501923, 213, 2130, 1704, 623, 5132448,
99657292,
105474683
472.15, 237, 501962, 237, 2370, 1896, 667, 5710752,
110924644,
122226004
411.32, 206, 500826, 206, 2060, 1648, 709, 4963776,
96491764,
95333273
458.87, 230, 501228, 230, 2300, 1840, 634, 5542080,
107758204,
123112090
475.01, 238, 501044, 238, 2380, 1904, 631, 5734848,
111477480,
122087426
471.08, 236, 500972, 236, 2360, 1888, 655, 5686656,
110498496,
114816162
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Sat, 2 Mar 2019 17:29:12 +0000 (17:29 +0000)]
vulkan/overlay: add a margin to the size of the window
Looks a bit better.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Sat, 2 Mar 2019 17:25:22 +0000 (17:25 +0000)]
vulkan/overlay: add no display option
In case you're just interested in data being record to the output
file.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Mon, 25 Feb 2019 16:59:11 +0000 (16:59 +0000)]
vulkan/overlay: add pipeline statistic & timestamps support
v2: switch to VkBase{In,Out}Structure
v3: Add timestamps at begin/end of primary command buffers to estimate
gpu time spent per submission (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Eric Engestrom <eric.engestrom@intel.com> (v2)
Lionel Landwerlin [Sat, 2 Mar 2019 17:15:41 +0000 (17:15 +0000)]
vulkan/overlay: record stats in command buffers and accumulate on exec/submit
This significantly reworks how numbers displayed are computed. We
accumulate operations written into command buffers and add those to
the device when submitted to a queue. These collected values are then
used to compute per frame overlay data.
We also accumulate the data over the sampling fps period to produce
numbers for that period of time.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Lionel Landwerlin [Thu, 28 Feb 2019 16:47:56 +0000 (16:47 +0000)]
vulkan/overlay: update help printout
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Lionel Landwerlin [Mon, 25 Feb 2019 16:43:15 +0000 (16:43 +0000)]
vulkan/util: generate a helper function to return pNext struct sizes
This will be used to copy chains of structures so that we can alterate
some of them.
v2: Drop vk_util.h include (Eric)
Use VkBaseInStructure directly (Eric)
v3: Drop --platforms= param to generator script, instead produce a
file with #ifdef based what platforms are compiled.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Tomeu Vizoso [Mon, 22 Apr 2019 15:06:24 +0000 (17:06 +0200)]
panfrost/midgard: Skip liveness analysis for instructions without dest
[Alyssa: Add comment explanation]
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Tomeu Vizoso [Mon, 22 Apr 2019 15:03:26 +0000 (17:03 +0200)]
panfrost/midgard: Skip register allocation if there's no work to do
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Eric Engestrom [Fri, 8 Mar 2019 12:47:23 +0000 (12:47 +0000)]
gitlab-ci: add scons windows build using mingw
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Eric Engestrom [Fri, 12 Apr 2019 16:47:59 +0000 (17:47 +0100)]
egl: hard-code destroy function instead of passing it around as a pointer
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Connor Abbott [Mon, 18 Feb 2019 16:28:32 +0000 (17:28 +0100)]
nir/search: Add debugging code to dump the pattern matched
This was useful while debugging the previous commit.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Connor Abbott [Mon, 18 Feb 2019 13:20:34 +0000 (14:20 +0100)]
nir/search: Add automaton-based pre-searching
nir_opt_algebraic is currently one of the most expensive NIR passes,
because of the many different patterns we've added over the years. Even
though patterns are already sorted by opcode, there are still way too
many patterns for common opcodes like bcsel and fadd, which means that
many patterns are tried but only a few actually match. One way to fix
this is to add a pre-pass over the code that scans it using an automaton
constructed beforehand, similar to the automatons produced by lex and
yacc for parsing source code. This automaton has to walk the SSA graph
and recognize possible pattern matches.
It turns out that the theory to do this is quite mature already, having
been developed for instruction selection as well as other non-compiler
things. I followed the presentation in the dissertation cited in the
code, "Tree algorithms: Two Taxonomies and a Toolkit," trying to keep
the naming similar. To create the automaton, we have to perform
something like the classical NFA to DFA subset construction used by lex,
but it turns out that actually computing the transition table for all
possible states would be way too expensive, with the dissertation
reporting times of almost half an hour for an example of size similar to
nir_opt_algebraic. Instead, we adopt one of the "filter" approaches
explained in the dissertation, which trade much faster table generation
and table size for a few more table lookups per instruction at runtime.
I chose the filter which resulted the fastest table generation time,
with medium table size. Right now, the table generation takes around .5
seconds, despite being implemented in pure Python, which I think is good
enough. Based on the numbers in the dissertation, the other choice might
make table compilation time 25x slower to get 4x smaller table size, but
I don't think that's worth it. As of now, we get the following binary
size before and after this patch:
text data bss dec hex filename
11979455 464720 730864
13175039 c908ff before i965_dri.so
text data bss dec hex filename
12037835 616244 791792
13445871 cd2aef after i965_dri.so
There are a number of places where I've simplified the automaton by
getting rid of details in the LHS patterns rather than complicate things
to deal with them. For example, right now the automaton doesn't
distinguish between constants with different values. This means that it
isn't as precise as it could be, but the decrease in compile time is
still worth it -- these are the compilation time numbers for a shader-db
run with my (admittedly old) database on Intel skylake:
Difference at 95.0% confidence
-42.3485 +/- 1.375
-7.20383% +/- 0.229926%
(Student's t, pooled s = 1.69843)
We can always experiment with making it more precise later.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Samuel Pitoiset [Thu, 2 May 2019 12:56:46 +0000 (14:56 +0200)]
radv: set WD_SWITCH_ON_EOP=1 when drawing primitives from a stream output buffer
According to RadeonSI, this seems to be required by the hardware
to avoid GPU hangs. I think I just forgot to set that bit when I
implemented VK_EXT_transform_feedback.
This fixes a GPU hang with Space Engineers and DXVK.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110291
Fixes:
b4eb029062a ("radv: implement VK_EXT_transform_feedback")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Brian Paul [Thu, 2 May 2019 12:32:56 +0000 (06:32 -0600)]
glsl: fix typo in #warning message
Trivial. Spotted by Eric Engestrom.
Brian Paul [Wed, 1 May 2019 21:49:44 +0000 (15:49 -0600)]
svga: add SVGA_NO_LOGGING env var (v2)
valgrind crashes when we try to initialize host logging. This
env var can be used to disable logging.
v2: rebase onto "svga: move host logging to winsys".
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Neha Bhende <bhenden@vmware.com>
Charmaine Lee [Wed, 1 May 2019 21:35:54 +0000 (15:35 -0600)]
svga: move host logging to winsys
This patch adds a host_log interface to svga_winsys and
moves the host logging code to the winsys layer.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Neha Bhende <bhenden@vmware.com>
Eric Engestrom [Fri, 5 Apr 2019 15:18:18 +0000 (16:18 +0100)]
wsi/wayland: document lack of vkAcquireNextImageKHR timeout support
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Daniel Stone [Tue, 30 Oct 2018 12:56:53 +0000 (12:56 +0000)]
vulkan/wsi/wayland: Respect non-blocking AcquireNextImage
If the client has requested that AcquireNextImage not block at all, with
a timeout of 0, then don't make any non-blocking calls.
This will still potentially block infinitely given a non-infinte
timeout, but the fix for that is much more involved.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Cc: mesa-stable@lists.freedesktop.org
Cc: Chad Versace <chadversary@chromium.org>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108540
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 11:52:15 +0000 (13:52 +0200)]
docs: reorder heading and notice
All other pages has the heading as ghe first thing in the article. Let's
clean this up for consistency.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 11:45:53 +0000 (13:45 +0200)]
docs: drop centered heading for faq
The FAQ is the only article we have that uses a centered heading, which
makes it look odd compared to the other articles. Let's drop the
centering for consistency.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 11:44:10 +0000 (13:44 +0200)]
docs: turn faq-index into an ordered list
HTML already have a way of doing automatically ordered lists, so let's
use that instead of open-coding one.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:41:08 +0000 (16:41 +0200)]
docs: replace empty list with a none-paragraph
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:39:25 +0000 (16:39 +0200)]
docs: fix closing of list-items
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:36:52 +0000 (16:36 +0200)]
docs: fixup list-item tags
The list items needs to contain everything part of the item, not just
the first paragraph.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:32:07 +0000 (16:32 +0200)]
docs: fix closing of paragraphs
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:31:26 +0000 (16:31 +0200)]
docs: add missing lists
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:27:26 +0000 (16:27 +0200)]
docs: fixup bad paragraphing
This markup seems to assume paragraphs survive across block-elements,
which isn't the case. Let's rectify that.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:27:10 +0000 (16:27 +0200)]
docs: remove stray list-start
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:25:39 +0000 (16:25 +0200)]
docs: don't pointlessly close and re-start definition lists
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:16:06 +0000 (16:16 +0200)]
docs: fix incorrectly closed paragraph
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:14:40 +0000 (16:14 +0200)]
docs: drop paragraph around preformatted text
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:13:44 +0000 (16:13 +0200)]
docs: start paragraph before closing it
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:12:08 +0000 (16:12 +0200)]
docs: close paragraphs before preformatted text
It's illegal to nest block-level elements such as <pre> inside <p> in
HTML. This means that when the paragraphs gets closed after a <pre>-tag,
we end up closing a non-existent tag, so the browser inserts a dummy
<p>-tag. This is entirely pointless, so let's just close these tags
before the <pre>-tag instead.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Erik Faye-Lund [Thu, 18 Apr 2019 14:05:54 +0000 (16:05 +0200)]
docs: remove stray paragraph-close
This isn't matching any paragraph-open tags, so let's get rid of it.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>