Tim Rowley [Mon, 27 Feb 2017 18:11:47 +0000 (10:11 -0800)]
swr: [rasterizer] Slight assert refactoring
Make asserts more robust.
Add SWR_INVALID(...) as a replacement for SWR_ASSERT(0, ...)
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Thu, 16 Feb 2017 18:53:01 +0000 (10:53 -0800)]
swr: [rasterizer] Backend code adjustments
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Wed, 22 Feb 2017 19:11:58 +0000 (11:11 -0800)]
swr: [rasterizer archrast] Fix the early and late depthstencil events
The coverage and stencil mask arguments were reversed.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Tue, 21 Feb 2017 19:20:38 +0000 (11:20 -0800)]
swr: [rasterizer core] Implement double pumped SIMD16 TESS
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Sat, 18 Feb 2017 08:29:06 +0000 (00:29 -0800)]
swr: [rasterizer archrast/core/scripts] Fix archrast multithreading issue
Per pixel stats are cached but were not always being flushed as threads
moved from one draw context to the next. Added an explicit flush to allow
all archrast objects to flush any cached events.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Fri, 17 Feb 2017 22:14:06 +0000 (14:14 -0800)]
swr: [rasterizer archrast] Remove redundant data from archrast files
If count can be derived from other counts then this can be done in
post processing scripts.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Fri, 17 Feb 2017 21:16:59 +0000 (13:16 -0800)]
swr: [rasterizer archrast/scripts] Further archrast cleanups
Removed redundant data being written out to file
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Fri, 17 Feb 2017 04:28:28 +0000 (20:28 -0800)]
swr: [rasterizer core] Fix RECT_LIST primitive assembly
The bug would make the 3rd component of attributes on the second
triangle of a RECT be invalid.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Fri, 17 Feb 2017 02:31:09 +0000 (18:31 -0800)]
swr: [rasterizer common] Add InterpolateComponentFlat utility
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Thu, 16 Feb 2017 22:48:28 +0000 (14:48 -0800)]
swr: [rasterizer archrast] Fix performance issue with archrast stats
Performance is now 50x faster with archrast now that we're properly
filtering out all of the rdtsc begin/end.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Thu, 16 Feb 2017 21:50:21 +0000 (13:50 -0800)]
swr: [rasterizer core] Implement SIMD16 GS and STREAMOUT
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Thu, 16 Feb 2017 07:57:50 +0000 (23:57 -0800)]
swr: [rasterizer archrast] Add additional API events
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Wed, 15 Feb 2017 21:45:16 +0000 (13:45 -0800)]
swr: [rasterizer core/scripts] Autogen backend initialization function(s)
Autogen functions that instantiates different BackendPixelRate templates.
Functions get split into separate files after reaching a user defined
threshold (currently 512 per file) to speed up compilation.
This change will enable the addition of more template flags in the pixel
back end.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Thu, 16 Mar 2017 17:00:15 +0000 (12:00 -0500)]
swr: [rasterizer core] backend.h declares gBackendPixelRateTable
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Fri, 10 Feb 2017 22:56:57 +0000 (14:56 -0800)]
swr: [rasterizer core] Finish SIMD16 PA OPT including tesselation
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Thu, 9 Feb 2017 21:43:32 +0000 (13:43 -0800)]
swr: [rasterizer core] Finish SIMD16 PA OPT except tesselation
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Mon, 6 Feb 2017 23:25:57 +0000 (15:25 -0800)]
swr: [rasterizer core] Support sparse numa id values on all OSes
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Kenneth Graunke [Fri, 3 Mar 2017 02:27:32 +0000 (18:27 -0800)]
i965: Skip register write detection when possible.
Detecting register write support by trial and error introduces a
stall at screen creation time, which it would be nice to avoid.
Certain command parser versions guarantee this will work (see the
giant comment in intelInitScreen2 below, or a few commits ago):
- Ivybridge: version >= 1 (kernel v3.16)
- Baytrail: version >= 2 (kernel v3.19)
- Haswell: version >= 7 (kernel v4.8)
For simplicity, we don't bother with version 1 in this patch.
This assumes that the user hasn't disabled aliasing PPGTT via a kernel
command line parameter. Don't do that - you're only breaking things.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Kenneth Graunke [Fri, 3 Mar 2017 02:21:31 +0000 (18:21 -0800)]
i965: Set screen->cmd_parser_version to 0 if we can't write registers.
If we can't write registers, then the effective command parser version
is 0 - it may exist, but it's not usefully enabling anything.
See kernel commit
1ca3712ca3429a617ed6c5f87718e4f6fe4ae0c6 (in v4.8)
where the kernel starts doing this for us. This makes us do more or
less the same thing on older kernels.
This should preserve a bit of sanity by allowing us to perform a
screen->cmd_parser_version > N check to determine that we really can
use the features promised by command parser version N.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Kenneth Graunke [Fri, 3 Mar 2017 02:12:28 +0000 (18:12 -0800)]
i965: Document the sad story of the kernel command parser.
This should help us figure out the complexities of which kernel
versions we need to get various features on various platforms.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Kenneth Graunke [Thu, 2 Mar 2017 19:33:37 +0000 (11:33 -0800)]
i965: Fall back to GL 4.2/4.3 on Haswell if the kernel isn't new enough.
In commit
d2590eb65ff28a9cbd592353d15d7e6cbd2c6fc6 I enabled GL 4.5
on Haswell...but failed to check if we could do indirect compute
shader dispatch...and query buffer objects.
Indirect compute shader dispatch requires command parser version 5
(kernel commit
7b9748cb513a6bef4af87b79f0da3ff7e8b56cd8, which is in
Linux v4.4). On earlier kernels we would have disabled
ARB_compute_shader, which is a mandatory part of OpenGL 4.3+.
Query buffer objects currently require MI_MATH and MI_LOAD_REGISTER_REG,
which mean command parser version 7 (Linux v4.8). On earlier kernels
we would have disabled ARB_query_buffer_object, which is a mandatory
part of OpenGL 4.4+.
The new version support looks like:
- Kernel 4.1 and older => OpenGL 3.3
- Kernel 4.2-4.3 => OpenGL 4.2
- Kernel 4.4-4.7 => OpenGL 4.3
- Kernel 4.8+ => OpenGL 4.5
Cc: "17.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Constantine Kharlamov [Mon, 20 Mar 2017 18:16:25 +0000 (21:16 +0300)]
r600g/sb: Fix memory leak by reworking uses list (rebased)
The author is Heiko Przybyl(CC'ing), the patch is rebased on top of Bartosz Tomczyk's one per Dieter Nützel's comment.
Tested-by: Constantine Charlamov <Hi-Angel@yandex.ru>
v2: Resend the patch again through git-email. The prev. rebase was sent
through Thunderbird, which screwed up tab characters, making the patch
not apply.
--------------
When fixing the stalls on evergreen I introduced leaking of the useinfo
structure(s). Sorry. Instead of allocating a new object to hold 3 values
where only one is actually used, rework the list to just store the node
pointer. Thus no allocating and deallocation is needed. Since use_info
and use_kind aren't used anywhere, drop them and reduce code complexity.
This might also save some small amount of cycles.
Thanks to Bartosz Tomczyk for finding the bug.
Reported-by: Bartosz Tomczyk <bartosz.tomczyk86 at gmail.com <https://lists.freedesktop.org/mailman/listinfo/mesa-dev>>
Signed-off-by: Heiko Przybyl <lil_tux at web.de <https://lists.freedesktop.org/mailman/listinfo/mesa-dev>>
Supersedes: https://patchwork.freedesktop.org/patch/135852
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Marek Olšák [Mon, 20 Mar 2017 15:39:02 +0000 (16:39 +0100)]
radeonsi: check the IR type before waiting for a compute compilation fence
This should fix OpenCL getting stuck.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100288
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Kenneth Graunke [Mon, 20 Mar 2017 05:11:52 +0000 (22:11 -0700)]
aubinator: Move the guts of decode_group() to decoder.c.
This lets us use it outside of the aubinator binary itself.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Mon, 20 Mar 2017 05:10:46 +0000 (22:10 -0700)]
aubinator: Drop spec parameter to decode_group().
No longer necessary - the iterator gets it from the group.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Mon, 20 Mar 2017 04:22:20 +0000 (21:22 -0700)]
aubinator: Make the iterator store a pointer to structure descriptions.
When the iterator encounters a structure field, it now looks up the
gen_group for that structure definition and saves a pointer to it.
This lets us drop a lot of ridiculous code in the caller, which looked
at item->value (<struct NAME dword>), strtok'd the structure name back
out, and looked it up itself.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Mon, 20 Mar 2017 04:45:20 +0000 (21:45 -0700)]
aubinator: Track the current field's starting dword offset.
The iterator code already computed this value, then we stored it in
the structure name, strtok'd it back out, and also manually computed
it when printing dword headers.
Just put the value in the struct and use it. Way simpler.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Mon, 20 Mar 2017 04:47:25 +0000 (21:47 -0700)]
aubinator: Drop decode_structure() helper.
It made more sense when decode_group() took a bunch of extra options,
but now that there's only one...we may as well pass 0 and call it a day.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Mon, 20 Mar 2017 04:30:37 +0000 (21:30 -0700)]
aubinator: Drop unused print_dword_headers flag.
I added this flag in
65a9d5eabb05e4925c1c9a17836cad57304210d6 but
it was completely unused. Both callers appear to have printed dword
headers, so we can just drop the flag and continue doing it
unconditionally.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Mon, 20 Mar 2017 04:24:24 +0000 (21:24 -0700)]
aubinator: Store a pointer from gen_group back to gen_spec.
When decoding a structure field within a group, we may want to look up
that structure type. Having a gen_spec pointer makes it easy to do so.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Kenneth Graunke [Mon, 20 Mar 2017 03:59:08 +0000 (20:59 -0700)]
aubinator: Store enum textual name in iter->value.
gen_field_iterator_next() produces a string representing the value of
the field. For enum values, it also produced a separate "description"
string containing the textual name of the enum.
The only caller of this function combines the two, printing enums as
"<numeric value> (<texture enum name>)". We may as well just store
that in item->value directly, eliminating the description field, and
a layer of wrapping.
v2: Use non-overlapping source and destination strings in snprintf.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Julien Isorce [Thu, 16 Mar 2017 13:09:21 +0000 (13:09 +0000)]
si_descriptor: move velems nullity check before dereference
CID 1399479: Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking velems suggests that it may be null,
but it has already been dereferenced on all paths leading to the check.
Signed-off-by: Julien Isorce <jisorce@oblong.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Julien Isorce [Wed, 15 Mar 2017 17:40:25 +0000 (17:40 +0000)]
radeon_drm_bo: explicitly check return value of drmCommandWriteRead
CID 1313492
Signed-off-by: Julien Isorce <jisorce@oblong.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Julien Isorce [Wed, 15 Mar 2017 17:31:40 +0000 (17:31 +0000)]
si_pipe: remove nullity check after dereference
sscreen cannot be NULL
CID 1354483
Signed-off-by: Julien Isorce <jisorce@oblong.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Julien Isorce [Mon, 27 Feb 2017 13:42:17 +0000 (13:42 +0000)]
radeon: initialize hole variable before calling container_of
Like in a few other places in that radeon_drm_bo.c file.
CID 715739.
Signed-off-by: Julien Isorce <jisorce@oblong.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nanley Chery [Tue, 7 Mar 2017 19:17:05 +0000 (11:17 -0800)]
intel: Correct the BDW surface state size
The PRMs state that this packet is 16 DWORDS long. Ensure that the last
three DWORDS are zeroed as required by the hardware when allocating a
null surface state.
Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Bartosz Tomczyk [Wed, 8 Feb 2017 16:16:13 +0000 (17:16 +0100)]
r600g: Fix out of bounds access
fc_sp variable should indicate number of elements in
fc_stack array, but fc_sp was increased at beginning of fc_pushlevel
function. It leads to situation where idx=0 was never used, and last
32 element was stored outside fs_stack array.
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Constantine Kharlamov [Mon, 20 Mar 2017 12:19:43 +0000 (15:19 +0300)]
r600g: update sb documentation
v2: s/r600/r600g in the title
Signed-off-by: Constantine Kharlamov <Hi-Angel@yandex.ru>
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Constantine Kharlamov [Mon, 20 Mar 2017 12:19:42 +0000 (15:19 +0300)]
r600g: make condition clearer
The second check in the old code looked pretty much unreachable, esp.
because it's not obvious that "max_entries" could be zero. To find out
that it was intentional I had to run some checks, and to dig into
the old versions of the file.
So, rewrite the check to make the intention clear.
v2: s/r600/r600g in the title, and per Dieter Nützel's comment wrap
lines of condition.
Signed-off-by: Constantine Kharlamov <Hi-Angel@yandex.ru>
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Emil Velikov [Mon, 20 Mar 2017 14:25:18 +0000 (14:25 +0000)]
docs: add news item and link release notes for 13.0.6/17.0.2
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Mon, 20 Mar 2017 14:17:20 +0000 (14:17 +0000)]
docs: add sha256 checksums for 17.0.2
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
(cherry picked from commit
9b66351f5b274f3d79cb2c48afa3b2fcc2bf3442)
Emil Velikov [Mon, 20 Mar 2017 14:07:38 +0000 (14:07 +0000)]
docs: add release notes for 17.0.2
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
(cherry picked from commit
373d88a7117150de984510453e1c30a455987686)
Emil Velikov [Mon, 20 Mar 2017 11:54:35 +0000 (11:54 +0000)]
docs: add sha256 checksums for 13.0.6
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
(cherry picked from commit
879d24c49727cfc6c62cbd5bca58efad4c914e40)
Emil Velikov [Mon, 20 Mar 2017 11:42:19 +0000 (11:42 +0000)]
docs: add release notes for 13.0.6
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
(cherry picked from commit
fcef88d13a9ebdcadc6a878e9284c55651785301)
Xu,Randy [Sat, 18 Mar 2017 11:20:17 +0000 (19:20 +0800)]
anv/genX: Solve the vkCreateGraphicsPipelines crash
The crash is due to NULL pColorBlendState, which is legal if the
pipeline has rasterization disabled or if the subpass of the render pass
the pipeline is created against does not use any color attachments.
Test: Sample subpasses from LunarG can run without crash
Signed-off-by: Xu,Randy <randy.xu@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: "17.0 13.0" <mesa-stable@lists.freedesktop.org>
Dave Airlie [Sun, 19 Mar 2017 23:00:36 +0000 (09:00 +1000)]
radv: fix logic for when to flush on multiple CS emission
The current code evaluated to always true, we only want to flush
on the first submit. Rename the variable to do_flush, and only
emit on the first iteration.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Jason Ekstrand [Fri, 17 Mar 2017 04:18:10 +0000 (21:18 -0700)]
spirv: Implement IsInf using an integer comparison
Since we already do fabs on the one source, we're guaranteed to get
positive infinity if we get any infinity at all. Since +inf only has
one IEEE 754 representation, we can use an integer comparison and avoid
all of the ordered/unordered issues.
Cc: Dave Airlie <airlied@redhat.com>
Reviewed-by: Elie Tournier <elie.tournier@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 17 Mar 2017 04:23:56 +0000 (14:23 +1000)]
radv/meta: fix image clears for r4g4 format.
This just uses an 8-bit clear and packs the values.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 20 Mar 2017 03:24:02 +0000 (13:24 +1000)]
Revert "radv: fallback to an in-memory cache when no pipline cache is provided"
This reverts commit
2845a108a9a8bd4b0e6e9b590c976452fb99eb10.
This break VK-GL-CTS randomly.
./deqp-vk --deqp-case=dEQP-VK.texture.filtering.3d.formats.r4g4b4a4*
bounces around here from 6/6 to 3/6 or 4/6 to hanging.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Timothy Arceri [Thu, 16 Mar 2017 06:01:26 +0000 (17:01 +1100)]
mesa: disable glthread when glNewList() is called
glNewList() swaps dispatch tables, and we don't have anything in
place to handle that in glthread.
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Dave Airlie [Sun, 19 Mar 2017 04:17:14 +0000 (14:17 +1000)]
radv: fix primitive reset index emission
This was meant to be checking the index type to get the correct
index not the last emitted one. This fixes:
dEQP-VK.pipeline.input_assembly.primitive_restart.index_type_uint32.triangle_strip_with_adjacency
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Cc: "13.0 17.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Grazvydas Ignotas [Sat, 18 Mar 2017 20:58:55 +0000 (22:58 +0200)]
util/disk_cache: check rename result
I haven't seen this causing problems in practice, but for correctness
we should also check if rename succeeded to avoid breaking accounting
and leaving a .tmp file behind.
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Grazvydas Ignotas [Sat, 18 Mar 2017 20:58:54 +0000 (22:58 +0200)]
util/disk_cache: delete .tmp if target exists
At the time of target file check, .tmp file is already created and file
lock is held, so we should remove the .tmp, like in other error paths.
With this, piglit no longer leaves large amount of empty .tmp files
behind, which waste directory entries and may interfere with eviction.
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Grazvydas Ignotas [Sat, 18 Mar 2017 22:46:39 +0000 (00:46 +0200)]
util/disk_cache: fix stored_keys index
It seems there is a bug because:
- 20 bytes are compared, but only 1 byte stored_keys step is used
- entries can overlap each other by 19 bytes
- index_mmap is ~1.3M in size, but only first 64K is used
With this fix for Deus Ex:
- startup time (from launch to Feral logo): ~38s -> ~16s
- disk_cache_has_key() hit rate: ~50% -> ~96%
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Ilia Mirkin [Sun, 19 Mar 2017 05:22:29 +0000 (01:22 -0400)]
nv30: create uploader after pipe->screen is set
Fixes crashes after recent upload rework.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Ilia Mirkin [Thu, 16 Mar 2017 03:29:47 +0000 (23:29 -0400)]
nv50,nvc0: enable TEX_LZ and TXF_LZ
There should be minimal gain, if any, for nvc0, but nv50 may end up
noticing more often that the lod argument is uniform. This, in turn,
will remove the need for some unnecessary transformations, which were
being hit due to the checks being done pre-ssa.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Ilia Mirkin [Sat, 4 Mar 2017 18:52:48 +0000 (13:52 -0500)]
st/mesa: set result writemask based on ir type
This prevents textureQueryLevels, which maps as LODQ, from ending up
with a xyzw writemask, which is illegal.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100061
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Karol Herbst [Sun, 19 Mar 2017 00:08:56 +0000 (01:08 +0100)]
nvc0/ir: treat FMA like MAD for operand propagation
Helps mainly Feral-ported games, due to their use of fma()
shader-db changes:
total instructions in shared programs : 3901147 -> 3842505 (-1.50%)
total gprs used in shared programs : 471258 -> 467359 (-0.83%)
total local used in shared programs : 27405 -> 27361 (-0.16%)
total bytes used in shared programs :
35749888 ->
35214176 (-1.50%)
local gpr inst bytes
helped 17 1829 4091 4091
hurt 4 44 3 3
Signed-off-by: Karol Herbst <karolherbst@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Alan Swanson [Fri, 17 Mar 2017 18:05:43 +0000 (18:05 +0000)]
util/disk_cache: pass predicate functions file stats directly (v4)
Since switching to LRU eviction the only user of these predicate
functions now resolves directory entry stats itself so pass them
directly saving calling fstat and strlen twice (and the
expensive strlen is skipped entirely if access time is newer).
v2: Update for empty cache dir detection changes
v3: Fix passing string length to predicate with the +1 for NULL
termination and also pass sb as pointer
v4: Missed ampersand for passing sb as pointer
Reviewed-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Timothy Arceri [Fri, 17 Mar 2017 13:40:56 +0000 (00:40 +1100)]
glsl: use set for copy propagation kills
Previously each time we saw a variable we just created a duplicate
entry in the list. This is particularly bad for loops were we add
everything twice, and then throw nested loops into the mix and the
list was growing expoentially.
This stops the glsl-vs-unroll-explosion test which has 16 nested
loops from reaching the tests mem usage limit in this pass. The
test now hits the mem limit in opt_copy_propagation_elements()
instead.
I suspect this was also part of the reason this pass can be so
slow with some shaders.
Reviewed-by: Thomas Helland <thomashelland90@gmail.com>
Timothy Arceri [Fri, 17 Mar 2017 00:02:14 +0000 (11:02 +1100)]
st/dri: wait for thread to finish before unbinding context
Fixes a bunch of piglit crashes that hit an assert() when trying
to delete the framebuffer. The assert() was triggered because
WinSysDrawBuffer was set to NULL before glDeleteFramebuffers()
was called.
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Timothy Arceri [Fri, 17 Mar 2017 10:53:35 +0000 (21:53 +1100)]
glsl: don't leak memory when trying to count loop iterations
Suggested-by: Damian Dixon <damian.dixon@gmail.com>
Reviewed-by: Elie Tournier <elie.tournier@collabora.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99789
Jason Ekstrand [Tue, 14 Mar 2017 23:22:00 +0000 (16:22 -0700)]
genxml: Make MI_STORE_DATA_IMM have a single 64-bit data field
This is way more convenient than having two separate dword fields.
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Thu, 16 Mar 2017 21:17:59 +0000 (14:17 -0700)]
anv: Turn on inherited queries
It all just works since it's just a hardware register so we might as
well turn it on.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Ilia Mirkin [Sun, 27 Nov 2016 19:23:23 +0000 (14:23 -0500)]
anv: Implement pipeline statistics queries
In the end, pipeline statistics queries look a lot like occlusion
queries only with between 1 and 11 begin/end pairs being generated
instead of just the one.
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Thu, 16 Mar 2017 21:12:03 +0000 (14:12 -0700)]
anv: Disable VF statistics for blorp and SOL memcpy
In order to get accurate statistics, we need to disable statistics for
blits, clears, and the surface state memcpy at the top of each secondary
command buffer. There are two possible approaches to this:
1) Disable before the blit/memcpy and re-enable afterwards
2) Move emitting 3DSTATE_VF_STATISTICS from initialization and make it
part of pipeline state and then just disabale statistics before
blits and memcpy operations.
Emitting 3DSTATE_VF_STATISTICS should be fairly cheap so it doesn't
really matter which path we take. We choose the second option as it's
more consistent with the way the rest of the statistics are enabled and
disabled.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Thu, 16 Mar 2017 19:32:16 +0000 (12:32 -0700)]
anv/pipeline: Enable clipper statistics
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Thu, 16 Mar 2017 19:31:42 +0000 (12:31 -0700)]
genxml: s/Clipper Statistics Enable/Statistics Enable/
It's in 3DSTATE_CLIP, so it doesn't really need the extra detail. This
matches what we do for VS, FS, etc.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 15 Mar 2017 22:29:13 +0000 (15:29 -0700)]
anv/query: Rework store_query_result
The new version is a nice GPU parallel to cpu_write_query_result and it
nicely handles things like dealing with 32 vs. 64-bit offsets in the
destination buffer.
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 15 Mar 2017 22:24:48 +0000 (15:24 -0700)]
anv/query: Break GPU query calculation into a helper
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 15 Mar 2017 22:18:21 +0000 (15:18 -0700)]
genxml: Add pipeline statistics registers on gen7+
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 15 Mar 2017 21:15:49 +0000 (14:15 -0700)]
anv/query: Add a helper for writing a query pool result
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 15 Mar 2017 00:58:44 +0000 (17:58 -0700)]
anv/query: Use a variable-length slot size
Not all queries are the same. Even the two queries we support today
require a different amount of data per slot. Once we introduce pipeline
statistics queries, the size will vary wildly.
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 15 Mar 2017 00:56:05 +0000 (17:56 -0700)]
anv/query: Move the available bits to the front
We're about to make slots variable-length and always having the
available bits at the front makes certain operations substantially
easier once we do that.
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Jason Ekstrand [Wed, 15 Mar 2017 23:43:54 +0000 (16:43 -0700)]
anv/query: Let 32-bit values wrap
From the Vulkan 1.0.39 Specification:
"If VK_QUERY_RESULT_64_BIT is not set and the result overflows a
32-bit value, the value may either wrap or saturate."
So we can either clamp or wrap. Wrapping is both easier and what the
user gets if they use vkCmdCopyQueryPoolResults and we should be
consistent. We could make vkCmdCopyQueryPoolResults clamp but it's
annoying and ends up burning extra batch for something the spec clearly
doesn't require.
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Alex Deucher [Fri, 17 Mar 2017 15:13:09 +0000 (11:13 -0400)]
radeonsi: add new polaris12 pci id
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Cc: 17.0 <mesa-stable@lists.freedesktop.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Wed, 1 Mar 2017 18:24:42 +0000 (19:24 +0100)]
gallium/radeon: formalize that create_batch_query doesn't need pipe_context
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Marek Olšák [Wed, 1 Mar 2017 18:24:42 +0000 (19:24 +0100)]
gallium/radeon: formalize that create_query doesn't need pipe_context
for threaded gallium
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Marek Olšák [Thu, 2 Mar 2017 00:06:19 +0000 (01:06 +0100)]
gallium/radeon: reference pipe_resource in pipe_transfer
for threaded gallium
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Marek Olšák [Sun, 5 Mar 2017 00:14:51 +0000 (01:14 +0100)]
radeonsi: compile all TGSI compute shaders asynchronously
required by threaded gallium
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Marek Olšák [Thu, 2 Mar 2017 23:24:03 +0000 (00:24 +0100)]
radeonsi: require that compiler threads are enabled
threaded gallium can't use pipe_context's LLVM target machine, because
create_shader_selector can be called from a non-driver thread.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Marek Olšák [Thu, 2 Mar 2017 18:01:35 +0000 (19:01 +0100)]
trace: remove leftover assertions after pipe_resource wrapping removal
Marek Olšák [Thu, 2 Mar 2017 00:42:37 +0000 (01:42 +0100)]
gallium/u_upload: make the first persistent mapping unsynchronized
This is simpler for drivers.
Robert Bragg [Fri, 17 Mar 2017 15:54:07 +0000 (15:54 +0000)]
anv/device: init timestampPeriod from devinfo
Now that there's a timebase_scale in gen_device_info which is
effectively the 'period' this switches anv_GetPhysicalDeviceProperties
to using this common device info to initialize the timestampPeriod
device limit.
Signed-off-by: Robert Bragg <robert@sixbynine.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Robert Bragg [Thu, 27 Oct 2016 21:08:19 +0000 (22:08 +0100)]
i965: Allow a per gen timebase scale factor
Prior to Skylake the Gen HW timestamps were driven by a 12.5MHz clock
with the convenient property of being able to scale by an integer (80)
to nanosecond units.
For Skylake the frequency is 12MHz or a scale factor of 83.333333
This updates gen_device_info to track a floating point timebase_scale
factor and makes corresponding _queryobj.c changes to no longer assume a
scale factor of 80 works across all gens.
Although the gen6_ code could have been been left alone, the changes
keep the code more comparable, and it now shares a few utility functions
for scaling raw timestamps and calculating deltas. The utility for
calculating deltas takes into account 32 or 36bit overflow depending on
the current kernel version.
Note: this leaves the timestamp handling of ARB_query_buffer_object
untouched, which continues to use an incorrect scale of 80 on Skylake
for now. This is more awkward to solve since the scaling is currently
done using a very limited uint64 ALU available to the command parser
that doesn't support multiply or divide where it's already taking a
large number of instructions just to effectively multiple by 80.
This fixes piglit arb_timer_query-timestamp-get on Skylake
v2: (Ken) Update timebase_scale for platforms past Skylake/Broxton too.
Signed-off-by: Robert Bragg <robert@sixbynine.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Thu, 16 Mar 2017 15:50:36 +0000 (08:50 -0700)]
anv/device: Remove a use of a compound literal
Older versions of GCC don't like compound literals in static const
variable declarations because they don't think it's an actual constant
value.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Robert Bragg [Thu, 16 Mar 2017 22:41:09 +0000 (22:41 +0000)]
i965: bounds checks while concatenating sysfs paths
This adds some missing return value checks for all uses of snprintf in
brw_performance_query.c. This also switches a use of strncpy + strncat
for snprintf for consistency and to avoid the chance of the strncpy
leaving an unterminated string in the dest buffer if the src is too
long.
This issue with strncpy was picked up by Coverity.
CID: 1402201
Signed-off-by: Robert Bragg <robert@sixbynine.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Emil Velikov [Thu, 16 Mar 2017 17:04:54 +0000 (17:04 +0000)]
mesa: automake: add all headers to the tarball.
Fixes:
d8d81fbc316 ("mesa: Add infrastructure for a worker thread to process GL commands.")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Thu, 16 Mar 2017 16:29:16 +0000 (16:29 +0000)]
mapi: automake: add all python scripts to EXTRA_DIST
Otherwise it'll be missing in the tarball and make distcheck will fail.
Fixes:
05dd4a1104e ("glapi: Generate GL API marshalling code from the XML.")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Jonathan Gray [Fri, 17 Mar 2017 06:10:22 +0000 (17:10 +1100)]
glapi: avoid using $< in non-suffix make rules
Using $< in non-suffix make rules is a GNU extension. Explicitly use
the name of the python script to fix the build on OpenBSD.
Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Reviewed-by: Emil Velikov <emil.velikov@collabore.com>
Alex Smith [Thu, 16 Mar 2017 16:40:07 +0000 (16:40 +0000)]
radv/ac: Fix shared memory offset calculation
The index passed to get_shared_memory_ptr is an attribute slot index,
i.e. the index of a vec4 within LDS. Therefore this must be scaled by
sizeof(vec4) to give the LDS byte offset.
Fixes:
f4e499ec791 ("radv: add initial non-conformant radv vulkan driver")
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
CC: <mesa-stable@lists.freedesktop.org>
James Legg [Thu, 16 Mar 2017 17:48:13 +0000 (17:48 +0000)]
radv: Fix using more than 4 bound descriptor sets
Avoid a buffer overflow in ac_nir_to_llvm.c's create_function when
using more than 4 descriptor sets. radv claims support for 8.
Cc: 17.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tapani Pälli [Thu, 16 Mar 2017 06:10:57 +0000 (08:10 +0200)]
util/build-id: check dlpi_name before strstr call
According to dl_iterate_phdr man page first object visited is the
main program where dlpi_name is an empty string. This fixes segfault
on Android when using build-id as identifier.
Fixes:
d4fa083e11f ("util: Add utility build-id code.")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Plamena Manolova <plamena.manolova@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Tapani Pälli [Thu, 16 Mar 2017 08:55:32 +0000 (10:55 +0200)]
android: fix segfault within swap_buffers
Function droid_swap_buffers may get called without dri2_surf->buffer set,
in these cases we don't have a back buffer set either. Patch fixes segfault
seen with 3DMark that uses android.opengl.GLSurfaceView for rendering it's UI.
backtrace:
#00 pc
00013f88 /system/lib/egl/libGLES_mesa.so (droid_swap_buffers+104)
#01 pc
000117b2 /system/lib/egl/libGLES_mesa.so (dri2_swap_buffers+50)
#02 pc
000058b2 /system/lib/egl/libGLES_mesa.so (eglSwapBuffers+386)
#03 pc
00011329 /system/lib/libEGL.so (eglSwapBuffersWithDamageKHR+553)
#04 pc
000118e7 /system/lib/libEGL.so (eglSwapBuffers+55)
#05 pc
000754dc /system/lib/libandroid_runtime.so
v2: do like other backends, call get_back_bo (Emil Velikov)
Fixes: 2acc69d ("EGL/Android: Add EGL_EXT_buffer_age extension")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Timothy Arceri [Wed, 15 Mar 2017 02:07:40 +0000 (13:07 +1100)]
radv: make sure gs copy shader is retrieved from the cache with the variant
Apps can limit the size of the cache via VkAllocationCallbacks so we
can't be sure that both are always in the cache.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Timothy Arceri [Wed, 15 Mar 2017 04:20:48 +0000 (15:20 +1100)]
radv: fallback to an in-memory cache when no pipline cache is provided
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Timothy Arceri [Wed, 15 Mar 2017 03:14:24 +0000 (14:14 +1100)]
radv: always create an fallback pipeline cache
This will be used as an in-memory cache when a pipeline cache is
not provided by the app.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Timothy Arceri [Wed, 15 Mar 2017 01:40:53 +0000 (12:40 +1100)]
radv: move cache check inside insert and search functions
This will allow us to use fallback in-memory and on-disk caches
should the app not provide a pipeline cache.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Timothy Arceri [Thu, 16 Mar 2017 12:09:19 +0000 (23:09 +1100)]
st/mesa: call glthread_destroy() before _vbo_DestroyContext()
Otherwise we have a race condition between vbo calls in the
glthread and the _vbo_DestroyContext() call.
This fixes a bunch of piglit crashes.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Jason Ekstrand [Wed, 15 Mar 2017 03:48:47 +0000 (20:48 -0700)]
anv/GetQueryPoolResults: Actually implement the spec
The Vulkan spec is fairly clear about when we should and should not
write query pool results. We're also supposed to return VK_NOT_READY if
VK_QUERY_RESULT_PARTIAL_BIT is not set and we come across any queries
which are not yet finished. This fixes rendering corruptions on The
Talos Principle where geometry flickers in and out due to bogus query
results being returned by the driver. These issues are most noticable
on Sky Lake GT4 2hen running on "ultra" settings.
Reviewed-By: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100182
Cc: "17.0 13.0" <mesa-stable@lists.freedesktop.org>