Tested with new piglit gl-3.2-adj-prims test.
v2: re-order trisadj and tristripadj code, per Roland.
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
The unfilled index translator/generator functions should only be
called when the primitive mode is one of the triangle types.
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
The original mode test was valid before we had GS support.
Regression tested with full piglit run. Though, I don't think we have
any piglit tests that exercise drawing unfilled adjacency primitives.
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Only one dEQP io_blocks test fails. This test fails for the same reason
as the match_different_member_struct_names test in a previous commit.
dEQP-GLES31.functional.separate_shader.validation.io_blocks.match_different_member_struct_names
v2: Add to release notes.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
v2: Also support GL_EXT_shader_io_blocks. It's pretty much identical to
the OES extension. Suggested by Ilia.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
v2: Move later in series to avoid issues with Gallium drivers and debug
contexts. Suggested by Ilia.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Suggested-by: Timothy Arceri <timothy.arceri@collabora.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
The new validate_io catches all of the cases (and many more) that the
old function caught.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Fixes the following dEQP tests on SKL:
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_qualifier_vertex_smooth_fragment_flat
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_implicit_explicit_location_1
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_array_element_type
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_qualifier_vertex_flat_fragment_none
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_struct_member_order
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_struct_member_type
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_qualifier_vertex_centroid_fragment_flat
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_array_length
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_type
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_struct_member_precision
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_explicit_location_type
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_qualifier_vertex_flat_fragment_centroid
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_explicit_location
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_qualifier_vertex_flat_fragment_smooth
dEQP-GLES31.functional.separate_shader.validation.varying.mismatch_struct_member_name
It regresses one test:
dEQP-GLES31.functional.separate_shader.validation.varying.match_different_struct_names
Hoever, this test is based on language in the OpenGL ES 3.1 spec that I
believe is incorrect. I have already submitted a spec bug:
https://www.khronos.org/bugzilla/show_bug.cgi?id=1500
v2: Move spec quote about built-in variables to the first place where
it's relevant. Suggested by Alejandro.
v3: Move patch earlier in series, fix rebase issues.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com> [v2]
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> [v2]
The interface type, interpolation mode, precision, the type of the
outermost structure, and whether or not the variable has an explicit
location will be used for SSO validation on OpenGL ES.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
There's no good reason for it to be a struct of an anonymous union.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96221
Tested-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
commit 7c8dfa78b9 (i965/draw: Use the real size for vertex
buffers) changed how we programmed the VERTEX_BUFFER_STATE size field.
Previously, we programmed it to the size of the actual underlying BO,
which is page-aligned, and potentially much larger than the GL buffer
object. This violated the ARB_robust_buffer_access spec.
With that change, we started programming it based on the range of data
we expect the draw call to actually access - which is based on the
min_index and max_index information provided to glDrawRangeElements().
Unfortunately, applications often provide inaccurate range information
to glDrawRangeElements(). For example, all the Unreal demos appear to
draw using a range of [0, 3] when the index buffer's actual index range
is [0, 5]. Such results are undefined, and we are absolutely allowed
to restrict access to the range they specified. However, the failure
mode is usually that nothing draws, or misrendering with wild geometry,
which is kind of bad for a common mistake. And people tend to assume
the range information isn't that important when data is in VBOs.
There's no real advantage, either. ARB_robust_buffer_access only
requires us to restrict access to the GL buffer object size, not
the range of data we think they should access. Doing that allows
buggy applications to still function. (Note that we still use this
information for busy-tracking, so if they try to overwrite the data
with glBufferSubData, they'll still hit a bug.) This seems to be
safer.
We may want to provide the more strict range as a debug option,
or scan the VBO and warn against bogus glDrawRangeElements in
debug contexts. That can be done as a later patch, though.
Makes Unreal demos draw again.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Like constant buffers, samplers and textures are aliased on Fermi and
we need to invalidate the state when switching from 3D to CP and vice
versa.
This fixes rendering issues in the UE4 demos.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
This brings the final size of an optimized non-debug build of the Vulkan
driver down to 2.9 MB as opposed to 8.7 MB for the dri driver.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Now that the compiler has been completely separated from libmesa, we no
longer need these. We can make the tests much smaller by not linking them
in. This also ensures that anyone who runs make check won't accidentally
put in any dependencies from the compiler to the rest of mesa core.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
They reference the compiler so they shouldn't go in libi965_compiler.la.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
None of them are actually using it. It's a relic of an older compiler
interface that required a gl_program.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
That's where brw_link_shader lives and they seem to go together. Also,
this gets it out of libi965_compiler.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
This way it's no longer part of libi965_compiler.la since it depends on
GLSL and ARB program stuff.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Right now libglsl.la depends on libnir.la so putting it in libnir.la
adds a dependency on libglsl.la that goes the wrong direction.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
The stated bug describes a scenario in which a post sync write operation for
depth or timestamp can be ignored. There are two workarounds suggested, the
first and easier is to simply do a cs stall when we do these type of writes.
The second option is to do a PIPE_CONTROL flush after the post sync but before
the data is required.
Generally, I believe the data written out is consumed by the application on the
CPU side and so doing the easier of the two is ideal. Furthermore, these queries
aren't tremendously common in the perf sensitive apps I have looked at. However,
there could be cases where a shader stage might directly consume the data, and
as a result option 2 may be desirable.
This patch goes with the easier solution for now.
gen9lp bug_de_id=2137196
By itself, this does *not* fix any of the GT4 hangs we're currently
experiencing.
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
The R_028B50_VGT_TESS_DISTRIBUTION value is copied from
amdgpu-pro. Smaller values in the ACCUM fields seem to
decrease the performance advantage from this patch, higher
values don't seem to matter.
v2: Add distribution mode field enums.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Using more than 1 wave per threadgroup does increase performance
generally. Not using too many patches per threadgroup also
increases performance. Both catalyst and amdgpu-pro seem to
use 40 patches as their maximum, but I haven't really seen
any performance increase from limiting the number of patches
to 40 instead of 64.
Note that the trick where we overlap the input and output LDS
does not work anymore as the insertion of the tess factors
changes the patch stride.
v2: - Add comment about LDS assumptions.
- Add constant for buffer size.
- Fix code style.
v3: - Correct limits for not splitting patches between waves.
- Set max num_patches to 40 as in the proprietary driver.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
The factors may be stored to LDs by another invocation than
the invocation for vertex 0.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
This allows running the TES on different CU's than the
TCS which results in performance improvements.
v2: Only write the control word from one invocation.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
They are unused.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
We always try to use 4-component loads, as LLVM does not combine loads
and they bypass the L1 cache.
We can't use a similar strategy for stores and this is especially
notable with the tess factors, as they are often set with separate
MOV's per component in the TGSI.
We keep storing to LDS and the LDS space, so we can load the outputs
later, either due to the shader, of for wrting the tess factors.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
We need to copy the VS outputs to memory. I decided to do this
using a shader key, as the value depends on other shaders.
I also switch the fixed function TCS over to monolithic, as
otherwisze many of the user SGPR's need to be passed to the
epilog, which increases register pressure, or complexity to
avoid that. The main body of the fixed function TCS is not
that interesting to precompile anyway, since we do it on
demand and it is very small.
v2: Use u_bit_scan64.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Instead of creating a memory area per patch and per vertex, we put
the same attribute of every vertex & patch together. Most loads
and stores access the same attribute across all lanes, only for
different patches and vertices.
For the TCS this results in tightly packed data for 4-component
stores.
For the TES this is not the case as within a patch the loads
often also access the same vertex. However if there are < 4
vertices/patch, this still results in a reduction of the number
of cache lines. In the LDS situation we only do better than worst
case if the data per patch < 64 bytes, which due to the
tessellation factors is pretty much never.
We do not use hardware swizzling for this. It would slightly reduce
the number of executed VALU instructions, but I had issues with
increased wait times that I haven't been able to solve yet.
Furthermore, the tbuffer_store intrinsic does not support both
VGPR offset and an index, so we have a problem storing
indirectly indexed outputs. This can be solved by temporarily
storing arrays in LDS and then copying them, but I don't think
that is worth the effort. The difference in VALU cycles
hardware swizzling gives is about 0.2% of total busy cycles.
That is without handling the array case.
I chose for attributes instead of components as they are often
accessed together, and the software swizzling takes VALU cycles
for calculating offsets.
v2: - Rename functions to get_tcs_tes_buffer_address.
- multiply by 16 as late as possible.
- Use tgsi_full_src_register_from_dst.
- Remove some bad comments.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
This happens to be in the right position, but that changes
when TCS/TES get new parameters.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
v2: - Use llvm.admgcn.buffer.load instrinsics for new LLVM.
- Code style fixes.
v3: - Code style fix.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
The buffer is quite large, but should only be allocated if the
application uses tessellation. Most non-games don't.
v2: - Use the correct register for SI.
- Add define for block size.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
CID 1271532 (#1 of 1): Out-of-bounds read (OVERRUN)34. overrun-local:
Overrunning array of 2 16-byte elements at element index 2 (byte offset
32) by dereferencing pointer &inst.Dst[i].
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
Not sure why coverity calls this an out-of-bounds read vs out-of-bounds
write.
CID 1358920 (#1 of 1): Out-of-bounds read (OVERRUN)9. overrun-local:
Overrunning array r of 3 16-byte elements at element index 3 (byte
offset 48) using index chan (which evaluates to 3).
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
XY_FAST_COPY_BLT command doesn't have a field for raster operation. So, fall
back to using XY_SRC_COPY_BLT to handle those cases.
Fixes piglit test gl-1.1-xor-copypixels when fast copy blit is enabled
for all tiling formats.
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Experimentation with different values of src/dst horizontal/vertical
alignment showed that these fileds are not used on gen9 hardware.
A recent update in graphics specs has removed these fields from
XY_FAST_COPY_BLT command.
Cc: Ben Widawsky <ben@bwidawsk.net>
Cc: Chad Versace <chad.versace@intel.com>
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
To read out MP perf counters we use a compute shader and need to upload
input data like a 64-bits addr used to store the values and a sequence
ID for synchronization. Currently, this input data is uploaded as user
uniforms which means that it's sticked to c0[], but if a compute shader
from a real application is used, monitoring those performance counters
will just overwrite some data and miserably crash.
Instead, sticking the 64-bits addr and the sequence into the driver
constant buffer seems like much better and will allow to monitor
counters with GL 4.3 apps.
Tested on GF119 and GK110, but should not hurt anything on GK104.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Brian Paul <brianp@vmware.com>
