AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
143,513 Commits 1 Branch 0 Tags
cb2e712b9e42bcbd4b91ec598f3b286c6a00a4a8
Commit Graph

143513 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Alyssa Rosenzweig
cb2e712b9e pan/bi: Unit test DISCARD+FCMP fusing 
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:26 -04:00
Alyssa Rosenzweig
d74ab1e4d9 pan/bi: Fuse DISCARD with conditions 
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:26 -04:00
Alyssa Rosenzweig
4f347e96b8 pan/bi: Add fclamp unit tests 
The negative cases here did not pass before this series, showing the bug
in the clamp optimization. By introducing the FCLAMP pseudo op, the bug
is fixed. Let's ensure we don't regress.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:26 -04:00
Alyssa Rosenzweig
ac636f5adb pan/bi: Use FCLAMP pseudo op for clamp prop 
Map nir_op_fsat/etc to FCLAMP pseudo ops, instead of FADD. There are
significantly fewer knobs on FCLAMP, meaning significantly fewer things
to get wrong.

This fixes two(!) classes of bugs:

* Swizzles (failing to lower/compose swizzles on clamps)
* Numerical bugs (incorrectly treating +0.0 as an additive identity)

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:26 -04:00
Alyssa Rosenzweig
a98790fa96 pan/bi: Add optimizer unit tests 
Writing these tests brought to light the cluster of bugs fixed in the
previous commits. Now that things work, let's ensure they stay working.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:26 -04:00
Alyssa Rosenzweig
89e452883a pan/bi: Use FABSNEG pseudo ops for modifier prop 
Simplifies pattern matching. This commit by itself fixes multiple
numerical issues -- the previous fabsneg check failed to check the round
mode or the sign of the zero. That will break Vulkan/OpenCL.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:26 -04:00
Alyssa Rosenzweig
b2b0dca4ce pan/bi: Add shader equality helper for unit tests 
Optimizer tests really are global.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:26 -04:00
Alyssa Rosenzweig
f3d8ec1484 pan/bi: Fuse abs/neg more on Valhall 
Some of these Bifrost restrictions may be skipped on Valhall.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:25 -04:00
Alyssa Rosenzweig
c27af4ad36 pan/bi: Simplify bi_compose_clamp 
Realized this trick when reversing Valhall.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:25 -04:00
Alyssa Rosenzweig
cfd73b4542 pan/bi: Unit test new constant folding patterns 
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:25 -04:00
Alyssa Rosenzweig
ec76119dfb pan/bi: Constant fold texturing lowerings 
This ensures we can constant fold the ALU ops used to lower:

* explicit LOD calculations
* array textures
* texture offsets
* multisample indices

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:59:24 -04:00
Alyssa Rosenzweig
6fa1479479 pan/va: Document IEEE 754 conformance of clamps 
These rules are not obvious. But they turn out to be exactly what's
required by the spec.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12205>
 2021-08-11 14:58:58 -04:00
Alyssa Rosenzweig
cdfb5d3788 panfrost: Test src*dst + dst*src blending 
Validates the prior commit.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12152>
 2021-08-11 18:15:52 +00:00
Alyssa Rosenzweig
0228ccf857 panfrost: Leverage Bifrost's 2*src blend factor 
Bifrost adds a value for the C factor equaling 2*src. This does not
correspond directly to API blend modes so it is not too useful in
general. However, it's required for src*dest + dest*src blending to be
done in hardware instead of a blend shader. GFXbench uses that blend
mode, so it must be important ;-)

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12152>
 2021-08-11 18:15:52 +00:00
Alyssa Rosenzweig
95e306dc84 panfrost: Add basic fixed-function blending tests 
Add unit tests for the fixed-function blending helpers in pan_blend.c.
Each test consists of a Porter-Duff blend mode and the associated
hardware state. In this commit, we add tests for the most common modes.

For motivation, this code has NOT been properly tested in CI. True,
functional correctness of the blend module as a whole is tested by
dEQP-GLES3.functional.fragment_ops.blend.* among other integration
tests. However, this testing is insufficient to check for regressions.
Crucially, the following broken patch would clear CI:

   bool pan_can_fixed_function(...) {
      return false;
   }

In that case, blend shaders are used 100% of the time, which will
regress performance horribly but still pass dEQP. The only clue
something went wrong would be some traces changing checksum due to the
fixed-function blender producing slightly different output than
equivalent blend shaders. By unit testing the fixed blend path, we
ensure we always use the fixed-function path when we expect it to.

Similarly, using incorrect values for the blend metadata may not affect
functional correctness but will increase power consumption. Let's check
all the data we export to drivers.

Note: due to additive commutativity, there are many pairs of equivalent
Mali blend modes. Unfortunately, the vendor is... inconsistent about how
to resolve ambiguous modes. Our algorithm for computing modes is
correct; the "preferred" values are left in comments since otherwise our
tests fail despite correct code. I want to blame Bifrost for this, but
Midgard was patient zero.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12152>
 2021-08-11 18:15:52 +00:00
Alyssa Rosenzweig
95187c03f1 panfrost: Simplify blend_factor_constant_mask 
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12152>
 2021-08-11 18:15:52 +00:00
Alyssa Rosenzweig
01a1b253b4 panfrost: Fix is_opaque when blend_enable=false 
Needed to pass the "replace" unit test.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12152>
 2021-08-11 18:15:52 +00:00
Alyssa Rosenzweig
b7af56f2ad panfrost: Add blend helper packing the equation 
This is more convenient for the Gallium driver and easier to test.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12152>
 2021-08-11 18:15:52 +00:00
Alyssa Rosenzweig
7eb2559198 panfrost: Use _PU for non-dithered formats 
This is required to disable dithering on a per-draw basis when OPAQUE
output is used (bypassing the blender which normally uses the
round_to_framebuffer_precision flag to do the same).

This functionally reverts:

   ebc07f4b2f ("panfrost: Remove padded unorm blendable formats")
   fae90a7940 ("panfrost: Always pick dithered tb formats")

while adding the functionality to make them useful.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12152>
 2021-08-11 18:15:52 +00:00
Alyssa Rosenzweig
07607c8443 panfrost: Remove unused #defines 
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12328>
 2021-08-11 18:00:45 +00:00
Alyssa Rosenzweig
3ec5e2b6a7 panfrost: Add LINEAR debug option 
Useful to cross off CPU texture tiling as the source of bugs.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12328>
 2021-08-11 18:00:45 +00:00
Alyssa Rosenzweig
3958f00215 pan/bi: Add a noopt debug option 
To rule out buggy optimization passes when debugging.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12328>
 2021-08-11 18:00:45 +00:00
Alyssa Rosenzweig
ff03f096bf pan/bi: Make bi_opt_push_ubo optional 
It's an optimization pass -- omitting it should not cause MMU faults
(!). Make sure the UBO push mask is set regardless of whether the pass
is called, and just call the pass when required.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12328>
 2021-08-11 18:00:45 +00:00
Lionel Landwerlin
01b0935d31 nir/lower_shader_calls: remove empty phis 
This is confusing opt_cse.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 8dfb240b1f ("nir: Add raytracing shader call lowering pass.")
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11953>
 2021-08-11 15:10:07 +03:00
Marcin Ślusarz
a1b7a5fad6 zink: use nir_shader_instructions_pass in nir_lower_dynamic_bo_access 
Changes:
- nir_metadata_preserve(..., nir_metadata_dominance)
  is called only when pass makes progress
- nir_metadata_preserve(..., nir_metadata_all) is called when pass doesn't
  make progress

Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
50cb70d38a zink: use nir_shader_instructions_pass in lower_discard_if 
Changes:
- nir_metadata_preserve(..., nir_metadata_dominance)
  is called only when pass makes progress
- nir_metadata_preserve(..., nir_metadata_all) is called when pass doesn't
  make progress

Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
65f9234f96 microsoft/compiler: use nir_shader_instructions_pass in dxil_nir_lower_double_math 
No functional changes.

Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
b8de41902e microsoft/compiler: use nir_shader_instructions_pass in dxil_nir_split_clip_cull_distance 
No functional changes.

v2: fix build

Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
9aafb91f0e microsoft/compiler: preserve all metadata when upcast_phi doesn't make progress 
Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
17a61ec541 microsoft/clc: use nir_shader_instructions_pass in clc_nir_dedupe_const_samplers 
Changes:
- nir_metadata_preserve(..., nir_metadata_all) is called when pass doesn't
  make progress

v2: fix build

Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
70723f278f microsoft/clc: preserve only valid metadata in clc_lower_printf_base 
Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
4d81226102 d3d12: use nir_metadata_none instead of its value 
Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
7810ca596c intel/compiler: use nir_shader_instructions_pass in brw_nir_apply_attribute_workarounds 
Changes:
- removal of attr_wa_state (it's passed directly)
- nir_metadata_preserve(..., nir_metadata_all) is called when pass doesn't
  make progress

Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Marcin Ślusarz
e1b325f587 nir/builder: invalidate metadata per function 
Fixes: a62098fff2 ("nir: Add a helper for general instruction-modifying passes.")
Signed-off-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12324>
 2021-08-11 11:23:30 +00:00
Danylo Piliaiev
a57bcc4394 freedreno/decode: print estimated crash location without colored output 
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12302>
 2021-08-11 09:24:19 +00:00
Pierre-Eric Pelloux-Prayer
7684d57a05 nir: add a pass to optimize "gl_FragDepth = gl_FragCoord.z" away 
gl_FragDepth default value is gl_FragCoord.z so if a shader does:

   gl_FragDepth = gl_FragCoord.z

we can drop this assignment.

v2: use nir_ssa_scalar_resolved and don't do this is gl_FragDepth
    is wrote multiple times (Jason)
v3: - move to its own pass (Jason)
    - handle var = NULL (Rhys)
v4: refactoring (Jason)

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10697>
 2021-08-11 11:00:11 +02:00
Kenneth Graunke
a6176881a6 iris: Drop dead drm_ioctl prototype 
We now use intel_ioctl instead.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12206>
 2021-08-11 08:05:00 +00:00
Kenneth Graunke
0707a1d842 iris: Improve the memory layout of iris_bo by fixing pahole issues 
We had a 4 byte hole and a 4-byte field breaking up a run of bools.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12206>
 2021-08-11 08:05:00 +00:00
Kenneth Graunke
2616e15c01 iris: Rename bo->gtt_offset to bo->address 
This is the virtual memory address of the buffer object.  Calling it the
BO's address is a lot more obvious than calling it an offset in one of
the now many graphics translation tables.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12206>
 2021-08-11 08:05:00 +00:00
Iago Toral Quiroga
c964e5f099 v3d,v3dv: add options to force 32-bit or 16-bit TMU precision 
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12303>
 2021-08-11 05:57:10 +00:00
Tapani Pälli
ff669ea93b anv/android: fix build error due refactoring 
Fixes: e08370dc37 ("anv: disable aux for exportable images without modifiers")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5208
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12300>
 2021-08-11 04:54:05 +00:00
Dave Airlie
c16f7e2a19 docs: add llvmpipe host memory extensions 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12316>
 2021-08-11 10:06:22 +10:00
Dave Airlie
c198adf718 lavapipe: add host ptr support. 
This actually doesn't need any backend support.

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12316>
 2021-08-11 09:58:17 +10:00
Dave Airlie
9a57dceeb7 llvmpipe: add support for user memory pointers 
This is useful for clover, but throw it at CI at least

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12316>
 2021-08-11 09:58:17 +10:00
Icecream95
ee2bb57f1e pan/bi: Use the computed scale for fexp NaN propagation 
This makes pow(NaN, x) return NaN rather than 1.0.

Fixes: 499397700c ("pan/bi: Don't lower fpow")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5189
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12269>
 2021-08-10 22:42:08 +00:00
Ian Romanick
84d2e53789 Revert "nir/algebraic: Convert some f2u to f2i" 
Per https://gitlab.freedesktop.org/mesa/mesa/-/issues/5178#note_1019666,
the assumption fundamental to this optimization is false.  Section
2.4.1 (Float to Integer) of Ivy Bridge PRMs describes the situation.
The wording of the section is somewhat confusing (because it doesn't
clearly delineate between signed and unsigned integers), but the last
two rows of the table make it clear that F->UD conversion clamps
negative float values to 0.

All other hardware mentioned in that thread seems to behave the same
way.

The real problem is that, with hardware that behaves in this ways,
converting f2u(2147483648.0) to f2i(2147483648.0) changes the bit pattern
that would be produced from 0x80000000 to 0x7fffffff.

This reverts commit ad05920258.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12297>
 2021-08-10 22:16:13 +00:00
Ian Romanick
3ba66ebbc8 nir/opcodes: Use u_intN_(min|max) 
uadd_sat was updated using sed, so I didn't even notice the surrounding
opcodes.  Oops.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12297>
 2021-08-10 22:16:13 +00:00
Dave Airlie
7fb9e78d09 clover: only return CLC version as 1.2 (even for 3.0) 
Fixes CTS compiler opencl_c_versions

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12286>
 2021-08-10 21:38:39 +00:00
Dave Airlie
e78d5bb8e2 clover/nir: don't convert to NIR on library link 
If just creating a library, just link the spir-v and store it.

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12286>
 2021-08-10 21:38:39 +00:00
Dave Airlie
6cc1568ff5 clover: fix compilation with clang + llvm 12. 
clang in llvm 12 no longer accepts "-cl-denorms-are-zero" as a cc1
options which is how this code uses it.

For now just pick the correct cc1 equivalent.

This fixes a crash with llvm master and CL conversions tests

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12286>
 2021-08-10 21:38:39 +00:00