AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
134,810 Commits 1 Branch 0 Tags
bd0ef080d03fa3a9eefb513aec8fee88339c33df
Commit Graph

134810 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Iago Toral Quiroga
bd0ef080d0 v3d/compiler: fix QPU scheduler TMU sequence shuffling 
The QPU scheduler allows to move certain TMU instructions around and
since we enabled pipelining, we need to protect against the case where
doing this might break a TMU sequence. For example, this test:

dEQP-VK.rasterization.line_continuity.line-strip

Was generating this VIR:

mov tmud, t187
mov.pushz null, t176
mov.ifa tmua, t9
nop null; wrtmuc (img[0].p0 | 0x0)
mov tmut, t185
mov tmud, t180
mov.ifa tmusf, t183
nop null; thrsw

where we have a general TMU access (tmud,tmua) followed by an image
access (wrtmuc, tmut, tmud, tmusf), which the QPU scheduler was turning
into:

nop            ; nop               ; ldunifrf.rf22 (0xffffff00 / -nan)
nop            ; nop               ; wrtmuc (img[0].p0 | 0x0)
nop            ; nop               ; ldtmu.r2
add  r0, r2, 1 ; nop               ; ldtmu.r3
nop            ; nop               ; ldtmu.r4
nop            ; mov  tmud, r0
nop            ; mov.ifa  tmua, rf15
nop            ; mov  tmut, r4     ; thrsw
nop            ; mov  tmud, rf22
nop            ; mov.ifa  tmusf, r3

where it allowed the wrtmuc to move up and before the general TMU access,
leading to an incorrect TMU sequence.

Fix this by flagging TMUA writes (which are the sequence terminators for
general TMU accessess) as writing new TMU configuration, like we do for all
other TMU sequence terminators for textures and images.

Fixes: 197090a3fc ('broadcom/compiler: implement pipelining for general TMU operations')

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8954>
 2021-02-10 13:18:25 +00:00
Mike Blumenkrantz
79fcea9475 zink: add a texture barrier hook 
this just flushes all the pending batches, which is enough since we
do our batch work synchronously and everything is awful

will be reworked when this is no longer the case

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8948>
 2021-02-10 13:12:12 +00:00
Mike Blumenkrantz
802c1afe9c zink: handle nir_texop_texture_samples 
this is just a SpvOpImageQuerySamples

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8947>
 2021-02-10 13:05:45 +00:00
Mike Blumenkrantz
efbd3584e3 zink: rework viewport handling 
viewport can be transformed at draw time based on clip_halfz, so we actually
want to do the conversion from gallium->vulkan at that time

as a nice bonus, we get to delete a struct member

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8946>
 2021-02-10 12:59:37 +00:00
Danylo Piliaiev
f0a76b2067 turnip: enable inheritedQueries 
Passes relevant CTS tests.

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8824>
 2021-02-10 12:38:44 +00:00
Samuel Pitoiset
ed7e6eb54b radv: fix memory leaks if a submission fails 
If the device isn't lost, the application might want to re-submit.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8931>
 2021-02-10 12:25:38 +00:00
Mike Blumenkrantz
bf5713d736 zink: handle vote intrinsics in ntv 
these are super simple and straightforward

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8950>
 2021-02-10 12:18:33 +00:00
Mike Blumenkrantz
e35612e45f zink: add spirv_builder wrapper for vote intrinsics 
these are all going to use the same mechanics, so we can simplify a bit

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8950>
 2021-02-10 12:18:33 +00:00
Michel Dänzer
cf6901809e ci: Add trailing slash to path for documentation preview 
Fixes the documentation preview link on the MR Overview tab. Found by
comparing with the same thing in weston.

(It still appears only after all jobs in the pipeline have at least
started once though)

Fixes: ae8f9584f4 "CI: always expose docs artifacts"
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8927>
 2021-02-10 12:04:22 +00:00
Alejandro Piñeiro
f758b1a25b v3dv: support for depthBiasClamp 
Gets tests like the following working:
dEQP-VK.dynamic_state.rs_state.depth_bias_clamp

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8928>
 2021-02-10 10:29:09 +00:00
Rhys Perry
f7575fa71f aco: fix adjust_vertex_fetch_alpha 
These offsets were wrong and didn't match the old behaviour.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: e8220e106b ("aco: optimize AC_FETCH_FORMAT_SNORM alpha adjust")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8935>
 2021-02-10 09:37:17 +00:00
Samuel Pitoiset
19711b0f44 radv: stop allocating useless ESGS scratch BO on GFX10+ 
On GFX10+, the ESGS ring is allocated in LDS. This saves one BO
when geometry shaders are used, and possibly two BOs if nothing
else (tess, scratch, etc) is used.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8926>
 2021-02-10 08:45:35 +00:00
Samuel Pitoiset
caafedf38f radv: fix double free when creating a fence failed 
It's already freed in radv_destroy_fence().

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8938>
 2021-02-10 08:45:49 +01:00
Eric Anholt
cd03f1c6f6 gallium: Document behavior of more lowering pipe caps. 
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
bcb5f9f94a v3d: Stop advertising support for flat shading. 
The GL frontend can lower this weird GL feature away for us.  This should
fix redeclaration of the gl_Color/SecondaryColor as centroid, since that
case had been missed in the !flat special case here.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
ff805f8ac7 v3d: Stop advertising support for PIPE_CAP_*_COLOR_CLAMPED. 
The GL frontend can lower away this deprecated GL feature for us.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
2992dc7386 v3d: Stop advertising support for PIPE_CAP_TWO_SIDED_COLOR. 
The GL frontend can lower away this deprecated GL feature for us.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
5ddc2f916f v3d: Clean up vestiges of alpha test lowering. 
We had an unnecessary case in our uniforms upload switch statement, since
we no longer advertise the cap.

Fixes: 8ad931808e ("v3d: do not report alpha-test as supported")
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
b9a9cd59ff vc4: Stop advertising support for VS color clamping. 
This is special GL behavior that the frontend is now able to handle for
us.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
e584d4f6dd vc4: Stop advertising support for PIPE_CAP_TWO_SIDED_COLOR. 
The GL frontend is perfectly able to handle this for us.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
cc0841c82a vc4: Remove vestiges of alpha test lowering. 
We stopped reporting the alpha test screen cap, and stopped using the
value in the key, so now shrink the key.  This gets another switch case
out of the hot uniforms upload path.

Fixes: 1404b8b1e5 ("vc4: do not report alpha-test as supported")
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
9aa7a2d10c mesa/st: Assume that the default variant is always first in the list. 
Since we always precompile now, we can remove some checks for presence of
the first variant or for it being the wrong key.

Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
1d5ce66b02 mesa/st: Always precompile the first shader variant. 
Every driver wants the chance to see the default variant of the shader in
the hopes of avoiding draw-time shader compiles, even if they have some
NOS compiled into the shader by st/mesa.  Before, every driver had to
avoid st/mesa lowering options to get this behavior, even if they would
like to have the shared lowering code, and some drivers were just missing
the behavior because nobody had debugged that it was missing yet.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8601>
 2021-02-09 20:06:48 -08:00
Eric Anholt
659916d674 freedreno/a5xx: Don't forget to count our custom blits against queries. 
Noticed when comparing the a5xx and a6xx paths.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8873>
 2021-02-10 03:46:25 +00:00
Eric Anholt
881ce0f0a4 freedreno: Remove FD_STAGE_* in favor of a "disable_all" flag. 
The only paths are "prepare queries for rendering" and "finish all
queries", everything else is handled by set_active_query_state().

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8873>
 2021-02-10 03:46:25 +00:00
Eric Anholt
4cc3956193 freedreno: Force updating active queries on batch reordering. 
The state of queries for the batch we've just switched to may not reflect
the current active flag, so make sure that we update at the start of the
next draw.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8873>
 2021-02-10 03:46:25 +00:00
Eric Anholt
1bab56e1ba ci/virgl: Fix GLES31 testing on desktop GL. 
The version overrides leaked into the virglrenderer server, breaking
compiling of all sorts of shaders.  We already get GLES3.1 when running
virgl on llvmpipe GL without doing any overrides.

Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8854>
 2021-02-10 03:31:56 +00:00
Mike Blumenkrantz
697fb11cfc features: mark off GL 4.3 for zink 
\o/

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8941>
 2021-02-09 19:26:47 -05:00
Mike Blumenkrantz
cb80609bb0 zink: GLSL 430 
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8941>
 2021-02-09 19:25:53 -05:00
Mike Blumenkrantz
7909c642b1 zink: enable compute 
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
c5cf5ad6f0 zink: export compute-specific shader/compute caps 
this doesn't enable the compute cap itself, it just handles all the other
caps

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
a2771967cb zink: add launch_grid pipe_context hook for compute handling 
we're reusing update_descriptors, so this is trivial

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
ce6d08a0e7 zink: flush gfx/compute batches when the other pipeline needs resource sync 
if compute or gfx use a resource which is already in use by the other pipeline,
we need to ensure tha we flush that pipeline now in order to have commands execute
in the expected order

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
dac99517ff zink: handle descriptor set updates for compute operations 
this is almost identical, the difference being that we need to explicitly
use the compute batch/program in a few places

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
e20a43ec35 zink: handle memory barriers for compute batch 
this is a bit different since the regular functions cycle the gfx batch
array, but the compute batch isn't in an array

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
e3633ed12b zink: setup compute batch and add handling 
this is separate to the gfx batch array,
so we need separate struct members in zink_context for it

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
fa555772b9 zink: break out descriptor updating into separate function 
this is a massive block of code that we need to reuse for compute shaders
because we absolutely don't want to be copying it around

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:39 +00:00
Mike Blumenkrantz
b2247a3724 zink: add compute programs and pipelines 
these are similar to the gfx variants, but mostly smaller

some things are implemented as placeholders for now to make the code
similar to gfx programs/pipelines and thus easier to read/review

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
6c50b12b9e zink: rename pipeline_cache_entry -> gfx_pipeline_cache_entry 
prep for compute pipelines

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
79613650b3 zink: explicitly get shader stage from shader during binding setup in draw 
the iterator matches the shader stage currently, but this won't always be the
case

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
184d773d0f zink: make allocate_descriptor_set() take more params instead of a gfx_program 
this makes the function reusable between different types of programs

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
c9aee1165b zink: refactor batch creation 
by moving this into its own function, we can reuse it for creating a compute
batch, which will not be part of the gfx batch array

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
dcab2ca6b1 zink: take a pipe_reference param in zink_batch_reference_program 
this lets us reuse the function for gfx and compute programs without much
churn, since in C we can use a pointer to the first member of a struct and
a pointer to the struct interchangeably

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
f32891b159 zink: ignore compute batch when starting/ending batches 
compute batches don't have queries

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
8f6ffc5543 zink: handle nir_intrinsic_memory_barrier_shared in ntv 
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
d2b6e10882 zink: add handling for shared atomic ops in ntv 
these are mostly the same as ssbo except they use the shared block variable

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
deff4c220b zink: implement shared load/store nir ops in ntv 
these access the data in the shared block variable at an offset

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
36cdcbe092 zink: handle COMPUTE glsl variables 
we can just use existing helpers for these

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
2d65c7cc0a zink: handle COMPUTE setup in ntv 
addressing mode, shared block, and execution modes all need to be handled
here

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00
Mike Blumenkrantz
d9122c0417 zink: handle COMPUTE bindings in compiler/ntv 
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8781>
 2021-02-10 00:19:38 +00:00