OpenHarmony/third_party_mesa3d
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
152,442 Commits 9 Branches 1 Tag
8e0e98a9af453db9c2e3b29231ba62bafc194c33
Commit Graph

77 Commits

Author SHA1 Message Date
Rhys Perry
8e0e98a9af radv: initialize meta shader options earlier 
We might need this for NIR builders.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15854>
 2022-04-15 23:56:11 +00:00
Samuel Pitoiset
576833507b radv: only declare dynamic states that are used by internal operations 
Initialize some default static PSO states instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15729>
 2022-04-05 07:54:52 +00:00
Samuel Pitoiset
7c14671535 radv: use the common vk_framebuffer 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15612>
 2022-04-01 08:21:59 +02:00
Samuel Pitoiset
b18afccb61 radv: convert the meta resolve depth/stencil FS path to dynamic rendering 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15612>
 2022-04-01 08:21:53 +02:00
Samuel Pitoiset
518c6d808e radv: convert the meta resolve color FS path to dynamic rendering 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15612>
 2022-04-01 08:21:53 +02:00
Samuel Pitoiset
9309c3d887 radv: rework the workaround that disables DCC for incompatible copies 
Rely on the image view to avoid using an extra structure for the
render pass. This will allow to convert the meta operations to
dynamic rendering.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15612>
 2022-04-01 08:21:52 +02:00
Samuel Pitoiset
2b18234e61 radv: drop EXT or KHR suffixes for stuff promoted in Vulkan 1.3 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15663>
 2022-03-31 07:13:17 +00:00
Jason Ekstrand
29a164b088 radv: Use vk_command_pool 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14917>
 2022-02-11 08:06:24 +00:00
Samuel Pitoiset
7b7debe8f9 radv: fix restoring subpass during hw/fs color resolves 
This fixes an stack-use-after-scope detect by ASAN because the
subpass is used after the loop by radv_mark_noncoherent_rb().

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/14317>
 2021-12-28 12:53:12 +00:00
Rhys Perry
d74498e617 radv: add radv_meta_init_shader 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14087>
 2021-12-08 11:07:40 +00:00
Samuel Pitoiset
52b4185012 radv: switch the remaining stages/access to VK_PIPELINE_STAGE_2/VK_ACCESS_2 
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/13549>
 2021-11-26 13:41:23 +00:00
Samuel Pitoiset
984091531e radv: remove unused parameter in radv_emit_subpass_barrier() 
It got introduced by "radv: optimize subpass barrier flushes for
imageless framebuffers" but never used in the final version.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13702>
 2021-11-08 18:10:49 +00:00
Samuel Pitoiset
e15e3a8e86 radv: optimize subpass barrier flushes for imageless framebuffers 
The driver should always know the attachments at this point. This
should reduce the number of L2 cache flushes for imageless framebuffers.

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/13291>
 2021-11-03 08:26:43 +01:00
Bas Nieuwenhuizen
64b237436a radv: Add hooks after in-renderpass meta operations. 
So that we can mark the rendertargets dirty.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13239>
 2021-10-13 14:07:05 +00:00
Samuel Pitoiset
f07e67272e radv: fix vk_object_base_init/finish for internal image views 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13102>
 2021-10-01 07:26:37 +00:00
Joshua Ashton
35897a1ac7 radv: Rename radv_subpass_barrier function to radv_emit_subpass_barrier 
Otherwise this shadows the name of the structure, which is problematic for C++ code due to constructors.

Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13103>
 2021-09-29 17:42:17 +00:00
Bas Nieuwenhuizen
59c501ca35 radv: Format. 
Using

find ./src/amd/vulkan -regex '.*/.*\.\(c\|h\|cpp\)' | xargs -P8 -n1 clang-format --style=file -i

Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10091>
 2021-04-10 03:31:58 +02:00
Mike Blumenkrantz
1d1c7a965c radv: use common interfaces for shader modules 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9508>
 2021-03-15 21:47:44 +00:00
Jason Ekstrand
8304b4eef7 radv/meta: Use CreateRenderPass2 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8857>
 2021-03-10 18:17:31 +00:00
Connor Abbott
219b84fd5f radv/meta: Use sized types for nir_tex_instr::dest_type 
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7989>
 2021-01-25 11:21:42 +01:00
Samuel Pitoiset
093b3f6e1f radv: only decompress the depth/stencil aspect that needs to be resolved 
For depth/stencil images, the driver was decompressing both aspects
while it should be enough to only decompress the one that's going
to be resolved.

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/8561>
 2021-01-22 18:16:19 +00:00
Bas Nieuwenhuizen
8f8d72af55 radv: Use access helpers for flushing with meta operations. 
This way we're properly using the vulkan barrier paradigm instead
of adhoc guessing what caches need to be flushed. This is more robust
for cache policy changes as we now don't have to revisit all the meta
operations all the time.

Note that a barrier has both a src and dst part though. So

barrier:
   flush src
   meta op
   flush dst

becomes

barrier:
  flush barrier src
  flush meta op dst
  meta op
  flush meta op src
  flush barrier dst

And there are some places where we've been able to replace a CB flush
with a shader flush because that is what we'd need according to vulkan rules
(and it turns out that in the cases the CB flush mattered the app will set the
bit in one of the relevant flushes or it was needed as a result of an optimization
that we counter-acted in the previous patch.)

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7202>
 2021-01-13 16:27:19 +00:00
Bas Nieuwenhuizen
f23eaf0db6 radv: Add option to disable DCC in renderpasses without layout. 
If DCC is enabled for GENERAL then we cannot disable DCC by going
to the GENERAL layout.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7004>
 2020-12-21 18:32:24 +00:00
Samuel Pitoiset
19e96d4566 radv: remove useless push constants data when resolving ds attachments 
Depth/stencil resolves are only allowed inside a subpass, which means
the offset is always 0 and the draw/dispatch covers the whole image.

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/8127>
 2020-12-18 17:58:54 +00:00
Rhys Perry
fb0385b57c radv: use intrinsic builders 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6587>
 2020-11-26 17:50:38 +00:00
Samuel Pitoiset
cd59c22325 ac,radv: use better export formats for 8-bit when RB+ isn't allowed 
When RB+ is enabled, R8_UINT/R8_SINT/R8_UNORM should use FP16_ABGR
for 2x exporting performance. Otherwise, use 32_R to remove useless
instructions needed for 16-bit compressed exports.

fossils-db (Vega10):
Totals from 8858 (6.35% of 139517) affected shaders:
SGPRs: 801248 -> 801210 (-0.00%); split: -0.01%, +0.00%
VGPRs: 596224 -> 596120 (-0.02%); split: -0.02%, +0.01%
CodeSize: 71462452 -> 71356684 (-0.15%); split: -0.15%, +0.00%
MaxWaves: 37097 -> 37105 (+0.02%); split: +0.04%, -0.02%
Instrs: 13963177 -> 13950809 (-0.09%); split: -0.09%, +0.00%
Cycles: 1476539360 -> 1476489996 (-0.00%); split: -0.00%, +0.00%
VMEM: 2363008 -> 2361349 (-0.07%); split: +0.04%, -0.11%
SMEM: 550362 -> 549977 (-0.07%); split: +0.01%, -0.08%
VClause: 245704 -> 245727 (+0.01%); split: -0.01%, +0.02%
SClause: 485161 -> 485104 (-0.01%); split: -0.01%, +0.00%
Copies: 1420034 -> 1422310 (+0.16%); split: -0.01%, +0.17%
Branches: 518710 -> 518705 (-0.00%)
PreSGPRs: 706633 -> 706584 (-0.01%)
PreVGPRs: 547163 -> 547007 (-0.03%); split: -0.03%, +0.01%

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/7512>
 2020-11-23 17:54:16 +00:00
Eric Anholt
eda3e4e055 nir/builder: Add a name format arg to nir_builder_init_simple_shader(). 
This cleans up a bunch of gross sprintfs and keeps the caller from needing
to remember to ralloc_strdup.  I added a couple of '"%s", name ? name :
""' to radv where I didn't fully trace through whether a non-null name was
being passed in.

I also took the liberty of adding a basic name to a few shaders (pan_blit,
unit tests)

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7323>
 2020-11-11 08:50:29 -08:00
Eric Anholt
5f992802f5 nir/builder: Drop the mem_ctx arg from nir_builder_init_simple_shader(). 
This looks a lot more simple now!

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7323>
 2020-11-11 08:50:29 -08:00
Eric Anholt
4e9328e3b6 nir_builder: Return a new builder from nir_builder_init_simple_shader(). 
It's a little inline function, so we can just RAII it for better
ergonomics.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7323>
 2020-11-11 08:49:49 -08:00
Samuel Pitoiset
0401507b77 radv: add support for CmdResolveImage2KHR() 
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/6813>
 2020-09-23 07:14:15 +00:00
Samuel Pitoiset
b6cebf6439 radv: do not recursively begin/end render pass for meta operations 
To avoid breaking SQTT user markers that are emitted to report
barriers and layout transitions to RGP.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4136>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4136>
 2020-03-11 07:54:43 +00:00
Samuel Pitoiset
0d2da2a8c0 radv: add explicit external subpass dependencies to meta operations 
No functional changes because a subpass dependency with dstStageMask
set to VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT is a no-op.

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/3330>
 2020-01-23 11:25:38 +01:00
Samuel Pitoiset
b537be4368 radv: update VK_KHR_depth_stencil_resolve for Vulkan 1.2 
Promoted to Vulkan 1.2 with the KHR suffix omitted.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2020-01-15 08:42:25 -06:00
Samuel Pitoiset
589bfcbde3 radv: init a default multisample state for the resolve FS path 
pMultisampleState must be a valid pointer.

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/3167>
 2019-12-20 08:20:44 +01:00
Eric Engestrom
30f639c181 radv: fix s/load/store/ copy-paste typo 
Fixes: cdc6efddf9 ("radv: implement all depth/stencil resolve modes using graphics")
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-09-24 19:18:54 +01:00
Bas Nieuwenhuizen
216a9d8871 radv: Add extra struct to image view creation. 
For extra args. Unlike image creation, I'm not embedding the vk
struct in there, so all the inline structs can be kept.

Reviewed-by: Dave Airlie <airlied@redhat.com>
 2019-08-07 02:13:07 +02:00
Bas Nieuwenhuizen
a7041f3b4e radv: Store image view also outside framebuffer. 
So we can use it with imageless framebuffers.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
 2019-08-02 22:19:16 +02:00
Bas Nieuwenhuizen
49e6c2fb78 radv: Store color/depth surface info in attachment info instead of framebuffer. 
That way we can use it for imageless framebuffers.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
 2019-08-02 22:18:51 +02:00
Connor Abbott
27f0c3c15e radv: Make FragCoord a sysval 
load_fragcoord is already handled in common code for radeonsi, so we
don't need to do anything to handle it. However, there were some passes
creating NIR with the varying, so we switch them over to the sysval. In
the case of nir_lower_input_attachments which is used by both radv and
anv, we add handling for both until intel switches to using a sysval.

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-07-08 14:14:53 +02:00
Samuel Pitoiset
cdc6efddf9 radv: implement all depth/stencil resolve modes using graphics 
When using graphics, the driver doesn't need to decompress HTILE
before resolving. This path currently doesn't support layers
so we have to fallback to the compute path.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-06-21 14:50:15 +02:00
Samuel Pitoiset
c39a1611ab radv: add radv_get_resolve_pipeline() helper in the graphics path 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-06-11 08:06:42 +02:00
Samuel Pitoiset
ff27eb509a radv: always dirty the framebuffer when restoring a subpass 
The old code was not wrong because the transitions performed
after the resolves should re-emit the framebuffer if needed.

This change is mostly a no-op but it improves consistency
regarding other meta operations that need to save/restore subpasses.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-05-27 13:52:36 +02:00
Samuel Pitoiset
9efa3405a7 radv: do not set preserveAttachments for internal render passes 
We don't use that.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-02-04 13:19:14 +01:00
Samuel Pitoiset
80e809d993 radv: drop useless checks when resolving subpass color attachments 
The Vulkan spec says:
   "If pResolveAttachments is not NULL, for each resolve attachment
    that does not have the value VK_ATTACHMENT_UNUSED, the
    corresponding color attachment must not have the value
    VK_ATTACHMENT_UNUSED."

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-02-04 13:19:14 +01:00
Samuel Pitoiset
a20c2e38d8 radv: store the list of attachments for every subpass 
This reworks how the depth stencil attachment is used for
simplicity. This also introduces radv_render_pass_compile()
helper that will be used for further optimizations.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-02-04 13:17:54 +01:00
Samuel Pitoiset
a7c7d811f1 radv: move subpass image transitions to radv_cmd_buffer_begin_subpass() 
Instead of doing them in radv_cmd_buffer_set_subpass().

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2019-02-04 13:17:52 +01:00
Bas Nieuwenhuizen
3308db2dd7 radv: Possible on-demand compilation fix. 
Seems that in a single case we use the renderpass before checking
the pipeline, so check the renderpass before we use it.

Fixes: fbcd167314 "radv: Add on-demand compilation of built-in shaders."
Tested-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
 2018-08-16 11:13:19 +02:00
Bas Nieuwenhuizen
fbcd167314 radv: Add on-demand compilation of built-in shaders. 
In environments where we cannot cache, e.g. Android (no homedir),
ChromeOS (readonly rootfs) or sandboxes (cannot open cache), the
startup cost of creating a device in radv is rather high, due
to compiling all possible built-in pipelines up front. This meant
depending on the CPU a 1-4 sec cost of creating a Device.

For CTS this cost is unacceptable, and likely for starting random
apps too.

So if there is no cache, with this patch radv will compile shaders
on demand. Once there is a cache from the first run, even if
incomplete, the driver knows that it can likely write the cache
and precompiles everything.

Note that I did not switch the buffer and itob/btoi compute pipelines
to on-demand, since you cannot really do anything in Vulkan without
them and there are only a few.

This reduces the CTS runtime for the no caches scenario on my
threadripper from 32 minutes to 8 minutes.

Reviewed-by: Dave Airlie <airlied@redhat.com>
 2018-08-14 10:26:24 +02:00
Bas Nieuwenhuizen
806a792b43 radv: Make fs key exemplars ordered to be a reverse fs_key lookup. 
While at it, share the exemplars and account for a non-occurring
fs key.

Reviewed-by: Dave Airlie <airlied@redhat.com>
 2018-08-14 10:26:06 +02:00
Samuel Pitoiset
1efc9094e0 radv: don't flush DB before subpass FS resolves 
That shouldn't be needed because the DB state is invalid.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
 2018-07-20 17:30:13 +02:00