AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
169,283 Commits 1 Branch 0 Tags
c19a3a6ddf463c93716aba2842346eef72f81df7
Commit Graph

169283 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Pierre-Eric Pelloux-Prayer c19a3a6ddf util/vbuf: clarify indirect draws handling 
It wasn't obvious from the u_vbuf code alone that num_draws
is guaranteed to be 1 in the indirect case.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22237>
 2023-04-04 11:50:58 +00:00
Pierre-Eric Pelloux-Prayer 2b717f01f7 mesa: fix invalid index_bo refcounting 
If take_index_buffer_ownership is true, then we should reserve
enough references of index_bo or we'll get an underflow later.

Fixes: 819627041e ("mesa: set pipe_draw_info::index::resource directly and remove gl_bo")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8747
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22237>
 2023-04-04 11:50:58 +00:00
Mike Blumenkrantz 09320705f0 zink: be explicit about separate shader dsl indexing during creation 
this fixes a minor issue where a fs without a vs dsl would have the wrong
index

Fixes: e3b746e3a3 ("zink: use GPL to handle (simple) separate shader objects")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22269>
 2023-04-04 11:29:01 +00:00
Tapani Pälli b0cdd5a1d2 anv: check for MESA_SHADER_TESS_CTRL with get_tcs_prog_data 
Fixes: 86d931724d ("anv: Implement Wa_14015297576")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22282>
 2023-04-04 10:42:12 +03:00
Tapani Pälli 98c6db3fc8 anv: take primitive ID override to account Wa_14015297576 
Commit ca4ec49b0e took primitive ID override fields in to use, this
has to be checked as part of Wa_14015297576.

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22282>
 2023-04-04 10:41:58 +03:00
Mike Blumenkrantz e84cf80fe1 zink: always store nir serialized 
this saves a ton of memory and has minimal, if any cpu impact

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz 4e1668384d zink: store num_inlinable_uniforms separately for cs programs 
compute is a special case because the zink_shader itself is created
in a thread, which means it cannot be accessed directly at bind time
since it may not have finished creating itself yet

to avoid prematurely waiting on an async fence, the one value needed
at bind time can instead be stored separately

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz 517146f540 zink: move nir cloning out to callers of zink_shader_compile 
this will make more sense with future changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz dfa8c15164 zink: break out nir blob deserializing 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz 0563499db0 zink: simplify assign_io() further 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz fd2714b5a9 zink: store nir as serialized on zink_shader structs 
nir_shader objects are hefty, and they really add up when there's a lot
of them. there's also not much use in keeping them around, as any time
they'll be used, they're always cloned first, and deserializing isn't
likely to be any slower than a clone

cuts driver memory usage by ~40% for tomb raider

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz e40134aea8 zink: streamline nir cloning for assign_io 
more relics from the time of pipe_shader_type

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz 2e1ad9f817 zink: directly return nir from zink_shader_tcs_create 
not currently used but will be

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz d3c9112404 zink: swap nir pointers when compiling compute shaders 
these are the same pointer, just use a different one

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz fdb9c91922 zink: pass nir directly to zink_shader_tcs_create() 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz 6524b9285f zink: simplify fbfetch output detection from fs 
this should be identical to the previous code, except it also
handles depth/stencil (not yet supported)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:41 +00:00
Mike Blumenkrantz edc804e50f zink: use zink_shader::info instead of zink_shader::nir::info 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:40 +00:00
Mike Blumenkrantz b33d5d1a98 zink: generate flat_flags during shader creation 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:40 +00:00
Mike Blumenkrantz f26526ff33 zink: pass nir_shader to update_so_info() 
no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:40 +00:00
Mike Blumenkrantz f7b76d681b zink: add and populate a shader_info struct to zink_shader 
this avoids the need to use the nir pointer to access shader info

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:40 +00:00
Mike Blumenkrantz eb30744562 zink: delete shader reordering in assign_io() 
this is a relic of when pipe_shader_type was different than gl_shader_stage

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22266>
 2023-04-04 01:37:40 +00:00
Alejandro Piñeiro bbd124fd00 v3dv/pipeline: use pipeline depth bias enabled to fill up CFG packet 
Even if the VkPipelineRasterizationStateCreateInfo sets
depthBiasEnable, internally we comput if it is really makes sense, and
use that to decide for example if we emit the Depth Offset packet.

But we were not using this to enable Depth Bias through the depth
offset enable field on the CFG packet.

So in some tests we were enabling depth bias, but not emitting the
packet to configure it, that seemed somewhat inconsistent.

This didn't cause any issue so far, but let's be conservative.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22252>
 2023-04-04 00:06:38 +00:00
Lionel Landwerlin a88aedbfa5 intel/perf: fix OA format selection on MTL 
Anything Gfx12.5+ has a different format.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 90c86fe63e ("intel: add MTL performance metrics")
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22257>
 2023-04-03 23:34:17 +00:00
Lionel Landwerlin c61101f151 intel/dev: set a default valid kmd_type 
src/intel/perf/intel_perf_query_layout.c needs a valid kmd type to
look at the metrics

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 757e2dd692 ("intel/perf: Disable it for Xe KMD")
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22257>
 2023-04-03 23:34:17 +00:00
Ian Romanick 2016d9f46c intel/fs: Rework the loop of opt_combine_constants that collects constants 
This is a bit more wordy, but it will greatly simplify some future
changes.

v2: Rebase on ADD3 changes.

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Tested-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22274>
 2023-04-03 21:50:06 +00:00
Ian Romanick 9e4bb4bfcf intel/fs: Refactor part of opt_combine_constants to a separate function 
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Tested-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22274>
 2023-04-03 21:50:06 +00:00
Ian Romanick 593cde0432 intel/fs: Output opt_combine_constants debug to stderr 
It's a lot more useful to have it in the same stream with the
INTEL_DEBUG=fs output.

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Tested-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22274>
 2023-04-03 21:50:06 +00:00
Iván Briano eadc64ab24 vulkan/wsi/display: do not dereference a NULL pointer 
Fixes dEQP-VK.wsi.direct_drm.swapchain.simulate_oom.*

Cc: mesa-stable

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19025>
 2023-04-03 20:21:17 +00:00
Faith Ekstrand 6ad5f885af vulkan: Drop VkRenderingSelfDependencyInfoMESA 
It's no longer needed as VK_EXT_attachment_feedback_loop_layout provides
everything we need.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Faith Ekstrand b4c18b9b76 vulkan: Drop vk_render_pass_state::*self_dependenc* 
ANV was the only driver using them and now it's on feedback loop flags.
Other drivers should convert instead of depending on the old version.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Faith Ekstrand 6d4e605378 hasvk/pipeline: Use feedback loop flags for self-dependencies 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Faith Ekstrand 8df8a3c19b anv/pipeline: Use feedback loop flags for self-dependencies 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Faith Ekstrand f3876db1ee vulkan: Plumb rendering flags through vk_graphics_pipeline_state 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Connor Abbott f4b534d50f vk/render_pass: Support VK_EXT_fragment_density_map 
Support emulating "classic" FDM with dynamic rendering.

Reviewed-by: Faith Ekstrand <faith@gfxstrand.net>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Connor Abbott df9694e98e tu: Use vk_pipeline_get_renderpass_flags() 
With this we can remove usage of VkSelfDependencyInfoMESA.

Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Jason Ekstrand bd26ae0086 vulkan: Record pipeline flags in the render pass 
This records any rendering pipeline flags in the render pass.  This
provides much-needed information for the VK_KHR_fragment_shading_rate
and VK_EXT_fragment_density_map extensions as well as provides an
alternative to VkRenderingSelfDependencyInfoMESA which is based on
VK_EXT_attachment_feedback_loop_layout.

v2 (Connor): Name something more general
v3 (Faith): Also add the FSR flag

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22191>
 2023-04-03 18:13:01 +00:00
Samuel Pitoiset 9582c1e52a radv/ci: add one more flake 
This one is randomly failing.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22258>
 2023-04-03 17:20:13 +00:00
Samuel Pitoiset d74790a865 zink/ci: remove primitive-id-no-gs-quads from the NAVI10 fail list 
It has been fixed too.

Fixes: 5fb8ba0eb8 ("zink/ci: remove `primitive-id-no-gs-quads` from radv-vangogh-fails")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22258>
 2023-04-03 17:20:13 +00:00
Rhys Perry f92994bb83 radv/gfx11: improve RT scratch allocation 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22232>
 2023-04-03 16:35:17 +00:00
Timur Kristóf 50e9a8f7c2 radv: Use linear_dispatch info in GFX11 task/mesh draw packet. 
This helps the CP (command processor) optimize mesh shader
dispatch when Y = Z = 1.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22222>
 2023-04-03 15:36:02 +00:00
Timur Kristóf e42d2bd534 nir: Gather compile time constant task->mesh dispatch size. 
Some GPUs such as AMD RDNA3 can use this information
to optimize mesh shader dispatches.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22222>
 2023-04-03 15:36:02 +00:00
Timur Kristóf 836204da25 aco: Better phi lowering for merge block when else-side is const. 
Add a new special case for binary merge blocks to boolean
phi lowerig. This special case benefits shaders that
have divergent branches with an empty else block,
for example all NGG culling shaders.

Fossil DB stats on Rembrandt (NGG culling enabled):

Totals from 61778 (45.79% of 134913) affected shaders:
SpillVGPRs: 2268 -> 2284 (+0.71%); split: -1.10%, +1.81%
CodeSize: 164317952 -> 162962772 (-0.82%); split: -0.83%, +0.00%
Instrs: 31249824 -> 30910686 (-1.09%); split: -1.09%, +0.00%
Latency: 154948555 -> 154781097 (-0.11%); split: -0.12%, +0.02%
InvThroughput: 30397664 -> 30370872 (-0.09%); split: -0.13%, +0.04%
VClause: 529239 -> 529229 (-0.00%); split: -0.00%, +0.00%
SClause: 783417 -> 783430 (+0.00%)
Copies: 2627570 -> 2595161 (-1.23%); split: -1.25%, +0.02%
Branches: 976506 -> 976508 (+0.00%); split: -0.00%, +0.00%

Fossil DB stats on GFX11 (NGG culling disabled):

Totals from 895 (0.66% of 134913) affected shaders:
SpillVGPRs: 2258 -> 2322 (+2.83%); split: -0.44%, +3.28%
CodeSize: 6229152 -> 6215880 (-0.21%); split: -0.37%, +0.16%
Scratch: 216576 -> 215808 (-0.35%); split: -0.47%, +0.12%
Instrs: 1202077 -> 1198396 (-0.31%); split: -0.43%, +0.13%
Latency: 15921336 -> 16000561 (+0.50%); split: -0.74%, +1.24%
InvThroughput: 7425765 -> 7474891 (+0.66%); split: -0.67%, +1.33%
VClause: 22976 -> 23008 (+0.14%); split: -0.03%, +0.17%
SClause: 38269 -> 38271 (+0.01%)
Copies: 123244 -> 123896 (+0.53%); split: -0.30%, +0.83%
Branches: 47570 -> 47574 (+0.01%); split: -0.00%, +0.01%

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21493>
 2023-04-03 14:36:07 +00:00
Timur Kristóf 81b4806d64 aco: Call dominator_tree before lower_phis. 
This just makes it possible to use the dominator
tree information during phi lowering.

No Fossil DB changes on GFX11.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21493>
 2023-04-03 14:36:07 +00:00
Timur Kristóf 0eb7c49c7f aco: Pop branch operands when targets are same in SSA elimination. 
The branch instruction is no longer conditional when the targets are the
same, so the operand is not necessary and can be removed.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21493>
 2023-04-03 14:36:07 +00:00
Timur Kristóf 739bd03c37 aco: Don't verify branch exec read when eliminating exec writes. 
Verifying that the branch instruction reads exec is not actually
necessary because the pattern that we look for already implies that.

This prepares for the next commit which will remove the exec operand
from branches that have the same target. These branches will no
longer read exec, but they should still get the same optimization.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21493>
 2023-04-03 14:36:07 +00:00
Timur Kristóf 9eb04d8f96 aco: Simplify get_phi_operand using Operand::c32_or_c64. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21493>
 2023-04-03 14:36:07 +00:00
Timur Kristóf 0211e66f65 aco: Don't remove exec writes that also write other registers. 
Don't eliminate an instruction that writes registers other than exec and scc.
It is possible that this is eg. an s_and_saveexec and the saved value is
used by a later branch.

Fixes: bc13049747
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21493>
 2023-04-03 14:36:07 +00:00
Timur Kristóf 54da863956 aco: Consider p_cbranch_nz as divergent branch too. 
A p_cbranch_nz instruction that reads exec is divergent too.

Fixes: f030b75b7d
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21493>
 2023-04-03 14:36:07 +00:00
David Heidelberg 78644c9bb0 ci: disable lima farm, currently out-of-space, needs to be fixed 
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22265>
 2023-04-03 14:41:15 +02:00
Tatsuyuki Ishi 7fe469d6c8 radv: Pre-compute descriptor set layout hash. 
While analyzing cache loading performance, hashing the pipeline layout was
surprisingly consuming around 4% of time, sometimes close to the cost of
hashing shader modules.

Turns out we were hashing the pipeline layout on every pipeline creation.
Considering that pipeline layouts are usually deduplicated by the
application, this was amplifying the hashing cost by a big margin.

With Graphics Pipeline Library, we do need to rebuild the pipeline layout
by combining those from each library, but we can memoize the hash of the
descriptor set layout. The cost of re-hashing hashes is negligible since
each descriptor set layout can amount to 1–2KB in size.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22254>
 2023-04-03 08:46:08 +00:00