AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
179,397 Commits 1 Branch 0 Tags
c29d8a9e68b7994cc5efd82b45038548eeaa2a41
Commit Graph

179397 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Rhys Perry c29d8a9e68 ac/nir,radv: pass workgroup size to ac_nir_lower_ngg_ms 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry c300a496ea ac/nir: add emit_ms_outputs helper 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry 23cb0b00db radv: implement mesh shader gs_fast_launch=2 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry c63ac28014 ac/nir: optimize mesh shader local_invocation_index 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry cddbe9a4c2 ac/nir: implement mesh shader gs_fast_launch=2 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry 75bc2e7149 ac/nir: refactor mesh vertex/primitive export 
To prepare for row export, move them into helpers with index and row
parameters.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry 24431419ce ac/nir: remove dead code 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry 6ec72806b7 ac/nir: add row parameter to helpers 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry 4c3677094e aco,nir: add export_row_amd intrinsic 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Rhys Perry 4bd4ff5d9b nir: improve ms_cross_invocation_output_access with local_invocation_id 
Since GFX11, RADV doesn't need to lower local_invocation_id.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25040>
 2023-10-24 21:36:06 +00:00
Caio Oliveira b91ed68fa0 intel/compiler: Don't emit calls to validate() in release build 
While the fs_visitor::validate() implementation is empty in release
build, we still emit calls to it since it is defined in a separate
compilation unit than its callers.  To fix this, just expose an inline
empty function in the header for the release mode.

Fossil run time differences in TGL laptop (difference at 95.0% confidence):

```
Rise of The Tomb Rider (Native) [n=7]
        -0.482857 +/- 0.010932
        -1.60608% +/- 0.0363621%

Cyberpunk 2077 (DXVK) [n=7]
        -0.987143 +/- 0.0904516
        -0.82996% +/- 0.076049%

Batman Arkham City (DXVK) [n=7]
        -7.74857 +/- 0.329561
        -1.46298% +/- 0.0622231%
```

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25847>
 2023-10-24 21:10:35 +00:00
Samuel Pitoiset 13cac671fd ac/gpu_info: fix querying the maximum number of IBs per ring 
The kernel uses 32-bit and Mesa was using 8-bit...

Fixes: 96345ae2ed ("ac/gpu_info: query the maximum number of IBs per submit from the kernel")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25867>
 2023-10-24 19:53:08 +00:00
Rohan Garg 3bf1b7deba anv: selectively enable FCV optimization for DG2 
Enabling FCV on MTL breaks a number of games and benchmarks. Let's
disable it for now till we can root cause the issue.

Closes: #9987
Fixes: 26c2c9 ('anv: enable FCV for Gen12.5')
Signed-off-by: Rohan Garg <rohan.garg@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25863>
 2023-10-24 19:27:14 +00:00
Rohan Garg 25a232238f anv: turn off non zero fast clears for CCS_E 
This helps fix a performance regression on games such as F1 22 and RDR2.
Turning on non zero fast clears causes additional partial resolves for
these games that degrades performance. Let's turn off non zero fast
clears till we can eliminate the partial resolves.

Signed-off-by: Rohan Garg <rohan.garg@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25863>
 2023-10-24 19:27:14 +00:00
Boyuan Zhang 2179cd129f radeonsi/vcn: disable tmz ctx buffer for VCN_2_2_0 
VCN_2_2_0 should not use tmz context buffer.

Fixes: ffbbf23e
A minor fix for above commit to use the original comparison logic "<"
instead of "<="

Cc: mesa-stable

Signed-off-by: Boyuan Zhang <boyuan.zhang@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25856>
 2023-10-24 18:48:45 +00:00
Juan A. Suarez Romero 209376f01a v3d/vc4/ci: add new fails/timeout 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25868>
 2023-10-24 18:27:34 +00:00
Timur Kristóf eff4db3718 radv: Add gang follower semaphore functions. 
These functions implement the possibility to have the
gang leader wait for its follower.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25796>
 2023-10-24 17:58:12 +00:00
Timur Kristóf b5e0a04d4e radv: Slightly refactor gang semaphore functions. 
This is in preparation for adding some new ones.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25796>
 2023-10-24 17:58:12 +00:00
Timur Kristóf cbbb85c907 radv: Allow gang submit use cases other than task shaders. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25796>
 2023-10-24 17:58:12 +00:00
Timur Kristóf 67ce3f11e8 radv: Simplify gang CS and semaphore initialization. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25796>
 2023-10-24 17:58:12 +00:00
Samuel Pitoiset 56a05b9251 radv: fix missing predicate bit for WRITE_DATA helper 
Fixes: 12a753f8d2 ("radv: Use new WRITE_DATA helper in more places.")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25850>
 2023-10-24 14:53:04 +00:00
Samuel Pitoiset f9cc2c5ba4 Revert "radv: fix finding shaders by PC" 
Shaders are no longer allocated contiguously in memory and the next
pointer is now the shader.

This fixes using the trap handler on GFX8.

This reverts commit 2dcd12f38b.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25852>
 2023-10-24 14:19:13 +00:00
Bas Nieuwenhuizen 653bcbcb09 radv: Expose VK_KHR_cooperative_matrix. 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24683>
 2023-10-24 13:24:18 +00:00
Bas Nieuwenhuizen 9df4703fbb radv: Add cooperative matrix lowering. 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24683>
 2023-10-24 13:24:18 +00:00
Bas Nieuwenhuizen 31863aa48c radv: Don't transparently use wave32 with cooperative matrices. 
The instruction has different regsizes for wave32 vs. wave64.

To ensure cases with cooperative matrix load/store without any
actual wmma instructions get handled correctly, also require
full subgroups if subgroup invocation/id are used. Not sure
those could be transparently changed anyway.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24683>
 2023-10-24 13:24:18 +00:00
Bas Nieuwenhuizen d8458c0559 aco: Make RA understand WMMA instructions. 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24683>
 2023-10-24 13:24:18 +00:00
Bas Nieuwenhuizen 5e7c828c0e aco: Add WMMA instructions. 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24683>
 2023-10-24 13:24:18 +00:00
Bas Nieuwenhuizen a29cd20d17 nir: Add AMD cooperative matrix intrinsics. 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24683>
 2023-10-24 13:24:18 +00:00
Helen Koike 663ad191ca ci/ci_run_n_monitor: print job duration time 
If the job is running, it prints the elapsed time so far, if it finished
already, it prints the time it took.

Signed-off-by: Helen Koike <helen.koike@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25822>
 2023-10-24 12:57:14 +00:00
Helen Koike 69012e355e ci/marge_queue: add pretty_dutation() 
Add pretty_duration() function that prints time in format 6m23s and use
it for marge_queue.

Signed-of-by: Helen Koike <helen.koike@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25822>
 2023-10-24 12:57:14 +00:00
Samuel Pitoiset 96345ae2ed ac/gpu_info: query the maximum number of IBs per submit from the kernel 
When the query fails (unsupported by the current kernel), we fallback
to some rough estimate.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10014
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25803>
 2023-10-24 11:05:26 +00:00
Rohan Garg f85d8d908c anv: cleanup includes 
Signed-off-by: Rohan Garg <rohan.garg@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25766>
 2023-10-24 10:33:57 +00:00
Rhys Perry fc342c897f Revert "radv: pre-init surface info" 
This reverts commit e15a4e6e1a.

The change wasn't thread safe.

Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9955
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10021
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25830>
 2023-10-24 10:11:34 +00:00
Samuel Pitoiset aba815c041 radv: initialize video decoder for GFX11.5 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25304>
 2023-10-24 08:20:47 +02:00
Samuel Pitoiset 50845f6fa4 radv: set ENABLE_PING_PONG_BIN_ORDER for GFX11.5 
Ported from RadeonSI.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25304>
 2023-10-24 08:20:47 +02:00
Yogesh Mohan Marimuthu 62a2ed8602 radeonsi: add radeonsi to GL_RENDERER string 
adding driver name to GL_RENDERER string will help to quickly
identify the driver being used.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25148>
 2023-10-24 04:46:27 +00:00
Yogesh Mohan Marimuthu 6cf58bc862 radeonsi: In gfx6_init_gfx_preamble_state() use gfx_level only from sctx 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25148>
 2023-10-24 04:46:27 +00:00
Yogesh Mohan Marimuthu b1111f35c6 radeonsi: correct old comment in si_emit_framebuffer_state() 
si_init_cs_preamble_state() function does not exist anymore and
PA_SC_WINDOW_SCISSOR_TL is set to 0,0. Update the comments with
this information.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25148>
 2023-10-24 04:46:27 +00:00
Yogesh Mohan Marimuthu e373b7c07e docs: remove document for unused variable dfsm from AMD_DEBUG 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25148>
 2023-10-24 04:46:26 +00:00
Yogesh Mohan Marimuthu cd2e3fc59a radeonsi: add more documentation for dpbb debug env variable 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25148>
 2023-10-24 04:46:26 +00:00
Mike Blumenkrantz 9de455bc43 zink: check for sampler view existence during zink_rebind_all_images() 
views with mutable formats do not get created immediately, which means
access of them must be checked first

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25829>
 2023-10-24 02:24:15 +00:00
Mike Blumenkrantz 12fc8cf4df zink: only increment image_rebind_counter on image export if binds exist 
rebinding all images on all contexts is only relevant for images which
have bindings, otherwise it's just pointless churn

fixes #10016

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25829>
 2023-10-24 02:24:15 +00:00
Lang Yu a65aab411f amd/radeonsi: add missing stuff for gfx11.5 
Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25836>
 2023-10-24 01:07:12 +00:00
Lang Yu f9629fa729 amd/common: add missing stuff for gfx11.5 
Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25836>
 2023-10-24 01:07:12 +00:00
Eric Anholt aed6a39c10 glsl: Retire dround lowering. 
We have competent lowering in NIR already available.

Drivers exposing CAP_DOUBLES but not SHADER_CAP_DROUND:
- d3d12 (NIR lowers ~0 if the underlying impl doesn't do floats)
- svga (Now sets the NIR lowering options)
- softpipe (Doesn't do GL4 so you can't use doubles anyway)
- llvmpipe (Lowers dround_even in NIR and passees the rest through
            successfully)
- zink (NIR lowers ~0 if the underlying impl doesn't do floats,
        otherwise passes things through successfully, except needed
        dround_even lowering to avoid lavapipe regression with
        native doubles)
- r600 (sets NIR rounding lowering flags, and lowers all fsign)

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25777>
 2023-10-24 00:16:30 +00:00
Eric Anholt b416248cb5 nir: Add nir_lower_dsign as 64-bit fsign lowering. 
Right now some drivers are doing dsign lowering in GLSL and haven't had to
have a NIR path due to there not being a corresponding vulkan driver.  We
want this in NIR now so that we can retire that batch of GLSL lowering
code.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25777>
 2023-10-24 00:16:30 +00:00
Eric Anholt 4c61de56c5 llvmpipe: Set nir_lower_dround_even. 
Avoids a regression when disabling GLSL-level lowering of the dround class
of ops, while passing through doubles to llvm for the other rounding ops.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25777>
 2023-10-24 00:16:30 +00:00
Eric Anholt b1b0ebba1e glsl: Remove int64 div/mod lowering. 
Most drivers that can expose GL4 were claiming the cap anyway (llvmpipe,
softpipe, zink, iris, nvc0, radeonsi, r600, freedreno, d3d12), and just
doing lowering in NIR if nessary.

crocus was only claiming the cap for gen8, but the backend compiler
enables NIR lowering regardless.

svga is the only other GL4 driver that didn't set it, and we can just set
the NIR lowering flag.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25777>
 2023-10-24 00:16:30 +00:00
José Roberto de Souza bd546f9e54 anv: Switch Xe KMD vm bind to sync 
It was never actually async as it was doing a DRM_IOCTL_SYNCOBJ_WAIT
right after DRM_IOCTL_XE_VM_BIND but it was required to allow the
partial binds required by sparse.
But it is now fixed and we can switch back to sync vm bind.

In future we will switch back to async vm bind to improve performance
but this time it will be properly implemented.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25300>
 2023-10-23 23:24:26 +00:00
José Roberto de Souza 531605accf intel: Sync xe_drm.h 
Sync xe_drm.h with commit xxxxx ("drm/xe/uapi: Fix naming of XE_QUERY_CONFIG_MAX_EXEC_QUEUE_PRIORITY").

One not so straght forward change is that sync VM binds now don't
require a syncobj anymore, the uAPI will return as soon the VM bind
operations are done.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25300>
 2023-10-23 23:24:26 +00:00