Mike Blumenkrantz [Fri, 21 Oct 2022 19:05:16 +0000 (15:05 -0400)]
zink: simplify some program descriptor helpers
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19306>
Karol Herbst [Tue, 25 Oct 2022 22:02:15 +0000 (00:02 +0200)]
zink: fix spirv_builder_spec_const_uint
We need to create the type upfront otherwise the spir-v binary gets
corrupted.
Fixes:
c305a2c9625 ("zink: move spec constant emission to the types/consts block")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19310>
Danylo Piliaiev [Wed, 19 Oct 2022 16:47:36 +0000 (18:47 +0200)]
freedreno: Add 'replay' tool which allows to replay cmdstreams
Replay command stream obtained from:
- /sys/kernel/debug/dri/0/rd
- /sys/kernel/debug/dri/0/hangrd
!!! Command stream capture should be done with ALL buffers:
- echo 1 > /sys/module/msm/parameters/rd_full
Requires kernel with MSM_INFO_SET_IOVA support.
This tool is intended for reproduction of various GPU issues:
- GPU hangs, note that command stream obtained from hangrd
may not reproduce (rarely) the hang, since the buffers are
snapshotted at the moment of the hang and not at the start
of the hanging command stream.
- TODO: Misrendering, would require marking framebuffer images
at each renderpass in order to fetch and decode them.
Code from Freedreno/Turnip is not re-used here since the relevant
pieces may introduce additional allocations which cannot be allowed
during the replay.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19171>
Samuel Pitoiset [Mon, 24 Oct 2022 02:15:06 +0000 (02:15 +0000)]
aco: add support for device clock on GFX11
According to LLVM, s_sendmsg_rtn(GET_REALTIME) should be used instead
of s_memrealtime.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19267>
Samuel Pitoiset [Tue, 25 Oct 2022 11:17:34 +0000 (13:17 +0200)]
aco: split the sendmsg enumeration into sendmsg_rtn
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19267>
Samuel Pitoiset [Mon, 24 Oct 2022 02:14:24 +0000 (02:14 +0000)]
aco: add support for s_sendmsg_rtn_b{32,64}
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19267>
Samuel Pitoiset [Fri, 21 Oct 2022 00:37:28 +0000 (00:37 +0000)]
ac/llvm: add support for device clock on GFX11
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19267>
Rhys Perry [Fri, 30 Sep 2022 18:29:43 +0000 (19:29 +0100)]
ac/nir: add legacy streamout and GS copy shader helpers
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19302>
Rhys Perry [Wed, 28 Sep 2022 18:32:26 +0000 (19:32 +0100)]
radv,nir: add intrinsics for streamout and GS copy shaders
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19302>
Emma Anholt [Tue, 25 Oct 2022 00:05:03 +0000 (17:05 -0700)]
turnip: Enable LRZ testing (not writing) in the presence of discards.
We can LRZ test to not rasterize maybe-discarded pixels, as long as we
don't LRZ write a Z that may be discarded. We can drop this check,
because LRZ writes are already disabled by TU_LRZ_FORCE_DISABLE_WRITE, and
tu_6_build_depth_plane_z_mode() uses has_kill to choose EARLY_LRZ_LATE_Z
(or just LATE_Z if LRZ isn't enabled).
gfxbench aztec ruins: +7.93104% +/- 0.117845% (n=4)
ANGLE trex_200: +11.7208% +/- 0.476166% (n=3)
ANGLE aztec_ruins_high: +11.9138% +/- 0.147586% (n=3)
Fixes: #6327 (now vk-5-normal is +0.712724% +/- 0.0655751% to gl-5-normal)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19286>
Eric Engestrom [Tue, 25 Oct 2022 08:20:37 +0000 (09:20 +0100)]
vc4: mark piglit copypixels-(draw-)sync as flaky
They sometimes fail when running all the tests together, but never when
running just them; not sure how to diagnose this, but for now jusk mark
them as flaky.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19295>
Eric Engestrom [Tue, 13 Sep 2022 14:56:45 +0000 (16:56 +0200)]
vc4: send shaderdb output through KHR_debug
This allows us to see the file names for the shaders.
Signed-off-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18599>
Jose Maria Casanova Crespo [Wed, 14 Apr 2021 18:03:21 +0000 (20:03 +0200)]
vc4: consolidate shader-db output
[Eric: update output string format for new shaderdb]
Signed-off-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18599>
Eric Engestrom [Fri, 21 Oct 2022 09:51:14 +0000 (10:51 +0100)]
vc4: use PIPE_MASK_RGBA name instead of its value 0xf
Signed-off-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18599>
Eric Engestrom [Wed, 24 Aug 2022 08:28:09 +0000 (09:28 +0100)]
vc4: pre-compile shaders to make up for the lack of draw calls in shader-db
This means we don't have the variant keys, and need to make up one
variant and pre-compile it.
Signed-off-by: Eric Engestrom <eric@igalia.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18599>
Brian Paul [Wed, 21 Sep 2022 21:40:48 +0000 (15:40 -0600)]
llvmpipe: improve some if/switch code in llvmpipe_get_shader_param()
Move the PIPE_SHADER_CAP_PREFERRED_IR case into the switch statement.
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:14 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_texture.[ch]
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:14 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_tex_sample.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_test_main.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_test_format.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_test_conv.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_test_blend.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_vertex.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_tess.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_surface.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:13 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_so.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:12 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_rasterizer.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:12 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_gs.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:12 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_cs.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:12 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_state_clip.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 04:23:53 +0000 (22:23 -0600)]
llvmpipe: asst. clean-ups in lp_state_derived.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 16 Sep 2022 16:14:05 +0000 (10:14 -0600)]
llvmpipe: asst. clean-ups in lp_state_sampler.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:12 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_scene.h
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:12 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_scene.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:11 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_linear_sampler.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:11 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_linear_interp.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:11 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_flush.[ch]
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:11 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_fence.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:10 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_draw_arrays.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:10 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_debug.h
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 16 Sep 2022 16:14:15 +0000 (10:14 -0600)]
llvmpipe: asst. clean-ups in lp_bld_interp.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:10 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_bld_depth.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:10 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_bld_blend_aos.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:10 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_bld_alpha.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 16:19:54 +0000 (10:19 -0600)]
llvmpipe: asst. clean-ups in lp_setup_tri.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 04:26:24 +0000 (22:26 -0600)]
llvmpipe: remove gotos in do_rect_ccw()
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 16:19:46 +0000 (10:19 -0600)]
llvmpipe: asst. clean-ups in lp_setup_point.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 16:19:34 +0000 (10:19 -0600)]
llvmpipe: asst. clean-ups in lp_setup_line.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 04:23:21 +0000 (22:23 -0600)]
llvmpipe: asst. clean-ups in lp_setup_vbuf.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 04:23:07 +0000 (22:23 -0600)]
llvmpipe: asst. clean-ups in lp_setup_rect.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 04:22:35 +0000 (22:22 -0600)]
llvmpipe: asst. clean-ups in lp_setup.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 16 Sep 2022 16:00:37 +0000 (10:00 -0600)]
llvmpipe: asst. clean-ups in lp_query.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 16 Sep 2022 15:55:27 +0000 (09:55 -0600)]
llvmpipe: asst. clean-ups in lp_screen.[ch]
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 16 Sep 2022 15:51:41 +0000 (09:51 -0600)]
llvmpipe: asst. clean-ups in lp_surface.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 26 Aug 2022 01:31:27 +0000 (19:31 -0600)]
llvmpipe: asst. clean-ups in lp_context.[ch]
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 26 Aug 2022 01:30:30 +0000 (19:30 -0600)]
gallium: whitespace fixes in p_context.h, p_screen.h
Use consistent whitespace near parens. Replace tabs w/ spaces, etc.
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:11 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_rast_linear_fallback.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:11 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_rast_linear.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Wed, 21 Sep 2022 17:30:11 +0000 (11:30 -0600)]
llvmpipe: asst. clean-ups in lp_rast.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 26 Aug 2022 01:12:02 +0000 (19:12 -0600)]
llvmpipe: asst. clean-ups in lp_rast_rect.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 26 Aug 2022 01:04:57 +0000 (19:04 -0600)]
llvmpipe: asst. clean-ups in lp_rast_priv.h
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 26 Aug 2022 01:01:49 +0000 (19:01 -0600)]
llvmpipe: asst. clean-ups in lp_rast_debug.c
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Brian Paul [Fri, 26 Aug 2022 01:01:25 +0000 (19:01 -0600)]
llvmpipe: asst. clean-ups in lp_rast.h
Signed-off-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19289>
Mike Blumenkrantz [Thu, 20 Oct 2022 16:32:18 +0000 (12:32 -0400)]
zink: add some docs for c++ pipeline state functions
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19205>
Mike Blumenkrantz [Thu, 20 Oct 2022 16:11:42 +0000 (12:11 -0400)]
zink: add some descriptor docs
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19205>
Mike Blumenkrantz [Thu, 20 Oct 2022 15:42:38 +0000 (11:42 -0400)]
zink: stop passing pointer to changed_sets to populate_sets()
I think this used to be a thing for caching but I don't remember
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19205>
Mike Blumenkrantz [Thu, 20 Oct 2022 15:40:58 +0000 (11:40 -0400)]
zink: delete unused descriptor function
this is leftover from caching
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19205>
Mike Blumenkrantz [Thu, 20 Oct 2022 14:35:14 +0000 (10:35 -0400)]
zink: add more zink_descriptor_type enum values to improve readability
the values are still the same, but now enum names can be used to more
accurately reflect their actual meaning
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19205>
Mike Blumenkrantz [Thu, 20 Oct 2022 14:26:48 +0000 (10:26 -0400)]
zink: ZINK_DESCRIPTOR_TYPES -> ZINK_DESCRIPTOR_BASE_TYPES
slightly more clear
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19205>
Mike Blumenkrantz [Thu, 20 Oct 2022 14:15:46 +0000 (10:15 -0400)]
zink: rename some variables in descriptor code
this is more accurate
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19205>
Qiang Yu [Thu, 16 Jun 2022 08:19:46 +0000 (16:19 +0800)]
radeonsi: implement nir shader query enabled intrinsics
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17457>
Qiang Yu [Thu, 16 Jun 2022 08:12:13 +0000 (16:12 +0800)]
radeonsi: implement llvm abi atomic_add_prim_count
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17457>
Qiang Yu [Fri, 1 Jul 2022 03:38:00 +0000 (11:38 +0800)]
ac/nir/ngg: add streamout emitted primitive query
For radeonsi to implement GL_TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17457>
Qiang Yu [Wed, 15 Jun 2022 08:50:13 +0000 (16:50 +0800)]
ac/nir/ngg: add query param to ac_nir_lower_ngg_gs
radeonsi may disable it. gfx_level will also be used by latter
vertex param export when gfx11.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17457>
Qiang Yu [Tue, 25 Oct 2022 06:20:04 +0000 (14:20 +0800)]
nir: add nir_load_prim_xfb_query_enabled_amd
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17457>
Qiang Yu [Thu, 16 Jun 2022 07:54:54 +0000 (15:54 +0800)]
nir,ac,radv: add primitive count add intrinsics
radeonsi use shader buffer, but radv use gds for the query
result storage.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17457>
Samuel Pitoiset [Mon, 24 Oct 2022 08:27:15 +0000 (08:27 +0000)]
radv: disable dual source blending in more situations
According to PAL, there is more restrictions that RADV doesn't have.
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/19278>
Pierre-Eric Pelloux-Prayer [Tue, 18 Oct 2022 09:37:22 +0000 (11:37 +0200)]
radeonsi: handle sqtt pipeline in shader prefetch
When sqtt is enabled, the shader code lives in the pipeline bo,
not in the shader bo.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Tue, 18 Oct 2022 09:36:58 +0000 (11:36 +0200)]
radeonsi: store the shader gpu adress in si_shader
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Tue, 18 Oct 2022 09:35:25 +0000 (11:35 +0200)]
radeonsi/sqtt: simplify condition to determine if sqtt is on
We don't need to load screen->debug_flags because sctx->thread_trace
is already telling us if sqtt is enabled.
Furthermore we can perform this check only for GFX9 because sqtt
isn't supported currently on older chips.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Tue, 18 Oct 2022 09:03:24 +0000 (11:03 +0200)]
radeonsi: simplify si_prefetch_shaders
Since
93cd96b5237 the only used value of si_L2_prefetch_mode
was PREFETCH_ALL so we can remove some dead code in si_prefetch_shaders.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Tue, 4 Oct 2022 08:33:52 +0000 (10:33 +0200)]
radeonsi/tests: update expected results for polaris11
Using the latest released LLVM version 15.0.1.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Mon, 3 Oct 2022 12:19:37 +0000 (14:19 +0200)]
radeonsi/tests: update expected results for navi10
Using the latest released LLVM version 15.0.1.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Wed, 28 Sep 2022 12:44:51 +0000 (14:44 +0200)]
radeonsi/tests: update expected results for navi21
Using the latest released LLVM version 15.0.1.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Thu, 29 Sep 2022 13:21:07 +0000 (15:21 +0200)]
radeonsi/tests: simplify the script
And use exit(0) on success, exit(1) on failure.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Thu, 29 Sep 2022 13:20:00 +0000 (15:20 +0200)]
radeonsi/tests: handle flakes file for all test suite
And a generic flakes file that applies to all GPUs.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Mon, 26 Sep 2022 12:43:24 +0000 (14:43 +0200)]
radeonsi/sqtt: allow to disable spm counters
And fix rgp capture on GFX9 (where spm was disabled by default).
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Mon, 26 Sep 2022 12:34:03 +0000 (14:34 +0200)]
radeonsi/sqtt: re-export shaders in a single bo
RGP expects a pipeline's shaders to be all stored sequentially, eg:
[vs][ps][gs]
As such, it assumes a single bo is dumped to the .rgp file, with
the following info:
* va of the bo
* offset to each shader inside the bo
For radeonsi, the shaders are stored individually, so we may have
a big gap between the shaders forming a pipeline => we can produce
very large file because the layout in the file must match the one
in memory (see the warning in ac_rgp_file_write_elf_text).
This commit implements a workaround: gfx shaders are re-exported as a
pipeline.
To update the shader address, a new state is created (sqtt_pipeline),
which will overwrite the needed _PGM_LO_* registers.
This reduces DeuxEX rgp captures from 150GB+ to less than 100MB.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Pierre-Eric Pelloux-Prayer [Mon, 26 Sep 2022 09:34:16 +0000 (11:34 +0200)]
radeonsi/sqtt: notify sqtt when we skip a flush
Otherwise we may not produce any data because
si_handle_thread_trace is never called.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18865>
Yonggang Luo [Thu, 20 Oct 2022 19:56:49 +0000 (03:56 +0800)]
util: #include "c11/threads.h" instead of #include "util/u_cpu_detect.h" u_format.c
As it does not access any symbols in util/u_cpu_detect.h but accessed symbols in "c11/threads.h"
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19266>
Yonggang Luo [Thu, 20 Oct 2022 19:18:35 +0000 (03:18 +0800)]
util: Move UTIL_MAX_CPUS from u_thread.h to u_cpu_detect.h
UTIL_MAX_CPUS is not used by u_thread.* anymore after commit
"util: replace UTIL_MAX_CPUS by util_cpu_caps.num_cpu_mask_bits"
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19266>
Yonggang Luo [Thu, 20 Oct 2022 19:16:14 +0000 (03:16 +0800)]
util: Merge duplicated function prototype in u_thread.h that named with util_barrier*
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19266>
Yonggang Luo [Thu, 20 Oct 2022 19:15:02 +0000 (03:15 +0800)]
util: Remove non used UTIL_MAX_L3_CACHES in u_thread.h
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19266>
Karmjit Mahil [Mon, 10 Oct 2022 09:26:24 +0000 (10:26 +0100)]
pvr: Change pvr_get_hw_clear_color() to pack based on accum formats.
`pvr_get_hw_clear_color()` now packs formats based on the accum
format which is how the hw deals with the formats internally and
might not line up exactly with the vk format representation.
E.g. R5G6B5_UNORM_PACK16 uses the U8 accum format so the USC will
internally use 3 bytes (1 per component) to deal with it instead
of 2 bytes which the vk format specifies. On USC EMITPIX, the PBE
will pack the results to 2 bytes using PACKMODE_R5G6B5 resulting
in the final value being in the vk format representation.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19218>
Karmjit Mahil [Mon, 5 Sep 2022 08:45:54 +0000 (09:45 +0100)]
pvr: Add PBE accum format size in bytes.
This will be used later on to implement
vkCmdClearAttachments() where we'll need to pick the
appropriate clear color shader based on the dwords used.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19218>
Erik Faye-Lund [Thu, 20 Oct 2022 10:09:59 +0000 (12:09 +0200)]
docs: vmware -> VMware
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19191>
Erik Faye-Lund [Thu, 20 Oct 2022 09:41:05 +0000 (11:41 +0200)]
docs: healthcheck -> health check
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19191>
Erik Faye-Lund [Thu, 20 Oct 2022 09:14:32 +0000 (11:14 +0200)]
docs: dependant -> dependent
We generally prefer US English in the docs, so let's correct this.
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19191>
Erik Faye-Lund [Thu, 20 Oct 2022 09:02:41 +0000 (11:02 +0200)]
docs: virtualisation -> virtualization
We generally perfer US English spelling in our docs, so let's fix up
this one word.
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19191>
Erik Faye-Lund [Thu, 20 Oct 2022 09:38:35 +0000 (11:38 +0200)]
docs: Linux fbdev -> Linux Framebuffer
It's either "fbdev" or "Linux Framebuffer", not "Linux fbdev". Since we
can spare the extra letters here, let's spell it out.
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19191>