71677 commits

Author	SHA1	Message	Date
Olivia Lee	dccb431254	lavapipe: replace vk_conv_topology with vk_topology_to_mesa from vulkan/util ... Signed-off-by: Olivia Lee <olivia.lee@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37038>	2025-09-03 22:04:14 -07:00
Yonggang Luo	949a056934	tgsi: Fixes ntt_should_vectorize_io parameters ... Fixes: 5f757bb95c ("nir: Make the load_store_vectorizer provide align_mul + align_offset.") This is found when I am trying to narrow bit_size and num_components to uint8_t Signed-off-by: Yonggang Luo <luoyonggang@gmail.com> Reviewed-by: Emma Anholt <emma@anholt.net> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37042>	2025-09-04 01:42:19 +00:00
Sagar Ghuge	bc8e29c04e	iris: Emit state cache invalidation after every compute dispatch ... Implement HSD 16028171704/14025112257: LSC state cache livelock:- Once state cache entries are full, subsequent walker dispatches with two threads per thread group maybe gets stuck infinitely because of state cache live lock. One thread continuously stuck in loop doing UGM fence + evict and UGM read is waiting on UGM read to have certain value. while other thread supposed to update the value that first thread is waiting for. But since entries are full in state cache, there is second thread never make progress. Closes: #12352 Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37128>	2025-09-04 00:14:48 +00:00
Valentine Burley	c6ce1a0caf	zink/ci: Disable zink-anv-cml-asan ... Too flaky at the moment with no quick fix. Signed-off-by: Valentine Burley <valentine.burley@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37154>	2025-09-03 22:52:43 +00:00
Christian Gmeiner	b4bac915f0	etnaviv: Add support for ARB_texture_gather ... The hardware support for tg4 is unclear from RE and feature databases. Enable this extension on halti5 GPUs as a conservative starting point. Support for 128 bit formats is missing, so it's gated behind ETNA_MESA_DEBUG=deqp. Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36622>	2025-09-03 22:07:26 +00:00
Christian Gmeiner	870252379c	etnaviv: nir: Add nir_texop_tg4 offset lowering ... Implement offset lowering by converting pixel offsets to normalized coordinate space and adjusting coordinates accordingly. Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36622>	2025-09-03 22:07:26 +00:00
Christian Gmeiner	aa91ece579	etnaviv: Enable texture_multisample for deqp testing ... This makes running GLES3.1 deqp easier. Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36622>	2025-09-03 22:07:26 +00:00
Mike Blumenkrantz	7984a16e27	tc: don't unset resolve resource in set_framebuffer_state ... this breaks the whole mechanism Fixes: 2eb45daa9c ("gallium: de-pointerize pipe_surface") Acked-by: Marek Olšák <marek.olsak@amd.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37166>	2025-09-03 19:42:52 +00:00
Qiang Yu	17647e8eeb	gallium/noop: add mesh shader callbacks ... Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36976>	2025-09-03 10:21:17 +08:00
Qiang Yu	6b52b371df	gallium/trace: dump mesh shader queries ... Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36976>	2025-09-03 10:21:17 +08:00
Qiang Yu	3bc4004ca9	gallium/ddebug: support mesh shader ... Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36976>	2025-09-03 10:21:17 +08:00
Qiang Yu	e1bd743c3a	gallium/u_blitter: save mesh shader ... Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36976>	2025-09-03 10:21:17 +08:00
Qiang Yu	5e35be338c	gallium: threaded context support mesh shader ... Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36976>	2025-09-03 10:21:17 +08:00
Qiang Yu	f03e49c609	mesa,gallium: handle mesh shader create and delete ... Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36976>	2025-09-03 10:21:17 +08:00
Qiang Yu	6459dacb8b	gallium: cso context support mesh shader ... Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36976>	2025-09-03 10:21:16 +08:00
sarbes	d0b134a71e	lima: clean up unused PP struct ... Reviewed-by: Erico Nunes <nunes.erico@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36541>	2025-09-02 18:44:36 +00:00
Pierre-Eric Pelloux-Prayer	9171461565	radeonsi/sqtt: retry a frame capture after reiszing the buffer ... Reviewed-by: Marek Olšák <marek.olsak@amd.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36930>	2025-09-02 17:46:47 +00:00
Pierre-Eric Pelloux-Prayer	cd5f77f6c2	radeonsi: sync harder on finish ... With the threaded submit code, there's a small window where the fence can be signalled but the buffer lists cleanup code from amdgpu_cs_submit_ib has not executed yet. This is harmless in general, because num_active_ioctls is an optimization. When the glFinish is called though, it's best to avoid this inconsistencies because it can create odd behavior as shown by the flakes being fixed by this commit. The test runs a query, uses glFinish and expect the results to be available. Since results availability depends on num_active_ioctls, it made the test flaky. Fix this by syncing when PIPE_FLUSH_HINT_FINISH is set. Reviewed-by: Marek Olšák <marek.olsak@amd.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36930>	2025-09-02 17:46:46 +00:00
Mike Blumenkrantz	b536e38607	kopper: unwrap screen before checking cpu flag ... this otherwise may access the trace screen and return garbage Fixes: 316bf3bd8a ("kopper, dri: remove trace_screen_unwrap") Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37140>	2025-09-02 17:01:38 +00:00
Mike Blumenkrantz	5cf9318ff2	zink: add some cml flakes ... Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 10:46:20 -04:00
Mike Blumenkrantz	4971b58c96	zink: don't increase db scale when resizing a db up to the current scale ... this otherwise triggers infinite db scaling cc: mesa-stable Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	a0e149ab1d	zink: reset batch descriptor states again before use on recycle ... this otherwise would lead to enlarging mid-batch again, which incurs unnecessary sync Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	335610c302	zink: don't init non-db batch stuff in db mode ... some minor memory savings Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	fbddc97b9e	zink: zero db offset on batch reset ... seems weird this hasn't been caught before cc: mesa-stable Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	a3630c865f	zink: once there are many outstanding submits, check for timeline updates ... this should ensure that there are never too many cmdbufs pending Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	86d0fc7b38	zink: stop deferring resource object unrefs ... now that batch state reset is threaded, this is unnecesary Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	cdb9bec202	zink: move buffer hashlist clear to normal batch state reset ... this is threaded now so it doesn't need to be in the submit thread Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	a184a4ca30	zink: rename zink_batch_state::unref_resources -> unref_resource_objs ... more accurate Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	3b39ecbdf8	zink: stop trying to oom prune batch states ... this should be regular enough now that such checks are no longer necessary Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	ffa5518c45	zink: stop using atomics to check fence submit/complete ... this isn't actually achieving anything since reading a "wrong" value here isn't harmful Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	d1f8cd54f4	zink: check ctx batch states first when finding a usable one ... this ensures the completed ones get queued for recycling faster Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	067873d4a4	zink: defer batch state resets more competently ... previously batch states would defer resource object destruction until "later", which was somewhat unreliable in the event that the batch state wasn't checked again for a long while instead, defer all "large" batch states and then reset them in the submit thread, which should yield more consistent results, including in scenarios with multiple contexts in use Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	6dd081ab1b	zink: null out zink_batch_state::next when reusing a batch state ... this is harmless at present Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	cec2b1327b	zink: split out batch state finding ... should be no functional changes Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:34 +00:00
Mike Blumenkrantz	6246590da1	zink: add a util function for appending a batch state ... these are a singly-linked list with a tail pointer Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37068>	2025-09-02 14:43:33 +00:00
Gert Wollny	c1611edc3a	r600/sfn: remove superfluous semicolon ... Signed-off-by: Gert Wollny <gert.wollny@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37123>	2025-09-02 13:50:31 +00:00
Gert Wollny	d41ac790d0	r600/sfn: Increase limit for lowering local arrays to scratch ... Signed-off-by: Gert Wollny <gert.wollny@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37123>	2025-09-02 13:50:31 +00:00
Gert Wollny	47c1880275	r600: remove hack to force a new CF if TEX grad is set ... The sfn scheduler takes care that we don't hit the slot limit for TEX instructions when doing this. Signed-off-by: Gert Wollny <gert.wollny@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37123>	2025-09-02 13:50:31 +00:00
Martin Roukala (né Peres)	4a42ea6785	nvk/ci: switch to default kernel to b2c's default kernel ... This brings linux v6.16.3 to all CI-tron jobs. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37085>	2025-09-02 07:48:53 +00:00
Martin Roukala (né Peres)	40a632cb18	radv/ci: switch to default kernel to b2c's default kernel ... This brings linux v6.16.3 to most jobs but the following ones: * navi21/31: still waiting for the zerovram bug fix * polaris10: 6.16 introduced an sdma-related hang that is being bisected Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37085>	2025-09-02 07:48:53 +00:00
Martin Roukala (né Peres)	20dfc94bf6	ci-tron: uprev b2c to v0.9.17 ... This brings also brings Linux v6.16.3 to r300 and etnaviv. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37085>	2025-09-02 07:48:53 +00:00
Rob Clark	c6aba9abfa	freedreno/a6xx: Mark tex and samp descriptors for dumping ... Fixes: c1aaef71da ("freedreno/a6xx: Make tex/samp descriptors a normal BO") Signed-off-by: Rob Clark <rob.clark@oss.qualcomm.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37129>	2025-09-02 01:07:31 +00:00
Rob Clark	39d8dcabf2	freedreno/a6xx: Disallow impossible image swizzles ... The image instructions can't do arbitrary swizzles, so L/LA/I formats would require shader lowering. Signed-off-by: Rob Clark <rob.clark@oss.qualcomm.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37129>	2025-09-02 01:07:31 +00:00
Eric Engestrom	a6bfb7e869	zink+lavapipe/ci: document recent flakes ... Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37127>	2025-09-01 20:46:42 +00:00
Eric Engestrom	9d73641b39	zink+radv/ci: document recent flakes ... Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37127>	2025-09-01 20:46:42 +00:00
Christian Gmeiner	ff008db792	etnaviv: blt: Add hardware based mipmap generation ... This provides a hardware-accelerated path for mipmap generation on supported formats. Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35991>	2025-09-01 19:10:34 +00:00
Christian Gmeiner	23b26c94f5	etnaviv: blt: Extend translate_blt_format(..) ... Add A8_UNORM, L8_UNORM, and L8A8_UNORM format translations to translate_blt_format(..) to enable BLT-based mipmap generation for these formats. This enables passing dEQP-GLES2/3 functional texture mipmap generation tests for these additional formats. Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35991>	2025-09-01 19:10:34 +00:00
David Rosca	2667db1114	radeonsi/vcn: Correctly set chroma location with EFC ... EFC supports horizontal left and vertical top/center. Reviewed-by: Ruijing Dong <ruijing.dong@amd.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36977>	2025-09-01 10:30:38 +00:00
Ashley Smith	b35bb9f5c7	panfrost: Enable shader_atomic_int64 for gallium ... Vulkan support was added previously, enable gallium support. Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Acked-by: Mary Guillemard <mary.guillemard@collabora.com> Signed-off-by: Ashley Smith <ashley.smith@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37051>	2025-09-01 08:22:10 +00:00
Karol Herbst	24c7e247af	rusticl: implement cl_khr_external_semaphore ... Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36007>	2025-08-31 13:12:17 +00:00