Commit Graph

  • c9e94f92a0 panvk: Don't allocate memory for a buffer descriptor in CreateBufferView() Boris Brezillon 2025-12-02 20:08:43 +01:00
  • b5e47ba894 pan/kmod: Add new helpers to sync BO CPU mappings Faith Ekstrand 2025-07-25 15:35:45 +00:00
  • af14c37bf1 pan/kmod: Enforce PAN_KMOD_BO_FLAG_NO_MMAP Boris Brezillon 2025-10-28 10:49:51 +01:00
  • 0f4f556229 pan/kmod: Expose the IO coherency property Boris Brezillon 2025-08-27 14:00:08 +02:00
  • cd8b8baf6e pan/kmod: Expose the BO flags supported by a pan_kmod_device Faith Ekstrand 2025-07-24 20:29:12 +00:00
  • 7e65322c93 pan/kmod: Add a panfrost_kmod_driver_version_at_least() helper Faith Ekstrand 2025-07-24 20:54:50 +00:00
  • ee172bb769 pan/kmod: Cache the device props at the pan_kmod_dev level Boris Brezillon 2025-08-27 16:20:11 +02:00
  • f43cff3728 util: Move STACK_ARRAY into util Faith Ekstrand 2025-07-29 03:10:41 -04:00
  • 61f7990b9a drm-uapi: Sync the panfrost header Faith Ekstrand 2025-07-29 06:29:20 +00:00
  • 8c8864baea drm-uapi: Sync the panthor header Faith Ekstrand 2025-07-24 17:53:45 +00:00
  • ef760d4c59 people: Add Christoph Pillmayer to the list Boris Brezillon 2025-12-12 10:00:44 +01:00
  • 0fe8250bf4 radv: optimize known front_face_fsign too Georg Lehmann 2025-12-11 14:03:34 +01:00
  • dd05c4a566 Uprev ANGLE to 2ed4b049c064add3109c7b1e0c954a0bce856df8 Collabora's Gfx CI Team 2025-12-12 00:08:50 +00:00
  • 9bd2c6dcb2 ac/nir: allow smaller workgroups for GS Marek Olšák 2025-11-08 12:25:50 -05:00
  • ff8df8712e gallium/frontends/dri: Don't force dri cursor buffers to be 64x64 stefan11111 2025-12-06 21:13:57 +02:00
  • 3f37740762 ac/llvm: workaround legacy fma intrinsic crash on gfx12 Qiang Yu 2025-12-10 16:57:17 +08:00
  • cb9d9b8a6e asahi: Set prefer_real_buffer_in_constbuf0 Rob Clark 2025-12-11 09:09:52 -08:00
  • 51605bfac2 gallium: Make upload_cb0 return a releasebuf Rob Clark 2025-12-10 14:44:20 -08:00
  • b2daaaec81 gallium/aux: Add debug option to force u_upload rollover Rob Clark 2025-12-10 16:46:25 -08:00
  • bdacab49c7 brw: Use LSC extended descriptor offsets for Xe2 URB messages Kenneth Graunke 2025-12-09 03:35:12 -08:00
  • 9482d392a1 brw: Fix outdated comments about urb->offset units Kenneth Graunke 2025-12-09 03:30:34 -08:00
  • bb06af3f9b panfrost/job: avoid shadowing variable name Alejandro Piñeiro 2025-12-09 18:10:20 +01:00
  • fecb9e0952 anv: switch shader heap placement to beginning of heap by default Lionel Landwerlin 2025-12-11 10:59:36 +02:00
  • 094f8f041f anv: enable fragmentShadingRateWithShaderSampleMask on Xe2+ Iván Briano 2025-11-24 13:31:15 -08:00
  • df15770785 anv: coarse_pixel doesn't require any InputCoverageMaskState Iván Briano 2025-11-24 13:29:51 -08:00
  • 978d4b2a99 anv: maxFragmentShadingRateCoverageSamples is 16 on all platforms Iván Briano 2025-11-24 13:24:28 -08:00
  • a7280ab590 nir: add nir_lower_single_sampled::lower_sample_mask_in option Iván Briano 2025-11-24 13:22:06 -08:00
  • ef31f07077 nir: clear SAMPLE_MASK_IN if we lowered it Iván Briano 2025-11-24 13:11:46 -08:00
  • 71b59563fe tu: Move tu_xs_config() to use the CRB builder. Emma Anholt 2025-12-08 11:28:07 -08:00
  • 780de476e7 tu: Split loading immediates for a program from the program config. Emma Anholt 2025-12-08 11:01:17 -08:00
  • 5f091af897 tu: Move a bunch of program config to CRB. Emma Anholt 2025-12-01 17:08:35 -08:00
  • c130c94bcb tu: move tu6_emit_msaa() to use CRB. Emma Anholt 2025-12-01 16:25:12 -08:00
  • f973ff307d tu: Move tu6_emit_window_offset() to use CRB. Emma Anholt 2025-12-01 16:23:51 -08:00
  • 5bfc950128 tu: Move tu6_emit_mrt() to use CRB. Emma Anholt 2025-08-13 15:08:08 -07:00
  • daa9c0e228 tu: Move VFD CRBs to the CRB builder. Emma Anholt 2025-12-08 10:16:54 -08:00
  • 02602cc3e4 tu: Move pipeline SO setup to the CRB builder. Emma Anholt 2025-12-08 10:08:10 -08:00
  • 1b93a4b778 tu: Add CRB builder. Emma Anholt 2025-08-13 14:34:58 -07:00
  • 2d6c15ad57 tu: remove magic bo reg packing (use iovas directly) Jonathan Marek 2020-09-17 11:31:33 -04:00
  • 67c6abb20b tu,freedreno: Drop the ".bo_write" flag. Emma Anholt 2025-08-13 12:33:37 -07:00
  • 0b3f0d1662 anv: Fix typo when checking if async rt scratch size changed Calder Young 2025-12-10 20:37:23 -08:00
  • b393518bdf treewide: Use wsi_common_is_swapchain_image() helper Christian Gmeiner 2025-12-08 23:54:25 +01:00
  • 9a89280d62 vulkan/wsi: Add wsi_common_is_swapchain_image() helper Christian Gmeiner 2025-12-08 22:33:52 +01:00
  • 840c80d80e pan/ci: Disable THP on panfrost-g52-piglit Boris Brezillon 2025-12-11 17:31:37 +01:00
  • 708d331a04 pvr: run clang-format Erik Faye-Lund 2025-12-08 11:18:44 +01:00
  • 306de6782c pvr: disable has_gs_rta_support for ge7800 as well Erik Faye-Lund 2025-12-11 15:33:54 +01:00
  • 39b8456a79 panvk: Drop compile_inputs from panvk_lower_nir() Faith Ekstrand 2025-11-06 11:07:35 -05:00
  • 4909e23d11 panvk: Pull multiview lowering out of panvk_lower_nir() Faith Ekstrand 2025-11-06 11:00:02 -05:00
  • 4f0386fbfb panvk: Restructure VS variant handling Faith Ekstrand 2025-11-06 11:38:55 -05:00
  • 0191e1f28a panvk: Make compile_inputs const in panvk_compile_nir() Faith Ekstrand 2025-11-06 11:35:18 -05:00
  • c161619a0c panvk: Only pass the panvk_shader_desc_info to panvk_lower_nir() Faith Ekstrand 2025-11-06 09:40:19 -05:00
  • 1c5ea90042 panvk: Store the varying attribute descriptor count in desc_info Faith Ekstrand 2025-11-05 16:52:19 -05:00
  • 153a35aa91 panvk: Move I/O lowering out of panvk_lower_nir() Faith Ekstrand 2025-11-07 11:46:30 -05:00
  • e694e2c352 panvk: Re-prefix panvk_shader_desc_info/map with lower_ Faith Ekstrand 2025-11-05 16:29:40 -05:00
  • 3747e3318a panvk: Call lower_input_attachment_loads() from compile_shader() Faith Ekstrand 2025-11-05 15:42:53 -05:00
  • cddfa4615b panvk: Break input attachment lowering into its own file Faith Ekstrand 2025-11-05 15:26:36 -05:00
  • 5ac0b11893 panvk: Stop using descriptor helpers in lower_input_attachments Faith Ekstrand 2025-11-05 15:16:06 -05:00
  • e518450661 panvk: Use nir_instr_clone() for input attachment loads Faith Ekstrand 2025-11-05 14:58:54 -05:00
  • c2e6133549 panvk: Add separate COMPUTE and FRAGMENT cases in compile_shader() Faith Ekstrand 2025-11-05 16:09:21 -05:00
  • 388a3cff5c panvk: Upload all variants at the end of compile_shader() Faith Ekstrand 2025-11-05 16:07:48 -05:00
  • aabdd0f5ca panvk: Move late lowering to panvk_compile_nir() Mary Guillemard 2025-11-05 14:46:39 -05:00
  • f4470dd7d7 panvk: Move FAU space info to panvk_compile_nir Mary Guillemard 2025-10-13 15:34:08 +02:00
  • 2bd282a968 pan/bi: Run nir_lower_all_phis_to_scalar() late Faith Ekstrand 2025-11-17 15:06:04 -05:00
  • 5f22dfa2e1 panvk: Run pan_preprocess_nir() in the preprocess step Faith Ekstrand 2025-12-03 17:45:39 -05:00
  • 7206572884 pan/bi: Move opt_sink and opt_move calls to postprocess Faith Ekstrand 2025-12-05 11:16:28 -05:00
  • 5841106027 pan/bi: Only delete function temp variables Faith Ekstrand 2025-12-05 11:15:00 -05:00
  • da969df092 ir3: Fix condition for using uniform predicates Connor Abbott 2025-12-10 17:16:43 -05:00
  • d0bae6c1b3 ir3: Fix early-preamble (sy) Rob Clark 2025-12-09 06:49:50 -08:00
  • 156ae6195e aco: print large p_parallelcopy using several lines Rhys Perry 2025-11-26 14:42:38 +00:00
  • 21414e0898 aco/ra: add first loop header phi operand to temp_to_phi_resources Rhys Perry 2025-11-26 15:43:11 +00:00
  • 43b3901362 aco/ra: copy vector_info to affinities Rhys Perry 2025-11-26 15:42:47 +00:00
  • 85e8f815e0 radv/nir: Use fmt_idx correctly Konstantin Seurer 2025-11-29 16:44:16 +01:00
  • c14eb415a2 radv/bvh: Avoid a slow case when compressing triangles Konstantin Seurer 2025-11-29 16:43:31 +01:00
  • 2749b5b713 radv/bvh: Fix calculating the vertex payload/prefix sizes Konstantin Seurer 2025-11-29 15:34:08 +01:00
  • 3a3810647e radv/bvh: Assert that indices_midpoint is valid Konstantin Seurer 2025-11-29 15:33:19 +01:00
  • 685d34a6d3 radv/rra/gfx12: Fix primitive/geometry index validation Konstantin Seurer 2025-11-15 09:33:22 +01:00
  • 07afe65beb radv: Use hw_leaf_node_count for computing BVH size Konstantin Seurer 2025-11-15 09:31:40 +01:00
  • 034f58c7e3 nir: Ignore ray query ranges that don't start with rq_initialize Konstantin Seurer 2025-12-08 15:34:38 +01:00
  • 5e03d09eb5 nir: Fix typo in nir_opt_ray_query_ranges Konstantin Seurer 2025-12-08 15:34:07 +01:00
  • be7f39a519 pvr: rename colliding symbol Erik Faye-Lund 2025-12-10 15:02:01 +01:00
  • 8fa6aabb33 pvr: break out pvr_free_list into a separate module Erik Faye-Lund 2025-11-27 16:03:22 +01:00
  • 17f99c870b pvr: replace constant-returning function with a macro Erik Faye-Lund 2025-11-14 14:26:43 +01:00
  • 745d576516 pvr: make queries arch-agnostic Erik Faye-Lund 2025-11-14 12:02:05 +01:00
  • 70270348e8 pvr: fixup some includes Erik Faye-Lund 2025-11-24 11:00:19 +01:00
  • 5dd76663a7 pvr: add helpers to query limits based on device-info Erik Faye-Lund 2025-12-05 15:39:05 +01:00
  • 0d38d22c02 pvr: move PVR_TEX_FORMAT_COUNT to pvr_limits.h Ashish Chauhan 2025-11-28 11:03:28 +00:00
  • ed83ee7659 pvr: do not store compressed pbe-formats Erik Faye-Lund 2025-12-04 12:18:17 +01:00
  • f716599d5f pvr: use strongly-typed enum instead of uint32_t Erik Faye-Lund 2025-12-04 15:29:30 +01:00
  • 298c14bbc4 pvr: rework format binding flags Erik Faye-Lund 2025-12-04 15:15:23 +01:00
  • 283c6bb19e pvr: split out pbe-details from main format-table Erik Faye-Lund 2025-12-04 12:08:02 +01:00
  • c1da96affe pvr: move border-specific format-code into pvr_border.c Erik Faye-Lund 2025-12-04 16:23:30 +01:00
  • cd6c4bf772 pvr: remove unused member Erik Faye-Lund 2025-12-04 14:22:21 +01:00
  • c11f5197f5 pvr: do not store VkFormat in pvr_format Erik Faye-Lund 2025-12-04 11:00:28 +01:00
  • de5d08eb72 pvr: move include to source-file Erik Faye-Lund 2025-12-04 13:03:51 +01:00
  • c00ae68585 egl/x11: Fix memory leak when querying translated coord. Hans-Kristian Arntzen 2025-12-05 16:29:30 +01:00
  • 1a2d0d3f31 nir: Optimistically unroll loops using induction var as a sample id. Emma Anholt 2025-11-21 16:52:40 -08:00
  • 10ba7675c8 nir/uub: Use an optional max_samples from drivers for sample counts. Emma Anholt 2025-11-24 16:28:44 -08:00
  • dc30e1a128 nir/loop_analyze: Use nir_unsigned_upper_bound for loop trip limits. Emma Anholt 2025-11-24 16:16:45 -08:00
  • 564bf2379e pvr: enable VK_KHR_dynamic_rendering Luigi Santivetti 2025-12-04 01:02:18 +00:00
  • 146364ab9f pvr: add support for VK_KHR_dynamic_rendering Luigi Santivetti 2025-11-13 10:54:49 +00:00
  • 5b752ccab0 pvr: move code for resolving attachments Luigi Santivetti 2025-12-08 18:34:17 +00:00