-
626c6b35f0
nak: add Movm
Karol Herbst
2025-11-12 00:00:30 +01:00
-
c4f07f3d79
nir: mark cmat_load_shared_nv as CAN_ELIMINATE
Karol Herbst
2025-11-26 15:03:46 +01:00
-
6e89dc33fe
nak/cmat: add optimisation to cmat load/store to do 32-bit load for f16vec2
Karol Herbst
2025-10-22 12:54:12 +02:00
-
79b3386810
nak/cmat: add alignment info to matrix load/stores
Karol Herbst
2025-11-04 12:37:03 +01:00
-
a643681dd5
broadcom/compiler: use nir_opt_uub
Iago Toral Quiroga
2025-11-13 09:48:01 +01:00
-
4234e7eed0
broadcom/compiler: enable umul24 and imul24 ALU opcodes
Jose Maria Casanova Crespo
2025-11-17 21:28:12 +01:00
-
29dff2fd75
radv: Check RADV_PERFTEST=sparse for image formats and sparse queue
Timur Kristóf
2025-11-25 14:59:58 +01:00
-
1884b03a10
radeonsi/vcn: Fix maybe uninitialized warning
David Rosca
2025-11-20 11:09:31 +01:00
-
e119091e7d
panfrost/ci: remove some out-of-date xfails
Erik Faye-Lund
2025-11-26 12:34:28 +01:00
-
f14e0d9f09
radv: add radv_hide_rebar_on_dgpu and enable for Red Dead Redemption 2
Samuel Pitoiset
2025-11-24 19:03:18 +01:00
-
9cca79d8f8
radv: fix resetting descriptor pool since the new descriptor sets allocator
Samuel Pitoiset
2025-11-25 09:59:17 +01:00
-
54077d9893
pvr: fix cleaning up failed CreateDevice
Icenowy Zheng
2025-11-18 18:49:06 +08:00
-
ab9e148bfb
pvr: enable samplerMirrorClampToEdge feature
Icenowy Zheng
2025-11-25 17:39:59 +08:00
-
1574a71438
nir/lower_wrmasks: clean up & deprecate pass
Alyssa Rosenzweig
2025-11-19 10:41:58 -05:00
-
2c2dd835af
nir/lower_wrmasks: drop callback
Alyssa Rosenzweig
2025-11-19 10:39:50 -05:00
-
5515160b55
nir/lower_wrmasks: drop support for I/O
Alyssa Rosenzweig
2025-11-19 10:36:10 -05:00
-
bdaff0b457
kk: Handle memory coherency for textures and buffers
Aitor Camacho
2025-11-23 15:37:54 +09:00
-
3aa4be52f3
poly: Make all heap allocations atomic
Faith Ekstrand
2025-11-24 12:02:48 -05:00
-
ed0998ca98
poly: Generalize unroll_restart() to arbitrary workgroup/subgroup sizes
Faith Ekstrand
2025-11-14 16:46:24 -05:00
-
d9f795e6d0
poly,asahi: Pull restart unrolling into libpoly
Faith Ekstrand
2025-11-14 14:10:23 -05:00
-
ddff3700a4
poly/asahi: Pull a bunch of vertex_id_for helpers into poly/prim.h
Faith Ekstrand
2025-11-14 16:19:48 -05:00
-
27b2290abe
poly: Move vs_grid to poly_vertex_params
Faith Ekstrand
2025-11-17 12:37:22 -05:00
-
70a1a4013f
agx: Use the new poly param helpers
Faith Ekstrand
2025-11-17 14:19:37 -05:00
-
5b677db712
hk: Use the new poly param helpers
Faith Ekstrand
2025-11-17 14:00:14 -05:00
-
96421876d7
poly: Add helpers for filling out poly_vertex_params
Faith Ekstrand
2025-11-17 13:08:59 -05:00
-
fd844e0ad4
poly: Add helpers for filling out poly_geometry_params
Faith Ekstrand
2025-11-17 12:16:40 -05:00
-
ad30ebccb6
poly,asahi: Put the indirect draw directly in the geometry params
Faith Ekstrand
2025-11-12 10:15:24 -05:00
-
fcb107accb
poly: Fetch the index size from a sysval
Faith Ekstrand
2025-11-07 15:31:24 -05:00
-
05aaa7df65
nir: Improve comments for a couple poly intrinsics
Faith Ekstrand
2025-11-13 16:38:23 -05:00
-
735b4ec9c7
poly: Add a poly_nir_lower_sysvals() pass
Faith Ekstrand
2025-11-13 22:15:28 -05:00
-
349219df6a
poly: Rename poly_nir_lower_gs.h to poly_nir.h
Faith Ekstrand
2025-11-13 22:05:17 -05:00
-
9ccac4e047
asahi: Reorder state uploads in agx_draw_patches()
Faith Ekstrand
2025-11-24 11:13:41 -05:00
-
5f5054a195
SQUASH: poly,asahi: Move the output mask to poly_vertex_state
Faith Ekstrand
2025-11-13 21:41:19 -05:00
-
05723bfa35
poly,asahi: Fetch directly from poly_vertex_state::output_buffer in GS
Faith Ekstrand
2025-11-13 18:24:04 -05:00
-
89fbb9cf84
poly,asahi: Move vertex_output_buffer to poly_vertex_param
Faith Ekstrand
2025-11-13 17:55:00 -05:00
-
8950efc006
hk: Expose the vertex param buffer to other stages
Faith Ekstrand
2025-11-18 17:49:18 -05:00
-
a96fd8f0a7
asahi: Upload vertex and geom/tess params together
Faith Ekstrand
2025-11-13 17:21:06 -05:00
-
f36465d574
poly,asahi: Rename poly_ia_state to poly_vertex_params
Faith Ekstrand
2025-11-13 16:11:20 -05:00
-
4716858a51
poly,asahi: Rename poly_tess_args to poly_tess_params
Faith Ekstrand
2025-11-13 15:30:26 -05:00
-
3182deaae1
brw: Combine output stores for TCS outputs even when unlinked
Kenneth Graunke
2025-11-16 14:38:50 -08:00
-
7e02738b63
brw: Drop check for legacy tess levels from remap_patch_urb_offsets
Kenneth Graunke
2025-11-06 00:45:15 -08:00
-
d95a9714c2
brw: Rewrite legacy tess level remapping
Kenneth Graunke
2025-10-27 15:45:45 -07:00
-
ee407481c2
brw: Switch to URB intrinsics for TCS inputs
Kenneth Graunke
2025-10-27 22:14:41 -07:00
-
943b2acf02
brw: Switch to NIR URB intrinsics for TES inputs
Kenneth Graunke
2025-11-15 01:59:41 -08:00
-
c0d69b2faf
brw: Switch to NIR URB intrinsics for TCS outputs
Kenneth Graunke
2025-10-27 01:23:08 -07:00
-
9aff3cac3c
brw: Add infrastructure for lowering to URB intrinsics
Kenneth Graunke
2025-10-16 13:50:02 -07:00
-
13acc889af
brw: Use io_sem.location instead of base to get varying slots
Kenneth Graunke
2025-11-17 13:06:11 -08:00
-
96d331766a
brw: Generalize read_attribute_payload_intel to handle more cases
Kenneth Graunke
2025-10-24 15:34:54 -07:00
-
792762617a
brw: Rename read_attribute_payload_intel to load_attribute_payload_intel
Kenneth Graunke
2025-10-27 00:54:55 -07:00
-
0f7590af81
brw, anv, iris: Switch to reversed patch header layouts
Kenneth Graunke
2025-10-22 14:54:15 -07:00
-
7d1dfc3468
brw: Lower tesslevel vars to vectors even for unlinked TCS/TES
Kenneth Graunke
2025-11-13 13:05:39 -08:00
-
7736e693b1
brw: Pass devinfo into remap_patch_urb_offsets
Kenneth Graunke
2025-11-06 00:39:12 -08:00
-
4dc6413de8
brw: Rename remap_non_header_patch_values to remap_patch_values
Kenneth Graunke
2025-11-05 22:18:18 -08:00
-
2b51963b8c
brw: Remap tesslevels before other patch remapping
Kenneth Graunke
2025-10-28 13:44:01 -07:00
-
e8669a8333
brw: Rework the tess level remapping interface
Kenneth Graunke
2025-11-05 22:14:14 -08:00
-
1995c879a9
brw: Flip the TESS_LEVEL_INNER/OUTER vue map slot assignments
Kenneth Graunke
2025-10-22 14:15:45 -07:00
-
e5c1d00faf
brw: Pass devinfo to brw_nir_lower_tes_inputs
Kenneth Graunke
2025-11-03 00:41:30 -08:00
-
a1c7ae9d15
brw: Implement URB handle intrinsics for TCS and TES stages
Kenneth Graunke
2025-10-27 01:25:25 -07:00
-
e290f9641d
brw: Implement load/store URB intrinsics
Lionel Landwerlin
2025-10-24 15:19:00 -07:00
-
0d8ee4ed23
brw: use default builder for urb handle adjustment
Lionel Landwerlin
2024-09-11 14:08:21 +03:00
-
f1ab64ad74
nir: add new intrinsics to load/store from URB on intel
Kenneth Graunke
2025-11-06 14:49:58 -08:00
-
8916b8a7f4
vulkan/wsi: Handle 0xFFFFFFFF special case in vk_wsi_force_swapchain_to_current_extent driconf
Joshua Ashton
2024-09-11 21:19:27 +01:00
-
5ee9a76058
nir: fix a typo in NIR_PASS_ASSERT_NO_PROGRESS for non-debug builds
Marek Olšák
2025-11-23 16:31:58 -05:00
-
d9d3f6703c
ac,winsys/amdgpu: report why ac_query_gpu_info failed
Marek Olšák
2025-11-21 12:34:13 -05:00
-
1c3e7e4ca0
ac: document RELEASE_MEM limitation with PS_DONE/CS_DONE on gfx6-11
Marek Olšák
2025-11-21 17:41:01 -05:00
-
406e6e094a
anv/rt: avoid out of bound access by clamping global id
Felix DeGrood
2025-11-24 19:16:21 +00:00
-
b1e74a1bb1
anv: shrink image opaque data
Lionel Landwerlin
2025-11-24 18:29:49 +02:00
-
6aabc3d5d2
ac/parse_ib: Implement VCN dec message parsing
Benjamin Cheng
2025-11-19 10:18:37 -05:00
-
28060c1e05
nvk/ci: add broken coop matrix CTS tests to skips
Karol Herbst
2025-11-25 13:57:56 +01:00
-
580ce1c911
tu: Handle mismatch in mip layouts for reinterpreted compressed images
Danylo Piliaiev
2025-08-26 19:32:35 +02:00
-
e347f82aeb
freedreno/layout: Use blocks for linear mipmap fallback where possible
Danylo Piliaiev
2025-08-19 15:20:06 +02:00
-
654b0dd548
freedreno: Remove use of FDL_MIN_UBWC_WIDTH
Rob Clark
2025-11-06 15:43:21 -08:00
-
c6ad0f781a
kk: Force attachment load as temp solution to preserve attachment
Aitor Camacho
2025-11-24 16:51:02 +09:00
-
fa420a8649
kk: Remove mem leaks in cmd buf destroy and residency set creation
Aitor Camacho
2025-11-25 03:31:26 +09:00
-
4f9d7c3385
gfxstream: codegen: don't generate custom protocols in function table
Gurchetan Singh
2025-11-24 10:37:02 -08:00
-
169f571f4f
gfxstream: delete createImmutableSamplersFilteredImageInfo
Gurchetan Singh
2025-11-24 09:49:30 -08:00
-
b6df034363
gfxstream: make functions static when needed
Gurchetan Singh
2025-11-24 09:41:42 -08:00
-
0a231dfb40
gfxstream: silence non-null Clang check on Android
Gurchetan Singh
2025-11-24 09:32:27 -08:00
-
7e72d392d7
brw: switch to load_(pixel_coord|frag_coord_z|frag_coord_w) intrinsics
Lionel Landwerlin
2025-07-25 10:01:42 +03:00
-
b7f011e653
radv/rt: Correctly copy culling flags when updating to separate AS
Natalie Vock
2025-11-17 16:41:58 +01:00
-
bc1eea90b9
radv/rt: Keep updated nodes always active
Natalie Vock
2025-11-17 16:41:19 +01:00
-
6d3be477ab
anv: enable application shader printfs with debug option
Lionel Landwerlin
2025-11-24 11:36:42 +02:00
-
4c3bf04dd0
anv: enable mesh/task shader hashes
Lionel Landwerlin
2025-11-24 11:36:08 +02:00
-
4b9aa9dc91
nir/lower_printf: fix missing singleton add
Lionel Landwerlin
2025-11-24 12:30:50 +02:00
-
d24633023f
nir/lower_printf: fix array alignment
Lionel Landwerlin
2025-11-24 12:30:22 +02:00
-
67faf6dfbd
spirv: fix printf generation
Lionel Landwerlin
2025-11-24 11:37:40 +02:00
-
6940b8fcd7
nir: fix lower_printf with no arguments
Lionel Landwerlin
2025-11-24 11:37:24 +02:00
-
c2c815afd9
nir: print out number of printfs
Lionel Landwerlin
2025-11-24 11:37:03 +02:00
-
b4e7c13ef4
meson: add libva wrap and fallback option
Thong Thai
2025-11-05 20:36:08 +00:00
-
f5eb3fe9cb
aco/optimizer: optimze cndmask(a, b, not(c)) to cndmask(b, a, c)
Georg Lehmann
2025-03-21 15:40:59 +01:00
-
752f1fb4ae
aco/optimizer: extend existing patterns to handle b2f/b2i(not(a))
Georg Lehmann
2025-03-21 16:12:51 +01:00
-
c538f47f03
aco/optimizer: create ff0/bcnt0
Georg Lehmann
2025-03-08 11:35:19 +01:00
-
0f7a1ce23e
aco/optimizer: some more mul opts
Georg Lehmann
2025-02-10 17:13:08 +01:00
-
92dbf42379
aco/optimizer: use cndmask for neg(b2i)
Georg Lehmann
2025-03-01 11:26:25 +01:00
-
0e4d4aeef7
aco/optimizer: add some bitop combining
Georg Lehmann
2025-03-01 11:26:02 +01:00
-
ee0354e0f1
aco/optimizer: use new helpers for bitwise n2 opts
Georg Lehmann
2024-12-16 12:27:10 +01:00
-
758fe79ad5
aco/optimizer: use new helpers for v_sub opts
Georg Lehmann
2024-12-16 12:22:23 +01:00
-
e42be7536c
aco/optimizer: use new helpers for remaining add opts
Georg Lehmann
2024-12-15 16:55:58 +01:00
-
adc55b1a1e
aco/optimizer: use new helpers for v_and opt
Georg Lehmann
2025-03-27 15:51:41 +01:00
-
7bc6d8e2ad
aco/optimizer: add more v_add_lshl_u32 opts
Georg Lehmann
2024-12-14 19:57:31 +01:00