ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-06-29 11:16:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • 1ae4925302 [CK_TILE] Separate PermuteN epilogue from CShuffle epilogue into standalone file (#5863) msaffari-amd 2026-04-14 22:22:18 +02:00
  • 5d2fce819d [rocm-libraries] ROCm/rocm-libraries#4769 (commit 72ae66e) arai713 2026-04-14 17:51:20 +00:00
  • 5f2517da31 [CK_TILE] Restructure Tile Engine's benchmarking and profiling (#4769) arai713 2026-04-14 10:50:24 -07:00
  • 12f3d646a0 [rocm-libraries] ROCm/rocm-libraries#4769 (commit 72ae66e) arai713 2026-04-14 10:50:24 -07:00
  • c810a01ec6 [CK_TILE] Restructure Tile Engine's benchmarking and profiling (#4769) arai713 2026-04-14 10:50:24 -07:00
  • f6eb5f0a6a [rocm-libraries] ROCm/rocm-libraries#6379 (commit b38b056) Estevan Vedovelli 2026-04-14 16:15:17 +00:00
  • 59ab9fdc3f [ck] Clamp negative kernel execution elapsed time to zero (#6379) Estevan Vedovelli 2026-04-14 12:14:26 -04:00
  • 0121f39b1f [rocm-libraries] ROCm/rocm-libraries#6379 (commit b38b056) Estevan Vedovelli 2026-04-14 12:14:26 -04:00
  • 43b33b9034 [ck] Clamp negative kernel execution elapsed time to zero (#6379) Estevan Vedovelli 2026-04-14 12:14:26 -04:00
  • c7e6e4f616 fix extra host side operations. Gino Lu 2026-04-14 10:11:00 -04:00
  • 14f7834a23 [rocm-libraries] ROCm/rocm-libraries#6342 (commit 31bcb51) Po Yen Chen 2026-04-14 14:08:44 +00:00
  • 5cab2c3565 [CK] Skip fp16 dropout d256 batch tests for compiler VGPR aliasing bug (#6342) Po Yen Chen 2026-04-14 22:07:20 +08:00
  • 0d53e3674b [rocm-libraries] ROCm/rocm-libraries#6342 (commit 31bcb51) Po Yen Chen 2026-04-14 22:07:20 +08:00
  • 470a48530b [CK] Skip fp16 dropout d256 batch tests for compiler VGPR aliasing bug (#6342) Po Yen Chen 2026-04-14 22:07:20 +08:00
  • 9491563725 [rocm-libraries] ROCm/rocm-libraries#6399 (commit 13bf528) Yaswanth Raparti 2026-04-14 07:45:14 +00:00
  • f603d75f20 [CK][CK TILE] Modify elementwise kernel template signature to accept independent type arguments (#6399) Yaswanth Raparti 2026-04-14 00:44:27 -07:00
  • b21f31c65c [rocm-libraries] ROCm/rocm-libraries#6399 (commit 13bf528) Yaswanth Raparti 2026-04-14 00:44:27 -07:00
  • d988d55275 [CK][CK TILE] Modify elementwise kernel template signature to accept independent type arguments (#6399) Yaswanth Raparti 2026-04-14 00:44:27 -07:00
  • e4bdc5d91a [CK Tile] Unification work - mma transformations pipeline (#5508) chris-tsiaousis-hpc 2026-04-14 09:25:01 +02:00
  • 1c04029c0e [rocm-libraries] ROCm/rocm-libraries#5508 (commit 0ad0aca) chris-tsiaousis-hpc 2026-04-14 09:25:01 +02:00
  • 89c5e67028 [CK Tile] Unification work - mma transformations pipeline (#5508) chris-tsiaousis-hpc 2026-04-14 09:25:01 +02:00
  • 918e8a1bd8 [rocm-libraries] ROCm/rocm-libraries#6400 (commit c0b3c95) Brock Hargreaves 2026-04-14 02:46:55 +00:00
  • e939299c69 [MIOPEN] [CK] Revert "[CK] Disable test cases affected by compiler codegen bugs on gfx90a" (#6400) Brock Hargreaves 2026-04-13 20:46:07 -06:00
  • a4e36c1b89 [rocm-libraries] ROCm/rocm-libraries#6400 (commit c0b3c95) Brock Hargreaves 2026-04-13 20:46:07 -06:00
  • 5eee93e67c [MIOPEN] [CK] Revert "[CK] Disable test cases affected by compiler codegen bugs on gfx90a" (#6400) Brock Hargreaves 2026-04-13 20:46:07 -06:00
  • a92fd0db0c [rocm-libraries] ROCm/rocm-libraries#6343 (commit 3604475) Ville Pietilä 2026-04-13 11:41:29 +00:00
  • 1b2a619107 [CK] Disable compilation of problematic bwd weight conv instances for gfx90a (#6343) Ville Pietilä 2026-04-13 14:40:27 +03:00
  • 0f2279920b [rocm-libraries] ROCm/rocm-libraries#6343 (commit 3604475) Ville Pietilä 2026-04-13 14:40:27 +03:00
  • 6e0454216d [CK] Disable compilation of problematic bwd weight conv instances for gfx90a (#6343) Ville Pietilä 2026-04-13 14:40:27 +03:00
  • 9fe98c864f [CK Tile] Add Tile Distribution Encoding Calculator (#5515) Kiefer van Teutem 2026-04-13 10:00:31 +02:00
  • b7156a8bbf [rocm-libraries] ROCm/rocm-libraries#5515 (commit 40f66c3) Kiefer van Teutem 2026-04-13 10:00:31 +02:00
  • 6cd016dde4 [CK Tile] Add Tile Distribution Encoding Calculator (#5515) Kiefer van Teutem 2026-04-13 10:00:31 +02:00
  • d1d457b82a Add sparge gpu pipeline in tile_example_sparge_vsa_sparse_attn Gino Lu 2026-04-13 03:34:08 -04:00
  • 643ad35de2 Merge remote-tracking branch 'origin/develop' into ginolu/sparge_attention Gino Lu 2026-04-13 03:27:43 -04:00
  • fa4473fde6 [rocm-libraries] ROCm/rocm-libraries#6323 (commit a668483) Aviral Goel 2026-04-11 10:01:30 +00:00
  • bfe574a430 CK: Extract shared boilerplate from 47 gemm_quant test files (#6323) Aviral Goel 2026-04-11 06:00:26 -04:00
  • 81a5826132 [rocm-libraries] ROCm/rocm-libraries#6323 (commit a668483) Aviral Goel 2026-04-11 06:00:26 -04:00
  • 160bc1363e CK: Extract shared boilerplate from 47 gemm_quant test files (#6323) Aviral Goel 2026-04-11 06:00:26 -04:00
  • ce099b7afd [rocm-libraries] ROCm/rocm-libraries#6303 (commit 784c268) Aviral Goel 2026-04-10 15:23:47 +00:00
  • b0e4472cf2 CK: Remove 4 orphaned files with verified replacements (~1,025 lines) (#6303) Aviral Goel 2026-04-10 11:22:31 -04:00
  • c2663ce9fd [rocm-libraries] ROCm/rocm-libraries#6303 (commit 784c268) Aviral Goel 2026-04-10 11:22:31 -04:00
  • 818704375c CK: Remove 4 orphaned files with verified replacements (~1,025 lines) (#6303) Aviral Goel 2026-04-10 11:22:31 -04:00
  • e0dfe58d66 [rocm-libraries] ROCm/rocm-libraries#6302 (commit 8d419e8) Aviral Goel 2026-04-10 15:18:02 +00:00
  • 2ff7ac5abc CK: Remove 41 commented-out dead code blocks (~200 lines) (#6302) Aviral Goel 2026-04-10 11:17:11 -04:00
  • c7eb33078c [rocm-libraries] ROCm/rocm-libraries#6302 (commit 8d419e8) Aviral Goel 2026-04-10 11:17:11 -04:00
  • 4ccbcbe0a4 CK: Remove 41 commented-out dead code blocks (~200 lines) (#6302) Aviral Goel 2026-04-10 11:17:11 -04:00
  • 4d0bbe5d17 [rocm-libraries] ROCm/rocm-libraries#5329 (commit 9c43062) Yi DING 2026-04-10 01:23:54 +00:00
  • 6cdc5bc3e2 [CK] Add flash_attn tests (#5329) Yi DING 2026-04-10 09:23:10 +08:00
  • 92dc99b713 [rocm-libraries] ROCm/rocm-libraries#5329 (commit 9c43062) Yi DING 2026-04-10 09:23:10 +08:00
  • 914f4e47a2 [CK] Add flash_attn tests (#5329) Yi DING 2026-04-10 09:23:10 +08:00
  • 59f8535bf9 [rocm-libraries] ROCm/rocm-libraries#6326 (commit c1b6c3e) alexxu-amd 2026-04-09 20:30:41 +00:00
  • f9737ad553 Correct .readthedocs.yml file path (#6326) alexxu-amd 2026-04-09 16:26:42 -04:00
  • 10028b783d [rocm-libraries] ROCm/rocm-libraries#6326 (commit c1b6c3e) alexxu-amd 2026-04-09 16:26:42 -04:00
  • e59e6a738a Correct .readthedocs.yml file path (#6326) alexxu-amd 2026-04-09 16:26:42 -04:00
  • 920acd2c12 [rocm-libraries] ROCm/rocm-libraries#5168 (commit 8b5afcb) Vidyasagar Ananthan 2026-04-09 17:39:35 +00:00
  • a2b844d335 [CK] [CK_Tile] Add GroupConv to Kernel Dispatcher (#5168) Vidyasagar Ananthan 2026-04-09 10:38:33 -07:00
  • ca28efac88 [rocm-libraries] ROCm/rocm-libraries#5168 (commit 8b5afcb) Vidyasagar Ananthan 2026-04-09 10:38:33 -07:00
  • 40290297cd [CK] [CK_Tile] Add GroupConv to Kernel Dispatcher (#5168) Vidyasagar Ananthan 2026-04-09 10:38:33 -07:00
  • 7182f6d139 Add missing gfx1033 to gfx103 group definition pytorch/release/2.11 Harkirat Gill 2026-04-03 19:45:41 +00:00
  • 1a9404ac96 [CK_TILE] Use Persistent Scheduling for FMHA BWD Group Deterministic Ding, Yi 2026-04-08 03:38:17 -05:00
  • 966302f00c Update CMakeLists.txt develop-test J PS 2026-04-08 11:56:25 -07:00
  • 0c9b6fe255 Update CMakeLists.txt J PS 2026-04-08 11:45:33 -07:00
  • d6e154a48c Update CMakeLists.txt J PS 2026-04-08 11:41:05 -07:00
  • 984ba36750 Update CMakeLists.txt J PS 2026-04-08 11:36:42 -07:00
  • c944c07230 Update CMakeLists.txt J PS 2026-04-08 11:30:54 -07:00
  • 711fb8d93d Update CMakeLists.txt J PS 2026-04-08 11:27:35 -07:00
  • 4c0e73ab12 [rocm-libraries] ROCm/rocm-libraries#6156 (commit 367565a) Hosang Yoon 2026-04-08 14:53:18 +00:00
  • fb22cd0c69 [CK_TILE] Optimize FMHA head-dim padded path on gfx11/gfx12 (#6156) Hosang Yoon 2026-04-08 10:51:53 -04:00
  • 00faecdfcf [rocm-libraries] ROCm/rocm-libraries#6156 (commit 367565a) Hosang Yoon 2026-04-08 10:51:53 -04:00
  • 65ad35becd [CK_TILE] Optimize FMHA head-dim padded path on gfx11/gfx12 (#6156) Hosang Yoon 2026-04-08 10:51:53 -04:00
  • 7d6c8e5afa [rocm-libraries] ROCm/rocm-libraries#6215 (commit bb1f765) Yaswanth Raparti 2026-04-08 09:55:56 +00:00
  • 2ed0791cf3 [CK] [CK Tile] Improved ci_safety_check in smart-build infrastructure (#6215) Yaswanth Raparti 2026-04-08 02:54:56 -07:00
  • 17fda8b176 [rocm-libraries] ROCm/rocm-libraries#6215 (commit bb1f765) Yaswanth Raparti 2026-04-08 02:54:56 -07:00
  • c953982434 [CK] [CK Tile] Improved ci_safety_check in smart-build infrastructure (#6215) Yaswanth Raparti 2026-04-08 02:54:56 -07:00
  • ab682f6b1c [CK_TILE] Refine FMHA Readme (#6003) Copilot 2026-04-08 14:58:53 +08:00
  • 31fd3fe9c1 [rocm-libraries] ROCm/rocm-libraries#6003 (commit b4c8c52) Copilot 2026-04-08 14:58:53 +08:00
  • ceddfcc13c [CK_TILE] Refine FMHA Readme (#6003) Copilot 2026-04-08 14:58:53 +08:00
  • bedd60a568 Merge remote-tracking branch 'origin/develop' into users/yiding12/fmha-bwd-workspace Ding, Yi 2026-04-07 23:13:09 -05:00
  • 5a63b343d0 Fix Ding, Yi 2026-04-07 23:12:39 -05:00
  • 5e1846914a Add missing gfx1033 to gfx103 group definition pytorch/release/2.10 Harkirat Gill 2026-04-03 19:45:41 +00:00
  • cdf6573171 Add missing gfx1033 to gfx103 group definition pytorch/release/2.9 Harkirat Gill 2026-04-03 19:45:41 +00:00
  • fa0d03c339 Add missing gfx1033 to gfx103 group definition Harkirat Gill 2026-04-03 19:45:41 +00:00
  • a170e2bd9d [rocm-libraries] ROCm/rocm-libraries#5939 (commit 6fb1791) Christopher Millette 2026-04-07 14:38:07 +00:00
  • 7816812ef8 [CK_TILE] Flatten nested static_for loops into static_ford (#5939) Christopher Millette 2026-04-07 08:36:45 -06:00
  • c6e5cd65d3 [rocm-libraries] ROCm/rocm-libraries#5939 (commit 6fb1791) Christopher Millette 2026-04-07 08:36:45 -06:00
  • 870112b861 [CK_TILE] Flatten nested static_for loops into static_ford (#5939) Christopher Millette 2026-04-07 08:36:45 -06:00
  • c2ac7aa7b0 [rocm-libraries] ROCm/rocm-libraries#6051 (commit f0838b2) Po Yen Chen 2026-04-07 14:20:43 +00:00
  • 341fb33386 [CK] Add FP8 per-tensor quantization support for FMHA V3 pipeline (#6051) Po Yen Chen 2026-04-07 22:19:28 +08:00
  • bf736dfa74 [rocm-libraries] ROCm/rocm-libraries#6051 (commit f0838b2) Po Yen Chen 2026-04-07 22:19:28 +08:00
  • 6dc44114ba [CK] Add FP8 per-tensor quantization support for FMHA V3 pipeline (#6051) Po Yen Chen 2026-04-07 22:19:28 +08:00
  • 020b6f435e [rocm-libraries] ROCm/rocm-libraries#6201 (commit 5c0697e) Jeff Huang 2026-04-07 12:42:08 +00:00
  • 8a29683326 [CK_TILLE] Temporarily remove batch prefill KV cache overflow asserts (#6201) Jeff Huang 2026-04-07 20:41:24 +08:00
  • 8287fe6c19 [rocm-libraries] ROCm/rocm-libraries#6201 (commit 5c0697e) Jeff Huang 2026-04-07 20:41:24 +08:00
  • 449844e3d3 [CK_TILLE] Temporarily remove batch prefill KV cache overflow asserts (#6201) Jeff Huang 2026-04-07 20:41:24 +08:00
  • 3848d2411a Merge origin/develop into users/yiding12/fmha-bwd-workspace Ding, Yi 2026-04-07 05:28:49 -05:00
  • a95f64601d Remove dropout=true instances to reduce compiling-time Qianfeng Zhang 2026-04-07 09:38:18 +00:00
  • 348c3e05be Rename default_policy to policy for hstu_attention forward Qianfeng Zhang 2026-04-07 08:41:14 +00:00
  • a9b3eaffb8 [CK][CK Tile] Conv Bwd Data flush cache and profiling improvements (#6090) Bartłomiej Kocot 2026-04-04 02:22:22 +02:00
  • dbdf0a6eca [rocm-libraries] ROCm/rocm-libraries#6090 (commit bd5709e) Bartłomiej Kocot 2026-04-04 02:22:22 +02:00
  • 4112e08d0c [CK][CK Tile] Conv Bwd Data flush cache and profiling improvements (#6090) Bartłomiej Kocot 2026-04-04 02:22:22 +02:00