Commit Graph

  • 590d95a7a9 Working Code after Restructure ThruptiRajLakshmanaGowda 2025-11-26 02:29:34 +00:00
  • cbc8959073 Partial Progress : Working GEMM Preshuffle ThruptiRajLakshmanaGowda 2025-11-26 02:25:40 +00:00
  • c0adc147a3 [CK_TILE] Fix Compilation of Flatmm Examples (#3285) Yi DING 2025-11-26 10:11:43 +08:00
  • e303358608 [CK_TILE] Fix Compilation of Flatmm Examples (#3285) Yi DING 2025-11-26 10:11:43 +08:00
  • c7dce2ac29 [CK_TILE] Fix Compilation of Flatmm Examples (#3285) Yi DING 2025-11-26 10:11:43 +08:00
  • b80f571425 Enable ck_builder in CI. (#3296) Illia Silin 2025-11-25 17:45:59 -08:00
  • 250fd6be12 Enable ck_builder in CI. (#3296) Illia Silin 2025-11-25 17:45:59 -08:00
  • a54f7b1138 Enable ck_builder in CI. (#3296) Illia Silin 2025-11-25 17:45:59 -08:00
  • 2f894b29bb Partial Progress : Working GEMM Multi D ThruptiRajLakshmanaGowda 2025-11-26 01:31:52 +00:00
  • 3447196c8a resolving merge conflicts khuagarw 2025-11-26 01:22:54 +00:00
  • f290428926 working code for preshuffleb khuagarw 2025-11-26 01:13:25 +00:00
  • 1a07c85301 Partial Progress : Jenkins change to run op_new ThruptiRajLakshmanaGowda 2025-11-25 23:46:00 +00:00
  • d78098498c Partial Progress : Working code for Universal GEMM ThruptiRajLakshmanaGowda 2025-11-25 23:40:44 +00:00
  • 7788979fbe initial commit khuagarw 2025-11-25 23:38:41 +00:00
  • 6d42c1d821 Merge commit 'cd4729386927c3d20b70fc9465614e9158524598' into develop assistant-librarian[bot] 2025-11-25 23:12:05 +00:00
  • f13e2e69cc chore(copyright): update copyright header for experimental & example directory (#3292) Aviral Goel 2025-11-26 03:09:39 +04:00
  • 9f94579e5c chore(copyright): update copyright header for experimental & example directory (#3292) Aviral Goel 2025-11-26 03:09:39 +04:00
  • cd47293869 chore(copyright): update copyright header for experimental & example directory (#3292) Aviral Goel 2025-11-26 03:09:39 +04:00
  • 2c2672ff0e [CK TILE] Grouped Conv Explicit Gemm (#3289) Bartłomiej Kocot 2025-11-25 23:28:35 +01:00
  • 91cc903d12 [CK TILE] Grouped Conv Explicit Gemm (#3289) Bartłomiej Kocot 2025-11-25 23:28:35 +01:00
  • 00dfa2f2ce [CK TILE] Grouped Conv Explicit Gemm (#3289) Bartłomiej Kocot 2025-11-25 23:28:35 +01:00
  • d3721db7d0 Update documentation requirements alexxu-amd 2025-11-25 16:31:22 -05:00
  • 8965f337ca Merge commit '37ea1600888f515e5dfb7153b75b2f06474d880d' into develop assistant-librarian[bot] 2025-11-25 21:12:15 +00:00
  • 192bb72244 [CK-Tile] fix block scale example for gfx1201 (#3283) Khushbu Agarwal 2025-11-25 13:10:28 -08:00
  • a6241c62cc [CK-Tile] fix block scale example for gfx1201 (#3283) Khushbu Agarwal 2025-11-25 13:10:28 -08:00
  • 37ea160088 [CK-Tile] fix block scale example for gfx1201 (#3283) Khushbu Agarwal 2025-11-25 13:10:28 -08:00
  • a80b2d924a Update documentation requirements alexxu-amd 2025-11-25 15:37:46 -05:00
  • 2b03f054f8 Partial Progress : Working GEMM Universal ThruptiRajLakshmanaGowda 2025-11-25 17:42:15 +00:00
  • c7afd540d9 correct the batch normalization algorithm to calc reduce over NxHxW Mohsen Saffari 2025-11-25 16:46:00 +00:00
  • 6c8cfa58c4 update yadaish 2025-11-25 16:21:37 +00:00
  • 083677b055 remove unused varable max_accumulated_value from example mohsen saffari 2025-11-21 14:51:39 +01:00
  • a4ac2629cd clang format correction mohsen saffari 2025-11-21 14:39:32 +01:00
  • 6a30ad9f25 removed unused rtol_atol variable from example code mohsen saffari 2025-11-21 14:36:24 +01:00
  • fe5ec43a51 correct clang-format mohsen saffari 2025-11-19 15:32:39 +01:00
  • eb7b099bf2 Add validity checks for MoE FlatMM scatter and enable bf16 hardware atomic Mohsen Saffari 2025-11-19 14:02:24 +00:00
  • 93aed9744c draft solin 2025-11-25 16:04:47 +00:00
  • 15345968ec Fix pre-commit error Manish Kumar 2025-11-25 14:59:16 +00:00
  • b649b364bf Remove commented code Manish Kumar 2025-11-25 14:55:19 +00:00
  • 8b6c11b490 Fix build errors Manish Kumar 2025-11-25 14:51:55 +00:00
  • a5b724bc4d Merge commit '9ac2666d5b48efc3743ce073aab0a68833accf5c' into develop assistant-librarian[bot] 2025-11-25 14:13:08 +00:00
  • 95ec5ccec0 [CK_BUILDER] Add grouped conv bwd ck tile traits (#3281) Bartłomiej Kocot 2025-11-25 14:57:43 +01:00
  • 083ea723a0 [CK_BUILDER] Add grouped conv bwd ck tile traits (#3281) Bartłomiej Kocot 2025-11-25 14:57:43 +01:00
  • 9ac2666d5b [CK_BUILDER] Add grouped conv bwd ck tile traits (#3281) Bartłomiej Kocot 2025-11-25 14:57:43 +01:00
  • 5fa3ec2b37 Some refactoring for simple batchnorm kernel and example Mohsen Saffari 2025-11-25 13:14:58 +00:00
  • 6880e11927 Missing comma Wojciech Laskowski 2025-11-25 12:02:42 +00:00
  • 3cb7f8db6e Update instances list Wojciech Laskowski 2025-11-25 11:37:22 +00:00
  • 0e81b84f4a A simple batchnorm kernel mohsen saffari 2025-11-25 11:19:34 +01:00
  • cc7caf4d7d correct results Tianxing Wu 2025-11-25 09:27:40 +00:00
  • 1885f68606 Partial Progress ThruptiRajLakshmanaGowda 2025-11-25 08:45:55 +00:00
  • 24dfa0dd48 draft of mxfp4 moe solin 2025-11-25 08:43:24 +00:00
  • bc152dcc4b Merge branch 'dev/yadai' of github.com:ROCm/composable_kernel into dev/moe_ffn dev/moe_ffn yadaish 2025-11-25 08:12:49 +00:00
  • 2848f99018 update yadaish 2025-11-25 08:11:41 +00:00
  • 42135da765 Merge branch 'moe-flatmm-scatter-validity' of github.com:ROCm/composable_kernel into dev/moe_ffn yadaish 2025-11-25 08:01:07 +00:00
  • 7aead9064b Remove changes from gemm example Manish Kumar 2025-11-25 06:28:46 +00:00
  • 984e3d4b71 Resolve PR comments Manish Kumar 2025-11-25 06:26:33 +00:00
  • dc2d4f39d1 enable int for moe ffn yadaish 2025-11-25 06:06:46 +00:00
  • 1575d849bd change endian seems working yadaish 2025-11-25 05:23:47 +00:00
  • 5375584eac Merge commit 'ab0101c59c6be6ad376ba668a51f0e38dca66aa2' into develop assistant-librarian[bot] 2025-11-25 02:43:48 +00:00
  • 52168e0481 chore(copyright): update copyright header for library directory (#3274) Aviral Goel 2025-11-25 06:10:26 +04:00
  • 804730c0f3 chore(copyright): update copyright header for library directory (#3274) Aviral Goel 2025-11-25 06:10:26 +04:00
  • ab0101c59c chore(copyright): update copyright header for library directory (#3274) Aviral Goel 2025-11-25 06:10:26 +04:00
  • a535de0f75 chore(copyright): update copyright header for example directory (#3273) Aviral Goel 2025-11-25 06:02:41 +04:00
  • 91ffc9dd1e chore(copyright): update copyright header for example directory (#3273) Aviral Goel 2025-11-25 06:02:41 +04:00
  • d85f065b15 chore(copyright): update copyright header for example directory (#3273) Aviral Goel 2025-11-25 06:02:41 +04:00
  • de08f43ef6 Merge commit '229d43ea0c8b9c94092ce001e411f82c3766b6fb' into develop assistant-librarian[bot] 2025-11-25 01:51:07 +00:00
  • f20f9dd453 Fix batch prefill compile fail in aiter (#3279) rocking 2025-11-25 09:46:32 +08:00
  • 9cb3d700da Fix batch prefill compile fail in aiter (#3279) rocking 2025-11-25 09:46:32 +08:00
  • 229d43ea0c Fix batch prefill compile fail in aiter (#3279) rocking 2025-11-25 09:46:32 +08:00
  • 4aaa8c92bb Merge commit 'de6a9590abe907283e189abba1b487f8e5562d1b' into develop assistant-librarian[bot] 2025-11-24 21:29:18 +00:00
  • a18901385b Reorganize of KPack in GEMM (#3247) Thomas Ning 2025-11-24 12:38:59 -08:00
  • 99e6b461db Reorganize of KPack in GEMM (#3247) Thomas Ning 2025-11-24 12:38:59 -08:00
  • de6a9590ab Reorganize of KPack in GEMM (#3247) Thomas Ning 2025-11-24 12:38:59 -08:00
  • 04aaf97192 debugging PermuteN khuagarw 2025-11-24 18:58:51 +00:00
  • 5297edb40c Merge commit 'e95337c58c00d12b5c947006836f9fb46964b35c' into develop assistant-librarian[bot] 2025-11-24 18:22:07 +00:00
  • ed24d3a8fa chore(copyright): update copyright header for codegen directory (#3266) Aviral Goel 2025-11-24 22:12:40 +04:00
  • f65f0820ca chore(copyright): update copyright header for codegen directory (#3266) Aviral Goel 2025-11-24 22:12:40 +04:00
  • e95337c58c chore(copyright): update copyright header for codegen directory (#3266) Aviral Goel 2025-11-24 22:12:40 +04:00
  • 39d9acab2e Guard a builder test to avoid gfx11 and gfx12 (#3268) John Shumway 2025-11-24 10:10:09 -08:00
  • 04f8fa2316 Guard a builder test to avoid gfx11 and gfx12 (#3268) John Shumway 2025-11-24 10:10:09 -08:00
  • 1bc7529977 Guard a builder test to avoid gfx11 and gfx12 (#3268) John Shumway 2025-11-24 10:10:09 -08:00
  • 10eb15416c First look at mfma / wmma unification (#2704) Christopher Millette 2025-11-24 10:39:59 -07:00
  • a049cdebba First look at mfma / wmma unification (#2704) Christopher Millette 2025-11-24 10:39:59 -07:00
  • b9c6cb1452 First look at mfma / wmma unification (#2704) Christopher Millette 2025-11-24 10:39:59 -07:00
  • c420d0386d Merge commit '8111572785d3de98457940f2b5ca6fe9cf7603af' into develop assistant-librarian[bot] 2025-11-24 16:13:04 +00:00
  • 7d6cd1f3c4 [CK_Tile] Support for preshuffle weight(B) quant tensor for block scale gemm (#3165) Khushbu Agarwal 2025-11-24 07:48:42 -08:00
  • 84b12586c6 [CK_Tile] Support for preshuffle weight(B) quant tensor for block scale gemm (#3165) Khushbu Agarwal 2025-11-24 07:48:42 -08:00
  • 8111572785 [CK_Tile] Support for preshuffle weight(B) quant tensor for block scale gemm (#3165) Khushbu Agarwal 2025-11-24 07:48:42 -08:00
  • b3c5cd0c76 Fixed the block_table Tianxing Wu 2025-11-24 15:32:33 +00:00
  • 6f3484eaa8 Merge commit 'e857e26bf64ab54dc6dcef0d89203982873a5fa8' into develop assistant-librarian[bot] 2025-11-24 15:13:49 +00:00
  • a1651a1b10 disable CI on gfx1010 by default (#3280) Illia Silin 2025-11-24 07:06:41 -08:00
  • 585a6a2048 disable CI on gfx1010 by default (#3280) Illia Silin 2025-11-24 07:06:41 -08:00
  • e857e26bf6 disable CI on gfx1010 by default (#3280) Illia Silin 2025-11-24 07:06:41 -08:00
  • 1a4543c060 Merge commit '81042ea5747d3e1e4a71c3f327556f3fb0655d99' into develop assistant-librarian[bot] 2025-11-24 14:13:16 +00:00
  • 3b341e4a16 Fix a bug for qr_ks_vs_async_trload pipeline (#3271) Qianfeng 2025-11-24 21:31:48 +08:00
  • 8ec85b7617 Fix a bug for qr_ks_vs_async_trload pipeline (#3271) Qianfeng 2025-11-24 21:31:48 +08:00
  • 81042ea574 Fix a bug for qr_ks_vs_async_trload pipeline (#3271) Qianfeng 2025-11-24 21:31:48 +08:00
  • a66a5bc75d fix develop merge ck_tile_fmha_block_scale ltqin 2025-11-24 11:07:12 +00:00
  • 843c092b63 update yadaish 2025-11-24 10:59:09 +00:00
  • ea45c116c7 update yadaish 2025-11-24 10:45:56 +00:00
  • 76d1866537 Pipeline minor fixes Tianxing Wu 2025-11-24 10:26:26 +00:00