Commit Graph

  • f2fbc44b7b fix Juuso Korhonen 2025-11-24 10:20:04 +00:00
  • b55d35d131 Merge branch 'develop' into ck_tile_fmha_block_scale ltqin 2025-11-24 10:15:43 +00:00
  • 92b7c1fcaf Remove dead code. vpietila/ckb-fwd-bwd-instances Ville Pietilä 2025-11-24 10:01:57 +00:00
  • fc70103f8b update yadaish 2025-11-24 09:57:31 +00:00
  • fcff617424 fix out of lds yadaish 2025-11-24 09:32:44 +00:00
  • f2425d427c Merge commit '5948dbffe4d0bbe4d1802a047bd8599ba662386e' into develop assistant-librarian[bot] 2025-11-24 09:15:05 +00:00
  • cdd72e57d3 Support fp8 dynamic quantization for fmha (#3206) rocking 2025-11-24 16:28:25 +08:00
  • ca1a0da0c3 Support fp8 dynamic quantization for fmha (#3206) rocking 2025-11-24 16:28:25 +08:00
  • 5948dbffe4 Support fp8 dynamic quantization for fmha (#3206) rocking 2025-11-24 16:28:25 +08:00
  • 7bd01a9f5f Merge commit '096f0a3b23a49ffaef1e2dbed74bf366e36ad15c' into develop assistant-librarian[bot] 2025-11-24 07:13:25 +00:00
  • 6636e0d8d2 Merge branch 'rocking/fmha-fp8-pertensor' into ck_tile_fmha_block_scale ltqin 2025-11-24 06:50:41 +00:00
  • 679699f32a [CK Tile] Fix example for conv fwd + bias + clamp (#3235) Johannes Graner 2025-11-24 07:36:26 +01:00
  • dd7a2d199f [CK Tile] Fix example for conv fwd + bias + clamp (#3235) Johannes Graner 2025-11-24 07:36:26 +01:00
  • 096f0a3b23 [CK Tile] Fix example for conv fwd + bias + clamp (#3235) Johannes Graner 2025-11-24 07:36:26 +01:00
  • f9e8c5539f Use explicit partition_index to ensure warp_id is allocated on vpgr when accessing LDS tile_window Qianfeng Zhang 2025-11-22 16:12:21 +00:00
  • e398aa6b4d Merge branch 'develop' into rocking/fmha-fp8-pertensor asleepzzz 2025-11-23 12:21:44 +08:00
  • 4f33eb5857 Merge branch 'develop' into hstu_attention_mi350_fwd_bwd Qianfeng Zhang 2025-11-23 04:20:53 +00:00
  • 8abfd83364 Merge commit 'f6c999bddb9e0ae468c7b45bc68cc1410472dcf5' into develop assistant-librarian[bot] 2025-11-23 00:40:28 +00:00
  • d171245c4b chore(copyright): update copyright header for test directory (#3265) Aviral Goel 2025-11-22 19:38:27 -05:00
  • 1bec1dd091 chore(copyright): update copyright header for test directory (#3265) Aviral Goel 2025-11-22 19:38:27 -05:00
  • f6c999bddb chore(copyright): update copyright header for test directory (#3265) Aviral Goel 2025-11-22 19:38:27 -05:00
  • d7685c394a Merge commit '02ab76c2cb47143b82743bcf9d86389c540a608b' into develop assistant-librarian[bot] 2025-11-22 04:13:58 +00:00
  • 0d6a0a3c2f Fix CK Tile DP + 2 Tile Stream-K Validation Errors (#3269) Emily Martins 2025-11-21 20:29:47 -07:00
  • ede105dd91 Fix CK Tile DP + 2 Tile Stream-K Validation Errors (#3269) Emily Martins 2025-11-21 20:29:47 -07:00
  • 02ab76c2cb Fix CK Tile DP + 2 Tile Stream-K Validation Errors (#3269) Emily Martins 2025-11-21 20:29:47 -07:00
  • 343e40d0e9 Merge commit '21ae743acd49c79913b3835236c5315983fa83ef' into develop assistant-librarian[bot] 2025-11-21 16:13:44 +00:00
  • d43b58b3cc Enable daily builds on gfx1010 (#3258) Illia Silin 2025-11-21 07:22:01 -08:00
  • 6d7d99f91b Enable daily builds on gfx1010 (#3258) Illia Silin 2025-11-21 07:22:01 -08:00
  • 21ae743acd Enable daily builds on gfx1010 (#3258) Illia Silin 2025-11-21 07:22:01 -08:00
  • 323c839a2b Merge commit 'ea6e4fcbbc0bd76a562f246f743f5554edc312e4' into develop assistant-librarian[bot] 2025-11-21 15:12:19 +00:00
  • 071fbaaf28 Fix builder errors. (#3260) John Shumway 2025-11-21 06:25:45 -08:00
  • 34c3e1f562 Fix builder errors. (#3260) John Shumway 2025-11-21 06:25:45 -08:00
  • ea6e4fcbbc Fix builder errors. (#3260) John Shumway 2025-11-21 06:25:45 -08:00
  • b8c99da8f6 group mode for block scale ltqin 2025-11-21 13:53:45 +00:00
  • caf82d1a4f remove unused varable max_accumulated_value from example mohsen saffari 2025-11-21 14:51:39 +01:00
  • 122a981483 clang format correction mohsen saffari 2025-11-21 14:39:32 +01:00
  • 8d56b0af6f removed unused rtol_atol variable from example code mohsen saffari 2025-11-21 14:36:24 +01:00
  • b83ea0b858 cascade to GEMM Multi D and GEMM Preshuffle operators philipm/fix-tile_engine-json-output Philip Maybank 2025-11-21 12:09:52 +00:00
  • 82cc3c3f95 run clang-format-18 Philip Maybank 2025-11-20 10:18:02 +00:00
  • 90fbaa777b use ofstream for writing json output Philip Maybank 2025-11-19 18:36:32 +00:00
  • 0304fd3cbb Normalize topk index for supporting python-liked negative indexing mhynag/moe-index-support MHYang 2025-11-21 18:18:33 +08:00
  • eb41811b5f Merge branch 'develop' into moe_xcd_remap Tianxing Wu 2025-11-21 11:31:31 +02:00
  • df848e6895 fix precommit ginolu/fix-fa-hdim160 Gino Lu 2025-11-21 03:30:19 -06:00
  • 612bedd3ab test if CI failed on hdim=160 Gino Lu 2025-11-21 02:48:13 -06:00
  • 449de985a7 remove max_token_id Tianxing Wu 2025-11-21 07:32:30 +00:00
  • 1829bc6596 Merge commit 'f38c3de9f9047e72429c796fd0445f36eceb142b' into develop assistant-librarian[bot] 2025-11-21 03:31:42 +00:00
  • a7048cf4d4 start group ltqin 2025-11-21 02:57:07 +00:00
  • 74e6e87603 Merge branch 'develop' into rocking/fmha-fp8-pertensor rocking 2025-11-21 10:18:55 +08:00
  • dfd9c2c7df Remove bias test case for fp8 rocking 2025-11-21 10:18:38 +08:00
  • 3f33037f60 Fix copyright messages in experimental/builder. (#3253) John Shumway 2025-11-20 17:40:55 -08:00
  • 345dbb25f8 Fix copyright messages in experimental/builder. (#3253) John Shumway 2025-11-20 17:40:55 -08:00
  • f38c3de9f9 Fix copyright messages in experimental/builder. (#3253) John Shumway 2025-11-20 17:40:55 -08:00
  • 967480c146 Merge commit 'c8563f2101d864ed0cc1f68f02763ee4ec6aa59d' into develop assistant-librarian[bot] 2025-11-21 01:40:40 +00:00
  • 7cee27c4a2 chore(copyright): update copyright header for test directory (#3252) Aviral Goel 2025-11-20 20:36:57 -05:00
  • 89e3931da8 chore(copyright): update copyright header for test directory (#3252) Aviral Goel 2025-11-20 20:36:57 -05:00
  • c8563f2101 chore(copyright): update copyright header for test directory (#3252) Aviral Goel 2025-11-20 20:36:57 -05:00
  • ad1f388f7f chore(copyright): update copyright header for cmake directory (#3254) Aviral Goel 2025-11-20 20:36:37 -05:00
  • a42ac42d00 chore(copyright): update copyright header for cmake directory (#3254) Aviral Goel 2025-11-20 20:36:37 -05:00
  • a960c9950b chore(copyright): update copyright header for cmake directory (#3254) Aviral Goel 2025-11-20 20:36:37 -05:00
  • ba44e7b7a4 fix static assert (#3178) lalala-sh 2025-11-21 09:27:05 +08:00
  • 391dfbb074 fix static assert (#3178) lalala-sh 2025-11-21 09:27:05 +08:00
  • f58bd56e6b fix static assert (#3178) lalala-sh 2025-11-21 09:27:05 +08:00
  • 03b79e2264 fix:bf16x3:enable all instances on gfx950 (#3248) yinglu 2025-11-21 09:09:43 +08:00
  • 14d3bfa1eb fix:bf16x3:enable all instances on gfx950 (#3248) yinglu 2025-11-21 09:09:43 +08:00
  • 4155eb24f9 fix:bf16x3:enable all instances on gfx950 (#3248) yinglu 2025-11-21 09:09:43 +08:00
  • 2c4d0dd289 Partial Progress : Generate Single Kernel until trait config ThruptiRajLakshmanaGowda 2025-11-20 22:36:59 +00:00
  • cf3f9b57b4 Merge branch 'develop' into lwpck-3985 khuagarw 2025-11-20 21:48:16 +00:00
  • a974a08c2e debugging khuagarw 2025-11-20 21:26:21 +00:00
  • 084d063087 Merge commit '938b8ed3bf40741176adbc897b66095c5453d15d' into develop assistant-librarian[bot] 2025-11-20 19:11:49 +00:00
  • 661a4f54a0 Merge branch 'develop' into moe_xcd_remap Illia Silin 2025-11-20 11:01:33 -08:00
  • e796265733 Merge branch 'develop' into rocking/fmha-fp8-pertensor Illia Silin 2025-11-20 10:52:21 -08:00
  • ff54ec9463 Spolifroni amd/update changelog 711 (#3211) spolifroni-amd 2025-11-20 13:51:18 -05:00
  • 3e09d4caf2 Spolifroni amd/update changelog 711 (#3211) spolifroni-amd 2025-11-20 13:51:18 -05:00
  • 938b8ed3bf Spolifroni amd/update changelog 711 (#3211) spolifroni-amd 2025-11-20 13:51:18 -05:00
  • ac4f4ffb79 [CK_TILE] Refine FP32 => FP16/BF16 Conversion (#3215) Yi DING 2025-11-21 02:50:26 +08:00
  • f0702c1636 [CK_TILE] Refine FP32 => FP16/BF16 Conversion (#3215) Yi DING 2025-11-21 02:50:26 +08:00
  • 8b284a63a4 [CK_TILE] Refine FP32 => FP16/BF16 Conversion (#3215) Yi DING 2025-11-21 02:50:26 +08:00
  • d80f38f77f Add support for RDNA1 GPUs (#3220) Gavin Zhao 2025-11-20 13:45:57 -05:00
  • 50e7d047f6 Add support for RDNA1 GPUs (#3220) Gavin Zhao 2025-11-20 13:45:57 -05:00
  • 07314ac543 Add support for RDNA1 GPUs (#3220) Gavin Zhao 2025-11-20 13:45:57 -05:00
  • eb3ebe3b38 ck-builder: add remaining ck factory tests (#3223) Robin Voetter 2025-11-20 19:42:36 +01:00
  • b558b6632c ck-builder: add remaining ck factory tests (#3223) Robin Voetter 2025-11-20 19:42:36 +01:00
  • bb155ef678 ck-builder: add remaining ck factory tests (#3223) Robin Voetter 2025-11-20 19:42:36 +01:00
  • fe6bb0e811 ck-builder: group transfer operations per tensor (#3217) Robin Voetter 2025-11-20 19:40:48 +01:00
  • ea6bc81dd1 ck-builder: group transfer operations per tensor (#3217) Robin Voetter 2025-11-20 19:40:48 +01:00
  • 245c6011cf ck-builder: group transfer operations per tensor (#3217) Robin Voetter 2025-11-20 19:40:48 +01:00
  • 635cf8df6c chore(copyright): update copyright header for library directory (#3239) Aviral Goel 2025-11-20 13:36:05 -05:00
  • 63a6703a81 chore(copyright): update copyright header for library directory (#3239) Aviral Goel 2025-11-20 13:36:05 -05:00
  • fb43760c66 chore(copyright): update copyright header for library directory (#3239) Aviral Goel 2025-11-20 13:36:05 -05:00
  • ef107dac80 chore(copyright): update copyright header for test directory (#3243) Aviral Goel 2025-11-20 13:33:34 -05:00
  • 0058fb65ff chore(copyright): update copyright header for test directory (#3243) Aviral Goel 2025-11-20 13:33:34 -05:00
  • 7dfc46d73d chore(copyright): update copyright header for test directory (#3243) Aviral Goel 2025-11-20 13:33:34 -05:00
  • d6db805e82 Partial Progress : Working till Listing kernels ThruptiRajLakshmanaGowda 2025-11-20 18:23:37 +00:00
  • eeeff3fbfe Partial Progress : Working till Listing kernels ThruptiRajLakshmanaGowda 2025-11-20 18:18:46 +00:00
  • 8b93b58bcd Merge commit '2e4b8a8fc455a14ad5cf89f7f750060ff20c40bb' into develop assistant-librarian[bot] 2025-11-20 17:12:11 +00:00
  • 4aa8d64c9a [CK_TILE] Remove Old CK Tile Stream-K Artifacts (#3202) Emily Martins 2025-11-20 09:32:32 -07:00
  • 2963649b29 [CK_TILE] Remove Old CK Tile Stream-K Artifacts (#3202) Emily Martins 2025-11-20 09:32:32 -07:00
  • 2e4b8a8fc4 [CK_TILE] Remove Old CK Tile Stream-K Artifacts (#3202) Emily Martins 2025-11-20 09:32:32 -07:00
  • b2e58aec1a Merge commit '5adaa201eda9337553459bc4321b11695e380832' into develop assistant-librarian[bot] 2025-11-20 16:14:36 +00:00
  • 06d2e609cd Revert "Add attn sink (#2892)" (#3250) asleepzzz 2025-11-20 23:55:15 +08:00