AI/rocm-systems
2
0
Форкнуть 0
Код Задачи Запросы на слияние Действия Пакеты Проекты Релизы Вики Активность

Граф коммитов

  • 84eca4cf9e Add -g compiler option for ADDRESS_SANITIZER Bill(Shuzhou) Liu 2021-08-03 13:49:58 -04:00
  • 6f95200387 Add -g compiler option for ADDRESS_SANITIZER Bill(Shuzhou) Liu 2021-08-03 13:49:58 -04:00
  • 1cbb056f4c SWDEV-297689 - Moving the vendor specific builtin function calls to hip_cooperative_helper.h header file Sourabh Betigeri 2021-08-02 16:11:51 -07:00
  • 8a4ca88477 SWDEV-297689 - Moving the vendor specific builtin function calls to hip_cooperative_helper.h header file Sourabh Betigeri 2021-08-02 16:11:51 -07:00
  • 872613f93f SWDEV-294319 & SWDEV-294321 Added support for missing functions Ammar ELWazir 2021-07-16 18:38:40 -04:00
  • 804e063eda SWDEV-294319 & SWDEV-294321 Added support for missing functions Ammar ELWazir 2021-07-16 18:38:40 -04:00
  • 14b57fe882 SWDEV-294597 - Fix hipNormalizedFloatValueTex Tao Sang 2021-07-22 21:46:05 -04:00
  • 148f1bd59d SWDEV-294597 - Fix hipNormalizedFloatValueTex Tao Sang 2021-07-22 21:46:05 -04:00
  • 6a62455270 SWDEV-297563 - Move surface types Rahul Garg 2021-08-02 18:50:33 +00:00
  • 03e205a1c8 SWDEV-297563 - Move surface types Rahul Garg 2021-08-02 18:50:33 +00:00
  • ab2cc17a2e SWDEV-297563 - Move surface types Rahul Garg 2021-08-02 18:48:15 +00:00
  • ec39f62ee1 SWDEV-297563 - Move surface types Rahul Garg 2021-08-02 18:48:15 +00:00
  • 53fe8f53b5 SWDEV-297563 - Update hip-common declarations Rahul Garg 2021-07-30 19:18:20 +00:00
  • 54e703b17e SWDEV-297563 - Update hip-common declarations Rahul Garg 2021-07-30 19:18:20 +00:00
  • 62a018191f SWDEV-297563 - Move declarations to hip Rahul Garg 2021-07-30 19:15:38 +00:00
  • 3dd9dbb359 SWDEV-297563 - Move declarations to hip Rahul Garg 2021-07-30 19:15:38 +00:00
  • 5be5b37e19 [TransferBench] Update to 2.10.3 Gilbert Lee 2021-08-02 05:53:20 -05:00
  • a44ed86b46 Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-09-15 10:33:25 -07:00
  • 1eaf495391 Remove extra L1 cache invalidate and restore __ATOMIC_SEQ_CST atomics (#426) Wenkai Du 2021-09-14 18:30:16 -07:00
  • d6064367f0 Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-09-13 17:19:07 -07:00
  • f4387b2954 Use relaxed atomics and add sleep and wakeup in barrier loop (#425) Wenkai Du 2021-09-13 17:03:49 -07:00
  • 8acdb77cc0 Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-09-13 15:32:46 -07:00
  • 9ffeb41fe1 Update tuning table (#424) Wenkai Du 2021-09-13 08:39:01 -07:00
  • 934885526d Merge pull request #423 from wenkaidu/prim-test Wenkai Du 2021-09-08 17:01:19 -07:00
  • d2580c8cf5 Improve barrier implementation Wenkai Du 2021-09-08 16:14:32 -05:00
  • d75504e9dc Remove atomic from profiling Wenkai Du 2021-09-08 14:20:32 -05:00
  • 310d51056f rccl-prim-test: enable 8p1h and 16p1h test Wenkai Du 2021-09-07 11:15:51 -05:00
  • 4f610a2239 Revert "rccl-prim-test: add all-to-all benchmark (#185)" Wenkai Du 2021-09-07 13:48:59 -05:00
  • 06b0e1c4e2 [TransferBench] ConfigFile parsing fixes, adding additional info (#422) gilbertlee-amd 2021-09-07 15:28:16 -06:00
  • 95b37d97cf Merge remote-tracking branch 'nccl/master' into 2.10.3 Wenkai Du 2021-09-01 16:31:44 -07:00
  • 04553b802a Fix to https://github.com/NVIDIA/nccl/issues/560 John Bachan 2021-08-31 14:33:48 -07:00
  • e0fd3c6ba3 Modifying ReduceOrCopyMulti to accept number of preOp source to support clique-based kernels Gilbert Lee 2021-08-11 11:00:34 -05:00
  • eaf54184bf Improve clique kernel performance by increasing unroll Wenkai Du 2021-08-26 18:06:09 -07:00
  • a5e85a66fd Fix typo that affects clique kernels Wenkai Du 2021-08-26 11:10:58 -07:00
  • bd03ec2a45 Unit Test: support ncclAvg Wenkai Du 2021-08-25 12:01:49 -07:00
  • da7ac55e6f Fix kernel data trace Wenkai Du 2021-08-24 14:02:53 -07:00
  • 4fd7a14087 Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-08-24 09:49:47 -07:00
  • b9508a6aba Implement NIC identification and remapping (#420) Wenkai Du 2021-08-24 09:42:04 -07:00
  • 57518da006 Add gfx908 VM model (#418) Wenkai Du 2021-08-10 08:55:11 -07:00
  • 51198b536d Use noinline for kernel functions Wenkai Du 2021-08-06 09:15:04 -07:00
  • 79bc1e3fde Fix incorrect network proxy received bytes reporting Wenkai Du 2021-08-05 17:45:48 -07:00
  • b0c3a1790f [TransferBench] Removing dependency on hip_fp16 header, fixing swapped output CSV header (#416) gilbertlee-amd 2021-08-04 10:53:41 -06:00
  • 6e997354e8 Merge branch 'develop' into 2.10.3 Wenkai Du 2021-08-04 09:45:22 -07:00
  • de0c586bad Sort IB devices based on device name (#413) Wenkai Du 2021-08-03 15:32:41 -07:00
  • 68ec3f84e6 [TransferBench] Update to 2.10.3 Gilbert Lee 2021-08-02 05:53:20 -05:00
  • 5b72727670 Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-09-15 10:33:25 -07:00
  • 5ae3f3f954 Remove extra L1 cache invalidate and restore __ATOMIC_SEQ_CST atomics (#426) Wenkai Du 2021-09-14 18:30:16 -07:00
  • 3667d308ab Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-09-13 17:19:07 -07:00
  • 020484bf40 Use relaxed atomics and add sleep and wakeup in barrier loop (#425) Wenkai Du 2021-09-13 17:03:49 -07:00
  • 8ee2b7932a Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-09-13 15:32:46 -07:00
  • ef432e48e1 Update tuning table (#424) Wenkai Du 2021-09-13 08:39:01 -07:00
  • a2421f8b4a Merge pull request #423 from wenkaidu/prim-test Wenkai Du 2021-09-08 17:01:19 -07:00
  • adb8d63352 Improve barrier implementation Wenkai Du 2021-09-08 16:14:32 -05:00
  • 31bd4236f1 Remove atomic from profiling Wenkai Du 2021-09-08 14:20:32 -05:00
  • 7558b5e2bf rccl-prim-test: enable 8p1h and 16p1h test Wenkai Du 2021-09-07 11:15:51 -05:00
  • b22d097524 Revert "rccl-prim-test: add all-to-all benchmark (#185)" Wenkai Du 2021-09-07 13:48:59 -05:00
  • 51d64894ff [TransferBench] ConfigFile parsing fixes, adding additional info (#422) gilbertlee-amd 2021-09-07 15:28:16 -06:00
  • 21969d7f89 Merge remote-tracking branch 'nccl/master' into 2.10.3 Wenkai Du 2021-09-01 16:31:44 -07:00
  • 5f2f2f670f Fix to https://github.com/NVIDIA/nccl/issues/560 John Bachan 2021-08-31 14:33:48 -07:00
  • ae13d2a354 Modifying ReduceOrCopyMulti to accept number of preOp source to support clique-based kernels Gilbert Lee 2021-08-11 11:00:34 -05:00
  • a4929465c5 Improve clique kernel performance by increasing unroll Wenkai Du 2021-08-26 18:06:09 -07:00
  • 574f0aca53 Fix typo that affects clique kernels Wenkai Du 2021-08-26 11:10:58 -07:00
  • 1faff323b4 Unit Test: support ncclAvg Wenkai Du 2021-08-25 12:01:49 -07:00
  • 60ca7484c0 Fix kernel data trace Wenkai Du 2021-08-24 14:02:53 -07:00
  • d5f93649ff Merge remote-tracking branch 'origin/develop' into 2.10.3 Wenkai Du 2021-08-24 09:49:47 -07:00
  • 5c8380ff5b Implement NIC identification and remapping (#420) Wenkai Du 2021-08-24 09:42:04 -07:00
  • 5f15ed6e3e Add gfx908 VM model (#418) Wenkai Du 2021-08-10 08:55:11 -07:00
  • 707c687090 Use noinline for kernel functions Wenkai Du 2021-08-06 09:15:04 -07:00
  • 01d3b20a66 Fix incorrect network proxy received bytes reporting Wenkai Du 2021-08-05 17:45:48 -07:00
  • 1ed272e5f0 [TransferBench] Removing dependency on hip_fp16 header, fixing swapped output CSV header (#416) gilbertlee-amd 2021-08-04 10:53:41 -06:00
  • babbd1047b Merge branch 'develop' into 2.10.3 Wenkai Du 2021-08-04 09:45:22 -07:00
  • 2d0ed8dff6 Sort IB devices based on device name (#413) Wenkai Du 2021-08-03 15:32:41 -07:00
  • 4b89e98675 Merge remote-tracking branch 'nccl/master' into 2.10.3 Wenkai Du 2021-07-28 13:27:06 -07:00
  • bf2339f93e Merge remote-tracking branch 'nccl/master' into 2.10.3 Wenkai Du 2021-07-28 13:27:06 -07:00
  • f1404ea392 CMakeLists: Re-enable pkgconfig, require libdrm Kent Russell 2021-07-30 12:50:08 -04:00
  • 6778da4460 CMakeLists: Re-enable pkgconfig, require libdrm Kent Russell 2021-07-30 12:50:08 -04:00
  • 95bf93f650 Update README and increment minor version number. Sean Keely 2021-07-30 17:13:17 -05:00
  • ed7eec14e1 Update README and increment minor version number. Sean Keely 2021-07-30 17:13:17 -05:00
  • e96500d901 Add optional dependency on rocm-core. Sean Keely 2021-07-21 21:38:44 -05:00
  • 4907d4577b Add optional dependency on rocm-core. Sean Keely 2021-07-21 21:38:44 -05:00
  • 2ddf2f8f82 Simplify RPM package building. Sean Keely 2021-07-21 22:46:12 -05:00
  • 471c1859ab Simplify RPM package building. Sean Keely 2021-07-21 22:46:12 -05:00
  • ed1e858da7 Remove unused branches from DEB packaging scripts. Sean Keely 2021-07-23 17:53:07 -05:00
  • e8439cca08 Remove unused branches from DEB packaging scripts. Sean Keely 2021-07-23 17:53:07 -05:00
  • 951ec33ae0 Correct hsa_status_string strings. Sean Keely 2021-07-15 15:04:38 -05:00
  • 081ab00f8e Correct hsa_status_string strings. Sean Keely 2021-07-15 15:04:38 -05:00
  • c549aa71c3 Add missing HSA_STATUS_ERROR_INVALID_MEMORY_POOL string. Sean Keely 2021-07-15 14:47:07 -05:00
  • 62b7c0ed3b Add missing HSA_STATUS_ERROR_INVALID_MEMORY_POOL string. Sean Keely 2021-07-15 14:47:07 -05:00
  • 4fb0096393 Minor correction to debug messages. Sean Keely 2021-07-15 13:52:34 -05:00
  • a0069904c8 Minor correction to debug messages. Sean Keely 2021-07-15 13:52:34 -05:00
  • ddef80a37e Correct single character transcription errors in license text. Sean Keely 2021-07-27 16:13:17 -05:00
  • 0778969e89 Correct single character transcription errors in license text. Sean Keely 2021-07-27 16:13:17 -05:00
  • dd7c30ec6b Correct data race in GpuAgent::GetXgmiBlit. Sean Keely 2021-07-29 17:39:13 -05:00
  • bb4dfbba1e Correct data race in GpuAgent::GetXgmiBlit. Sean Keely 2021-07-29 17:39:13 -05:00
  • 63d9a19053 [rocm_smi.py] --showpower error bugfix Elena Sakhnovitch 2021-07-29 11:56:56 -04:00
  • 6a01b6b2ec [rocm_smi.py] --showpower error bugfix Elena Sakhnovitch 2021-07-29 11:56:56 -04:00
  • 6496488d1d [rocm_smi.py] add secondary die check. Elena Sakhnovitch 2021-07-29 11:39:34 -04:00
  • 2c39e6cf51 [rocm_smi.py] add secondary die check. Elena Sakhnovitch 2021-07-29 11:39:34 -04:00
  • 42b0e1c549 rocm_smi.py: Remove extraneous line during process termination Harish Kasiviswanathan 2021-07-27 14:22:22 -04:00
  • cef19745d1 rocm_smi.py: Remove extraneous line during process termination Harish Kasiviswanathan 2021-07-27 14:22:22 -04:00