AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1 410 Commits 3 Branches 0 Tags
955347bab499eb4d263e46a089fdb3fcaebf93e8
Commit Graph

1410 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Nusrat Islam 955347bab4 Merge pull request #1184 from nusislam/256-channel-2 
add 256 channels support
 2024-06-04 08:25:34 -05:00
Nusrat Islam 9746d8ca3f set MIN_NCHANNEL limit to 64 for multi-node 2024-06-03 13:05:05 -05:00
Nusrat Islam 0634c5c8e1 doubling debug buffer size with increased channels 2024-06-03 13:05:05 -05:00
Nusrat Islam ef442f8f92 set MAXCHANNELS to 128 2024-06-03 13:05:05 -05:00
Nusrat Islam 9f654f6cf5 graph: restrict MAXCHANNELS for certain platforms 2024-06-03 13:05:01 -05:00
Nusrat Islam 48859a97b1 device: update the logic for channelId assignment 2024-06-03 13:03:18 -05:00
Nusrat Islam 506f16c506 add 256 channels support 2024-06-03 13:03:18 -05:00
akolliasAMD 6475da2ed9 fixed typo on BFD linkage (#1192) 2024-06-03 10:05:47 -06:00
gilbertlee-amd 0948eecbba Changing channel stride for MI300X multinode (#1196) 
* Shuffling MI300X multi-node channels
* Updating tree channel logic
 2024-06-03 10:00:55 -06:00
srawat 3301cdf59a doc organization (#1197) 
* doc organization

* removing what is rccl file

* Update index.rst
 2024-06-03 18:38:45 +05:30
ClementLinCF cab25f919e Optimize NCHANNELS and MSCCL config for gfx942 80CUs (#1195) 
* Optimize NCHANNELS and MSCCL config for gfx942 80CUs

Set appropriately for different NCCL_MIN_NCHANNELS and MSCCL config,
potentially improving communication perf on the MI300x 80CUs

* Delete tools/msccl-algorithms/allreduce_1step_mccl_8_2_16777216_LL.xml

* Change the factor of gfx94 and update msccl config
 2024-06-01 07:07:46 -07:00
Nilesh M Negi 5aaf7121d9 [BUILD] Update install.sh for RCCL build (#1191) 
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com>
 2024-05-31 17:58:34 -05:00
Nilesh M Negi 1249a6c3fd [MSCCL]: Move scratch buffer debug msgs to TRACE (#1189) 
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com>
 2024-05-31 17:54:23 -05:00
gilbertlee-amd 354e0b29a6 Addressing possible out-of-bounds mem access during channel duplication (#1193) 2024-05-30 14:02:14 -06:00
Wenkai Du 73221b4230 Add ring simple chunk size tuning (#1180) 
* Add ring simple chunk size tuning

* modifying the tuning table to improve the performance of broadcast for 8MB to 32MB for single-node MI300X after ring simple chunk size tuning

* modifying the tuning table to improve the performance of reduce for 1MB to 4MB for single-node MI300X after ring simple chunk size tuning

---------

Co-authored-by: PedramAlizadeh <pmohamma@amd.com>
 2024-05-29 07:59:47 -07:00
Wenkai Du 8f099b1adb Make WSL detection thread safe (#1178) 
* Make WSL detection thread safe

* Change static to beginning

* Switch to use atomics
 2024-05-28 17:23:50 -07:00
Edgar Gabriel a78c4f5e88 Merge pull request #1175 from edgargabriel/topic/alt_rsmi 
add alternative to rocm_smi_lib
 2024-05-28 09:36:55 -05:00
Joseph Macaranas e4c10e4438 Merge pull request #1188 from ROCm/amd/jmacaran/externalCIEnablement 
Enable external CI pipeline triggers
 2024-05-27 15:52:49 -04:00
amd-jmacaran 125f841c5f Enable external CI pipeline triggers 2024-05-23 16:29:05 -04:00
dependabot[bot] b06e617bea --- (#1183) 
updated-dependencies:
- dependency-name: requests
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2024-05-21 09:34:41 -06:00
Wenkai Du eeea3b693b Report error when collective is not enabled in build (#1177) 
* Report error when collective is not enabled in build

* Fix typo
 2024-05-16 10:11:12 -07:00
Edgar Gabriel 9ad913bfa8 add alternative to rocm_smi_lib 2024-05-14 13:51:41 -07:00
gilbertlee-amd 52fa5d1178 RCCL Replayer - multi communicator support (#1176) 2024-05-13 10:56:32 -06:00
Wenkai Du ecafc1969c Support WSL2 (#1173) 2024-05-10 07:31:12 -07:00
Tim f078db5998 Upload npkit_trace_analysis.py (#1152) 
script for parsing json trace, generating heatmap, throughput series, etc.
 2024-05-09 16:27:49 -04:00
mberenjk 87d01e6bf5 Changing the failure print to reflect a bad installation of MSCCL (#1160) 
Co-authored-by: Marzieh Berenjkoub <mberenjk@amd.com>
 2024-05-08 16:56:26 -05:00
Wenkai Du a64aab5f63 Use rocm-smi thread only mutex when available (#1169) 2024-05-08 14:32:24 -07:00
Pedram Alizadeh 73acf3eeec modifying the tuning table to improve the performance of broadcast for 1MB to 64MB for single-node MI300X (#1172) 2024-05-08 15:49:33 -04:00
mberenjk 408278209d Adding ASAN changes to address memory leak issue" (#1170) 
Co-authored-by: akolliasAMD <akollias@amd.com>
 2024-05-08 09:16:00 -05:00
Wenkai Du b18784d8b8 Add compiler warning for uninitialized variable and fix (#1163) 
* Add compiler warning for uninitialized variable and fix

* Add -Wsometimes-uninitialized

* Convert warning to error
 2024-05-08 07:00:25 -07:00
Wenkai Du f679db6ff6 Use normal permute path when one NIC per GPU (#1171) 2024-05-08 06:59:57 -07:00
Wenkai Du a0cef69110 npkit: add broadcast trace (#1166) 2024-05-07 14:00:16 -07:00
Pak Nin Lui 92a4fc6204 Merge pull request #1167 from paklui/dmabuf 
fix typo for DMABUF_ENABLE
 2024-05-07 08:48:44 -07:00
dependabot[bot] eb562e7b22 Bump jinja2 from 3.1.3 to 3.1.4 in /docs/sphinx (#1168) 
Bumps [jinja2](https://github.com/pallets/jinja) from 3.1.3 to 3.1.4.
- [Release notes](https://github.com/pallets/jinja/releases)
- [Changelog](https://github.com/pallets/jinja/blob/main/CHANGES.rst)
- [Commits](https://github.com/pallets/jinja/compare/3.1.3...3.1.4)

---
updated-dependencies:
- dependency-name: jinja2
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2024-05-06 15:35:34 -06:00
paklui 140b7dd40f fix typo for DMABUF_ENABLE 2024-05-06 13:27:50 -07:00
Wenkai Du b513c3970a Bypass NVIDIA Ampere related tuning (#1165) 2024-05-03 17:57:16 -07:00
Wenkai Du bb58b1c258 Fix ignore NUMA not being observed for NICs during model matching (#1164) 2024-05-03 16:42:07 -07:00
Wenkai Du 6f5a8ce1fb Fix build error when roctracer-dev package is not installed (#1161) 2024-05-01 13:55:09 -07:00
Wenkai Du 4e1b8c1cbb MSCCL: add support for out-of-place all reduce (#1156) 2024-04-28 19:49:09 -07:00
Wenkai Du cd6e840e0b Add back tree simple chunk size tuning (#1157) 2024-04-28 19:48:53 -07:00
Nilesh M Negi b90436d292 [GRAPH] Reduce NCCL_TOPO_MAX_NODES to 64 (#1153) 
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com>
 2024-04-27 23:41:11 -05:00
Tim cc39e91c6f Merge pull request #1158 from AtlantaPepsi/NPKit_fix 
Prevent segfault from npkit-enabled rccl build
 2024-04-26 12:44:04 -04:00
AtlantaPepsi 67246649ac prevent segfault from npkit-enabled rccl build 
Signed-off-by: AtlantaPepsi <timhu102@amd.com>
 2024-04-26 10:54:27 -05:00
Wenkai Du f330b82985 Revert "Use relaxed atomics for LL on GFX11 (#859)" (#1148) 
This reverts commit 6a0a6a37d9.

Use inline asm for 128b load on GFX11 for better peformance.
 2024-04-26 07:49:55 -07:00
Bertan Dogancay 0ec41f1386 [UT] Start supporting multiple group calls and graphs (#1151) 
* Start supporting multiple group calls UT
 2024-04-25 11:11:16 -06:00
Shilei Tian efe99057b0 SWDEV-455705: Fix an UB that could lead to miscompilation (#1155) 2024-04-25 10:10:01 -07:00
Wenkai Du 9e0c9b4ed8 Replace __HIP_PLATFORM_HCC__ with __HIP_PLATFORM_AMD__ (#1154) 2024-04-25 07:19:18 -07:00
Bertan Dogancay dcc75797a1 Update CHANGELOG.md for RCCL 2.20.5 (#1150) 2024-04-24 09:07:49 -06:00
Bertan Dogancay 8753bec3ea Merge pull request #1111 from BertanDogancay/2.20 
2.20.5 Sync
 2024-04-24 09:05:41 -06:00
BertanDogancay e1a835910e Merge remote-tracking branch 'nccl/master' into develop 2024-04-23 13:34:00 -07:00