The Wayback Machine - https://web.archive.org/web/20200918080801/https://github.com/NVIDIA/nccl/issues
Skip to content

/ nccl

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

55 Open 251 Closed
55 Open 251 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels.
Projects
Filter by project
Milestones
Filter by milestone
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀
Specify client ports for back-connects
#388 opened Sep 17, 2020 by vadimkantorov
4
How to determine a optimal value of NCCL_BUFFSIZE
#387 opened Sep 17, 2020 by firejq
1
problem to build nccl
#386 opened Sep 16, 2020 by Sin-searchman
pytorch Multi-GPU program hangs with unblance threads?
#383 opened Sep 3, 2020 by Approximetal
1
How to use p2p over PCIe between different containers on the same node?
#382 opened Sep 3, 2020 by weberxie
1
ncclReduce is inconsistent when seeing different gpu numbers
#381 opened Sep 2, 2020 by Richie-yan
8
Missing CPU node attributes with topology injection
#379 opened Aug 28, 2020 by rashikakheria
nccl v2.7.6 slow down 30% than nccl v2.5.7 when run tf_cnn_benchmarks.py vgg16 models on 8*v100 4nodes.
#378 opened Aug 28, 2020 by qianzhang613
13
Suspected NCCL hang issue with Horovod
#377 opened Aug 25, 2020 by nachtsky1077
3
Is NCCL creating blocking threads?
#374 opened Aug 18, 2020 by frankseide
8
why nccl2.7.8 is lower than nccl 2.5.6 for all_reduce_perf in 8V100 + IB environment?
#371 opened Aug 17, 2020 by weizaixl
3
extremely long ncclAllGather
#370 opened Aug 13, 2020 by Michael-JY-He
2
How does the LL affect model covergence?
#369 opened Aug 13, 2020 by weberxie
NCCL backend hangs but not sure why
#367 opened Aug 11, 2020 by dorucia
6
Installation failed : make pkg.debian.build
#363 opened Aug 2, 2020 by Skylixia
2
Is it ok if I don't run "make install" ?
#362 opened Jul 31, 2020 by gorou-gnm
2
How to figure out where the big jitter for ncclAllReduce come from?
#361 opened Jul 31, 2020 by kehuanfeng
6
Computation gets slower when overlapped with NCCL communication
#357 opened Jul 21, 2020 by MonicaGu
CUDA_ERROR_TOO_MANY_PEERS after export NCCL_P2P_DISABLE=1
#356 opened Jul 21, 2020 by chuanli11
3
NCCL receive bug when NCCL_NTHREADS=64
#355 opened Jul 16, 2020 by MonicaGu
2
Info on collectives plugin
#353 opened Jul 15, 2020 by AmedeoSapio
9
nccl2.7.6 p2p api send/recv using same buff, result not correct
#351 opened Jul 8, 2020 by legatoo
How to monitor slow nodes in ringallreduce
#349 opened Jun 30, 2020 by Richie-yan
4
Bug: ncclSend/ncclRecv: mixed (but matching) sized messages or connectivity result in invalid argument cuda errors and fatal crashes
#345 opened Jun 24, 2020 by nevion
12
v2.7.5 source?
#343 opened Jun 20, 2020 by chsigg
1
ProTip! Exclude everything labeled bug with -label:bug.
You can’t perform that action at this time.