Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / ROCmSoftwarePlatform/rccl issues and pull requests

#858 - Use relaxed atomics for LL

Pull Request - State: open - Opened by wenkaidu over 1 year ago

#857 - p2p/ll-latency-test: convert to single thread tests

Pull Request - State: open - Opened by wenkaidu over 1 year ago
Labels: noCI

#856 - NCCL_TREES variable rome model fixes

Pull Request - State: open - Opened by akolliasAMD over 1 year ago

#855 - Re-enable LL128 for gfx90a

Pull Request - State: open - Opened by wenkaidu over 1 year ago - 2 comments

#854 - gfx11: don't use LL for sendrecv (#853)

Pull Request - State: open - Opened by wenkaidu over 1 year ago

#853 - gfx11: don't use LL for sendrecv

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#852 - Add ncclCommSplit test

Pull Request - State: open - Opened by BertanDogancay over 1 year ago

#851 - Topo/tree set

Pull Request - State: open - Opened by akolliasAMD over 1 year ago

#850 - Bump gitpython from 3.1.31 to 3.1.32 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#848 - [WIP] Add APIs to export flow information

Pull Request - State: open - Opened by sreeram-arista over 1 year ago - 3 comments

#847 - Add new model support

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#846 - Fix mscclLoadAlgo error

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago

#845 - Make full use of NIC

Pull Request - State: open - Opened by clearsky07 over 1 year ago - 2 comments

#844 - NPKit update

Pull Request - State: closed - Opened by yzygitzh over 1 year ago

#843 - Detect HIP_UNCACHED_MEMORY support from HIP version (#842)

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#842 - Detect HIP_UNCACHED_MEMORY support from HIP version

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#841 - gfx11xx: disable LL protocol to workaround mtype issue (#840)

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#840 - gfx11xx: disable LL protocol to workaround mtype issue

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#839 - Detect HIP_UNCACHED_MEMORY support from HIP version

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#838 - Fix merge error and replace inline asm

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#837 - Enable LL128 on gfx90a

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#836 - Bump cryptography from 41.0.2 to 41.0.3 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#835 - Improve collective trace

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#834 - Disable msccl at compile time

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago

#833 - Bump rocm-docs-core from 0.19.0 to 0.20.0 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#832 - p2p_latency_test: clean up IPC temp files at exit

Pull Request - State: closed - Opened by wenkaidu over 1 year ago
Labels: noCI

#831 - Updating Doxygen documentation

Pull Request - State: closed - Opened by gilbertlee-amd over 1 year ago

#830 - Protocol selection needs to follow ENABLE_LL128

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#829 - Revert "Enable Ll128 on gfx90a (#823)"

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#828 - Change default number of parallel jobs for linking

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#827 - Change default number of parallel for linking

Pull Request - State: closed - Opened by wenkaidu over 1 year ago - 1 comment

#826 - Automatically add channels and select appropriate NIC through environ…

Pull Request - State: closed - Opened by clearsky07 over 1 year ago - 4 comments

#825 - ll_latency_test: fix time calculation

Pull Request - State: closed - Opened by wenkaidu over 1 year ago
Labels: noCI

#824 - Enable LL128 on gfx90a

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#823 - Enable LL128 on gfx90a

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#822 - Replace atomicExch with __atomic_store_n (#818)

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#821 - Fix MSCCL proxy number of chunks calculation

Pull Request - State: closed - Opened by yzygitzh over 1 year ago

#820 - tools: Add LL latency test

Pull Request - State: closed - Opened by wenkaidu over 1 year ago
Labels: noCI

#819 - Bump certifi from 2022.12.7 to 2023.7.22 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#818 - Replace atomicExch with __atomic_store_n

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#817 - Rccl replayer

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago
Labels: noCI

#816 - Enable gfx94x (#808)

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#815 - removed codeowners file

Pull Request - State: closed - Opened by akolliasAMD over 1 year ago

#814 - stream sync between cuda memcpy async

Pull Request - State: closed - Opened by akolliasAMD over 1 year ago

#813 - added codeowners file

Pull Request - State: closed - Opened by akolliasAMD over 1 year ago

#812 - Bump pygments from 2.14.0 to 2.15.0 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#811 - Removing unnecessary chrpath check for unit tests

Pull Request - State: closed - Opened by gilbertlee-amd over 1 year ago

#810 - device: fine-tune RCCL send-recv on MI250/MI200

Pull Request - State: closed - Opened by nusislam over 1 year ago - 1 comment

#809 - Bump cryptography from 41.0.0 to 41.0.2 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#808 - Enable gfx94x

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#807 - Enable gfx94x

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#806 - Enable gfx940

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#805 - Enable gfx940

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#804 - Add GPU P2P ping-pong latency test tool

Pull Request - State: closed - Opened by yzygitzh over 1 year ago
Labels: noCI

#803 - rccl-prim-test: calculate iterations' standard deviation

Pull Request - State: closed - Opened by wenkaidu over 1 year ago
Labels: noCI

#802 - rccl-prim-test: calculate throughput standard deviations

Pull Request - State: closed - Opened by wenkaidu over 1 year ago
Labels: noCI

#801 - Bump rocm-docs-core from 0.18.4 to 0.19.0 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#800 - Fix path finding in msccl internal scheduler

Pull Request - State: closed - Opened by yzygitzh over 1 year ago

#799 - device: fine tune MI200/MI250 simple protocol performance

Pull Request - State: closed - Opened by nusislam over 1 year ago

#798 - npkit: separate network timing between send and test

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#797 - device: fine tune MI200/MI250 simple protocol performance

Pull Request - State: closed - Opened by nusislam over 1 year ago - 1 comment

#796 - Bump rocm-docs-core from 0.18.2 to 0.18.4 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#795 - Sync latest update from NCCL

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#794 - Fix path finding in msccl internal scheduler

Pull Request - State: closed - Opened by yzygitzh over 1 year ago

#793 - temporarily added npkit compilation to ci

Pull Request - State: open - Opened by akolliasAMD over 1 year ago

#792 - Bump rocm-docs-core from 0.18.2 to 0.18.3 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: noCI, dependencies

#791 - Bump rocm-docs-core from 0.18.1 to 0.18.2 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#790 - added npkit into the all_gather run ring algorithm

Pull Request - State: closed - Opened by akolliasAMD over 1 year ago

#789 - Report unit test environment variable values as part of output

Pull Request - State: closed - Opened by gilbertlee-amd over 1 year ago

#788 - Bump rocm-docs-core from 0.17.2 to 0.18.1 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#787 - Bump rocm-docs-core from 0.16.0 to 0.17.2 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#785 - Limiting # parallel jobs in install script to 16 by default

Pull Request - State: closed - Opened by gilbertlee-amd over 1 year ago

#784 - Bump rocm-docs-core from 0.15.0 to 0.16.0 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#783 - Revert "Disable Colltrace for --fast option (#778)"

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#782 - Sync up with NCCL 2.18.3

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#781 - Bump rocm-docs-core from 0.14.0 to 0.15.0 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#779 - fixed npkit size to never be a negative number

Pull Request - State: closed - Opened by akolliasAMD over 1 year ago

#778 - Disable Colltrace for --fast option

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago

#777 - Bump rocm-docs-core from 0.13.4 to 0.14.0 in /docs/sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: noCI, dependencies

#776 - ASAN build excluding additional files

Pull Request - State: closed - Opened by arvindcheru over 1 year ago - 2 comments

#775 - device: use unroll factor based on platforms

Pull Request - State: closed - Opened by nusislam over 1 year ago

#774 - Enable --fast

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago

#773 - improve compilation time and create timetrace plot

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago - 1 comment

#772 - Update Read the Docs, documentation, and dependabot

Pull Request - State: closed - Opened by samjwu over 1 year ago
Labels: noCI, dependencies

#771 - Wall clock update and npkit trace script Update

Pull Request - State: closed - Opened by akolliasAMD over 1 year ago

#770 - Updating NOTICES.txt and LICENSE.txt

Pull Request - State: closed - Opened by gilbertlee-amd over 1 year ago

#769 - Cherry pick documentation dependency updates for rocm release 5.6 branch

Pull Request - State: closed - Opened by samjwu over 1 year ago
Labels: noCI, noExtendedCI, dependencies

#768 - resolving the pthread-gtest linking issue for rccl-UnitTests

Pull Request - State: closed - Opened by PedramAlizadeh over 1 year ago

#767 - Add NCCL_NCHANNELS_PER_PEER override

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#766 - Bump cryptography from 40.0.2 to 41.0.0 in /docs/.sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#765 - Bump rocm-docs-core from 0.11.0 to 0.13.3 in /docs/.sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#764 - Bump rocm-docs-core from 0.11.0 to 0.13.2 in /docs/.sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#763 - add DMA_BUF support

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago

#762 - Removing init_nvtx.cc from source list

Pull Request - State: closed - Opened by gilbertlee-amd over 1 year ago

#761 - Rework barrier and event code

Pull Request - State: closed - Opened by wenkaidu over 1 year ago

#760 - Bump rocm-docs-core from 0.11.0 to 0.13.1 in /docs/.sphinx

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#759 - add DMA_BUF support

Pull Request - State: closed - Opened by BertanDogancay over 1 year ago - 1 comment