Skip to content

Issues: gpudirect/libgdsync

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.

Sign up for GitHub

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

25 Open 32 Closed
25 Open 32 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

could testcase run on one node?
#108 opened Jan 6, 2023 by hongbilu
2
Support spectrum MPI enhancement
#94 opened Feb 17, 2021 by pakmarkthub
Optimize gds_mlx5_post_send enhancement
#90 opened Feb 3, 2020 by pakmarkthub
gds_prepare_send should support all types of send_flags enhancement
#89 opened Feb 3, 2020 by pakmarkthub
gds_prepare_send should support all types of opcode enhancement
#88 opened Feb 3, 2020 by pakmarkthub
Code style violation enhancement
#83 opened May 20, 2019 by pakmarkthub
@pakmarkthub
get rid of _exp_ types in core.h enhancement
#79 opened May 1, 2019 by drossetti
@pakmarkthub
1
detect and enable IBV_EXP_PEER_OP_POLL_NOR_DWORD_CAP
#73 opened Jun 7, 2018 by drossetti
fix assumption in gds_ordinal_from_device() implementation bug
#67 opened May 22, 2018 by drossetti
provide optimized APIs accepting GPU device handles or ordinals enhancement
#66 opened May 22, 2018 by drossetti
in gds_kernel_latency !peersync case does not work bug
#64 opened May 17, 2018 by drossetti
1
hpgmg and gds_kernel_latency errors on DGX TeslaV100 bug
#61 opened May 11, 2018 by e-ago
4
registration cache is not correctly maintained bug
#54 opened Apr 18, 2018 by drossetti
flusher can be optimized away in certain cases enhancement
#53 opened Apr 14, 2018 by drossetti
1
GPU context bug in gds_fill_poll when GDS_ENABLE_FLUSHER=1 bug
#52 opened Apr 14, 2018 by drossetti
@drossetti
Question about multi streams operate the same QP and CQ question
#48 opened Apr 9, 2018 by Hannah-xxl
6
Improve, in case of wait on stream (SA model), multiple receives from different peers (CQs) using CUDA threads enhancement
#39 opened Mar 5, 2018 by e-ago
Improve, in case of send on stream (SA model), multiple sends to different peers (QPs) using CUDA threads enhancement
#38 opened Mar 5, 2018 by e-ago
properly fence sends to same QP bug
#28 opened Jun 16, 2017 by drossetti
efficiently support COPY_BLOCK for kernel-side API enhancement
#24 opened May 31, 2017 by drossetti
provide a service to long-running GPU kernels to flush incoming RDMA traffic
#23 opened May 31, 2017 by drossetti
@e-ago
use verbs res_domain to obtain QPs with dedicated blueflame space
#22 opened May 31, 2017 by drossetti
3
lack of context initialization in case of host memory bug
#18 opened May 24, 2017 by e-ago
add APIs to abort outstanding work related to QP and CQ enhancement
#12 opened May 19, 2017 by drossetti
Support RoCE protocol in tests
#10 opened May 17, 2017 by ksang
2
ProTip! no:milestone will show everything without a milestone.