3ac98e7d39
Merge pull request #188 from wenkaidu/prim_test
...
rccl-prim-test: auto-detect rings in 4P and 8P configurations
2020-04-14 09:52:49 -07:00
ef7064ba9b
rccl-prim-test: auto-detect rings in 4P and 8P configurations
2020-04-10 18:17:21 +00:00
fa52d4f0aa
Merge pull request #187 from aaronenyeshi/fix-hip-vdi-hsa-ext
...
Fix HIP-Clang build with HSA headers
2020-04-03 19:06:38 -04:00
a95090d981
Fix HIP-Clang build with HSA headers
...
HIP-Clang does not include these HSA headers, and they need to be explicitly added in RCCL.
2020-04-03 17:58:23 -04:00
3cbe5c8a40
Merge pull request #186 from wenkaidu/v2.6.4
...
Merge with NCCL 2.6.4
2020-04-02 10:42:01 -07:00
6f54b23503
topo_expl: update to 2.6
2020-04-01 13:37:08 -07:00
fa36fd9ef9
Merge remote-tracking branch 'nccl/master' into v2.6.4_merge
2020-04-01 13:35:12 -07:00
533e3702cf
Merge pull request #314 from NVIDIA/v2.6
...
2.6.4-1
2020-03-26 17:31:24 -07:00
b221128eca
2.6.4-1
...
Add support for network collectives.
Add support for XML topology dump/injection.
Add text values for GDR and P2P Levels, including "NVL".
Add speed detection for PCI, Infiniband and Ethernet cards.
Add CPU detection for ARM and AMD CPUs.
Add support for adaptive routing on Infiniband.
Change NET plugin API to v3 : merge PCI path and GPU pointer
capability into a single structure and add other properties.
2020-03-20 14:58:36 -07:00
6c61492eba
Check return code for Flush operation
...
Current NCCL code does not abort for failed Flush operations by
underlying network. This may compromise data integrity.
Signed-off-by: Rashika Kheria <rashika@amazon.com >
2020-03-16 20:40:59 -07:00
ebc823e603
rccl-prim-test: add all-to-all benchmark ( #185 )
...
For gfx908, support simple detection of ring topology.
Call ReduceOrCopyMulti directly from kernel.
Also simplify code by removing kernel start synchronization option
which has no effect on throughput measurements.
2020-03-16 10:00:54 -07:00
b9fb0cd808
set hip::host and hip::device and remove some deprecated targets ( #184 )
2020-03-05 13:36:55 -07:00
0976e47b06
Merge pull request #183 from wenkaidu/dup_rings
...
Remove condition for ring duplication
2020-03-02 17:12:42 -08:00
88752f9173
Merge pull request #182 from wenkaidu/topo_expl
...
Topo expl
2020-03-02 15:44:09 -08:00
d27ae9407c
Merge pull request #181 from wenkaidu/rdma
...
Check fine grained memory before enabling RDMA
2020-03-02 13:24:55 -08:00
62dc28bd2e
Remove condition for ring duplication
...
Fix insufficent number of rings on single node after pull #179
2020-03-02 12:55:06 -08:00
32388d60a9
topo_expl: add a few more single node models
2020-03-02 11:43:03 -08:00
fb59328a7b
Check fine grained memory before enabling RDMA
...
Adding back the check which was lost from 2.5 merge.
2020-03-02 11:18:27 -08:00
8b5bc8bca2
Merge pull request #179 from wenkaidu/search
...
Use fraction of system maxWidth as steps for searching
2020-02-28 11:05:46 -08:00
bbbe3dd628
Merge pull request #180 from wenkaidu/2.5.7
...
Sync up with NCCL 2.5.7
2020-02-27 15:36:01 -08:00
8e73a2ad60
Merge remote-tracking branch 'remotes/nccl/master'
2020-02-27 12:53:03 -08:00
63af7ab5cb
Merge pull request #178 from wenkaidu/topo_visual
...
Add topology visualizer tool
2020-02-27 09:31:34 -08:00
498d5029ad
Add topology visualizer tool
2020-02-26 15:23:34 -08:00
ae834349ec
Merge pull request #177 from wenkaidu/topo_expl
...
topo_expl: use bandwidth numbers defined in graph in CPU models
2020-02-26 14:37:45 -08:00
934b6de557
topo_expl: use bandwidth numbers defined in graph in CPU models
2020-02-26 14:17:36 -08:00
67b4bea41f
Merge pull request #176 from wenkaidu/rome_bw
...
Revise PCI BW numbers on Rome
2020-02-26 13:47:39 -08:00
d2adc61bf6
Revise PCI BW numbers on Rome
2020-02-26 13:17:49 -08:00
8391637613
Use fraction of system maxWidth as steps for searching
...
This reverts previous workaround of deducting only half of width
from paths.
2020-02-26 09:14:35 -08:00
9bf1edb554
Merge pull request #175 from wenkaidu/ll_abort
...
Fix abort handling in LL primitives
2020-02-25 14:12:33 -08:00
077c3cda74
Fix abort handling in LL primitives
2020-02-25 13:42:54 -08:00
1569f93f26
Merge pull request #174 from wenkaidu/max_speed
...
Fix system maxSpeed and maxWidth calculation
2020-02-24 15:42:24 -08:00
9b80b3633f
Fix system maxSpeed and maxWidth calculation
2020-02-24 15:18:57 -08:00
8fb18e0ae7
Merge pull request #173 from wenkaidu/cr8
...
Fix incorrect CR8 detection
2020-02-21 10:24:41 -08:00
f54dc58113
Fix incorrect CR8 detection
...
Also change level of ring graph print to help debugging
2020-02-21 10:09:49 -08:00
5b3856f2ed
Merge pull request #172 from wenkaidu/topo_expl
...
Add topology explorer
2020-02-20 15:16:55 -08:00
55f8e2dec7
Add topology explorer
2020-02-19 14:42:06 -08:00
9110820470
Merge pull request #167 from wenkaidu/cr8
...
Generate 8G6L chordal ring from reference
2020-02-18 14:59:23 -08:00
428f1f1555
Fix hipclang argument in CI ( #171 )
2020-02-18 13:17:52 -07:00
edb863de62
Refactor Jenkinsfiles to allow use of new docker containers ( #170 )
2020-02-18 11:25:29 -07:00
c38f174bd4
Fix Allgather operations above 4G with multiple GPUs per process.
...
Fixes nccl-tests#37.
Direct offsets were still on 32 bits in the low-level primitives.
2020-02-12 11:11:55 -08:00
abcfbf1231
Generate 8G6L chordal ring from reference
2020-02-11 22:01:12 +00:00
f4ba41e80a
Merge pull request #166 from wenkaidu/hcc_ver
...
Bump up HCC version for -hc-function-calls switch
2020-02-11 13:20:50 -08:00
3d092f32b8
Bump up HCC version for -hc-function-calls switch
2020-02-11 19:37:13 +00:00
d1dae2721d
Add ring bandwidth correction factor
2020-01-30 09:52:27 -08:00
20fa04d9b6
Updating copyright notices for 2020.
2020-01-29 15:28:08 -08:00
fe6d012eb0
Merge remote-tracking branch 'remotes/rccl/master' into rccl_2.5.6_cleanup
2020-01-29 15:28:03 -08:00
486fd436af
Split primitive class to smaller structures
2020-01-29 15:27:23 -08:00
1e55645d97
Misc fixes and improvements for 2.5.6
...
1. Fix RCCL unit test
2. Add ROME detection and tuning
3. Change default P2P level
4. Fix search algorithm for XGMI
5. Remove explicit channel duplication with implicit by using half of link speed
6. Add collective trace support
7. Correct Intel Skylake CPU detection and bandwidth
8. Fix topo connect function
9. Disable GDR read and remove unreachable code
10. Disable LL128 kernels
11. Add tuning parameters
12. Use original clock64() implementation which returns RTC counter value
13. Print out timestamp of collective trace
14. Do not use struct ncclColl in kernel launch parameter
15. Fix abort handling and add tracing
17. Add __launch_bounds__ to kernel functions
18. Remove unused abortCount
19. Unset default MIN_NRINGS and MIN_NCHANNELS
20. Do not allocate shared memory when not using LL128 kernels
21. Correct time print out in tuning log
2020-01-29 15:27:05 -08:00
3701130b3c
2.5.7-1
2020-01-16 15:40:57 -08:00
44c34e5d10
Merge pull request #283 from lukeyeager/topo-trim-net-links
...
Topo trim net links
2020-01-16 15:40:36 -08:00
Wenkai Du
Wenkai Du
Aaron Enye Shi
Sylvain Jeaugey
Rashika Kheria
amdkila
Eiden Yoshida
Stanley Tsang