yezhengmao/nccl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
232 Commits 1 Branch 63 Tags
Commit Graph

232 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sylvain Jeaugey
469b69a5d0 Fix #163 : remove warnings 2018-12-11 09:19:16 -08:00
Ke Wen
8606cdb8b2 Fix dummy plugin 2018-12-05 17:25:23 -08:00
Sylvain Jeaugey
57368189e1 Remove error logging from a normal path 
When initNet fails, we should not print the backtrace as it is
supposed to be normal operation (falling back to sockets)
 2018-12-04 14:47:41 -08:00
Sylvain Jeaugey
4b39a4cf91 Fix GPU Direct RDMA detection. 
Whether the network supported GPU Direct RDMA or not was ignored,
causing sockets to break when cards were local enough that NCCL
tried to use it.
 2018-12-04 14:42:28 -08:00
Sylvain Jeaugey
b8a9a32ccb Add NCCL_NET flag to many debug lines. 2018-12-04 13:10:19 -08:00
Sylvain Jeaugey
cdae05b277 Improve INFO message when external network is not found. 
Fix #162
 2018-12-04 12:10:58 -08:00
David Addison
5fe2618c0e Fixed some compilation errors when TRACE=1 set 2018-11-29 14:12:14 -08:00
Sylvain Jeaugey
eed8218e17 Rework shared memory code to use SYSCHECK macros. 
This is to handle EINTR/EGAIN properly (issue #137), and also
make the code consistent with the rest.

Unfortunately posix_fallocate and mmap do not follow the classic
return code/errno pattern, so we need to write wrappers around those
functions.
 2018-11-29 12:52:13 -08:00
Sylvain Jeaugey
302d538b73 Rework SYSCHECK macros to better handle retries. 
SYSCHECKVAL was not retrying when a retry was needed. Since not all
calls are inside a loop, that means we could silently miss an
EINTR/EAGAIN return code.

Also rework the socket connection code and improve error reporting.
 2018-11-29 12:52:13 -08:00
Sylvain Jeaugey
61b50a63ef Improve net API description 2018-11-26 16:24:31 -08:00
Sylvain Jeaugey
98adf2fe11 Make network isend/irecv non blocking 2018-11-26 16:24:31 -08:00
Sylvain Jeaugey
0d3a20f96d Add support for external network. 
Dynamically load external network from libnccl-net.so.
Add init function in networks.
Move PCI scoring to net.cu, only ask transport to provide a path.
Simplify CUDA PCI path detection.
Add dummy external network
 2018-11-26 16:24:31 -08:00
Alex Sergeev
d7a58cfa58 Generate host-hash for P2P and SHM based on $(readlink /proc/self/ns/uts) + $(readlink /proc/self/ns/mnt) (#156) 2018-11-19 17:39:44 -08:00
Sylvain Jeaugey
3c6e25210b
Generate nccl.h in build instead of src 
Generating nccl.h in src makes source directories dirty after builds.
 2018-11-09 14:00:41 -08:00
Ke Wen
21d9a877be Add official builds download link 2018-11-08 11:22:28 -08:00
Sylvain Jeaugey
f7d31919d7
Add instructions to install packaging toolchain 
Address #143 and #150 : debuild not installed.
 2018-11-05 11:42:33 -08:00
Sylvain Jeaugey
bed43524cc Add install target 
Fix issue #145
 2018-11-05 09:53:59 -08:00
David Addison
b56650c7f5 2.3.7-1 
Improved LL tuning for multi-node jobs.
Improved bootstrap for large job scaling.
Fixed a hang during bootstrap due to socket reuse.
Added operation name to the COLL INFO logging.
v2.3.7-1 		2018-10-24 14:44:59 -07:00
Obihörnchen
3202d6b393 Fix nccl-tests all_reduce_perf path 
It's `all_reduce_perf` not `allreduce_perf`
 2018-10-14 00:53:17 -07:00
Sylvain Jeaugey
f93fe9bfd9 2.3.5-5 
Add support for inter-node communication using sockets and InfiniBand/RoCE.
Improve latency.
Add support for aggregation.
Improve LL/regular tuning.
Remove tests as those are now at github.com/nvidia/nccl-tests .
v2.3.5-5 		2018-09-25 14:12:01 -07:00
Sylvain Jeaugey
286916a1a3
Merge pull request #119 from sclarkson/master 
Fix tests: call cudaHostUnregister on the host pointer instead of the device pointer.
 2017-11-28 18:41:26 -08:00
sclarkson
680a35c6b7 fix tests on maxwell 2017-11-11 19:22:06 -08:00
Sylvain Jeaugey
03d856977e Update README to link to NCCL2 2017-08-04 09:44:37 -07:00
Sylvain Jeaugey
4a33f66e27 Update README to link to NCCL2 part 3 2017-08-04 09:44:09 -07:00
Sylvain Jeaugey
d66fb63679 Update README to link to NCCL2 #2 2017-08-04 09:43:29 -07:00
Sylvain Jeaugey
80ae43b443 Update README to link to NCCL2 2017-08-04 09:42:25 -07:00
Sylvain Jeaugey
29a1a916dc Add support for CUDA9 half semantics 2017-06-14 11:20:24 -07:00
Sylvain Jeaugey
ccfc4567dc Merge pull request #78 from ilya-biryukov/master 
Fix compilation error when compiling with 'clang -x cuda'.
 2017-04-04 09:47:52 -07:00
Boris Fomitchev
649f04d077 Added Pascal nvcc flags, bumped version v1.3.4-1 2017-03-24 11:58:14 -07:00
Ilya Biryukov
8241cd7b6e Fix compilation error when compiling with 'clang -x cuda'. 
Functions vFetch and vStore are not found by ADL with clang,
so they need to be declared before usage in ReduceCopy.
 2017-03-16 12:01:11 +01:00
Sylvain Jeaugey
7fef264bfa Bumping version to 1.3.3 2017-03-01 16:44:27 -08:00
Nathan Luehr
8996811936 Only enable peer access for ring neighbors. 
This enables support for systems with more than 9 GPUs attached to a single PCIe root complex.
 2017-03-01 16:42:38 -08:00
Sylvain Jeaugey
c219a183d0 Fix copy/paste typo in error message 2017-03-01 16:42:38 -08:00
Sylvain Jeaugey
8e1d6f9b60 Fix crash in Reduce when non-root ranks have invalid recvbuff 2017-03-01 16:42:38 -08:00
Sylvain Jeaugey
024d1e2678 Merge pull request #69 from cwhipkey/master 
Qualify nullptr_t with std::
 2017-02-08 09:17:50 -08:00
Chad Whipkey
5eab428294 Qualify nullptr_t with std::. 2017-02-08 07:06:31 -08:00
Sylvain Jeaugey
2a974f5ca2 Fix 1.3.2 compilation 2016-12-08 09:11:43 -08:00
Sylvain Jeaugey
648e9fbb58 Adding missing file 2016-12-05 18:06:24 -08:00
Sylvain Jeaugey
34d27771c6 1.3.2 release 
Broadcast tuning
Better checking of inputs
Copy/reduce code simplification
 2016-12-01 15:17:50 -08:00
Sylvain Jeaugey
1093821c33 Replace min BW by average BW in tests 2016-12-01 15:16:35 -08:00
Sylvain Jeaugey
ddddfba1c0 Merge pull request #54 from peterhj/peterhj-staticlib 
Add a static library target "staticlib" to the Makefile.
 2016-11-28 09:15:39 -08:00
Peter Jin
5765d608cc Add a static library target "staticlib" to the Makefile. 
Rename the static library "libnccl_static.a" to disambiguate from the
dynamic libraries.
 2016-11-24 11:31:03 -08:00
Kyle Fernandes, ne Jacobs
c2c515516b Remove irrelevant output from ncclReduce Fortran tests 2016-11-21 10:18:04 -08:00
Kyle Fernandes, ne Jacobs
9c18468fe2 Add Copyright header to Fortran bindings source files 2016-11-21 10:17:58 -08:00
Kyle Fernandes, ne Jacobs
5f2b32e45b Add Fortran bindings 2016-11-17 15:33:34 -08:00
Sylvain Jeaugey
534b9a1697 Bump to 1.3.1 2016-10-13 10:33:05 -07:00
Sylvain Jeaugey
b2781d0501 Fix primitives function prototype 2016-10-13 10:32:42 -07:00
Sylvain Jeaugey
bf7d1514f7 NVML (libwrap) : import the needed definitions 2016-10-13 10:28:59 -07:00
Sylvain Jeaugey
8bb06c94be Improved allreduce segmentation for small sizes 2016-10-07 12:42:23 -07:00
Sylvain Jeaugey
ca330b110a Add scan tests v1.3.0-1 2016-09-22 11:58:33 -07:00