yezhengmao/nccl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • ca330b110a Add scan tests v1.3.0-1 Sylvain Jeaugey 2016-09-20 11:52:15 -07:00
  • 6c77476cc1 Make tests check for deltas and report bandwidth Sylvain Jeaugey 2016-09-20 11:48:34 -07:00
  • cabd6848e4 Heavy code refactoring to remove a lot of code in collectives (~1000 lines). Sylvain Jeaugey 2016-09-22 11:57:56 -07:00
  • e3dbc6110e Add profiling API Sylvain Jeaugey 2016-09-22 11:56:51 -07:00
  • 1d6715fe20 Fix MPI test path Sylvain Jeaugey 2016-09-20 11:43:31 -07:00
  • 9ee6189bf9 Merge pull request #41 from jia-kai/master Sylvain Jeaugey 2016-09-15 09:45:52 -07:00
  • 939b0a4297 Merge pull request #45 from NVIDIA/cw-update-copyright-year Sylvain Jeaugey 2016-08-26 15:44:00 -07:00
  • 234c8c9ef3 Update LICENSE.txt Cliff Woolley 2016-08-26 15:39:21 -07:00
  • 75bad643bd Updated LICENCE.txt Sylvain Jeaugey 2016-08-26 15:08:20 -07:00
  • 47b0797fe1 pass devlist as const int* rather than int* in ncclCommInitAll jiakai 2016-08-19 19:00:14 +08:00
  • ed401cc29b link library with -lrt; otherwise there is undefined reference to shm_open jiakai 2016-08-19 18:55:41 +08:00
  • b3a9e1333d Remove unneeded deb build script Sylvain Jeaugey 2016-07-27 17:58:00 -07:00
  • 428ec5b2a3 Merge remote-tracking branch 'github/master' into public Sylvain Jeaugey 2016-07-25 10:53:01 -07:00
  • 55c42ad681 Fixed redundant contexts in multi-process apps Nathan Luehr 2016-07-22 17:29:13 -07:00
  • 7a1aa6b563 Improved Deb generation Sylvain Jeaugey 2016-07-07 16:13:52 +02:00
  • 9ae84f5d6b Fix version number Sylvain Jeaugey 2016-06-16 17:07:42 -07:00
  • e51e922924 Add a debug level to NCCL and CUDA versions at init Sylvain Jeaugey 2016-06-16 16:50:14 -07:00
  • 9fcc523485 Increased version to 1.2.3 Sylvain Jeaugey 2016-06-15 19:18:13 -07:00
  • 67d1ab9106 Packaging : Generate shlibs.local Sylvain Jeaugey 2016-06-15 19:01:29 -07:00
  • da6d2009e0 Move deb to build directory Sylvain Jeaugey 2016-06-15 18:02:03 -07:00
  • 155132d336 Fix make install to use BUILDDIR Sylvain Jeaugey 2016-06-15 17:26:47 -07:00
  • 08ddfe03d2 Rework debian packaging Sylvain Jeaugey 2016-06-15 16:54:57 -07:00
  • 5d4716a8a3 Include link to blog post in README.md Sylvain Jeaugey 2016-06-15 10:53:43 -07:00
  • 4d9188a818 Preparing for 1.2.3 rebuild v1.2.3-1+cuda8.0 Boris Fomitchev 2016-06-13 02:35:30 -07:00
  • aa8f669a3d Updating for .deb rebuild v1.2.3-1+cuda7.5 Boris Fomitchev 2016-06-13 01:48:59 -07:00
  • d5e507fc7f Only call the CUDA runtime. That may fix #27. Sylvain Jeaugey 2016-06-07 16:27:51 -07:00
  • 620491a649 Merge remote-tracking branch 'github/master' into HEAD Sylvain Jeaugey 2016-06-06 14:35:57 -07:00
  • 7edfc57228 Make NCCL collectives work on communicators with only one rank Sylvain Jeaugey 2016-06-06 14:35:00 -07:00
  • bd3cf73e6e Changed CURAND generator to work on a wider set of platforms. Sylvain Jeaugey 2016-06-06 14:34:03 -07:00
  • 560f3ffc31 Gencodes changed to NV recommended v1.2.2-1+cuda8.0 Boris Fomitchev 2016-06-05 23:51:02 -07:00
  • 63ecbff473 Fixed version in ChangeLog Boris Fomitchev 2016-04-21 16:41:25 -07:00
  • 177505b757 Gencodes changed to NV recommended v1.2.2-1+cuda7.5 v1.2.1-2+cuda7.5 Boris Fomitchev 2016-06-05 23:51:02 -07:00
  • 9d9d8cd59f Bump to 1.2.2 Sylvain Jeaugey 2016-06-03 17:21:53 -07:00
  • 1657af1567 Better name for GENCODE Sylvain Jeaugey 2016-06-03 10:22:46 -07:00
  • acb93d1aed Removing unneeded includes Sylvain Jeaugey 2016-06-02 17:33:43 -07:00
  • 889ad3d4e6 Makefile improvements Sylvain Jeaugey 2016-06-02 15:01:03 -07:00
  • 93538def65 Merge pull request #22 from borisfom/master v1.2.1-1+cuda7.5 Boris Fomitchev 2016-04-21 18:58:44 -07:00
  • 0a0bbd6064 Fixed version in ChangeLog v1.2.1-1+cuda8.0 Boris Fomitchev 2016-04-21 16:41:25 -07:00
  • e5067b6611 Fixed version in ChangeLog Boris Fomitchev 2016-04-21 16:28:13 -07:00
  • 0629fb62d7 Merge pull request #21 from borisfom/master Boris Fomitchev 2016-04-21 14:46:41 -07:00
  • 4eef2ae63b version bumped v1.2.1-1+cuda8.0-1 Boris Fomitchev 2016-04-21 14:21:38 -07:00
  • f2ed4b0f2d conf added Boris Fomitchev 2016-03-17 18:27:03 -07:00
  • f9f8c71b7c Started 8.0 branch Boris Fomitchev 2016-03-17 18:21:07 -07:00
  • 0177cf3ea4 Fixed install location, new .deb version Boris Fomitchev 2016-04-21 10:00:24 -07:00
  • 658aca1469 Merge pull request #17 from Hopobcn/master Nathan Luehr 2016-04-21 13:25:18 -07:00
  • 03df4c7759 Moved no-as-needed flag to link rule. Nathan Luehr 2016-04-19 14:51:03 -07:00
  • 0d4f8f4e95 Merge pull request #18 from apaszke/master Nathan Luehr 2016-04-19 11:11:39 -07:00
  • ddd3f2084d Fix readme to reflect the new test paths Sylvain Jeaugey 2016-04-19 11:09:25 -07:00
  • dba3ec9428 Fix random deadlock during ncclCommInitRank. Sylvain Jeaugey 2016-04-19 10:47:27 -07:00
  • 9de361a1b9 Fix MPI test usage Sylvain Jeaugey 2016-04-19 10:43:38 -07:00
  • c0c959b1be Add --no-as-needed to make sure that cudart library gets liked Adam Paszke 2016-04-13 10:04:38 -04:00
  • e30bf95989 Enable compilation with old g++ when the default g++ is not supported (+5.0) Pau Farré 2016-04-12 12:49:13 +02:00
  • b16cc5d197 Merge pull request #16 from borisfom/master v1.2.1-1+cuda7.5-1 Boris Fomitchev 2016-03-17 17:35:04 -07:00
  • e6f4a83da6 Removing Tegra Boris Fomitchev 2016-03-17 17:25:27 -07:00
  • 1a8bae5b2f fixed version format Boris Fomitchev 2016-03-17 17:13:45 -07:00
  • e8eb285a59 Merge pull request #15 from borisfom/master Boris Fomitchev 2016-03-17 16:03:05 -07:00
  • b508d28123 Version with . 7.5 Boris Fomitchev 2016-03-17 15:48:48 -07:00
  • 62b551798f Use arch=5.3 as well Boris Fomitchev 2016-03-16 23:09:36 -07:00
  • dfbebe395c Delete libnccl1_1.1.1+cuda75_amd64.deb Boris Fomitchev 2016-03-16 21:44:13 -07:00
  • 85280b5bf4 Delete libnccl-dev_1.1.1+cuda75_amd64.deb Boris Fomitchev 2016-03-16 21:44:04 -07:00
  • fb53cfd9b0 Added files via upload Boris Fomitchev 2016-03-16 21:42:47 -07:00
  • 92d2123d8d Added compute 5.3 Boris Fomitchev 2016-03-16 19:24:48 -07:00
  • ec3de28ae5 Preparing for pbuild Boris Fomitchev 2016-03-16 19:23:49 -07:00
  • 86dc136fa9 Moved to pbuilder Boris Fomitchev 2016-03-16 18:32:53 -07:00
  • 172f316ac2 Moved release files to proper area Boris Fomitchev 2016-02-29 21:55:43 -08:00
  • 941d9da08c Updated package version, added manpage Boris Fomitchev 2016-02-29 12:10:34 -08:00
  • 5554a4c9f0 Fixed useRemoteRecv consistency issue. Nathan Luehr 2016-02-18 11:59:54 -08:00
  • 9442285526 Fixed buffer overflow in ReduceOrCopy Nathan Luehr 2016-02-11 12:59:31 -08:00
  • caa40b8dd3 Libwrap checks for LIB.so.1 if LIB.so not found Nathan Luehr 2016-01-21 16:45:19 -08:00
  • 2758353380 Added NCCL error checking to tests. Nathan Luehr 2016-01-21 16:30:05 -08:00
  • fe1a956715 Enabled support for char type to be unsigned. Nathan Luehr 2016-01-21 15:23:28 -08:00
  • c05312f151 Moved tests to separate dir and improved MPI test Sylvain Jeaugey 2016-01-20 14:18:25 -05:00
  • 5966316771 Added support for more than 8 GPUs. Nathan Luehr 2016-01-20 18:19:45 -08:00
  • 130ee246e2 Fixed deadlock in back-to-back reduce_scatters. Nathan Luehr 2016-01-20 17:58:25 -08:00
  • 90af7c73ef Merge pull request #6 from lukeyeager/deb Nathan Luehr 2016-01-07 13:06:28 -08:00
  • 3251681207 Merge branch 'yangky11-patch-1' Nathan Luehr 2016-01-06 16:46:22 -08:00
  • d332c41e71 fix a typo in README.md Kaiyu Yang 2015-12-24 00:01:02 +08:00
  • c9da89254b Update deb packaging scripts Luke Yeager 2015-12-18 14:05:37 -08:00
  • eb2d869f71 Merge pull request #5 from lukeyeager/tests-nvml Nathan Luehr 2015-12-18 13:36:20 -08:00
  • f1e92fe2a3 Added Debian packaging files Boris Fomitchev 2015-12-18 13:35:22 -08:00
  • b5400c54df Don't link tests with NVML Boris Fomitchev 2015-12-18 13:26:51 -08:00
  • a4de6016f8 Merge pull request #4 from lukeyeager/build-sm50 Nathan Luehr 2015-12-18 13:23:48 -08:00
  • 4807909e3f Merge pull request #3 from lukeyeager/semver Nathan Luehr 2015-12-18 13:22:19 -08:00
  • dd0884b707 Build SM 5.0 code Luke Yeager 2015-12-18 13:12:09 -08:00
  • e1634ca6cb Use semantic versioning Luke Yeager 2015-12-18 12:00:30 -08:00
  • 651a6edc5c Fixed bug in MPI initialization. Nathan Luehr 2015-12-10 17:54:41 -08:00
  • ada5edce88 Merge pull request #1 from slayton58/int64_uint64 Nathan Luehr 2015-12-10 17:22:50 -08:00
  • 41ce4ca9fc Add int64 and uint64 types for all algorithms and tests Simon Layton 2015-12-03 16:35:54 -05:00
  • 27d32ac5d9 Fixed a race condition in reduce and braodcast. Nathan Luehr 2015-11-19 11:11:52 -08:00
  • 0673d5f44f Initial release. Nathan Luehr 2015-11-17 11:30:40 -08:00