Olli Saarikivi
|
9fbb0debdd
|
C++ API changes
|
2023-04-19 22:02:23 +00:00 |
|
Olli Saarikivi
|
83c7ba1afb
|
C++ API working, allgather_test_cpp passing
|
2023-04-19 17:11:21 +00:00 |
|
Olli Saarikivi
|
46790d79e8
|
Implement C API buffer registration support
|
2023-04-14 23:20:42 +00:00 |
|
Olli Saarikivi
|
a0f1d36026
|
Start HostConnection implementation
Add declarations on the C-side for functions to enable multiple buffer
registrations per connection.
|
2023-04-14 15:57:47 +00:00 |
|
Olli Saarikivi
|
45172bec88
|
Implement mscclpp::Communicator using C-style API
|
2023-04-14 14:21:53 +00:00 |
|
Madan Musuvathi
|
c042d9af54
|
Merge branch 'cpp-api' into saemal/api-extension
|
2023-04-13 22:32:38 +00:00 |
|
Olli Saarikivi
|
0eec1d438b
|
Move over C++ API work to new branch
|
2023-04-13 18:38:38 +00:00 |
|
Changho Hwang
|
dd0883b84f
|
Lint
|
2023-04-12 09:25:35 +00:00 |
|
Changho Hwang
|
63a5be6953
|
Move ibQp to mscclppHostIBConn
|
2023-04-12 09:20:05 +00:00 |
|
Changho Hwang
|
bc729cd481
|
Move MRs / MR infos to mscclppHostIBConn & cleanup
|
2023-04-12 09:05:42 +00:00 |
|
Changho Hwang
|
fd3f928108
|
remove hostFifo & rename devFifo to just fifo
|
2023-04-12 08:08:19 +00:00 |
|
Madan Musuvathi
|
9124856ea4
|
first version hostConn
|
2023-04-12 01:36:06 +00:00 |
|
Saeed Maleki
|
459c56d3cb
|
not compiling -- wip for hostfunctions
|
2023-04-11 22:52:47 +00:00 |
|
Changho Hwang
|
7a0e64813a
|
Add fifo for host connections
|
2023-04-11 12:28:45 +00:00 |
|
Changho Hwang
|
35acdf796c
|
Add mscclppProxyFifo
|
2023-04-11 11:28:40 +00:00 |
|
Changho Hwang
|
d2c2ae72a7
|
Some cleanup
|
2023-04-11 08:45:22 +00:00 |
|
Changho Hwang
|
69b5bdfd13
|
minor fix
|
2023-04-11 05:01:39 +00:00 |
|
Saeed Maleki
|
1bfa8d4034
|
creating a base conn class
|
2023-04-11 04:05:44 +00:00 |
|
Saeed Maleki
|
ee6c2deb44
|
Merge branch 'main' into saemal/api-extension
|
2023-04-11 01:43:13 +00:00 |
|
Saeed Maleki
|
b6179224aa
|
lint
|
2023-04-11 01:36:37 +00:00 |
|
Saeed Maleki
|
48102a0858
|
removing unnecessary flags
|
2023-04-11 01:22:40 +00:00 |
|
Changho Hwang
|
a1ae982c61
|
Merge signalEpochId with proxySignalEpochId
|
2023-04-10 14:05:25 +00:00 |
|
Saeed Maleki
|
426e78997c
|
name changes + documentation for clarity
|
2023-04-09 02:20:54 +00:00 |
|
Ziyue Yang
|
3796f5251e
|
Merge branch 'ziyyang/npkit-fix-numa' of https://github.com/microsoft/mscclpp into ziyyang/npkit-fix-numa
|
2023-04-08 07:13:08 +00:00 |
|
Ziyue Yang
|
748d3d1596
|
separate flag and data
|
2023-04-08 07:12:46 +00:00 |
|
Saeed Maleki
|
3d2a3a3b3a
|
Merge branch 'main' into ziyyang/npkit-fix-numa
|
2023-04-08 06:37:58 +00:00 |
|
Saeed Maleki
|
f3f53a4148
|
lint
|
2023-04-08 06:32:57 +00:00 |
|
Saeed Maleki
|
e336c93dc9
|
Merge branch 'main' into binyli/mscclpp-test
|
2023-04-08 06:30:48 +00:00 |
|
Ziyue Yang
|
f68eeba2d4
|
change clock collection approach
|
2023-04-08 05:29:34 +00:00 |
|
Crutcher Dunnavant
|
34464b40bb
|
register buffers
|
2023-04-07 19:11:50 -07:00 |
|
Saeed Maleki
|
ec83a27e83
|
wip
|
2023-04-08 01:57:22 +00:00 |
|
Felipe Petroski Such
|
38cd87cdcc
|
add memory region functions
|
2023-04-07 15:38:48 -07:00 |
|
Changho Hwang
|
949a9cd0a3
|
Optional use of gdrcopy (#48)
Co-authored-by: Saeed Maleki <saemal@microsoft.com>
|
2023-04-07 13:36:59 +08:00 |
|
Ziyue Yang
|
352a10a33d
|
NPKit: improve event collection for async requests (#45)
|
2023-04-06 16:21:34 +08:00 |
|
Saeed Maleki
|
08275e93d7
|
added barrier API + pushed one after mscclppsetup
|
2023-04-06 03:15:54 +00:00 |
|
Saeed Maleki
|
0887cfe768
|
no need for remapping anymore
|
2023-04-02 02:35:08 +00:00 |
|
Saeed Maleki
|
4c6616e7b9
|
lint
|
2023-04-01 19:20:50 +00:00 |
|
Saeed Maleki
|
8927dd4d72
|
great allgather numbers with the current binding mechanism
|
2023-04-01 18:54:42 +00:00 |
|
Binyang Li
|
8e4edd4d83
|
fix lint
|
2023-03-31 08:32:45 +00:00 |
|
Binyang Li
|
af5825b474
|
bind numa node to communicator
|
2023-03-31 08:05:49 +00:00 |
|
Saeed Maleki
|
44e8760af7
|
allgather kernel2
|
2023-03-31 06:31:25 +00:00 |
|
Changho Hwang
|
b58eae4037
|
Minor changes
|
2023-03-30 07:11:41 +00:00 |
|
Saeed Maleki
|
e2cfd5ac83
|
a lot of documentation
|
2023-03-30 00:37:33 +00:00 |
|
Saeed Maleki
|
be5e422021
|
merged with main
|
2023-03-29 23:03:12 +00:00 |
|
Saeed Maleki
|
629d59a9c0
|
bug fix -- flush doesn't need to increment the epoch
|
2023-03-29 22:21:24 +00:00 |
|
Saeed Maleki
|
debd110874
|
fused flush instructions
|
2023-03-29 22:17:02 +00:00 |
|
Saeed Maleki
|
42b11c5c9a
|
fast flush
|
2023-03-29 20:50:01 +00:00 |
|
Saeed Maleki
|
d97bee6973
|
flush mechanism
|
2023-03-29 17:31:20 +00:00 |
|
Binyang2014
|
62279b0063
|
Add mscclppSetBootstrapConnTimeout (#34)
|
2023-03-28 14:01:56 +08:00 |
|
Saeed Maleki
|
fa26bdd9fc
|
no gdr copy anywhere in the code except for the files that are not compiled
|
2023-03-28 05:40:40 +00:00 |
|