293f0fb752
Use a map to host scratch buffers ( #1004 )
...
* Use a map to host scratch buffers
* Address review feedbacks. Deliberately keep mscclSetupScratch function.
2023-12-05 13:15:28 -06:00
bc44e3faa7
Fix gcnArch bug in IFC mix build ( #998 ) ( #1002 )
...
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com >
2023-12-04 16:20:22 -06:00
7c0f49a878
IFC mix build ( #998 )
2023-12-02 18:49:52 -07:00
4ba65d1d6a
Increase max channles to 64 ( #993 )
2023-12-01 16:01:11 -08:00
0b53f79196
Fix GPU to NIC mapping for peertopeer ( #994 )
2023-12-01 08:00:17 -08:00
e44e112a17
Fix mscclAlgoHandle not initialized issue ( #995 )
2023-12-01 07:58:01 -08:00
4bb0b4a380
Move MSCCL algorithm loading to initialization to workaround HIP graph conflict ( #982 )
...
* MSCCL: pre-specify channels and pre-load algorithms
* add mutex
* fix bug
* clean include
* disable all-gathers temporarily
2023-11-30 09:47:20 -08:00
56ce9ef05f
recreated pr 914 to work with current develop branch ( #979 )
2023-11-28 16:33:47 -07:00
50b2dd9fd7
Add special handling of gfx940 ( #976 )
...
* Add special handling of gfx940
* Update ring base
2023-11-22 15:07:36 -08:00
b6d7438d31
Merge remote-tracking branch 'origin/master'
2023-11-20 05:07:23 -08:00
569d3f7d59
msccl: allocate scratch as ext-scope fine-grained ( #968 )
2023-11-16 09:57:25 -06:00
bc8661f092
Fix kernel command line warnings ( #961 )
...
* Fix kernel command line warnings
* Remove while loop
2023-11-15 18:01:12 -08:00
7fc891bc8d
Fix MSCCL work FIFO allocation with HIP graph enabled ( #967 )
2023-11-15 16:43:28 -08:00
198f14923b
Check to support older ROCm versions ( #963 )
2023-11-15 12:36:31 -07:00
7ae95db5b8
Optimize MSCCL all-gather algorithms for gfx942 ( #964 )
2023-11-15 08:18:59 -08:00
df128879a6
Optimize MSCCL reduce primitive switching for gfx942 ( #962 )
...
* Optimize reduce primitive switching for gfx942
* address comment
2023-11-15 08:18:44 -08:00
cece6415b0
Fix use of CPUID overwriting registers in use.
...
CPUID writes to EAX, EBX, ECX, and EDX so the inline-asm must state that.
Otherwise currently in-use register might get overwritten which may
cause all kinds of failures like segfaults or wrong results.
Alternatively `__cpuid` can be used which avoids this and related issues.
So do that as suggested in the GCC issue https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112513
2023-11-14 12:38:02 +01:00
88d44d777f
2.19.4-1
...
Split transport connect phase into multiple steps to avoid port
exhaustion when connecting alltoall at large scale. Defaults to 128
peers per round.
Fix memory leaks on CUDA graph capture.
Fix alltoallv crash on self-sendrecv.
Make topology detection more deterministic when PCI speeds are not
available (fix issue #1020 ).
Properly close shared memory in NVLS resources.
Revert proxy detach after 5 seconds.
Add option to print progress during transport connect.
Add option to set NCCL_DEBUG to INFO on first WARN.
2023-11-13 10:36:12 -08:00
5a800e00cd
msccl: enable basic collective trace ( #959 )
...
To avoid increasing number of kernels, colltrace is only enabled with
RCCL_MSCCL_FORCE_FULLOPS=1
2023-11-08 20:14:28 -08:00
efc42d9045
Use send instead of sendWithBarrier. ( #727 )
2023-11-07 13:47:24 -06:00
022735d208
Merge pull request #950 from nusislam/msccl-red2
...
msccl: remove cases from numReduction switch statement
2023-11-04 02:48:03 -05:00
dbcba2923b
Use parallel init of LDS and adjust P2P channels for gfx94x ( #943 )
...
* Use parallel init of LDS and adjust P2P channels for gfx94x
* Move another init to parallel
* Fix NCCL_NCHANNELS_PER_PEER setting
2023-11-03 16:06:49 -07:00
f545b94d4b
msccl: remove cases from numReduction switch statement
2023-11-03 16:56:51 -05:00
bb84345943
msccl: use 32-bit LDS access and add RCCL_MSCCL_FORCE_FULLOPS ( #953 )
2023-11-03 10:38:02 -07:00
988efe605a
MSCCL stream fix ( #948 )
2023-11-03 09:10:52 -06:00
f484ff17b9
msccl: add templated kernel ( #945 )
...
* msccl: add templated kernel
* Use defines to improve code readability
* Fix kernel indexing and review feedback
2023-11-02 17:21:53 -07:00
6b80a0d0d4
msccl: remove dereference of reduce args
...
It can be removed because the msccl kernel will never execute this code
according to the current msccl setup.
2023-11-02 13:20:00 -05:00
a7400218a2
msccl: use atomic to set dependency flags ( #941 )
2023-10-31 14:46:57 -07:00
a497722894
NPkit: misc fixes for MSCCL ( #936 )
...
* msccl: add xcc_id to timestamp sync
* NPKit: add timestamp for rrc operator
* NPKit: add timestamp for MSCCL init
2023-10-30 10:00:12 -07:00
1e5ca6820b
Fix gcnArchName bug in topology dump ( #937 )
...
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com >
2023-10-28 12:30:36 -05:00
4c117e5335
Fix MSCCL work FIFO out-of-bound issue ( #935 )
2023-10-27 11:24:52 -07:00
96ec3ffe2e
SRC/INIT: fix typo for ENABLE_PROFILING ( #934 )
...
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com >
2023-10-26 23:52:46 -05:00
f22df90e5c
remove gcnArch support ( #920 )
...
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com >
2023-10-26 12:09:15 -05:00
fb0eccb57b
msccl: reduce debug output when using NCCL_DEBUG=INFO ( #932 )
2023-10-25 08:05:19 -07:00
c4e65fd382
Add missing gfx942 support ( #927 )
2023-10-23 12:04:37 -07:00
dbb5611a3a
Remove LDS based software barriers from MSCCL ( #923 )
2023-10-19 16:39:41 -05:00
4278a9918b
Update rome models ( #922 )
2023-10-18 17:28:01 -07:00
39812ce757
NPKit: add xcc_id field ( #918 )
2023-10-13 15:24:59 -07:00
1b80d041cb
Fix incorrect arch name parsing ( #916 )
2023-10-13 10:01:11 -07:00
6d0b5c1e89
Port init_once fix from NCCL ( #915 )
2023-10-13 08:01:12 -07:00
7ee5c1c28b
Change MSCCL kernel signature to allow kernel arguments be preloaded via SGPR ( #911 )
...
* Adding a script that will download/compile/run TransferBench/RCCL/UCX/RCCL-tests/RCCL-Unittests/hip-mpi-testsuite (#895 )
Co-authored-by: Pedram Alizadeh <pmohamma@banff-pla-r27-05.pla.dcgpu >
* Only build gfx941
* demo
* fine tune malloc
* Fix merge errors
* Fix merge errors
* Disable parallel build
* Adopt --amdgpu-kernarg-preload-count
* Revert "Adding a script that will download/compile/run TransferBench/RCCL/UCX/RCCL-tests/RCCL-Unittests/hip-mpi-testsuite (#895 )"
This reverts commit f5e252dddf02a41b4d1bc512f306f45f97166304.
* Revert CMake changes.
* NPKIT changes.
* Remove some license declarations.
* Address code review feedbacks on msccl_kernel_impl.h
* Update CMakeLists.txt
* Add CMake logic to check the existence of --amdgpu-kernarg-preload-count
* Fix NPKIT trace logic.
---------
Co-authored-by: Pedram Alizadeh <pmohamma@amd.com >
Co-authored-by: Pedram Alizadeh <pmohamma@banff-pla-r27-05.pla.dcgpu >
Co-authored-by: Ziyue Yang <ziyyang@microsoft.com >
2023-10-12 20:17:08 -05:00
a6ff4618c7
Revert "Remove 2H4P condition from P2P channels adjustment ( #890 )" ( #904 )
...
This reverts commit 16dd05a58a
2023-10-04 09:46:11 -06:00
28d7fe5629
Dma buf support optin ( #905 )
...
* dmaBufSupport Optin added on every part of the code that should invoke it
2023-10-03 03:17:48 -06:00
c1f57a7041
Modify All-To-All doc ( #896 )
...
* Modify All-To-All doc
* Update nccl.h.in
* update unit-tests
---------
Co-authored-by: gilbertlee-amd <44450918+gilbertlee-amd@users.noreply.github.com >
2023-09-27 12:45:21 -04:00
8c6c595185
2.19.3-1
...
H800/H100 fixes and tuning.
Re-enable intra-process direct pointer buffer access when CUMEM is
enabled.
2023-09-26 05:57:15 -07:00
3435178b6c
Merge remote-tracking branch 'origin/master' into v2.19
2023-09-26 05:55:56 -07:00
f9c3dc251e
2.19.1-1
...
Add local user buffer registration for NVLink SHARP.
Add tuning plugin support.
Increase net API to v7 to allow for device-side packet reordering;
remove support for v4 plugins.
Add support for RoCE ECE.
Add support for C2C links.
Better detect SHM allocation failures to avoid crash with Bus Error.
Fix missing thread unlocks in bootstrap (Fixes #936 ).
Disable network flush by default on H100.
Move device code from src/collectives/device to src/device.
2023-09-26 05:50:33 -07:00
4365458757
Fix cudaMemcpyAsync bug
...
We are trying to use the copy result of first cudaMemcpyAsync in the
second cudaMemcpyAsync without sync in between. This patch fixes it
by allocating a CPU side array to cache device side addr so that we
can avoid this consecutive cuda mem copy.
Fixes #957
2023-09-20 05:51:14 -07:00
b85d73c02e
changed the form that RCCL_TREE uses ( #888 )
...
* changed the form that RCCL_TREE uses
2023-09-15 15:01:33 -06:00
26e982d913
Reduce NPKit latency overhead in MSCCL kernel ( #893 )
...
* Reduce NPKit latency overhead in MSCCL kernel
* Fix build error without NPKit enable
2023-09-15 13:28:26 -07:00
Wen-Heng (Jack) Chung
Nilesh M Negi
Bertan Dogancay
Wenkai Du
pradeep-ramanna
Ziyue Yang
akolliasAMD
Sylvain Jeaugey
Alexander Grund
Nusrat Islam
Kaiming Ouyang