This repository was archived by the owner on Jan 15, 2026. It is now read-only.

mulit-GPU and CUDA Stream Support - #60

Open

rubber-duck-debug wants to merge 32 commits into

mainfrom

multigpu_and_streams

Commits on Apr 29, 2024

Add CI config files
Prashanth Kanduri
committed
72ce285

Commits on May 2, 2024

added stream management to OPSA
rubber-duck-debug
committed
d463d21
x
rubber-duck-debug
committed
d81ec9c
added support for other ops.
rubber-duck-debug
committed
c8d8776
formatting.
rubber-duck-debug
committed
0b06e79
Merge branch 'main' into multigpu_and_streams
rubber-duck-debug
committed
de5a3da

Commits on May 3, 2024

fixed build issues.
rubber-duck-debug
committed
b2c170a
fixed issue with opsaw instantiation
rubber-duck-debug
committed
f78161c
formatting.
rubber-duck-debug
committed
e89350b
add comments on the test job script
Prashanth Kanduri
committed
ea54e9a
insignificant commit for retriggering pipeline
Prashanth Kanduri
committed
1bb35aa

Commits on May 4, 2024

Merge branch 'main' into cscs_ci
frostedoyster
authored
497d811
Install tox
frostedoyster
authored
b9be358

Commits on May 6, 2024

attempt addressing CI compiler errors and warnings
Prashanth Kanduri
committed
d9fc283
Merge branch 'cscs_ci' of github.com:lab-cosmo/mops into cscs_ci
Prashanth Kanduri
committed
5b04a18
remove polynomial order zero case to avoid divide by zero issue
Prashanth Kanduri
committed
af2ec60
correct address of the atomic add operations in sap
Prashanth Kanduri
committed
9fee2a1
specify addresses with indices for atomic add operations to address CI errors
Prashanth Kanduri
committed
fc80766

Commits on May 7, 2024

dummy implementation for polynomial order zero
Prashanth Kanduri
committed
2ed5771

Commits on May 10, 2024

removed bug in CUDA HPE.
rubber-duck-debug
committed
62b73d8
header change
rubber-duck-debug
committed
b6d4795
HPE divide by zero fix.
rubber-duck-debug
committed
68223e2
added in code for pre-sm60 atomicAdd(doubles)
rubber-duck-debug
committed
3b4527f
macro to switch out the atomicAdds depending on ARCH
rubber-duck-debug
committed
e3c70c2
documentation.
rubber-duck-debug
committed
ade1cd3
changed macro to device function.
rubber-duck-debug
committed
263984d
Merge branch 'cscs_ci' into multigpu_and_streams
rubber-duck-debug
committed
e654289
fixed sap cstream
rubber-duck-debug
committed
899ea94
formatting.
rubber-duck-debug
committed
99cedf5
missing device guard in SAP
rubber-duck-debug
committed
21b6235
missing stream.
rubber-duck-debug
committed
dd29ac0

Commits on May 12, 2024

Add stream benchmark
frostedoyster
committed
7f92034