Kernel mapping error resolve #42466
Kernel mapping error resolve #42466
Conversation
Aaraviitkgp
force-pushed
the
kernel_mapping_error_resolve
branch
from
9623bfe to
04e27cb
Compare
Aaraviitkgp
force-pushed
the
kernel_mapping_error_resolve
branch
from
5a4a08a to
e328d0c
Compare
|
@Rocketknight1 I noticed that the run test issues that is occuring ColQwen2ForRetrieval: Tensor vlm.lm_head.weight seems to have nothing to do with the changes made in pr. |
|
cc @MekkCyber for kernels I think! also @Aaraviitkgp that |
|
@Rocketknight1 the test_processor and test_tokenization have similar issue they fail but no changes in pr related to them can you look into that aswell 😁. |
MekkCyber
left a comment
MekkCyber
left a comment
There was a problem hiding this comment.
Thanks for fixing this !
Aaraviitkgp
force-pushed
the
kernel_mapping_error_resolve
branch
from
47249c9 to
21eda55
Compare
MekkCyber
left a comment
MekkCyber
left a comment
There was a problem hiding this comment.
Thanks a lot! just one nit and good to go
Aaraviitkgp
force-pushed
the
kernel_mapping_error_resolve
branch
from
0ea638d to
b744a09
Compare
MekkCyber
left a comment
MekkCyber
left a comment
There was a problem hiding this comment.
Thanks a lot for iterating !
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
4 participants
issue: #42451
This PR fixes a bug in KernelConfig.create_compatible_mapping where defining a kernel mapping for multiple devices (e.g., both cuda and rocm) caused the last defined device to overwrite the active device configuration.
@Rocketknight1 @MekkCyber