This repository was archived by the owner on Nov 17, 2023. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 6.7k
[MXNET-115] USE_LAPACK is forced on all platforms with OpenBLAS and c… #11094
Merged
Merged
Changes from all commits
Commits
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There is no separate liblapack for openblas as far as I understood. Is there for other libraries other than OpenBLAS? As far as I can tell CI checks only for OpenBLAS.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't get your comment, I tried to fix the logic as I understood the previous logic was working around adding these flags in the windows case and breaking other platforms. Can you elaborate?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
list(APPEND mxnet_LINKER_LIBS lapack)line adds liblapack to required linker libraries. But there is no such think as liblapack for OpenBLAS, at least not for our builds. The CI passed in this case because with this change-DUSE_LAPACK=OFFbecomes effective.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
And there is a chance that liblapack exists for other blas libraries (atlas?) but we are not testing them in CI currently
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
With your change you need to set
-DUSE_LAPACK=ONin arm builds not to loose lapack functionality.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just create symlinks such as libcblas and liblapack to the compiled libopenblas. There’s no need to hack around.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok, this might be an option
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the clarification, I got now what was the concern. We need better ARM / ANDROID architecture detection when cross compiling in CMake. This might be something we can definitely refine later. For the moment I'm fine with either solution proposed by @lebeg or @szha I think we will need to refine and improve this logic on future PRS.