-
Notifications
You must be signed in to change notification settings - Fork 567
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[OpenBLAS] Add v0.3.20 #4485
[OpenBLAS] Add v0.3.20 #4485
Conversation
55555dc
to
7c48cda
Compare
7c48cda
to
6a2ab20
Compare
@@ -0,0 +1,19 @@ | |||
diff --git a/kernel/arm64/KERNEL.NEOVERSEN1 b/kernel/arm64/KERNEL.NEOVERSEN1 |
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.
@Keno: OpenMathLib/OpenBLAS#3052 supposedly fixed the issue you reported at OpenMathLib/OpenBLAS#2998, did you ever get a chance to verify whether we can drop this patch now?
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.
No idea
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.
For the record, I verified the patch should now be redundant, as vanilla OpenBLAS now produces the right result: #7202 (comment).
When building for x86_64 with GCC and libgfortran3 (so GCC 6):
OpenMathLib/OpenBLAS#3530 looks relevant, it appears GCC headers in v6 were wrong. This wouldn't even be the first time we hit similar issues when building OpenBLAS, see #2142, #2385, #2389, #2415. Sigh. Edit: for reference, this should be the change which introduced the new intrinsics in GCC: gcc-mirror/gcc@167a5b7 |
No description provided.