-
Notifications
You must be signed in to change notification settings - Fork 868
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
README: Note about ld issue for XL and PGI on PPC #3886
README: Note about ld issue for XL and PGI on PPC #3886
Conversation
* Related to Issue open-mpi#2606 and Issue open-mpi#3075 * The core problem in those two issues is related to a regression in ld upstream. Add a note in the README about this issue. Signed-off-by: Joshua Hursey <[email protected]> (cherry picked from commit 1c6a253) Signed-off-by: Joshua Hursey <[email protected]>
README
Outdated
@@ -188,6 +188,16 @@ Compiler Notes | |||
for more details: | |||
https://github.com/open-mpi/ompi/issues/3612 | |||
|
|||
- Compiling Fortran programs using the mpi_f08 module on PowerPC with | |||
the PGI (tested 17.5) or XL (tested v15.1.5) Fortran compilers and GNU | |||
linker after 2.25.1 and before 2.28 will likely experience runtime failures. |
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.
Per IM discussion with @jjhursey, I think that this above sentence is trying to say that compiling MPI applications that utilize the mpi_f08
module with the ... compilers on the ... platforms with the GNU linker ... will likely experience runtime failures when running the MPI applications. Even that language is a little awkward -- the key points here are:
- compiling Fortran MPI applications that utilize
mpi_f08
(vs. compiling Open MPI itself) - problems occur when you run the MPI application (vs. when compiling or linking Open MPI and/or the application)
README
Outdated
version of ld. This GNU linker regression will be fixed in version 2.28. | ||
Below is a link to the GNU bug on this issue: | ||
https://sourceware.org/bugzilla/show_bug.cgi?id=21306 | ||
The XL compiler will have a fix for this issue in their next release. |
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.
- s/fix/workaround/ because you just said the bug is in the GNU ld, so XL isn't fixing the bug -- it's working around the bug
- s/their/its/ because you're not anthropomorphizing the XL compiler 😄
Can you cite a specific version number of the next XL release? "Next" is an relative (i.e., amorphous) word to use in a README file that will continue to live on the web for years.
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.
How about instead of "in their next release" I just say "in a future release"?
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.
👍 That works -- that's a little more intentionally saying "some future version, but I'm specifically not telling you which version" (Vs. "next", which might force a reader to look at the date on this README/release and try to reverse-map it back to XL versions).
@jjhursey Any progress, perchance? |
Signed-off-by: Joshua Hursey <[email protected]>
@jsquyres I pushed a new commit to clarify the language. Let me know what you think. I figure we can iterate on the additional commit until we get the language set to how we like it. Then we can discuss how to update the language in all of the necessary release branches (this was already merged into |
ld upstream. Add a note in the README about this issue.
[skip ci] bot:notest