-
Notifications
You must be signed in to change notification settings - Fork 901
configure: do look for sed #10400
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
configure: do look for sed #10400
Conversation
Interesting, I'm not sure how this ever worked. |
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.
This must be getting invoked implicitly somehow in at least some cases. With AC 2.71 on my Mac, I see this in configure output (before this PR is applied):
checking for a sed that does not truncate output... /usr/bin/sed
Maybe that doesn't always happen in all versions of Autoconf...? 🤷♂️
Regardless, it does not hurt to explicitly invoke this to ensure that $SED
is set.
Answering my own question: I see what's happening. Yes, something is invoking this implicitly: Putting |
That being said, putting the
While that's certainly correct, it would be 17.2% nicer if that sed test output was below the |
We could just move the |
Refs. open-mpi#10392 Signed-off-by: Gilles Gouaillardet <[email protected]>
aee141f
to
2012196
Compare
Agreed, BTW, how did you end up with 17.2%? Is there a reference I missed? |
It was a complex set of calculations involving many mathematical disciplines, including linear algebra, calculus, hokus pokus, and imaginarium. Mostly the last one. Actually, entirely the last one. 😜 |
Should this also go to v4.0.x and v4.1.x? |
I just checked: it does not appear to be necessary on v4.0.x or v4.1.x. 🎉 |
Interesting - is sed not used on those branches? |
I didn't track down why it isn't necessary. But I checked both branches, and the same erroneous output (indicating that Something must be different on the v4.x branches -- either the Fortran tests don't use |
Refs. #10392
Signed-off-by: Gilles Gouaillardet [email protected]