Skip to content
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

install script doesn't recognize --docdir option #36451

Closed
kamathba opened this issue Sep 13, 2016 · 13 comments
Closed

install script doesn't recognize --docdir option #36451

kamathba opened this issue Sep 13, 2016 · 13 comments
Assignees
Labels
P-high High priority regression-from-stable-to-beta Performance or correctness regression from stable to beta.

Comments

@kamathba
Copy link

ran ./configure && make && sudo make install

...
gen-install-script: CFG_LEGACY_MANIFEST_DIRS := rustlib,cargo
gen-install-script:
gen-install-script: validating arguments
gen-install-script:
install: error: Option '--docdir=/usr/local/share/doc/rust' is not recognized
/home/banjo/Work/dev/rust/mk/install.mk:21: recipe for target 'install' failed
make: *** [install] Error 1

Seems to be introduced by this commit, which was merged yesterday.

commit 5e9149d
Author: orbea ovariegata@yahoo.com
Date: Thu Sep 8 23:18:20 2016 -0700

Allow setting --docdir
@nodakai
Copy link
Contributor

nodakai commented Sep 14, 2016

Had we tested make install on buildbot... #19698

@m10e
Copy link

m10e commented Sep 14, 2016

This is preventing installs for me too. Not sure if it matters, but I got it building a cross compiler from master branch with custom install dir (configure --prefix).

@andoriyu
Copy link
Contributor

@m10e you can always apply patch to rust-installer before it's merged upstream.

@m10e
Copy link

m10e commented Sep 14, 2016

@andoriyu Sure, so it's definitely not urgent. I was just weighing in with a +1. Thanks for the tip though!

@orbea
Copy link

orbea commented Sep 15, 2016

Ideally the merged pull request should of only been merged when both when both pull requests were ready to be merged followed by updating the submodule. Unfortunately it did not work out that way, sorry for any issues I have caused...

@cesarb
Copy link
Contributor

cesarb commented Sep 26, 2016

I believe this should be tagged with the label regression-from-stable-to-nightly. @alexcrichton ?

@alexcrichton alexcrichton added the regression-from-stable-to-nightly Performance or correctness regression from stable to nightly. label Sep 27, 2016
@alexcrichton
Copy link
Member

Ah, didn't realize it was a regression, but in that case indeed!

@cesarb
Copy link
Contributor

cesarb commented Sep 27, 2016

@alexcrichton Yes, make install on nightly is completely broken. If like me you follow Rust's master by doing ./configure --prefix=... && make && make install, you have to revert commit 5e9149d, otherwise the make install step will fail.

The issue is that, since that commit, install.sh is called by make install with the nonexistent option --docdir, and Rust's install.sh doesn't ignore unknown options.

@cesarb
Copy link
Contributor

cesarb commented Sep 27, 2016

Related: #29632

@alexcrichton
Copy link
Member

cc @orbea this may be caused by #36360, mind helping to investigate?

@pnkfelix pnkfelix added regression-from-stable-to-beta Performance or correctness regression from stable to beta. and removed regression-from-stable-to-nightly Performance or correctness regression from stable to nightly. labels Sep 29, 2016
@tikue
Copy link
Contributor

tikue commented Sep 30, 2016

Is there a workaround for this currently? The latest nightly ICEs on code that previously compiled (I think due to #36816) so I wanted to compile from source with the merged fix, but now I'm hitting this.

@cesarb
Copy link
Contributor

cesarb commented Sep 30, 2016

@tikue Yes, the workaround is to revert commit 5e9149d. There will be a conflict when reverting, but it's easy to resolve. Nothing depends on that commit (and in fact it's broken), so reverting it causes no problems.

@brson brson self-assigned this Oct 4, 2016
@brson brson added P-high High priority and removed I-nominated labels Oct 4, 2016
@brson
Copy link
Contributor

brson commented Oct 5, 2016

#36971

bors added a commit that referenced this issue Oct 5, 2016
@bors bors closed this as completed in ee0d833 Oct 5, 2016
alexcrichton pushed a commit to alexcrichton/rust that referenced this issue Oct 10, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
P-high High priority regression-from-stable-to-beta Performance or correctness regression from stable to beta.
Projects
None yet
Development

No branches or pull requests

10 participants