Fix OpenSSL linkage by using the final install-directory in the build #1065
Conversation
This file contains 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
This causes linking failures on most platforms because the pkg-configs are still pointing to the location specified in prefix (as they should). This should address rust-lang#1051 but I'll add a regression test in the next commit
This is the test suggested by @malbarbo in his PR: rust-lang#1054
|
@jelford I checked the test it it does not print the message when it fails. The script uses # check that rustup-init was built with ssl support
# see https://github.com/rust-lang-nursery/rustup.rs/issues/1051
if ! (nm target/$TARGET/release/rustup-init | grep Curl_ssl_version &> /dev/null); then
echo "Missing ssl support!!!!"
exit 1
fi |
|
Thanks @malbarbo , incorporated your change. |
|
We want this!!! |
|
📌 Commit 3b7ada1 has been approved by |
bors
added a commit
that referenced
this pull request
Apr 15, 2017
Fix OpenSSL linkage by using the final install-directory in the build This PR addresses #1051 by avoiding moving OpenSSL's install-target directory after it's been configured. It also encorporates the regression test suggested by @malbarbo on #1054. It still makes sense to move directories about to avoid getting a partially-built copy of openssl, and I think it makes sense to cache the finished product rather than the src/build directory. I haven't been able to test the output on one of the affected platforms (although the check on symbols passes) as I don't know a good way to get artefacts out of the travis build.
|
💔 Test failed - status-appveyor |
|
Could we get this merged soon, hopefully? It's quite annoying not to be able to update, right now. |
|
@Diggsey this is going to start to get more urgent with 1.17 shipping. Any timeline on when we might be able to get something pushed out? Sorry to nag. |
|
Okay, thanks for the update, no problem. Sorry for all this.
…On Fri, 28 Apr 2017, 10:38 Diggory Blake, ***@***.***> wrote:
@jelford <https://github.com/jelford> There's nothing I can do about it
unfortunately. @brson <https://github.com/brson> is aware of the issue
and is going to make a new release ASAP, but he's been away and I imagine
has had a ton of stuff to catch up since getting back.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1065 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAMhKcEonTTUtJNLHU4NyU1p_Tf_FHIgks5r0bObgaJpZM4M9n2L>
.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This PR addresses #1051 by avoiding moving OpenSSL's install-target directory after it's been configured. It also encorporates the regression test suggested by @malbarbo on #1054.
It still makes sense to move directories about to avoid getting a partially-built copy of openssl, and I think it makes sense to cache the finished product rather than the src/build directory.
I haven't been able to test the output on one of the affected platforms (although the check on symbols passes) as I don't know a good way to get artefacts out of the travis build.