-
Notifications
You must be signed in to change notification settings - Fork 352
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
Windows support in the test suite + appveyor #334
Conversation
Indeed they're getting out of hand. I am not very in favor of this PR. What is even the problem with |
It tries to execute the final binary. Which doesn't work in the cross compile case. We hacked this away by adding a binary runner |
compiletest just mirrors upstream rustc. Do you think they would accept a patch adding a mode they don't use? A less invasive solution may be to change the Or we just fork compiletest-rs... |
nonono, it breaks every now and then. We don't want yet another randomly breaking dependency. If we scrap the |
What's it missing to handle dependencies? |
|
Oh oh... now the fullmir tests are failing... This will be a pain... no mir for |
I will prepare a quick PR against upstream Rust to see what they think about a |
waiting for rust-lang/rust#44537 |
With #343 landed, it should now be possible to do this without new hacks, right? |
Yes, I rebased over master to do those changes myself, and saw that you already did them XD Then I started getting the github mail notfications. I'll probably disable the fullmir tests on windows for now though |
What's the matter with those? |
At minimum rust-lang/rust#44537 This isn't an issue on linux, becaus memchr is a libc function there, but on windows libstd has its own Rust impl. |
|
Will this be entirely fixed by your PR? I can imagine that it may still do byte-wise accesses which would still break when there are pointers in the buffer. |
Strings usually don't contain pointers ;) It'll fix it in the use cases we have in CI |
appveyor.yml
Outdated
test_script: | ||
- set RUST_BACKTRACE=1 | ||
- cargo build | ||
- cargo test |
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.
What about using --locked --release
like we do for Travis?
(Maybe --locked
fails due to platform-specific dependencies, but at least --release
should help speeding things up a little. If it doesn't slow down compliation too much.)
Yay, everything passes. We don't have appveyor CI automatically running though |
Let's see if appveyor works |
🎉 it does |
We seriously need to give compiletest a
compile-pass
test kind... These hacks are getting out of hand