-
Notifications
You must be signed in to change notification settings - Fork 12.7k
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
x86_64 to i686 cross compile: down-with-thread-dtors.stage2 SEGVs only under valgrind #31328
Comments
This is also reproducible on a completely i686 toolchain and environment (only the processor is x64, edit: also the kernel), see Debian bug 812825 and some build logs: I currently have an uncleaned build tree with the failing compiled test files intact, in case you have further ideas on how to debug this. (I already tried setting |
I can reproduce this outside of the bootstrap as well. By downloading the i686 linux nightlies and compiling this tests I get the same valgrind error. I'm not really sure if it's legitimate or not, as it only causes weird behavior in valgrind... |
Here's the proc-mem map from vgdb https://paste.debian.net/394367/ |
It's unclear to me whether this test failing under valgrind is actually legit. The test only fails in valgrind when everything is dynamically linked, and it appears to work when statically linked. For now just add the `// no-prefer-dynamic` directive and let's just chalk it up to a weird valgrind issue. Closes rust-lang#31328
It's unclear to me whether this test failing under valgrind is actually legit. The test only fails in valgrind when everything is dynamically linked, and it appears to work when statically linked. For now just add the `// no-prefer-dynamic` directive and let's just chalk it up to a weird valgrind issue. Closes #31328
Hm, shouldn't we at least file a bug upstream to valgrind? |
Perhaps, yeah, but I'm also not confident that it's a bug in valgrind, so it may be premature to do so |
I can confirm that it fixed the test failure on Debian i386 and the build is now green on this arch. |
Also fixed on my rust-nightly ubuntu builds. However, I'd suggest that we file this to valgrind even if we're "not sure" if it's a bug, since otherwise this is going to get lost to history - the issue is "closed" and nobody looks at these afterwards. |
Sure, go ahead @infinity0 ! |
Alright I filed https://bugs.kde.org/show_bug.cgi?id=359705 feel free to add details, I'm not an expert in this topic. |
This applies the same fix as added in 595d5b2 which is to just compile all valgrind tests statically instead of dynamically. It looks like this is a resurgence of either rust-lang#30383 or rust-lang#31328 in some weird fashion. I'm not actually sure what's going on with the bots, and it's unclear whether this is a valgrind bug or a libstd bug, but for now let's just get things landing again. Closes rust-lang#30383
Not even sure if this is a real bug or not, but it's 100% reproducible in my environment, even after
make clean
.The text was updated successfully, but these errors were encountered: