It seems that rusti is not stable enough to be part of normal
integration tests; completely unrelated changes are causing it to fail
more often than not, and bors is backing up with changes that are
otherwise fine.
I discussed with folks on IRC and the consensus seemed to be disabling
rusti tests until LLVM gets upgraded, since it's going to bring a lot of
JIT fixes. If everyone else is ok with that, I'll land a change that
disables them.
(I'm a little concerned about rusti bitrotting again, but am not sure
what else we can do in the short term. The failures all seem pretty
opaque -- lockups and crashes.)
-Graydon
_______________________________________________
Rust-dev mailing list
[email protected]
https://mail.mozilla.org/listinfo/rust-dev
