You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I assert that if we had just one autobuild instance, on our fastest platform (linux?), that tested this scenario, then we could keep developers like me happy (as well as a number of other developers; I'll gather a list if I have to)
The text was updated successfully, but these errors were encountered:
pnkfelix
added
the
T-infra
Relevant to the infrastructure team, which will review and decide on the PR/issue.
label
Feb 16, 2018
This should effectively be done so long as we don't remove this commit: e78ecd2. Most of our docker images (i.e., Travis) build out of tree in obj dir anyway so they all test building outside root dir. However, only a few pass paths to x.py invocations. Those should test this; I'm going to close as such.
I'm hoping to come up with a better approach to testing rustbuild soon, which will then include something for this too directly in rustbuild. However, for now, I think we can close this as fixed.
Supposedly we support bootstrapping rust from directories other than the root of the rust.git checkout.
At least, in the past we have rejected PR's based on the desirability for such support. See for example: #42689 (comment)
But bugs like #48268 nonetheless creep through.
I assert that if we had just one autobuild instance, on our fastest platform (linux?), that tested this scenario, then we could keep developers like me happy (as well as a number of other developers; I'll gather a list if I have to)
The text was updated successfully, but these errors were encountered: