-
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
"Could not compile advapi32-sys
on a network-mapped drive
#28002
Comments
Can you get a backtrace with |
|
The code was being compiled on a network-mapped drive in PowerShell if that helps at all. |
Ah, that would do it. Network mapped drives mean you don't see the results of your actions until some time later. Which also means that after doing something, like deleting a file, there is a short period in which that file hangs in limbo until the action is processed. In this case Rust is probably assuming the file is immediately available, but that's clearly false, and so it panics. The way to fix this would be to find every place where rustc works with files and make it loop a few times to try to open the file while sleeping between attempts. |
What does I-ICE mean? Sorry I'm new to Rust and not very knowledgeable about GitHub in general. |
Internal Compiler Error. There used to be a wiki about such things, not sure where that knowledge lives now. Edit: Here's a backup anyway: https://github.com/rust-lang/rust-wiki-backup/blob/7097b584ebcfe514df646347fcda1f0e12390614/Note-tag-label-names-and-definitions.md |
@horatius83 are you still seeing this? |
This issue should probably remain open until someone has thoroughly tested building rust code on a network mapped drive and verified that there are no issues. |
I am still observing, yes. Is there any particular build or anything I should test for my particular use-case? |
advapi32-sys
advapi32-sys
on a network-mapped drive
I have been reviewing many older issues and trying to ascertain their applicability, whether they should be redirected, closed, left open, etc. Usually, I feel comfortable making a quick investigation and then deciding to leave it open or closed based on findings plus what seems to have been emerging consensus, and trying to document consensus. But with this one, I am at a bit of a loss. So here, I formally am requesting a judgement from T-compiler on whether this issue, and indeed this class of issue, "Let's keep this open until we can be certain no issues are present in a known-faulty environment", is actionable or desirable, whether it should be broken down into smaller steps or simply cast to the winds, or even whether the changes over the past seven and a half years have made this issue functionally impossible to follow up on (e.g. due to implementing the incremental compilation system, which will tend to dominate over other kinds of filesystem issues). Or, perhaps, if those changes have given this one increased relevance. |
I should also note there are, at a glance, many other network filesystem, shared partition, or ramdisk issues that have caused other interesting ICEs, some of which may be related and some of which are probably not, e.g. |
Discussed at T-compiler triage. We hypothesize:
In the absence of data to guide us into which one of the above four cases we are in, we are going to speculative close this issue. If you see this comment, and are someone that is running into this issue, please do open a new issue (with lots of info to help us know more about your specific scenario), or leave a comment and reopen this issue. |
I attempted to run the command:
cargo build --verbose
and got the following error:
This is my Cargo.toml file:
Meta
rustc --version --verbose
:Backtrace:
The text was updated successfully, but these errors were encountered: