flock: Use fcntl constants directly from libc crate on Unix targets #57121

glaubitz · 2018-12-25T12:02:13Z

Since the values for the fcntl constants can vary from architecture
to architecture, it is better to use the values defined in the libc
crate instead of assigning literals in the flock code which would
make the assumption that all architectures use the same values.

Fixes #57007

rust-highfive · 2018-12-25T12:02:17Z

r? @varkor

(rust_highfive has picked a reviewer for you, use r? to override)

glaubitz · 2018-12-25T20:30:25Z

This also requires two small fixes in the libc crate: rust-lang/libc#1186

nagisa · 2018-12-26T00:05:40Z

This also requires two small fixes in the libc crate: rust-lang/libc#1186

liblibc is a submodule in rust tree, so you will have to update the submodule after the referenced PR lands and before this can/should land.

glaubitz · 2018-12-26T00:07:13Z

Thanks for the heads-up! I will open a PR for that, too.

nagisa · 2018-12-26T00:16:54Z

Would be fine to just have it all be in the same PR (i.e. this PR), unless you’re in a hurry to land this regardless of whether the liblibc change langs.

glaubitz · 2018-12-26T00:19:29Z

You mean this PR and the one for updating liblibc? And I'm not in a hurry, I'll prepare the PR whatever suits you best. Let's just wait for the libc update first.

varkor · 2018-12-26T20:58:30Z

And I'm not in a hurry, I'll prepare the PR whatever suits you best.

I think it would be most straightforward just to update the submodule in this PR, seeing as how they're directly related.

glaubitz · 2018-12-26T21:00:10Z

I think it would be most straightforward just to update the submodule in this PR, seeing as how they're directly related.

And I'm just trying to find out how to do that properly. I have updated a submodule in a PR before.

Care to explain?

nagisa · 2018-12-26T21:08:00Z

Oh, it seems that we have recently stopped using the submodule approach and are now pulling the dependency in via cargo from crates.io instead... Which essentially means that all you need to do is to update the relevant Cargo.toml files to use the new version from crates.io and it should work just as one would expect.

cc @alexcrichton liblibc probably needs a new release, right?

glaubitz · 2018-12-26T21:09:12Z

Ah, right. I was already wondering why git submodule didn't list liblibc. That explains it :).

glaubitz · 2018-12-26T21:09:55Z

@alexcrichton Oh, before you release. I think @karcherm has two fixes for sparc64 in libc ready.

karcherm · 2018-12-26T21:55:31Z

One of the patches @glaubitz is talking about is [https://github.com/rust-lang/libc/pull/1187]. I'm unsure whether it should be merged or not. The second one is work in progress to enable reliable stack overflow detection on sparc64. It is not ready yet, but possibly in a couple of hours.

glaubitz · 2018-12-27T11:35:19Z

@alexcrichton I would suggest to do a new libc now, so we can get this PR merged.

We can do another release once @karcherm has pushed his fixes.

Bump version to 0.2.46 New release required for rust-lang/rust#57121

glaubitz · 2019-01-04T08:43:44Z

I'm trying to update the libc crate now for this PR but that fails with:

glaubitz@epyc:..rust/rust> cargo update -p libc
    Updating crates.io index
error: no matching package named `core` found                                                                                                                                                
location searched: registry `https://github.com/rust-lang/crates.io-index`
required by package `backtrace-sys v0.1.27`
    ... which is depended on by `backtrace v0.3.11`
    ... which is depended on by `error-chain v0.12.0`
    ... which is depended on by `cargo_metadata v0.6.2`
    ... which is depended on by `miri v0.1.0 (/local_scratch/glaubitz/rust/rust/src/tools/miri)`
glaubitz@epyc:..rust/rust>

I made a fresh checkout to make sure my repo isn't corrupted but that didn't help.

Anyone got an idea what's wrong?

alexcrichton · 2019-01-04T16:04:20Z

Hm I think you may need to make sure to use the nightly cargo to update libc, but you can also just update one of the version requirements in this repository and then ./x.py build

glaubitz · 2019-01-04T21:28:30Z

@alexcrichton Updating cargo to nightly did the trick. I just updated libc and pushed a new revision.

Should be all set now :).

bors · 2019-01-06T07:58:58Z

☔ The latest upstream changes (presumably #57263) made this pull request unmergeable. Please resolve the merge conflicts.

glaubitz · 2019-01-06T10:07:50Z

@alexcrichton @nagisa r?

nagisa · 2019-01-06T10:33:49Z

src/librustc_data_structures/flock.rs

@@ -16,6 +16,12 @@ cfg_if! {
        use std::os::unix::prelude::*;
        use libc;

+        pub const F_RDLCK: libc::c_short = libc::F_RDLCK as libc::c_short;


Why is it necessary to have these as constants here?

Good point. I'll push an updated version in a second.

Since the values for the fcntl constants can vary from architecture to architecture, it is better to use the values defined in the libc crate instead of assigning literals in the flock code which would make the assumption that all architectures use the same values. Fixes rust-lang#57007

nagisa · 2019-01-06T11:24:42Z

@bors r+

bors · 2019-01-06T11:24:43Z

📌 Commit 1c4823e has been approved by nagisa

bors · 2019-01-06T13:05:18Z

⌛ Testing commit 1c4823e with merge 6b2c311...

flock: Use fcntl constants directly from libc crate on Unix targets Since the values for the fcntl constants can vary from architecture to architecture, it is better to use the values defined in the libc crate instead of assigning literals in the flock code which would make the assumption that all architectures use the same values. Fixes #57007

bors · 2019-01-06T15:40:53Z

☀️ Test successful - status-appveyor, status-travis
Approved by: nagisa
Pushing 6b2c311 to master...

rust-highfive assigned varkor Dec 25, 2018

rust-highfive added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Dec 25, 2018

glaubitz mentioned this pull request Dec 29, 2018

Define missing F_RDLCK, F_WRLCK and F_UNLCK constants on linux-mips -and s390x rust-lang/libc#1186

Merged

varkor added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jan 1, 2019

glaubitz mentioned this pull request Jan 2, 2019

Bump version to 0.2.46 rust-lang/libc#1195

Merged

glaubitz force-pushed the flock-fixes branch from 9343b14 to 182af30 Compare January 2, 2019 20:50

bors added a commit to rust-lang/libc that referenced this pull request Jan 2, 2019

Auto merge of #1195 - glaubitz:new-release, r=alexcrichton

39ae9b9

Bump version to 0.2.46 New release required for rust-lang/rust#57121

bors added a commit to rust-lang/libc that referenced this pull request Jan 3, 2019

Auto merge of #1195 - glaubitz:new-release, r=alexcrichton

ba17deb

Bump version to 0.2.46 New release required for rust-lang/rust#57121

bors added a commit to rust-lang/libc that referenced this pull request Jan 3, 2019

Auto merge of #1195 - glaubitz:new-release, r=alexcrichton

a9e3cc6

Bump version to 0.2.46 New release required for rust-lang/rust#57121

glaubitz force-pushed the flock-fixes branch from 182af30 to 1677fd8 Compare January 4, 2019 21:27

Update libc to 0.2.46

1e903c3

glaubitz force-pushed the flock-fixes branch from 1677fd8 to 5bf9500 Compare January 6, 2019 10:03

nagisa reviewed Jan 6, 2019

View reviewed changes

glaubitz force-pushed the flock-fixes branch from 5bf9500 to 1c4823e Compare January 6, 2019 10:58

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jan 6, 2019

bors merged commit 1c4823e into rust-lang:master Jan 6, 2019

bors mentioned this pull request Jan 6, 2019

Update compiler_builtins 0.1.2 -> 0.1.4 #57327

Merged

glaubitz deleted the flock-fixes branch January 6, 2019 17:25

glaubitz restored the flock-fixes branch January 15, 2019 15:51

ehuss mentioned this pull request May 29, 2021

incremental compilation: could not create session directory lock file: No locks available (os error 37) #49773

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

flock: Use fcntl constants directly from libc crate on Unix targets #57121

flock: Use fcntl constants directly from libc crate on Unix targets #57121

glaubitz commented Dec 25, 2018

rust-highfive commented Dec 25, 2018

glaubitz commented Dec 25, 2018

nagisa commented Dec 26, 2018

glaubitz commented Dec 26, 2018

nagisa commented Dec 26, 2018

glaubitz commented Dec 26, 2018

varkor commented Dec 26, 2018

glaubitz commented Dec 26, 2018

nagisa commented Dec 26, 2018

glaubitz commented Dec 26, 2018

glaubitz commented Dec 26, 2018

karcherm commented Dec 26, 2018

glaubitz commented Dec 27, 2018

glaubitz commented Jan 4, 2019

alexcrichton commented Jan 4, 2019

glaubitz commented Jan 4, 2019

bors commented Jan 6, 2019

glaubitz commented Jan 6, 2019

nagisa Jan 6, 2019

glaubitz Jan 6, 2019

glaubitz Jan 6, 2019

nagisa commented Jan 6, 2019

bors commented Jan 6, 2019

bors commented Jan 6, 2019

bors commented Jan 6, 2019

flock: Use fcntl constants directly from libc crate on Unix targets #57121

flock: Use fcntl constants directly from libc crate on Unix targets #57121

Conversation

glaubitz commented Dec 25, 2018

rust-highfive commented Dec 25, 2018

glaubitz commented Dec 25, 2018

nagisa commented Dec 26, 2018

glaubitz commented Dec 26, 2018

nagisa commented Dec 26, 2018

glaubitz commented Dec 26, 2018

varkor commented Dec 26, 2018

glaubitz commented Dec 26, 2018

nagisa commented Dec 26, 2018

glaubitz commented Dec 26, 2018

glaubitz commented Dec 26, 2018

karcherm commented Dec 26, 2018

glaubitz commented Dec 27, 2018

glaubitz commented Jan 4, 2019

alexcrichton commented Jan 4, 2019

glaubitz commented Jan 4, 2019

bors commented Jan 6, 2019

glaubitz commented Jan 6, 2019

nagisa Jan 6, 2019

Choose a reason for hiding this comment

glaubitz Jan 6, 2019

Choose a reason for hiding this comment

glaubitz Jan 6, 2019

Choose a reason for hiding this comment

nagisa commented Jan 6, 2019

bors commented Jan 6, 2019

bors commented Jan 6, 2019

bors commented Jan 6, 2019