Update libc and some fixes for x86_64-unknown-linux-gnux32

[malbarbo]

No description provided.

[rust-highfive]

r? @alexcrichton

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

[malbarbo]

The libstd tests fails in my machine, let's see on travis.

[alexcrichton]

Looks great!

[alexcrichton]

We've got existing tests for x86/x86_64, perhaps those could just switch to being two tests based on target_pointer_width?

[malbarbo]

Done.

[alexcrichton]

Is this necessary b/c the values are always < 1e9? Doesn't that mean we can basically cast to "anything reasonable" here and expect it to work?

[malbarbo]

Should we use i32?

[alexcrichton]

I think that could work? It just seemed like the extra casting wasn't necessary here given the various constraints.

[malbarbo]

As subsec_nanos is u32, I opted to use u32. Both i32 and u32 are fine.

[alexcrichton]

Do you plan on leaving this permanently? It looks like we don't have a lot of extra time on this builder unfortunately (the last one clocked in at 1h40m)...

I'd imagine though that once we get the test suite passing once it's probably pretty unlikely to regress?

[alexcrichton]

Looks like maybe something related to stack walking is odd?

[01:12:28]     [run-pass] run-pass/backtrace-debuginfo.rs

[01:12:28]     [run-pass] run-pass/backtrace.rs

[01:12:28]     [run-pass] run-pass/command-before-exec.rs

[01:12:28]     [run-pass] run-pass/command-exec.rs

[01:12:28]     [run-pass] run-pass/out-of-stack.rs

[01:12:28]     [run-pass] run-pass/panic-runtime/lto-unwind.rs

[01:12:28]     [run-pass] run-pass/stack-probes-lto.rs

[01:12:28]     [run-pass] run-pass/stack-probes.rs

[01:12:28]     [run-pass] run-pass/wait-forked-but-failed-child.rs

[malbarbo]

@alexcrichton I squashed the commits and disable the gnux32 tests. I also opened the issue #45416 to track the failures.

[alexcrichton]

Ok! Is this ready to go? (did you mean to add ignore annotations to all affected tests?)

[malbarbo]

I think this is ready to go. This will allows others to test the target with the fixed libc.

I think you are correct about the stack walk odds. I will be out for some time now, but it would be great if someone could take a look at it.

[alexcrichton]

@bors: r+

[bors]

📌 Commit e57ee3d has been approved by alexcrichton

[Mark-Simulacrum]

Is this intended to fix #45417, #45416 and perhaps others?

[bors]

⌛ Testing commit e57ee3d with merge d532ba7...

[bors]

☀️ Test successful - status-appveyor, status-travis
Approved by: alexcrichton
Pushing d532ba7 to master...

[malbarbo]

Is this intended to fix #45417, #45416 and perhaps others?

No. 😞