-
Notifications
You must be signed in to change notification settings - Fork 110
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
On s390x, test_kernel_stat and test_meminfo fail #324
Comments
Actually, this is not quite right. We don’t have a |
I don't have access to an s390x environment (though if you have any suggestions about how to easily emulate it, let me know), so I'll need some help here. For the assert_eq!(cpuinfo.num_cores(), stat.cpu_time.len()); So can you please share your For the assert!(meminfo.hardware_corrupted.is_some()); So can you please let me know if |
The following are from a “scratch” build in Fedora infrastructure on real
|
For those working on Fedora, installing
and I can do arbitrary interactive debugging in the chroot with:
Using Alternatively, you can use |
Podman also supports this (not sure if Docker does as well) so long as there's a container image available:
Unfortunately, being an emulated container, I think some of this information may be less-truthful than a real one. I get permission denied for |
I originally saw this while working on an update for the
rust-procfs
package in Fedora; the package build still happens in amock
chroot, but on reals390x
hardware rather than underqemu-user-static
emulation. I’ve tried to reproduce it in emulation, below, since I suspect that may be more useful to you.Working in a Fedora 41
mock
chroot withgit
,cargo
, andrust
installed, and with network access enabled:On
x86_64
, skipping:process::tests::test_proc_fd_count_runsinglethread
because of On at least Fedora 41, test_proc_fd_count_runsinglethread fails #322sys::kernel::random::tests::test_write_wakeup_threshold
,process::tests::test_proc_auxv
, andprocess::tests::test_proc_status_for_kthreadd
because they fail in a mock chroot for various reasonsThis is our “control” build.
Now, repeating in an
aarch64
chroot, emulated viaqemu-user-static
:Okay, the
tests::test_cpuinfo
failure is already reported as #323, theprocess::tests::test_proc_exe
failure seems to be specific to running in emulation, and I don’t know what to say aboutprocess::tests::test_procinfo
. I didn’t see it in a “real” Fedora build, and none of these are what I am trying to report here.Repeating the above on
ppc64le
gives the same result, except thattests::test_cpuinfo
doesn’t fail – although it did fail on realppc64le
hardware.Now, for the subject of this report, on
s390x
:The failures in
process::tests::test_proc_exe
,process::tests::test_procinfo
, andtests::test_cpuinfo
are consistent with those onaarch64
andpp664le
;tests::test_kernel_stat
is one of the failures I was trying to report in this issue. The output on real hardware is similar:Unfortunately, I wasn’t able to reproduce the
test_meminfo
failure in emulation, but it looks like this on real hardware:The text was updated successfully, but these errors were encountered: