Created attachment 677887 [details] build.log (ppc64) ================================================================================== /src/vfs/extfs/helpers-list: tests/src/vfs/extfs/helpers-list/test-suite.log ================================================================================== # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 .. contents:: :depth: 2 FAIL: run ========= Testing /var/tmp/portage/app-misc/mc-4.8.25/work/mc-4.8.25/tests/src/vfs/extfs/helpers-list/data/hp48+.i nput /var/tmp/portage/app-misc/mc-4.8.25/work/mc-4.8.25/tests/src/vfs/extfs/helpers-list/data/hp48+.output /v ar/tmp/portage/app-misc/mc-4.8.25/temp/hp48+.actual-parsed-output.rZ3XaA2e differ: char 17, line 1 ERROR: hp48+ has produced output that's different than the expected output.
Created attachment 677890 [details] test-suite.log
Created attachment 677893 [details] emerge --info
diffs claim to have some uid/gid parsing problem: """ ------------ diff of the expected output vs the actual output: ------------- --- /var/tmp/portage/app-misc/mc-4.8.25/work/mc-4.8.25/tests/src/vfs/extfs/helpers-list/data/hp48+.output 2020-01-01 16:52:34.000000000 +0100 +++ /var/tmp/portage/app-misc/mc-4.8.25/temp/hp48+.actual-parsed-output.rZ3XaA2e 2020-12-11 13:23:55.995461256 +0100 @@ -1,3 +1,3 @@ --rw-r--r-- 1 0 0 30 /YEN --rw-r--r-- 1 0 0 21848 /JYTLIGHT --rw-r--r-- 1 0 0 37 /IOPAR +-rw-r--r-- 1 <<uid>> <<gid>> 30 /YEN +-rw-r--r-- 1 <<uid>> <<gid>> 21848 /JYTLIGHT +-rw-r--r-- 1 <<uid>> <<gid>> 37 /IOPAR """ Can't reproduce on timberdoodle ppc64 machine. Probably related to differences in environment.
Could you check if this still happens with 4.8.26, ernsteiswuerfel? According to src/vfs/extfs/helpers/hp48+.in in the source tarball, or /usr/libexec/mc/extfs.d/hp48+ on my box, this version should have the UID and the GID hard-coded to 0 0 for any output of this helper.
Created attachment 723874 [details] build.log (ppc64, 4.8.26-r3) The test still fails on 4.8.26-r2 and -r3.
Created attachment 723877 [details] test-suite.log (ppc64, 4.8.26-r3)
(In reply to ernsteiswuerfel from comment #5) > The test still fails on 4.8.26-r2 and -r3. I think I've figured it out... Could you please try again with FEATURES=userpriv?
...and it turns out upstream has already known about this for several years.
(In reply to Marek Szuba from comment #7) > I think I've figured it out... Could you please try again with > FEATURES=userpriv? Yes, I can confirm this!
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=c5ff312037a07683efb5513f26759fa3fd9b4fce commit c5ff312037a07683efb5513f26759fa3fd9b4fce Author: Marek Szuba <marecki@gentoo.org> AuthorDate: 2021-07-14 23:30:52 +0000 Commit: Marek Szuba <marecki@gentoo.org> CommitDate: 2021-07-14 23:32:05 +0000 app-misc/mc: warn about test failures if executing them as UID 0 Closes: https://bugs.gentoo.org/759466 Signed-off-by: Marek Szuba <marecki@gentoo.org> app-misc/mc/mc-4.8.26-r3.ebuild | 6 ++++++ 1 file changed, 6 insertions(+)
