On 25.11.2015 02:07, Pádraig Brady wrote: > On 25/11/15 00:40, Joachim Schiele wrote: >> On 02.07.2015 09:45, Pádraig Brady wrote: >>> On 02/07/15 03:38, Pádraig Brady wrote: >>>> On 02/07/15 02:55, Assaf Gordon wrote: >>>>> Hello, >>>>> >>>>> few results for coreutils-8.23.245-8bf2af : >>> >>> >>>>> On OpenSUSE 13.2 x86_64, "tests/tail-2/assert" fail, log attached. >>>> >>>> Interesting. Reproduced 100% of the time here too. >>>> BTW that VM gives a different version? >>>> $ head -n1 /etc/issue >>>> Welcome to openSUSE 12.3 ... >>> >>> It's not 100% reproducible actually, though once it starts happening >>> it continues to, no matter how many times the file is replaced and removed. >>> What seems to be happening is that the IN_ATTRIB event is delivered >>> for the file on unlink() (the link count decreasing is an attribute change). >>> However the subsequent open() in recheck() succeeds! >>> I.E. the kernel dentry (cache) is still accessible through that name. >>> Adding a xnanosleep(.1) before the open() is enough to make the test >>> succeed. >>> >>> We could avoid this issue by adding IN_DELETE on the parent directory, >>> though I'm loath to change this at the moment. >> >> i've hit the same issue: >> what can i do about it? >> >> my system specs: >> - the host is a odroid xu4 octa-core >> - uname -a >> Linux nixos 4.2.0 #1-NixOS SMP PREEMPT Thu Jan 1 00:00:01 UTC 1970 >> armv7l GNU/Linux >> >> the complete log is at: >> - https://lastlog.de/misc/konsole.txt is the logfile >> >> look at the bottom, you will find: >> >> ============================================================================ >> Testsuite summary for GNU coreutils 8.24 >> ============================================================================ > > This issue should be addressed with: > http://git.sv.gnu.org/gitweb/?p=coreutils.git;a=commitdiff;h=v8.24-35-g3ba68f9
i've already tried this patch but then i hit this issue: http://paste.ubuntu.com/13498376/ i don't have packaged makeinfo yet and i don't really care for the documentation at the moment. i didn't find a ./configure switch to disable it either. and there is a patch to make it compile without 'makeinfo' which changes doc/local.mk but it seems already applied in 8.24 so what should i do now? thanks for your help! -- Joachim Schiele blog: http://blog.lastlog.de wiki: http://lastlog.de jabber: [email protected] GPG: C6AC8770
signature.asc
Description: OpenPGP digital signature
