Red Hat Bugzilla – Bug 426024
nx uses obsolete (and non-POSIX) sa_restorer
Last modified: 2008-03-10 06:23:43 EDT
I cloned this bug from libwvstreams, as it's quite the same. Please find a build.log from nx here: http://buildsys.zero42.at/koji/getfile?taskID=82832&name=build.log Some relevant comments from my libwvstreams bug: +++ This bug was initially created as a clone of Bug #402531 +++ From my build.log: [oliver@gosa libwvstreams]$ cat .build-4.4-1.fc8.log |tail utils/wvcrash.cc:96: warning: ignoring return value of 'ssize_t write(int, const void*, size_t)', declared with attribute warn_unused_result utils/wvcrash.cc: In function 'void wvcrash_add_signal(int)': utils/wvcrash.cc:388: error: 'struct sigaction' has no member named 'sa_restorer' make[1]: *** [utils/wvcrash.o] Error 1 make[1]: Leaving directory `/home/oliver/cvs/F-8/libwvstreams/wvstreams-4.4' error: Bad exit status from /var/tmp/rpm-tmp.99427 (%build) RPM build errors: Bad exit status from /var/tmp/rpm-tmp.99427 (%build) From sigaction manpage: <snip> The sa_restorer element is obsolete and should not be used. POSIX does not specify a sa_restorer element. </snip>
Thanks for the report. I have two questions: o what is the suggested action wrt to the missing sa_restorer filed? o I just tried a build on F9 - while it failed at the very end when computing the dependencies for other reasons, it did manage to build everything. How could that be, shouldn't F9 choke on this as well?
nx-3.1.0-25.1.fc8 has been submitted as an update for Fedora 8
http://buildsys.zero42.at/koji/getfile?taskID=91317&name=build.log still fails. sa_restorer seems to be still available in x86 and others, while alpha is more strict and it is disabled... The best thing is to just delete the sa_restorer lines!
nx-3.1.0-25.1.fc8 has been pushed to the Fedora 8 stable repository. If problems still persist, please make note of it in this bug report.
I don't feel quite that confident on removing sa_restorer in x86 w/o investigating and understanding what impact it may have. Maybe the best thing is to have a conditional patch on whether sa_restorer exists in the source or failing that a specfile conditional for alpha that patches it out? Accepting both kind of patches :)
Anything I find on sa_restorer is, that it was dangerous, that it should not be used (anymore) and that code should be removed... Well, I'm not an expert, but for me this sounds like: Simply don't set sa_restorer and everything should be fine.