> On Jan. 8, 2016, 6:50 p.m., Martin Klapetek wrote: > > Heads-up: this commit actually broke kwallet when opened from pam; > > kwallet is basically blocked forever and all requests time out. > > > > I'll try to break down this patch to pinpoint which change it was > > exactly. > > Martin Klapetek wrote: > This is the change that breaks kwallet: > > ```diff > + fclose(s3); > + > ``` > > I'm not entirely sure (yet) why though. > > Christoph Cullmann wrote: > Hi, don't you need to do f = fdopen (dup(filehandle)) to have not > fclose(f) close the handle, too, and not only the FILE *? > > Martin Klapetek wrote: > You're right; `FILE *s3 = fdopen(dup(s2), "r");` /does/ fix the issue.
:=) nice - Christoph ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://git.reviewboard.kde.org/r/126410/#review90801 ----------------------------------------------------------- On Dec. 18, 2015, 10:50 p.m., Michael Pyne wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://git.reviewboard.kde.org/r/126410/ > ----------------------------------------------------------- > > (Updated Dec. 18, 2015, 10:50 p.m.) > > > Review request for KDE Frameworks, Martin Klapetek and Valentin Rusu. > > > Repository: kwallet > > > Description > ------- > > Fix a leak of a `FILE*` in kwalletd, spotted by Coverity (CID 1335090), and a > couple of minor fixes/cleanups since I was there. > > 1. No need to memset a `char[]`, you can let compiler default-initialize it > (and likely even be able to move the whole array into .bss). > 2. The chop after `fgets` can easily be confused for a null terminator check, > and `fgets` isn't required to return a newline anyways, so add sanity > checking (which helps document better at the same time). > > > Diffs > ----- > > src/runtime/kwalletd/main.cpp db8d55c > > Diff: https://git.reviewboard.kde.org/r/126410/diff/ > > > Testing > ------- > > Everything compiles. Not much beyond done. > > > Thanks, > > Michael Pyne > >
_______________________________________________ Kde-frameworks-devel mailing list Kde-frameworks-devel@kde.org https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
