On Mon, Jul 4, 2011 at 2:39 AM, Phil Blundell <ph...@gnu.org> wrote: > On Sat, 2011-07-02 at 17:47 -0700, Khem Raj wrote: >> +- if ((fd = mkostemp(temp, O_CLOEXEC|O_CREAT|O_WRONLY)) < 0) { >> ++ if ((fd = mkstemp(temp)) < 0) { > > This change looks like it will effectively remove O_CLOEXEC without > reinstating it anywhere else. Have you verified this is a safe thing to > do?
Initial tests on uclibc seems to work well. but this needs more testing thats why the patch is only applied for uclibc based systems > >> +- fd = mkostemp(path, O_RDWR|O_CLOEXEC); >> ++ fd = mkstemp(path); > > Likewise. > >> ++ char dev[20]; >> + int prio = 0, k; >> + >> + if ((k = fscanf(m->proc_swaps, >> +- "%ms " /* device/file */ >> ++ "%19s " /* device/file */ >> + "%*s " /* type of swap */ > > Is 19 characters definitely enough? If this is potentially a pathname > then it could be much longer than that. > it just reads /dev/* so should be ok. > p. > > > > _______________________________________________ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel > _______________________________________________ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel