Eric Blake wrote:
> Eric Blake byu.net> writes:
>> Here's the latest draft of my patch.
>
> While we're at it, I noticed via findutils that fts leaks fds into child
> processes. This plugs the fts leak (but completely fixing find also requires
> a
> patch to findutils).
>
> Hmm - POSIX states th
Eric Blake byu.net> writes:
> Here's the latest draft of my patch.
While we're at it, I noticed via findutils that fts leaks fds into child
processes. This plugs the fts leak (but completely fixing find also requires a
patch to findutils).
Hmm - POSIX states that fdopendir can, but not must,
Jim Meyering meyering.net> writes:
> > While we're visiting fts, how about this patch? POSIX 2008 is clear that
> > opendir should use O_DIRECTORY, so opendirat should probably do likewise.
>
> There is no need for O_DIRECTORY in that function, since the only thing
> it does with the resulting
Eric Blake wrote:
> Jim Meyering meyering.net> writes:
>> I suspect that it'd be very hard to trigger these close failures,
>> but it's better to be safe.
>
> While we're visiting fts, how about this patch? POSIX 2008 is clear that
> opendir should use O_DIRECTORY, so opendirat should probably do
Jim Meyering meyering.net> writes:
>
> I suspect that it'd be very hard to trigger these close failures,
> but it's better to be safe.
While we're visiting fts, how about this patch? POSIX 2008 is clear that
opendir should use O_DIRECTORY, so opendirat should probably do likewise. Do
we wan
I suspect that it'd be very hard to trigger these close failures,
but it's better to be safe.
And along the same lines, since there are more way in which fts_close
can fail, encourage callers not to ignore its return value.
There were four offenders in coreutils, before today.
I chose not to make
