On Thu, Feb 02, 2012 at 10:29:00PM -0700, Philip Guenther wrote:
> On Thu, 2 Feb 2012, Philip Guenther wrote:
> > I also think readdir() should set errno if it detects an invalid
> > seekdir(). EINVAL seems correct.
>
> Here's a diff for this bit.
>
> oks?
>
>
> Philip Guenther
>
>
> Index: gen/readdir.c
> ===================================================================
> RCS file: /cvs/src/lib/libc/gen/readdir.c,v
> retrieving revision 1.15
> diff -u -p -r1.15 readdir.c
> --- gen/readdir.c 18 Nov 2009 07:43:22 -0000 1.15
> +++ gen/readdir.c 3 Feb 2012 05:21:58 -0000
> @@ -29,6 +29,7 @@
> */
>
> #include <dirent.h>
> +#include <errno.h>
> #include "thread_private.h"
>
> /*
> @@ -52,12 +53,14 @@ _readdir_unlocked(DIR *dirp, struct dire
> return (-1);
> }
> dp = (struct dirent *)(dirp->dd_buf + dirp->dd_loc);
> - if ((long)dp & 03) /* bogus pointer check */
> - return (-1);
> - if (dp->d_reclen <= 0 ||
> - dp->d_reclen > dirp->dd_len + 1 - dirp->dd_loc)
> + if ((long)dp & 03 || /* bogus pointer check */
> + dp->d_reclen <= 0 ||
> + dp->d_reclen > dirp->dd_len + 1 - dirp->dd_loc) {
> + errno = EINVAL;
> return (-1);
> + }
> dirp->dd_loc += dp->d_reclen;
> +
> /*
> * When called from seekdir(), we let it decide on
> * the end condition to avoid overshooting: the next
>
I like this diff per se, just not enough of a POSIX lawyer to say if the
subtly different behaviour will set off alarms in Austin or wherever POSIX
hangs out these days. :-)
.... Ken
