Re: [PATCH] Avoid overlapping addresses to stpcpy calls in nscd (BZ #16760)

[Siddhesh Poyarekar <siddhesh at redhat dot com>]

On Thu, Mar 27, 2014 at 03:34:11AM -0400, Mike Frysinger wrote:
> On Thu 27 Mar 2014 09:34:06 Siddhesh Poyarekar wrote:
> > Calls to stpcpy from nscd netgroups code will have overlapping source
> > and destination when all three values in the returned triplet are
> > non-NULL and in the expected (host,user,domain) order.  This is seen
> > in valgrind as:
> > 
> > ==3181== Source and destination overlap in stpcpy(0x19973b48, 0x19973b48)
> > ==3181==    at 0x4C2F30A: stpcpy (in
> > /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) ==3181==    by
> > 0x12567A: addgetnetgrentX (string3.h:111)
> > ==3181==    by 0x12722D: addgetnetgrent (netgroupcache.c:665)
> > ==3181==    by 0x11114C: nscd_run_worker (connections.c:1338)
> > ==3181==    by 0x4E3C102: start_thread (pthread_create.c:309)
> > ==3181==    by 0x59B81AC: clone (clone.S:111)
> > ==3181==
> > 
> > Fix this by using memmove instead of stpcpy.  Tested x86_64 using
> > various combinations of triplets (including NULL and non-NULL ones) to
> > verify that this works correctly and there are no regressions.
> 
> i feel like we've wanted an equivalent of stpcpy/memccpy for memmove.  good 
> time to add it ? :)

Seems like a useful thing to have.

> we do the ?: thing a lot in this code.  time to assign a local var for it ?

Yeah, I was also thinking of breaking the entire logic out into a
function of its own to improve readability, but I didn't because I
wanted the change to be minimal.  It would definitely be a good
cleanup in future.

Siddhesh

Attachment: pgp4NncCs4yKZ.pgp
Description: PGP signature