[PATCH] apply: fix access to an uninitialized mode variable, found by valgrind

[PATCH] apply: fix access to an uninitialized mode variable, found by valgrind

[Johannes Schindelin]

When 'tpatch' was initialized successfully, st_mode was already taken
from the previous diff.  We should not try to override it with data
from an lstat() that was never called.

This is a companion patch to 7a07841(git-apply: handle a patch that
touches the same path more than once better).

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
 builtin-apply.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/builtin-apply.c b/builtin-apply.c
index 1e7f917..f312798 100644
--- a/builtin-apply.c
+++ b/builtin-apply.c
@@ -2441,7 +2441,7 @@ static int check_preimage(struct patch *patch, struct cache_entry **ce, struct s
 		return error("%s: %s", old_name, strerror(errno));
 	}
 
-	if (!cached)
+	if (!cached && !tpatch)
 		st_mode = ce_mode_from_stat(*ce, st->st_mode);
 
 	if (patch->is_new < 0)
-- 
1.6.1.2.582.g3fdd5

Re: [PATCH] apply: fix access to an uninitialized mode variable, found by valgrind

[Junio C Hamano]

Looks correct; thanks.

Re: [PATCH] apply: fix access to an uninitialized mode variable, found by valgrind

[Don Zickus]

On Wed, Feb 04, 2009 at 02:50:15AM +0100, Johannes Schindelin wrote:
> When 'tpatch' was initialized successfully, st_mode was already taken
> from the previous diff.  We should not try to override it with data
> from an lstat() that was never called.

heh, oops.  thanks.

Acked-by:  Don Zickus <dzickus@redhat.com>

Re: [PATCH] apply: fix access to an uninitialized mode variable, found by valgrind

[Jeff King]

On Wed, Feb 04, 2009 at 02:50:15AM +0100, Johannes Schindelin wrote:

> When 'tpatch' was initialized successfully, st_mode was already taken
> from the previous diff.  We should not try to override it with data
> from an lstat() that was never called.

Yay. Valgrind For The Win, as the kids are saying these days.

Was this being triggered by t4121 and t4127? If so, then I think it
solves the mystery valgrind errors I reported in the last paragraph
here:

  http://article.gmane.org/gmane.comp.version-control.git/98898

-Peff

Re: [PATCH] apply: fix access to an uninitialized mode variable, found by valgrind

[Johannes Schindelin]

Hi,

On Wed, 4 Feb 2009, Jeff King wrote:

> On Wed, Feb 04, 2009 at 02:50:15AM +0100, Johannes Schindelin wrote:
> 
> > When 'tpatch' was initialized successfully, st_mode was already taken
> > from the previous diff.  We should not try to override it with data
> > from an lstat() that was never called.
> 
> Yay. Valgrind For The Win, as the kids are saying these days.

Yes, indeed.

> Was this being triggered by t4121 and t4127? If so, then I think it 
> solves the mystery valgrind errors I reported in the last paragraph 
> here:
> 
>   http://article.gmane.org/gmane.comp.version-control.git/98898

Yep, t4121 and t4127.  But I did not have time to follow the link.

Ciao,
Dscho