Re: [PATCH] shallow: verify shallow file after taking lock
On Thu, Feb 27, 2014 at 04:10:12AM -0500, Jeff King wrote:
> I also notice that check_shallow_file_for_update returns early if
> !is_shallow. Is that safe? Is it possible for another process to have
> made us shallow since the program began? In that case, we would have to
> stat() the file always, then complain if it exists and !is_shallow.
That patch would look like this:
diff --git a/shallow.c b/shallow.c
index 75da07a..e05a241 100644
--- a/shallow.c
+++ b/shallow.c
@@ -139,13 +139,13 @@ void check_shallow_file_for_update(void)
{
struct stat st;
- if (!is_shallow)
- return;
- else if (is_shallow == -1)
+ if (is_shallow == -1)
die("BUG: shallow must be initialized by now");
if (stat(git_path("shallow"), &st))
die("shallow file was removed during fetch");
+ else if (!is_shallow)
+ die("shallow file appeared during fetch");
else if (st.st_mtime != shallow_stat.st_mtime
#ifdef USE_NSEC
|| ST_MTIME_NSEC(st) != ST_MTIME_NSEC(shallow_stat)
but again, I'm not really clear on whether this is possible.
-Peff
Jeff King <peff@peff.net>