Re: git bugs

From: Jeff King <peff@peff.net>
Date: Jun 13, 2008, 10:10 UTC
Message-ID: <20080613101024.GA8221@sigill.intra.peff.net>
In-Reply-To: <alpine.LFD.1.10.0806101028040.3101@woody.linux-foundation.org>

On Tue, Jun 10, 2008 at 10:44:43AM -0700, Linus Torvalds wrote:

> You can fix this multiple ways. One would be to pick another size that is 
> simply less likely (eg ~0 instead), which leaves the theoretical race, and 
> just makes it practically impossible to hit (not that I think it's very 
> practical to hit already).

Hmm. I may have just hit it in the test suite.

Try this:

  cd git/t
  for i in `seq 1 1000`; do
    ./t4126-apply-empty.sh -v -i || break
  done
  echo made it to $i

Most of the time it works, but somewhere in that thousand (generally within a few hundred), I end up with a failed test. The failure looks like:

   * expecting success:
           git reset --hard &&
           >empty &&
           rm -f missing &&
           git apply --index patch0 &&
           test_cmp expect empty &&
           git diff --exit-code

   HEAD is now at e3f79fd initial
   error: empty: does not match index
   * FAIL 3: apply --index empty

> The other approach is to know that an empty blob always has a very 
> specific SHA1. Here's an trial patch.

However, I can still trigger the failure with your patch, so I wonder if it is some other race entirely...

-Peff