diff options
| author | dpranke@google.com <dpranke@google.com@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-01-14 01:27:58 +0000 |
|---|---|---|
| committer | dpranke@google.com <dpranke@google.com@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-01-14 01:27:58 +0000 |
| commit | 56ee3109efdf0eae868fc75fc52c64a226d40579 (patch) | |
| tree | 5abad8621e196eb549fcbafa7d402c16cdbb295c /net/http/http_cache_transaction.cc | |
| parent | d8c0ef76d07d5973d2a2354a410e31170aa4d73d (diff) | |
| download | chromium_src-56ee3109efdf0eae868fc75fc52c64a226d40579.zip chromium_src-56ee3109efdf0eae868fc75fc52c64a226d40579.tar.gz chromium_src-56ee3109efdf0eae868fc75fc52c64a226d40579.tar.bz2 | |
There appears to be a race where we were diffing the image output before we
actually copied the output file into the right place. Presumably this means
we were diffing the previous image file instead. I think we didn't notice this
more often because this only happens if the hash check fails. Still, a bit
puzzling that this wasn't breaking more things.
BUG=none
TEST=no more "image_diff: unable to open file" messages on screen.
R=ojan@chromium.org
Review URL: http://codereview.chromium.org/542046
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@36210 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'net/http/http_cache_transaction.cc')
0 files changed, 0 insertions, 0 deletions