Rolling to leveldb rev which reuses manifest/log.

Rolling to leveldb rev which reuses manifest/log.

This leveldb rev is after 1.18, but is not yet an official leveldb release.

Aside from the manifest/log reuse change, all other leveldb changes are minor:

* 251ebf5 LevelDB now attempts to reuse the preceding MANIFEST and log file when re-opened.
* 77948e7 Add benchmark that measures cost of repeatedly opening the database.
* 34ad72e Move header guard below copyright banner.
* a75d435 Clean up layering of storage/leveldb/...
* b234f65 Added a new fault injection test.
* c4c38f9 Add arm64 support to leveldb.
* cea9b10 Fixed incorrect comment wording for Iterator::Seek.
* c00c569 Deleted old README file.

BUG=402980, 395799
Committed: https://crrev.com/fecd7c1299b730359bfb39bd801cc1ab1d34b84b
Cr-Commit-Position: refs/heads/master@{#307976}

[cmumford, 2014-12-11 17:38:12]

cmumford@chromium.org changed reviewers:
+ dgrogan@chromium.org, jsbell@chromium.org

[cmumford, 2014-12-11 17:38:12]

dgrogan@, jsbell@: I'll take the first lgtm.

FYI the commit reference by this change is from the
https://github.com/google/leveldb/commits/reuse-manifest branch, and not master
(like prior DEP rolls). This was discussed with Sanjay. After a few weeks of
(hopefully) no problems in Chrome I will merge the branch back to master and do
an official 1.19 leveldb release and a second DEPS roll.

[dgrogan, 2014-12-11 19:59:32]

lgtm

[cmumford, 2014-12-11 20:00:48]

The CQ bit was checked by cmumford@chromium.org

[commit-bot: I haz the power, 2014-12-11 20:01:41]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/793423002/1

[dgrogan, 2014-12-11 20:01:53]

https://codereview.chromium.org/793423002/diff/1/third_party/leveldatabase/RE...
File third_party/leveldatabase/README.chromium (right):

https://codereview.chromium.org/793423002/diff/1/third_party/leveldatabase/RE...
third_party/leveldatabase/README.chromium:4: Version: r80
This doesn't make sense now that we're pulling intermediate-versioned git
commits. I don't understand the point of the Version field, it just seems like
it's ripe for being inconsistent with what's in DEPS. But last time I tried to
leave it out I got a presubmit error saying it was necessary.

[commit-bot: I haz the power, 2014-12-11 21:11:17]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2014-12-11 21:12:06]

Patchset 1 (id:??) landed as
https://crrev.com/fecd7c1299b730359bfb39bd801cc1ab1d34b84b
Cr-Commit-Position: refs/heads/master@{#307976}

[cmumford, 2014-12-12 00:37:46]

https://codereview.chromium.org/793423002/diff/1/third_party/leveldatabase/RE...
File third_party/leveldatabase/README.chromium (right):

https://codereview.chromium.org/793423002/diff/1/third_party/leveldatabase/RE...
third_party/leveldatabase/README.chromium:4: Version: r80
On 2014/12/11 20:01:53, dgrogan wrote:
> This doesn't make sense now that we're pulling intermediate-versioned git
> commits. I don't understand the point of the Version field, it just seems like
> it's ripe for being inconsistent with what's in DEPS. But last time I tried to
> leave it out I got a presubmit error saying it was necessary.

I agree. I will try to get to the bottom of this when I roll DEPS again in a
week or two. It looks like the rNumber is one that we made up and doesn't
correspond (at least not that I can see) with any tag or revision number in any
repo.