IndexedDB: Optimize IDBCursor.advance() if it occurs within prefetched range

IndexedDB: Optimize IDBCursor.advance() if it occurs within prefetched range

To reduce the IPC overhead, repeated IDBCursor.continue()
calls trigger pre-fetching where multiple values are
fetched at once. As long as only simple continue() calls
are made, the pre-fetch cache can be used to service them.
Other calls invalidate the cache.

IDBCursor.advance() is like a continue() call that skips
over values; it is safe to use the cache as long as it is skipping within the pre-fetched range, so do so.

(TBR for a unittest 'friend' change)

TBR=jam

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=242045

[jsbell, 2013-12-12 00:18:44]

alecflett@ - please take a look?

https://codereview.chromium.org/104663007/diff/20001/content/child/indexed_db...
File content/child/indexed_db/proxy_webidbcursor_impl.cc (right):

https://codereview.chromium.org/104663007/diff/20001/content/child/indexed_db...
content/child/indexed_db/proxy_webidbcursor_impl.cc:140: while (count > 1) {
If we think this will be common we could replace the std::deque usage here with
std::vector so we can just slice off the front of the queue.

[alecflett, 2013-12-19 00:41:38]

lgtm

lgtm with one suggestion

https://codereview.chromium.org/104663007/diff/20001/content/child/indexed_db...
File content/child/indexed_db/proxy_webidbcursor_impl.cc (right):

https://codereview.chromium.org/104663007/diff/20001/content/child/indexed_db...
content/child/indexed_db/proxy_webidbcursor_impl.cc:140: while (count > 1) {
On 2013/12/12 00:18:44, jsbell wrote:
> If we think this will be common we could replace the std::deque usage here
with
> std::vector so we can just slice off the front of the queue.

I'd do that in another patch under the umbrella of code cleanup.. (which I'd
happily review too) I'd be surprised if it made any difference for the size of
prefetches we do..

https://codereview.chromium.org/104663007/diff/120001/content/child/indexed_d...
File content/child/indexed_db/indexed_db_dispatcher.h (right):

https://codereview.chromium.org/104663007/diff/120001/content/child/indexed_d...
content/child/indexed_db/indexed_db_dispatcher.h:85: virtual void
RequestIDBCursorAdvance(unsigned long count,
I have been thrown off by this before, being virtual only for tests. Maybe a
comment (and in the next 2) indicating why they are virtual?

[jsbell, 2013-12-19 18:31:18]

https://codereview.chromium.org/104663007/diff/120001/content/child/indexed_d...
File content/child/indexed_db/indexed_db_dispatcher.h (right):

https://codereview.chromium.org/104663007/diff/120001/content/child/indexed_d...
content/child/indexed_db/indexed_db_dispatcher.h:85: virtual void
RequestIDBCursorAdvance(unsigned long count,
On 2013/12/19 00:41:39, alecflett wrote:
> I have been thrown off by this before, being virtual only for tests. Maybe a
> comment (and in the next 2) indicating why they are virtual?

Done.

[commit-bot: I haz the power, 2013-12-19 18:32:47]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jsbell@chromium.org/104663007/140001

[commit-bot: I haz the power, 2013-12-19 18:49:17]

Retried try job too often on chromium_presubmit for step(s) presubmit
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

[commit-bot: I haz the power, 2013-12-19 19:04:47]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jsbell@chromium.org/104663007/140001

[commit-bot: I haz the power, 2013-12-20 00:23:31]

Retried try job too often on linux_aura for step(s) browser_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_aura...

[commit-bot: I haz the power, 2013-12-20 00:31:24]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jsbell@chromium.org/104663007/140001

[commit-bot: I haz the power, 2013-12-20 08:53:15]

Change committed as 242045