Further instrumentations to locate the jank source

Further instrumentations to locate the jank source

Prior instrumentations identified that these are the jankiest places:
* HttpCache::Transaction::OnIOComplete
* DiskCacheBasedQuicServerInfo::OnIOComplete
* ShaderDiskReadHelper::OnOpComplete

Instrumenting their internals.

BUG=422516
Committed: https://crrev.com/8a5d6423139ecc006f90f63c15086bf12b8c1cec
Cr-Commit-Position: refs/heads/master@{#306665}

[vadimt, 2014-12-02 00:10:27]

vadimt@chromium.org changed reviewers:
+ avi@chromium.org, mmenke@chromium.org

[vadimt, 2014-12-02 00:10:28]

mmenke@chromium.org, please review net/

avi@chromium.org, content/

[mmenke, 2014-12-02 02:56:11]

Per earlier request, could you have someone who knows the cache (Like gavinp)
review cache-related measurements?  If the issue is in the cache code, then any
solution will lie there, so better to have someone with some knowledge in that
direction keep up with these.

[vadimt, 2014-12-02 02:59:01]

vadimt@chromium.org changed reviewers:
+ gavinp@chromium.org

[vadimt, 2014-12-02 02:59:37]

gavinp@, please review cache code; thanks!

[gavinp, 2014-12-03 01:17:13]

This is LGTM with the noted coding style violation fixed. I have not done a
thorough review to see if you're missing interesting points to review, but it
does look very good at a first impression.

Will you be updating the referenced bug with tracking information?

https://codereview.chromium.org/773463004/diff/1/net/http/http_cache_transact...
File net/http/http_cache_transaction.cc (right):

https://codereview.chromium.org/773463004/diff/1/net/http/http_cache_transact...
net/http/http_cache_transaction.cc:444: if (rv == ERR_IO_PENDING)
You need curly braces here now, see
http://google-styleguide.googlecode.com/svn/trunk/cppguide.html#Conditionals .

(likewise elsewhere throughout this CL)

[vadimt, 2014-12-03 01:41:53]

> Will you be updating the referenced bug with tracking information?
I do this. It will be either a separate comment, or a part of the next CL
description.

https://codereview.chromium.org/773463004/diff/1/net/http/http_cache_transact...
File net/http/http_cache_transaction.cc (right):

https://codereview.chromium.org/773463004/diff/1/net/http/http_cache_transact...
net/http/http_cache_transaction.cc:444: if (rv == ERR_IO_PENDING)
On 2014/12/03 01:17:13, gavinp wrote:
> You need curly braces here now, see
> http://google-styleguide.googlecode.com/svn/trunk/cppguide.html#Conditionals .
> 
> (likewise elsewhere throughout this CL)

Done.

[gavinp, 2014-12-03 05:17:00]

This new upload lgtm.

As a commentary; I love the style of interposing instrumentation that's being
used here. However, I don't like how verbose using it is; it makes the diffs
annoying. Could we maybe come up with macros that make this instrumentation
cheaper?

[Avi (use Gerrit), 2014-12-03 16:27:03]

stampity stamp

[Avi (use Gerrit), 2014-12-03 16:27:09]

On 2014/12/03 16:27:03, Avi wrote:
> stampity stamp

LGTM that is

[vadimt, 2014-12-03 18:36:12]

The CQ bit was checked by vadimt@chromium.org

[commit-bot: I haz the power, 2014-12-03 18:37:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/773463004/20001

[vadimt, 2014-12-03 18:40:39]

On 2014/12/03 05:17:00, gavinp wrote:
> This new upload lgtm.
> 
> As a commentary; I love the style of interposing instrumentation that's being
> used here. However, I don't like how verbose using it is; it makes the diffs
> annoying. Could we maybe come up with macros that make this instrumentation
> cheaper?

This was how I originally wanted to do this; however, see the discussion started
with comment #12:
https://codereview.chromium.org/600213002#msg12

[commit-bot: I haz the power, 2014-12-03 20:29:51]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2014-12-03 20:30:50]

Patchset 2 (id:??) landed as
https://crrev.com/8a5d6423139ecc006f90f63c15086bf12b8c1cec
Cr-Commit-Position: refs/heads/master@{#306665}