Remove all code related to the PreRead field trial.

Remove all code related to the PreRead field trial.

This field trial has expired.

BUG=557984
Committed: https://crrev.com/f888dced1e8295c076d99bbf57d62a217594b555
Cr-Commit-Position: refs/heads/master@{#415773}

[fdoray, 2016-08-26 20:08:16]

fdoray@chromium.org changed reviewers:
+ gab@chromium.org, jochen@chromium.org

[fdoray, 2016-08-26 20:08:17]

gab@chromium.org: Please review changes in components/startup_metric_utils

jochen@chromium.org: Please review everything else

[jochen (gone - plz use gerrit), 2016-08-29 09:22:00]

lgtm

[gab, 2016-08-31 16:39:58]

lgtm % question

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
File chrome/app/file_pre_reader_win.cc (left):

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
chrome/app/file_pre_reader_win.cc:44: ::GetModuleHandle(L"kernel32.dll"),
"PrefetchVirtualMemory"));
Interesting, so the built-in Windows prefetcher wasn't as good as a basic
buffered file read?

[fdoray, 2016-08-31 19:50:52]

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
File chrome/app/file_pre_reader_win.cc (left):

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
chrome/app/file_pre_reader_win.cc:44: ::GetModuleHandle(L"kernel32.dll"),
"PrefetchVirtualMemory"));
On 2016/08/31 16:39:57, gab wrote:
> Interesting, so the built-in Windows prefetcher wasn't as good as a basic
> buffered file read?

From ETW traces, it looks like PrefetchVirtualMemory does the same thing as us.
At the 75th percentile, I got a 1% +/- 6% improvement when using
PrefetchVirtualMemory (week ending on February 8th) vs buffered file reads. I'm
not convinced that this justifies the added complexity of having 2 code paths
(Win7 vs Win8+).

[fdoray, 2016-08-31 19:53:40]

The CQ bit was checked by fdoray@chromium.org

[fdoray, 2016-08-31 19:53:41]

The patchset sent to the CQ was uploaded after l-g-t-m from gab@chromium.org,
jochen@chromium.org
Link to the patchset: https://codereview.chromium.org/2280183002/#ps40001
(title: "rebase")

[commit-bot: I haz the power, 2016-08-31 19:54:26]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[gab, 2016-08-31 19:54:47]

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
File chrome/app/file_pre_reader_win.cc (left):

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
chrome/app/file_pre_reader_win.cc:44: ::GetModuleHandle(L"kernel32.dll"),
"PrefetchVirtualMemory"));
On 2016/08/31 19:50:52, fdoray wrote:
> On 2016/08/31 16:39:57, gab wrote:
> > Interesting, so the built-in Windows prefetcher wasn't as good as a basic
> > buffered file read?
> 
> From ETW traces, it looks like PrefetchVirtualMemory does the same thing as
us.
> At the 75th percentile, I got a 1% +/- 6% improvement when using
> PrefetchVirtualMemory (week ending on February 8th) vs buffered file reads.
I'm
> not convinced that this justifies the added complexity of having 2 code paths
> (Win7 vs Win8+).

Ok, then I'd say add a comment mentioning that this would be a good replacement
when we drop support for Win7 with an explanation of the results (so someone
doesn't come in and redo the same experiment later).

[fdoray, 2016-08-31 19:57:34]

The CQ bit was unchecked by fdoray@chromium.org

[fdoray, 2016-08-31 20:00:33]

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
File chrome/app/file_pre_reader_win.cc (left):

https://codereview.chromium.org/2280183002/diff/1/chrome/app/file_pre_reader_...
chrome/app/file_pre_reader_win.cc:44: ::GetModuleHandle(L"kernel32.dll"),
"PrefetchVirtualMemory"));
On 2016/08/31 19:54:47, gab wrote:
> On 2016/08/31 19:50:52, fdoray wrote:
> > On 2016/08/31 16:39:57, gab wrote:
> > > Interesting, so the built-in Windows prefetcher wasn't as good as a basic
> > > buffered file read?
> > 
> > From ETW traces, it looks like PrefetchVirtualMemory does the same thing as
> us.
> > At the 75th percentile, I got a 1% +/- 6% improvement when using
> > PrefetchVirtualMemory (week ending on February 8th) vs buffered file reads.
> I'm
> > not convinced that this justifies the added complexity of having 2 code
paths
> > (Win7 vs Win8+).
> 
> Ok, then I'd say add a comment mentioning that this would be a good
replacement
> when we drop support for Win7 with an explanation of the results (so someone
> doesn't come in and redo the same experiment later).

Done.

[fdoray, 2016-08-31 20:00:37]

The CQ bit was checked by fdoray@chromium.org

[fdoray, 2016-08-31 20:00:38]

The patchset sent to the CQ was uploaded after l-g-t-m from gab@chromium.org,
jochen@chromium.org
Link to the patchset: https://codereview.chromium.org/2280183002/#ps60001
(title: "add comment")

[commit-bot: I haz the power, 2016-08-31 20:01:28]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-31 20:12:52]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-31 20:12:53]

Try jobs failed on following builders:
  linux_chromium_chromeos_ozone_rel_ng on master.tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[fdoray, 2016-08-31 20:58:41]

The CQ bit was checked by fdoray@chromium.org

[fdoray, 2016-08-31 20:58:42]

The patchset sent to the CQ was uploaded after l-g-t-m from gab@chromium.org,
jochen@chromium.org
Link to the patchset: https://codereview.chromium.org/2280183002/#ps80001
(title: "fix include")

[commit-bot: I haz the power, 2016-08-31 20:59:09]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-31 22:08:34]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-08-31 22:11:21]

Description was changed from

==========
Remove all code related to the PreRead field trial.

This field trial has expired.

BUG=557984
==========

to

==========
Remove all code related to the PreRead field trial.

This field trial has expired.

BUG=557984

Committed: https://crrev.com/f888dced1e8295c076d99bbf57d62a217594b555
Cr-Commit-Position: refs/heads/master@{#415773}
==========

[commit-bot: I haz the power, 2016-08-31 22:11:22]

Patchset 5 (id:??) landed as
https://crrev.com/f888dced1e8295c076d99bbf57d62a217594b555
Cr-Commit-Position: refs/heads/master@{#415773}

[chrisha, 2016-09-01 13:13:58]

chrisha@chromium.org changed reviewers:
+ chrisha@chromium.org

[chrisha, 2016-09-01 13:13:58]

This broke the win-asan build, by removing the chrome_metrics_service_accessor.h
include. Fix here:

https://codereview.chromium.org/2302843002/