Detect seek penalties on Linux/ChromeOS.

Detect seek penalties on Linux/ChromeOS.

This wont work any time some fancy /dev setup is going on.

R=thestig@chromium.org
BUG=463209
Committed: https://crrev.com/022b19cb4b2f4e7a0f56ad92198b050b76e4415f
Cr-Commit-Position: refs/heads/master@{#323865}

[Dan Beam, 2015-04-03 04:06:39]

Patchset #1 (id:1) has been deleted

[Lei Zhang, 2015-04-03 05:12:08]

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:24: const char
kRotationalFormat[] = "/sys/block/sd%c/queue/rotational";
I believe after sdz comes sdaa.

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:28: #if
!defined(OS_ANDROID)
just exclude the file for android entirely?

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:33: if
(base::SysInfo::GetLsbReleaseBoard() != "parrot") {
What if this returns "unknown" ?

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:34: // All ChromeOS
devices that have ever shipped have SSDs. Except parrots.
some parrots...

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:56: // Old Linux kernel?
How old? Kernels that are too old won't run Chrome properly.

[Dan Beam, 2015-04-03 18:25:06]

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:24: const char
kRotationalFormat[] = "/sys/block/sd%c/queue/rotational";
On 2015/04/03 05:12:08, Lei Zhang wrote:
> I believe after sdz comes sdaa.

Done.

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:28: #if
!defined(OS_ANDROID)
On 2015/04/03 05:12:07, Lei Zhang wrote:
> just exclude the file for android entirely?

Done.

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:33: if
(base::SysInfo::GetLsbReleaseBoard() != "parrot") {
On 2015/04/03 05:12:07, Lei Zhang wrote:
> What if this returns "unknown" ?

Done.

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:34: // All ChromeOS
devices that have ever shipped have SSDs. Except parrots.
On 2015/04/03 05:12:08, Lei Zhang wrote:
> some parrots...

Done.

https://codereview.chromium.org/1057093003/diff/20001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:56: // Old Linux kernel?
On 2015/04/03 05:12:07, Lei Zhang wrote:
> How old? Kernels that are too old won't run Chrome properly.

don't know. removed.

[Dan Beam, 2015-04-03 19:28:26]

https://codereview.chromium.org/1057093003/diff/40001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/40001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:58: }
annnd this will never work

[Dan Beam, 2015-04-03 22:25:31]

Patchset #3 (id:60001) has been deleted

[Dan Beam, 2015-04-03 22:27:05]

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:55: sdXX += 'a' + (minor
/ 26) - 1;
maybe this will

[Lei Zhang, 2015-04-03 22:45:57]

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:52: uint16 minor =
MINOR(path_stat.st_dev) / 16;
Why is this divided by 16?

https://codereview.chromium.org/1057093003/diff/80001/chrome/chrome_browser.gypi
File chrome/chrome_browser.gypi (right):

https://codereview.chromium.org/1057093003/diff/80001/chrome/chrome_browser.g...
chrome/chrome_browser.gypi:3185: ['OS!="android"', {
GN

[Dan Beam, 2015-04-03 22:47:10]

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:52: uint16 minor =
MINOR(path_stat.st_dev) / 16;
On 2015/04/03 22:45:57, Lei Zhang wrote:
> Why is this divided by 16?

16 partitions per drive

https://codereview.chromium.org/1057093003/diff/80001/chrome/chrome_browser.gypi
File chrome/chrome_browser.gypi (right):

https://codereview.chromium.org/1057093003/diff/80001/chrome/chrome_browser.g...
chrome/chrome_browser.gypi:11: # These variables are so the file lists can be
shared with the GN build.
^

[Lei Zhang, 2015-04-03 22:51:10]

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:52: uint16 minor =
MINOR(path_stat.st_dev) / 16;
On 2015/04/03 22:47:09, Dan Beam wrote:
> On 2015/04/03 22:45:57, Lei Zhang wrote:
> > Why is this divided by 16?
> 
> 16 partitions per drive

Oh, I forgot about that. That's why I was thinking device 17 is sdaa, but if the
max minor value is 255, and we have to divide by 16... how do we get to sdaa?
.... http://www.tldp.org/HOWTO/SCSI-2.4-HOWTO/dnames.html

[Dan Beam, 2015-04-03 23:01:40]

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/80001/chrome/browser/metrics/...
chrome/browser/metrics/drive_metrics_provider_linux.cc:52: uint16 minor =
MINOR(path_stat.st_dev) / 16;
On 2015/04/03 22:51:10, Lei Zhang wrote:
> On 2015/04/03 22:47:09, Dan Beam wrote:
> > On 2015/04/03 22:45:57, Lei Zhang wrote:
> > > Why is this divided by 16?
> > 
> > 16 partitions per drive
> 
> Oh, I forgot about that. That's why I was thinking device 17 is sdaa, but if
the
> max minor value is 255, and we have to divide by 16... how do we get to sdaa?
> .... http://www.tldp.org/HOWTO/SCSI-2.4-HOWTO/dnames.html

ignoring /dev/sdq+ for now

[Lei Zhang, 2015-04-03 23:06:48]

lgtm

https://codereview.chromium.org/1057093003/diff/120001/chrome/browser/metrics...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/120001/chrome/browser/metrics...
chrome/browser/metrics/drive_metrics_provider_linux.cc:52: char sdX = 'a' +
MINOR(path_stat.st_dev) / 16;
Can you mention what the 16 is?

https://codereview.chromium.org/1057093003/diff/120001/chrome/browser/metrics...
chrome/browser/metrics/drive_metrics_provider_linux.cc:53: std::string
rotational = base::StringPrintf(kRotationalFormat, sdX);
nit: rotational_path

[Dan Beam, 2015-04-03 23:36:19]

Patchset #4 (id:100001) has been deleted

[Dan Beam, 2015-04-03 23:55:17]

Patchset #5 (id:140001) has been deleted

[Dan Beam, 2015-04-03 23:55:45]

still working on gyp/gn voodoo

https://codereview.chromium.org/1057093003/diff/120001/chrome/browser/metrics...
File chrome/browser/metrics/drive_metrics_provider_linux.cc (right):

https://codereview.chromium.org/1057093003/diff/120001/chrome/browser/metrics...
chrome/browser/metrics/drive_metrics_provider_linux.cc:52: char sdX = 'a' +
MINOR(path_stat.st_dev) / 16;
On 2015/04/03 23:06:47, Lei Zhang wrote:
> Can you mention what the 16 is?

Done.

https://codereview.chromium.org/1057093003/diff/120001/chrome/browser/metrics...
chrome/browser/metrics/drive_metrics_provider_linux.cc:53: std::string
rotational = base::StringPrintf(kRotationalFormat, sdX);
On 2015/04/03 23:06:47, Lei Zhang wrote:
> nit: rotational_path

Done.

[Dan Beam, 2015-04-04 00:19:23]

Patchset #5 (id:160001) has been deleted

[Dan Beam, 2015-04-04 01:17:46]

The CQ bit was checked by dbeam@chromium.org

[Dan Beam, 2015-04-04 01:17:46]

The patchset sent to the CQ was uploaded after l-g-t-m from thestig@chromium.org
Link to the patchset: https://codereview.chromium.org/1057093003/#ps180001
(title: "mac stub")

[commit-bot: I haz the power, 2015-04-04 01:17:52]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1057093003/180001

[commit-bot: I haz the power, 2015-04-04 01:24:21]

Committed patchset #5 (id:180001)

[commit-bot: I haz the power, 2015-04-04 01:25:03]

Patchset 5 (id:??) landed as
https://crrev.com/022b19cb4b2f4e7a0f56ad92198b050b76e4415f
Cr-Commit-Position: refs/heads/master@{#323865}