Revert of Flip remaining CQ Win builders to GN.

Revert of Flip remaining CQ Win builders to GN. (patchset #11 id:200001 of https://codereview.chromium.org/1908733002/ )

Reason for revert:
reverting this until after the branch to maintain GYP coverage ...

Original issue's description:
> Flip remaining CQ Win builders to GN.
>
> This CL flips the remaining CQ builders on tryserver.chromium.win to GN:
>
>   - win_chromium_compile_dbg_ng
>   - win_chromium_rel_ng
>
> and some matching bots on chromium.win:
>
>   - Win Builder
>   - Win Builder (dbg)
>
> and tryserver.chromium.win:
>
>   - win_chromium_compile_rel_ng
>   - win_chromium_dbg_ng
>
> and the affected GPU bots that are also tested by win_chromium_rel_ng:
>
>   - GPU Win Builder
>   - GPU Win Builder (dbg)
>
> R=brucedawson@chromium.org, brettw@chromium.org
> BUG=605318
>
> Committed: https://crrev.com/77405408a41233916670d7c4cbf0e18ad451a802
> Cr-Commit-Position: refs/heads/master@{#393560}

TBR=brettw@chromium.org,brucedawson@chromium.org,kbr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=605318
Committed: https://crrev.com/fe79b4b7bd484f806217fd8668de721306ff376a
Cr-Commit-Position: refs/heads/master@{#394300}

[Dirk Pranke, 2016-05-17 21:42:33]

Created Revert of Flip remaining CQ Win builders to GN.

[Dirk Pranke, 2016-05-17 21:42:44]

The CQ bit was checked by dpranke@chromium.org to run a CQ dry run

[Dirk Pranke, 2016-05-17 21:42:57]

Description was changed from

==========
Revert of Flip remaining CQ Win builders to GN. (patchset #11 id:200001 of
https://codereview.chromium.org/1908733002/ )

Reason for revert:
reverting this until after the branch to maintain coverage ...

Original issue's description:
> Flip remaining CQ Win builders to GN.
> 
> This CL flips the remaining CQ builders on tryserver.chromium.win to GN:
> 
>   - win_chromium_compile_dbg_ng
>   - win_chromium_rel_ng
> 
> and some matching bots on chromium.win:
> 
>   - Win Builder
>   - Win Builder (dbg)
> 
> and tryserver.chromium.win:
> 
>   - win_chromium_compile_rel_ng
>   - win_chromium_dbg_ng
> 
> and the affected GPU bots that are also tested by win_chromium_rel_ng:
> 
>   - GPU Win Builder
>   - GPU Win Builder (dbg)
> 
> R=brucedawson@chromium.org, brettw@chromium.org
> BUG=605318
> 
> Committed: https://crrev.com/77405408a41233916670d7c4cbf0e18ad451a802
> Cr-Commit-Position: refs/heads/master@{#393560}

TBR=brettw@chromium.org,brucedawson@chromium.org,kbr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=605318
==========

to

==========
Revert of Flip remaining CQ Win builders to GN. (patchset #11 id:200001 of
https://codereview.chromium.org/1908733002/ )

Reason for revert:
reverting this until after the branch to maintain GYP coverage ...

Original issue's description:
> Flip remaining CQ Win builders to GN.
> 
> This CL flips the remaining CQ builders on tryserver.chromium.win to GN:
> 
>   - win_chromium_compile_dbg_ng
>   - win_chromium_rel_ng
> 
> and some matching bots on chromium.win:
> 
>   - Win Builder
>   - Win Builder (dbg)
> 
> and tryserver.chromium.win:
> 
>   - win_chromium_compile_rel_ng
>   - win_chromium_dbg_ng
> 
> and the affected GPU bots that are also tested by win_chromium_rel_ng:
> 
>   - GPU Win Builder
>   - GPU Win Builder (dbg)
> 
> R=brucedawson@chromium.org, brettw@chromium.org
> BUG=605318
> 
> Committed: https://crrev.com/77405408a41233916670d7c4cbf0e18ad451a802
> Cr-Commit-Position: refs/heads/master@{#393560}

TBR=brettw@chromium.org,brucedawson@chromium.org,kbr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=605318
==========

[commit-bot: I haz the power, 2016-05-17 21:43:38]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1981313003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1981313003/1

[Dirk Pranke, 2016-05-17 21:44:13]

It turns out that simply flipping the waterfall GN builders to GYP is
complicated because of all of the GN-only targets (like mus and mash) that are
currently tested (only) there.

Figuring out which test targets should be moved from the (old) GN bots to the
(old) GYP bots is more thinking than I want to do at the moment, so I'm just
going to flip the (old) GYP bots back to GYP instead.

[brucedawson, 2016-05-17 23:08:21]

lgtm

Sounds like the safe thing to do.

[commit-bot: I haz the power, 2016-05-17 23:47:14]

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

[commit-bot: I haz the power, 2016-05-17 23:47:15]

Dry run: This issue passed the CQ dry run.

[Dirk Pranke, 2016-05-18 00:45:58]

The CQ bit was checked by dpranke@chromium.org

[commit-bot: I haz the power, 2016-05-18 00:46:52]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1981313003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1981313003/1

[commit-bot: I haz the power, 2016-05-18 00:52:34]

Description was changed from

==========
Revert of Flip remaining CQ Win builders to GN. (patchset #11 id:200001 of
https://codereview.chromium.org/1908733002/ )

Reason for revert:
reverting this until after the branch to maintain GYP coverage ...

Original issue's description:
> Flip remaining CQ Win builders to GN.
> 
> This CL flips the remaining CQ builders on tryserver.chromium.win to GN:
> 
>   - win_chromium_compile_dbg_ng
>   - win_chromium_rel_ng
> 
> and some matching bots on chromium.win:
> 
>   - Win Builder
>   - Win Builder (dbg)
> 
> and tryserver.chromium.win:
> 
>   - win_chromium_compile_rel_ng
>   - win_chromium_dbg_ng
> 
> and the affected GPU bots that are also tested by win_chromium_rel_ng:
> 
>   - GPU Win Builder
>   - GPU Win Builder (dbg)
> 
> R=brucedawson@chromium.org, brettw@chromium.org
> BUG=605318
> 
> Committed: https://crrev.com/77405408a41233916670d7c4cbf0e18ad451a802
> Cr-Commit-Position: refs/heads/master@{#393560}

TBR=brettw@chromium.org,brucedawson@chromium.org,kbr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=605318
==========

to

==========
Revert of Flip remaining CQ Win builders to GN. (patchset #11 id:200001 of
https://codereview.chromium.org/1908733002/ )

Reason for revert:
reverting this until after the branch to maintain GYP coverage ...

Original issue's description:
> Flip remaining CQ Win builders to GN.
> 
> This CL flips the remaining CQ builders on tryserver.chromium.win to GN:
> 
>   - win_chromium_compile_dbg_ng
>   - win_chromium_rel_ng
> 
> and some matching bots on chromium.win:
> 
>   - Win Builder
>   - Win Builder (dbg)
> 
> and tryserver.chromium.win:
> 
>   - win_chromium_compile_rel_ng
>   - win_chromium_dbg_ng
> 
> and the affected GPU bots that are also tested by win_chromium_rel_ng:
> 
>   - GPU Win Builder
>   - GPU Win Builder (dbg)
> 
> R=brucedawson@chromium.org, brettw@chromium.org
> BUG=605318
> 
> Committed: https://crrev.com/77405408a41233916670d7c4cbf0e18ad451a802
> Cr-Commit-Position: refs/heads/master@{#393560}

TBR=brettw@chromium.org,brucedawson@chromium.org,kbr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=605318
==========

[commit-bot: I haz the power, 2016-05-18 00:52:35]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-05-18 00:53:51]

Description was changed from

==========
Revert of Flip remaining CQ Win builders to GN. (patchset #11 id:200001 of
https://codereview.chromium.org/1908733002/ )

Reason for revert:
reverting this until after the branch to maintain GYP coverage ...

Original issue's description:
> Flip remaining CQ Win builders to GN.
> 
> This CL flips the remaining CQ builders on tryserver.chromium.win to GN:
> 
>   - win_chromium_compile_dbg_ng
>   - win_chromium_rel_ng
> 
> and some matching bots on chromium.win:
> 
>   - Win Builder
>   - Win Builder (dbg)
> 
> and tryserver.chromium.win:
> 
>   - win_chromium_compile_rel_ng
>   - win_chromium_dbg_ng
> 
> and the affected GPU bots that are also tested by win_chromium_rel_ng:
> 
>   - GPU Win Builder
>   - GPU Win Builder (dbg)
> 
> R=brucedawson@chromium.org, brettw@chromium.org
> BUG=605318
> 
> Committed: https://crrev.com/77405408a41233916670d7c4cbf0e18ad451a802
> Cr-Commit-Position: refs/heads/master@{#393560}

TBR=brettw@chromium.org,brucedawson@chromium.org,kbr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=605318
==========

to

==========
Revert of Flip remaining CQ Win builders to GN. (patchset #11 id:200001 of
https://codereview.chromium.org/1908733002/ )

Reason for revert:
reverting this until after the branch to maintain GYP coverage ...

Original issue's description:
> Flip remaining CQ Win builders to GN.
>
> This CL flips the remaining CQ builders on tryserver.chromium.win to GN:
>
>   - win_chromium_compile_dbg_ng
>   - win_chromium_rel_ng
>
> and some matching bots on chromium.win:
>
>   - Win Builder
>   - Win Builder (dbg)
>
> and tryserver.chromium.win:
>
>   - win_chromium_compile_rel_ng
>   - win_chromium_dbg_ng
>
> and the affected GPU bots that are also tested by win_chromium_rel_ng:
>
>   - GPU Win Builder
>   - GPU Win Builder (dbg)
>
> R=brucedawson@chromium.org, brettw@chromium.org
> BUG=605318
>
> Committed: https://crrev.com/77405408a41233916670d7c4cbf0e18ad451a802
> Cr-Commit-Position: refs/heads/master@{#393560}

TBR=brettw@chromium.org,brucedawson@chromium.org,kbr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=605318

Committed: https://crrev.com/fe79b4b7bd484f806217fd8668de721306ff376a
Cr-Commit-Position: refs/heads/master@{#394300}
==========

[commit-bot: I haz the power, 2016-05-18 00:53:52]

Patchset 1 (id:??) landed as
https://crrev.com/fe79b4b7bd484f806217fd8668de721306ff376a
Cr-Commit-Position: refs/heads/master@{#394300}

[Nico, 2016-05-18 02:34:17]

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

[Nico, 2016-05-18 02:34:17]

win_chromium_rel_ng had lots and lots of pending jobs today, and job lengths
over 2h. It seems like this is getting better now.
(https://build.chromium.org/p/tryserver.chromium.win/stats/win_chromium_rel_ng -
but i think there are better dashboards elsewhere.) This is probably something
to watch out for before relanding.

[Dirk Pranke, 2016-05-18 03:09:26]

On 2016/05/18 02:34:17, Nico (vacation May 19 to 22) wrote:
> win_chromium_rel_ng had lots and lots of pending jobs today, and job lengths
> over 2h. It seems like this is getting better now.
> (https://build.chromium.org/p/tryserver.chromium.win/stats/win_chromium_rel_ng
-
> but i think there are better dashboards elsewhere.) This is probably something
> to watch out for before relanding.

Good observation. 

From scanning 

https://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel...

it looks like we are still seeing webkit_tests failing with and without the
patch, which will definitely
increase the load.

Also, once we re-land this, we can shut off the GN bots, which'll free up a
bunch of resources.

[Nico, 2016-05-18 03:37:45]

But jobs taking > 2h is unusual, right? Neither webkit_tests nor more
slaves will help with peak latency for individual jobs
On May 17, 2016 11:09 PM, <dpranke@chromium.org> wrote:

> On 2016/05/18 02:34:17, Nico (vacation May 19 to 22) wrote:
> > win_chromium_rel_ng had lots and lots of pending jobs today, and job
> lengths
> > over 2h. It seems like this is getting better now.
> > (
> https://build.chromium.org/p/tryserver.chromium.win/stats/win_chromium_rel_ng
> -
> > but i think there are better dashboards elsewhere.) This is probably
> something
> > to watch out for before relanding.
>
> Good observation.
>
> From scanning
>
>
>
https://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel...
>
> it looks like we are still seeing webkit_tests failing with and without the
> patch, which will definitely
> increase the load.
>
> Also, once we re-land this, we can shut off the GN bots, which'll free up a
> bunch of resources.
>
> https://codereview.chromium.org/1981313003/
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Dirk Pranke, 2016-05-18 04:17:57]

On 2016/05/18 03:37:45, Nico (vacation May 19 to 22) wrote:
> But jobs taking > 2h is unusual, right? Neither webkit_tests nor more
> slaves will help with peak latency for individual jobs

If webkit_tests fails, then the patch gets de-applied and we have to do
another full compile. That can certainly push things over 2h depending
on what else is going on.

But, yes, having more slaves would only help to reduce pending jobs,
and that assumes we're not bottlenecked on, e.g., goma cycles.

[Nico, 2016-05-18 11:24:07]

On Wed, May 18, 2016 at 12:17 AM, <dpranke@chromium.org> wrote:

> On 2016/05/18 03:37:45, Nico (vacation May 19 to 22) wrote:
> > But jobs taking > 2h is unusual, right? Neither webkit_tests nor more
> > slaves will help with peak latency for individual jobs
>
> If webkit_tests fails, then the patch gets de-applied and we have to do
> another full compile. That can certainly push things over 2h depending
> on what else is going on.
>

Ah, right. (I still feel we shouldn't retry without patch, the benefits
seem too small for the gain to me still.)


>
> But, yes, having more slaves would only help to reduce pending jobs,
> and that assumes we're not bottlenecked on, e.g., goma cycles.
>
> https://codereview.chromium.org/1981313003/
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Nico, 2016-05-18 11:24:35]

On Wed, May 18, 2016 at 7:24 AM, Nico Weber <thakis@chromium.org> wrote:

> On Wed, May 18, 2016 at 12:17 AM, <dpranke@chromium.org> wrote:
>
>> On 2016/05/18 03:37:45, Nico (vacation May 19 to 22) wrote:
>> > But jobs taking > 2h is unusual, right? Neither webkit_tests nor more
>> > slaves will help with peak latency for individual jobs
>>
>> If webkit_tests fails, then the patch gets de-applied and we have to do
>> another full compile. That can certainly push things over 2h depending
>> on what else is going on.
>>
>
> Ah, right. (I still feel we shouldn't retry without patch, the benefits
> seem too small for the gain to me still.)
>

...too small for the _cost_. /me gets coffee


>
>
>>
>> But, yes, having more slaves would only help to reduce pending jobs,
>> and that assumes we're not bottlenecked on, e.g., goma cycles.
>>
>> https://codereview.chromium.org/1981313003/
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Nico, 2016-05-18 11:56:03]

Actually,
https://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel...
took over 2h without retrying any tests. So this isn't due just to
webkit_tests.
On May 18, 2016 7:24 AM, "Nico Weber" <thakis@chromium.org> wrote:

> On Wed, May 18, 2016 at 7:24 AM, Nico Weber <thakis@chromium.org> wrote:
>
>> On Wed, May 18, 2016 at 12:17 AM, <dpranke@chromium.org> wrote:
>>
>>> On 2016/05/18 03:37:45, Nico (vacation May 19 to 22) wrote:
>>> > But jobs taking > 2h is unusual, right? Neither webkit_tests nor more
>>> > slaves will help with peak latency for individual jobs
>>>
>>> If webkit_tests fails, then the patch gets de-applied and we have to do
>>> another full compile. That can certainly push things over 2h depending
>>> on what else is going on.
>>>
>>
>> Ah, right. (I still feel we shouldn't retry without patch, the benefits
>> seem too small for the gain to me still.)
>>
>
> ...too small for the _cost_. /me gets coffee
>
>
>>
>>
>>>
>>> But, yes, having more slaves would only help to reduce pending jobs,
>>> and that assumes we're not bottlenecked on, e.g., goma cycles.
>>>
>>> https://codereview.chromium.org/1981313003/
>>>
>>
>>
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.