Add VR test APKs hooks to DEPS

Add VR test APKs hooks to DEPS

Adds the hooks for the VR test APKs into src/DEPS since it looks like
there isn't a consensus on how to add custom hooks to particular bots.

BUG=671373
Review-Url: https://codereview.chromium.org/2712733002
Cr-Commit-Position: refs/heads/master@{#452174}
Committed: https://chromium.googlesource.com/chromium/src/+/828f55534c88e66ac7d9ae7ed7c6ac3d5ed6172f

[bsheedy, 2017-02-22 17:21:07]

bsheedy@chromium.org changed reviewers:
+ jbudorick@chromium.org

[jbudorick, 2017-02-22 17:28:15]

lgtm

[bsheedy, 2017-02-22 17:29:40]

The CQ bit was checked by bsheedy@chromium.org

[commit-bot: I haz the power, 2017-02-22 17:30:15]

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, 2017-02-22 20:01:27]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1487784580526700, "parent_rev":
"8b8e7a6af44c9e563eb2584a9d43c483051d41c1", "commit_rev":
"828f55534c88e66ac7d9ae7ed7c6ac3d5ed6172f"}

[commit-bot: I haz the power, 2017-02-22 20:05:20]

Description was changed from

==========
Add VR test APKs hooks to DEPS

Adds the hooks for the VR test APKs into src/DEPS since it looks like
there isn't a consensus on how to add custom hooks to particular bots.

BUG=671373
==========

to

==========
Add VR test APKs hooks to DEPS

Adds the hooks for the VR test APKs into src/DEPS since it looks like
there isn't a consensus on how to add custom hooks to particular bots.

BUG=671373

Review-Url: https://codereview.chromium.org/2712733002
Cr-Commit-Position: refs/heads/master@{#452174}
Committed:
https://chromium.googlesource.com/chromium/src/+/828f55534c88e66ac7d9ae7ed7c6...
==========

[commit-bot: I haz the power, 2017-02-22 20:05:21]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/828f55534c88e66ac7d9ae7ed7c6...

[Nico, 2017-02-24 16:08:56]

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

[Nico, 2017-02-24 16:08:56]

Most devs don't need this. It added like 20s to runhooks time on my linux corp
box, which is on super fast internet. Can we keep these opt-in please?

[Nico, 2017-02-24 16:09:18]

(The bot problem can be solved differently.)

[bsheedy, 2017-02-24 17:19:12]

On 2017/02/24 16:09:18, Nico wrote:
> (The bot problem can be solved differently.)

If this is causing issues, feel free to revert it/change it to be opt-in.

[jbudorick, 2017-03-01 02:24:48]

On 2017/02/24 16:09:18, Nico wrote:
> (The bot problem can be solved differently.)

We discussed a few of these different options w/ infra folks prior to landing
this. This was the implementation that we settled on *despite* the vast majority
of devs not needing it, as it:
 - doesn't require additional highly specific hooks in the chromium recipe
 - doesn't require modifications to bot_update
 - isn't exorbitantly large (~100MB+) and isn't updated frequently (~once a
month, iirc)

[bsheedy, 2017-03-01 02:28:35]

On 2017/03/01 02:24:48, jbudorick wrote:
> On 2017/02/24 16:09:18, Nico wrote:
> > (The bot problem can be solved differently.)
> 
> We discussed a few of these different options w/ infra folks prior to landing
> this. This was the implementation that we settled on *despite* the vast
majority
> of devs not needing it, as it:
>  - doesn't require additional highly specific hooks in the chromium recipe
>  - doesn't require modifications to bot_update
>  - isn't exorbitantly large (~100MB+) and isn't updated frequently (~once a
> month, iirc)

On a related note, I have this out which will help with the download size a bit
https://codereview.chromium.org/2724643002.

The removal of most of the Daydream Home APKs is particularly helpful since they
were the largest (~25 MB vs ~8 MB for a VR Services APK).

[Nico, 2017-03-01 02:54:10]

I feel pretty strongly it's the wrong decision. (100MB/month is quite a
lot, especially given that about 0 devs use this, and doing this just for
bots sets a really bad precedent). What are next steps? Do you have a
doc/bug with other approaches considered?

On Feb 28, 2017 9:24 PM, <jbudorick@chromium.org> wrote:

On 2017/02/24 16:09:18, Nico wrote:
> (The bot problem can be solved differently.)

We discussed a few of these different options w/ infra folks prior to
landing
this. This was the implementation that we settled on *despite* the vast
majority
of devs not needing it, as it:
- doesn't require additional highly specific hooks in the chromium recipe
- doesn't require modifications to bot_update
- isn't exorbitantly large (~100MB+) and isn't updated frequently (~once a
month, iirc)

https://codereview.chromium.org/2712733002/

-- 
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.

[chromium-reviews, 2017-03-01 02:58:32]

Here's a 3 line suggestion that's imho much better: make the step only
download this stuff if some env var is set, set it on the bots that need
this. (Might require a small py deps hook wrapper, but still easy to
implement)

On Feb 28, 2017 9:54 PM, "Nico Weber" <thakis@chromium.org> wrote:

I feel pretty strongly it's the wrong decision. (100MB/month is quite a
lot, especially given that about 0 devs use this, and doing this just for
bots sets a really bad precedent). What are next steps? Do you have a
doc/bug with other approaches considered?


On Feb 28, 2017 9:24 PM, <jbudorick@chromium.org> wrote:

On 2017/02/24 16:09:18, Nico wrote:
> (The bot problem can be solved differently.)

We discussed a few of these different options w/ infra folks prior to
landing
this. This was the implementation that we settled on *despite* the vast
majority
of devs not needing it, as it:
- doesn't require additional highly specific hooks in the chromium recipe
- doesn't require modifications to bot_update
- isn't exorbitantly large (~100MB+) and isn't updated frequently (~once a
month, iirc)

https://codereview.chromium.org/2712733002/

-- 
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.

[jbudorick, 2017-03-01 15:58:55]

On 2017/03/01 02:54:10, Nico wrote:
> I feel pretty strongly it's the wrong decision. (100MB/month is quite a
> lot, especially given that about 0 devs use this, and doing this just for
> bots sets a really bad precedent). What are next steps? Do you have a
> doc/bug with other approaches considered?

No; this was an in-person discussion I had w/ dpranke, iannucci, and hinoka.
I've filed https://bugs.chromium.org/p/chromium/issues/detail?id=697469 for
discussing this further.

> 
> On Feb 28, 2017 9:24 PM, <mailto:jbudorick@chromium.org> wrote:
> 
> On 2017/02/24 16:09:18, Nico wrote:
> > (The bot problem can be solved differently.)
> 
> We discussed a few of these different options w/ infra folks prior to
> landing
> this. This was the implementation that we settled on *despite* the vast
> majority
> of devs not needing it, as it:
> - doesn't require additional highly specific hooks in the chromium recipe
> - doesn't require modifications to bot_update
> - isn't exorbitantly large (~100MB+) and isn't updated frequently (~once a
> month, iirc)
> 
> https://codereview.chromium.org/2712733002/
> 
> -- 
> 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 mailto:chromium-reviews+unsubscribe@chromium.org.