1249102 - Make overrides of WorkerRunnable::PostRun a bit more consistent

Status: RESOLVED FIXED

(Core :: DOM: Workers, defect)

Description

[Boris Zbarsky [:bzbarsky]]

Right now they do similar things in somewhat different ways.  It would really be better if they didn't do that.

Comment 1

[Boris Zbarsky [:bzbarsky]]

Attachment: Make overrides of WorkerRunnable::PostRun a bit more consistent

Specifically we make the following changes:

1)  Remove WorkerSameThreadRunnable::PostRun, because it does exactly the same
things as WorkerRunnable::PostRun.

2)  Always treat ModifyBusyCountFromWorker as infallible in terms of throwing
JS exceptions.

3)  Change ExtendableFunctionalEventWorkerRunnable::PostRun to properly call
its superclass PostRun so we will correctly decrement the busy count our
PreDispatch incremented.

Comment 2

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

Comment on attachment 8720635 [details] [diff] [review]
Make overrides of WorkerRunnable::PostRun a bit more consistent

Review of attachment 8720635 [details] [diff] [review]:
-----------------------------------------------------------------

Can we go ahead and nerf ModifyBusyCountFromWorker's return value too?

You might want the commit message to mention Pre/PostDispatch too.

::: dom/base/WebSocket.cpp
@@ +2703,5 @@
>    PreDispatch(JSContext* aCx, WorkerPrivate* aWorkerPrivate)
>    {
> +    // We don't call WorkerRunnable::PreDispatch because it would assert the
> +    // wrong thing about which thread we're on.  Which thread _are_ we on,
> +    // exactly?

That's an *excellent* question.  I think you're on either the main thread or the STS thread.  See WebSocketImpl::InitializeConnection and the retargeting, and note that WebSocketImpl is an nsIEventTarget.

Comment 3

[Boris Zbarsky [:bzbarsky]]

> Can we go ahead and nerf ModifyBusyCountFromWorker's return value too?

I _think_ so, but would really rather do that in a followup or something.  

> You might want the commit message to mention Pre/PostDispatch too.

Will do.

> I think you're on either the main thread or the STS thread.

Ah, indeed.  Will document that we're on the thread the channel implementation is running on, which I agree should generally be main or STS.

Comment 4

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/298964285e1f

Comment 5

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/298964285e1f