Primordial message pipe

[Sam McNally]

Hi,

I've put together a doc about the primordial browser-renderer message pipe for supporting the migration from Chrome IPC: https://docs.google.com/document/d/18mqeSuru517BPRkA-t0Q68aCzS3dzxsDnKLZdtd2VL4/edit?usp=sharing. Please take a look.

[Yuzhu Shen]

(reposted to a bigger group)

I added a comment in the design doc about how associated interface endpoints not listening could affect message order. I think a concrete example may help to explain my thoughts: Example

Ben, John and I had a chat about this. One possible solution is to let mojo's router (MultiplexRouter) also manage the IPC message dispatching from IO to UI thread, so that we can guarantee they arrive in the right order. That requires exposing some API from MultiplexRouter. But maybe it is also helpful to take a step back and consider whether we do need to strictly preserve message order.

On Wed, Feb 3, 2016 at 4:00 PM, Sam McNally <sa...@chromium.org> wrote:

Hi,

I've put together a doc about the primordial browser-renderer message pipe for supporting the migration from Chrome IPC: https://docs.google.com/document/d/18mqeSuru517BPRkA-t0Q68aCzS3dzxsDnKLZdtd2VL4/edit?usp=sharing. Please take a look.

--
You received this message because you are subscribed to the Google Groups "chromium-mojo" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chromium-moj...@chromium.org.
To post to this group, send email to chromi...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-mojo/CAJqEsoAZ%3Dzi4dr_jcyyPinVP6_cPHxXpoK%2B3FukOn%3DNTKmLUKg%40mail.gmail.com.

[Sam McNally]

I've taken some benchmarks of the current primordial pipe channel: https://docs.google.com/document/d/18mqeSuru517BPRkA-t0Q68aCzS3dzxsDnKLZdtd2VL4/edit?pref=2&pli=1#heading=h.nbdod4egkxys. HandleWatcher is slow.