836033 - Event - (2013-02-01 10:30 UTC-8) - Ten Forward - Part I : Higgs, a Monitoring JIT for JavaScript Part II: In Praise of Metacircular Virtual Machine Layering

Status: RESOLVED FIXED

(Air Mozilla :: Events, task)

Description

[Azita]

Topic: Part I : Higgs, a Monitoring JIT for JavaScript   Part II: In Praise of Metacircular Virtual Machine Layering
Presenter: Maxime Chevalier-Boisvert and Erick Lavoie
Date: 2013-02-01
Time: 10:00 UTC-8
Duration: 2 hours
Privacy: The Public
Large Screen Location: MTV, 
Description: 
There are 2 parts to this presentation.  Total presentation time will be about 1.5 hours with 1/2 hour for questions.
Part I: Higgs, a Monitoring JIT for JavaScript
Part II: In Praise of Metacircular Virtual Machine Layering

Do I need to list the detail description of each section here?
Special Requirements:

Comment 1

[Azita]

Update to the description:
2 presentations:
Part I: Maxime Chevalier-Boisvert will present Higgs, a new JIT compiler architecture under development in which monitoring is used to maintain accurate type information at a minimal cost, pushing the limits of dynamic language optimization.

Part II: Erick Lavoie will show how to provide a flexible run-time instrumentation of the JavaScript object-model and function-call protocol with a performance comparable to a state-of-the-art interpreter, without having to modify the VM
source code.

Comment 2

[Patrick Dang [:pdang]]

Updated start time to 10:30AM PST

Comment 3

[Richard A Milewski[:richard]]

Azita,

Please confirm these presenters are from outside Mozilla.

Comment 4

[Richard A Milewski[:richard]]

Patrick,

If you can, arrange to stop and restart the recording between speakers.  That way we can archive this as two separate presentations.

Comment 5

[Azita]

Presenters are not Mozilla employees, but will be at Mozilla MV to present in person.  Does this answer your questions?

Comment 6

[Patrick Dang [:pdang]]

(In reply to Richard A Milewski[:richard] from comment #4)
> Patrick,
> 
> If you can, arrange to stop and restart the recording between speakers. 
> That way we can archive this as two separate presentations.

Hey Richard, sure thing.