793911 - Implement jwcrypto.js natively

Status: RESOLVED WONTFIX

(Core Graveyard :: Identity, enhancement)

Description

[Ben Adida [:benadida]]

so it can be fast!

Comment 1

[Ben Adida [:benadida]]

we have existing navigator.id related crypto functions in IdentityCryptoService. We should expose these to the server-provided identity selector (and invisible assertion generator), using JS injection via loadFrameScript, so that the process is much faster on low-powered phones.

Comment 2

[Jason Smith [:jsmith]]

Is this a feature or a bug? If it's a feature, this needs to be a minus per today's discussion in the b2g meeting.

Comment 3

[Ben Adida [:benadida]]

Jason: we need this, unless you want 8-second login times on the phone, which clee marked as blocker somewhere else.

This is on the implementation plan, I would recommend checking in with us before relabeling these, we're going to waste quite a bit of time in the churn otherwise.

Comment 4

[Ben Adida [:benadida]]

we've done some initial performance testing on the unagi device, and the crypto component does not appear to be the bottleneck. We'll continue testing here, but this bug may be worth lowering in priority.

Comment 5

[Jason Smith [:jsmith]]

(In reply to Ben Adida [:benadida] from comment #4)
> we've done some initial performance testing on the unagi device, and the
> crypto component does not appear to be the bottleneck. We'll continue
> testing here, but this bug may be worth lowering in priority.

What about the Otoro device? If you need one for testing, please let me know (we've got a bunch in the QA area).

Comment 6

[Ben Adida [:benadida]]

(In reply to Jason Smith [:jsmith] from comment #5)
> What about the Otoro device? If you need one for testing, please let me know
> (we've got a bunch in the QA area).

The Otoro device builds we have are pretty slow for just about everything. Do you have one that is usable in general? And is that device still a target for performance?

Comment 7

[Jason Smith [:jsmith]]

(In reply to Ben Adida [:benadida] from comment #6)
> (In reply to Jason Smith [:jsmith] from comment #5)
> > What about the Otoro device? If you need one for testing, please let me know
> > (we've got a bunch in the QA area).
> 
> The Otoro device builds we have are pretty slow for just about everything.
> Do you have one that is usable in general? And is that device still a target
> for performance?

To my understanding Otoro was the key target point for perf (Unagi devices are just dogfooding devices). 

As for the builds, usable depends on the daily build quality state (which can be anything from down right broken to smoke tests passing). I could talk with John H. to see if you want a golden build specifically if you want that (high quality state). If you just need a device to borrow to test something, I could get that as well from our QA stash if need be (I'll have to check what extras we have).

Comment 8

[Jason Smith [:jsmith]]

(In reply to Jason Smith [:jsmith] from comment #7)
> (In reply to Ben Adida [:benadida] from comment #6)
> > (In reply to Jason Smith [:jsmith] from comment #5)
> > > What about the Otoro device? If you need one for testing, please let me know
> > > (we've got a bunch in the QA area).
> > 
> > The Otoro device builds we have are pretty slow for just about everything.
> > Do you have one that is usable in general? And is that device still a target
> > for performance?
> 
> To my understanding Otoro was the key target point for perf (Unagi devices
> are just dogfooding devices). 
> 
> As for the builds, usable depends on the daily build quality state (which
> can be anything from down right broken to smoke tests passing). I could talk
> with John H. to see if you want a golden build specifically if you want that
> (high quality state). If you just need a device to borrow to test something,
> I could get that as well from our QA stash if need be (I'll have to check
> what extras we have).

Change that - I just got an update from Tony. Otoro isn't a target device anymore - Unagi is number #1. So ignore my comments above.

Comment 9

[Ben Adida [:benadida]]

thanks Jason, I'm marking this no-longer b2g blocking then.

Comment 10

[Ryan Kelly [:rfkelly]]

I'm WONTFIXing persona-related bugs now that we've committed to decommissioning in the persona.or service.