855839 - Remove --disable-ion from armv6 mozconfigs

Status: RESOLVED FIXED

(Firefox for Android Graveyard :: General, defect)

Description

[Kartikaya Gupta (email:[e-mail])]

Once bug 836486 is landed (real soon now) we should turn on ionmonkey for our armv6 builds by updating the armv6 mozconfigs.

Comment 1

[Kartikaya Gupta (email:[e-mail])]

Attachment: Patch

Try run is clean at https://tbpl.mozilla.org/?tree=Try&rev=cb653cd963d3. However that may not mean much, as mjrosenb said on IRC:

10:17:31 mjrosenb: before we had a specific armv6 option for tbpl because we statically did all of the hardware compatability checks
10:17:34 mjrosenb: now they are dynamic
10:17:59 mjrosenb: which means tha we aren't actually testing what it would be like to run on an armv6 device
10:18:58 kats: mjrosenb: i'm not sure i understand. do you mean that if we take an armv6 build and run it on an armv7 device it will not use the armv6 codepaths?
10:19:49 mjrosenb: it will use some of them
10:20:15 mjrosenb: actually, on an android device, it may use none of them.
10:21:59 kats: mjrosenb: ok, so on tbpl it won't be testing the armv6 codepaths, but if i take the tbpl build and run it on an actual armv6 device it should be fine, right?
10:22:21 mjrosenb: kats: yes.
10:22:51 kats: ok
10:23:03 kats: i'll make a note of that in the bug, i guess
10:23:12 kats: i still feel we should enable ion on armv6
10:23:52 mjrosenb: yeah, it shouldn't be that hard to switch it back from dynamic checking to static checking
10:24:01 mjrosenb: or just get some armv6 devices

Comment 2

[Kartikaya Gupta (email:[e-mail])]

Also CC'ing some people from bug 789373 in case there's still some reason we need this turned off.

Comment 3

[Kartikaya Gupta (email:[e-mail])]

https://hg.mozilla.org/integration/mozilla-inbound/rev/9fe83576713e

Comment 4

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/9fe83576713e