Closed Bug 787483 Opened 12 years ago Closed 12 years ago

SecReview: Armv6 support

Categories

(mozilla.org :: Security Assurance: Review Request, task)

Product:
mozilla.org
Component:
Security Assurance: Review Request
Type:
task
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
VERIFIED FIXED

People

(Reporter: curtisk, Assigned: decoder)

References

(
URL
)

Details

(Whiteboard: [completed secreview])

SecReview tracking bug Actions regarding the review of the dependent bug should be tracked here.
Inspection of the armv6 support bug (bug 697205) and discussion with the mobile team suggest that there's little that can be secreviewed (it's mostly releng work). There are some outstanding questions around JS changes (there's a minor change for soft-float support) and it may be worth including ARMv6 in our fuzzing work. Assigning to decoder to deal with once discussions on ARMv6 fuzzing have taken place.
Assignee: mgoodwin → choller
Whiteboard: [pending secreview][start mm/dd/yyyy][target mm/dd/yyyy] → [completed secreview]
According to dvander, the JS engine behaves on ARMv6 almost the same way as on ARMv7 when running with the --no-ti option. The only remaining differences are subtle instruction changes, but those are unlikely to cause trouble compared to the untested paths that we hit with --no-ti. I'll give that --no-ti configuration a try on X86 and ARM and maybe gkw can do the same with jsfunfuzz. With that in place, we should be more than good from the JS side.
Status: NEW → RESOLVED
Closed: 12 years ago
Resolution: --- → FIXED
We already fuzz with random flags including --no-ti on x86. ARM will have to wait till my pandaboard ES arrives, which has already been ordered.
But of course, pandaboards are ARMv7, so we have to think of ARMv6 hardware another day..
Status: RESOLVED → VERIFIED
Blocks: 785073
You need to log in before you can comment on or make changes to this bug.