1213351 - Intermittent aurora build fails with Lowering-shared-inl.h:188:5: error: no matching function for call to 'js::jit::LFloatReg::LFloatReg(const ARMFPRegister&)'

Status: RESOLVED FIXED

(Core :: JavaScript Engine: JIT, defect, P3)

Description

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://treeherder.mozilla.org/logviewer.html#?job_id=1295370&repo=mozilla-aurora

Comment 1

[Benjamin Bouvier [:bbouvier] (inactive)]

It's an ARM64 build => cc'ing Jakob, in case he knows what happens here.

Comment 2

[Jakob Stoklund Olesen [:jolesen]]

+sfink

This is fallout from the patches in bug 984018 - sincos optimization.

Victor, could you check if only some of your patches made it to the aurora branch, please?

You may want to the the ARM build too.

Comment 3

[Victor Carlquist]

Hi guys,
The last patch had not landed in aurora (bug 984018 attachment 8666111 [details] [diff] [review]).

Should I repost the patch here and fill out the approval request to aurora or should I write the approval request in bug 984018?

thanks.

Comment 4

[Nicolas B. Pierron [:nbp]]

(In reply to Victor Carlquist from comment #3)
> The last patch had not landed in aurora (bug 984018 attachment 8666111 [details] [diff] [review]
> [details] [diff] [review]).

You should request uplift approval on the attachment of the original bug.

Comment 5

[Jakob Stoklund Olesen [:jolesen]]

Thanks, Victor.

If you get any resistance, make sure to mention that the sincos optimization is already in aurora, it is only the small build error fix for ARM64 that is missing.

Comment 6

[Victor Carlquist]

Thanks guys :)

Comment 7

[Jan de Mooij [:jandem]]

I think ARM64 fixes don't need approval and can land with a=npotb (not part of the build), because we aren't shipping an ARM64 JIT atm.

Comment 8

[BMO Automation]

Bulk assigning P3 to all open intermittent bugs without a priority set in Firefox components per bug 1298978.

Comment 9

[Jon Coppeard (:jonco)]

Fixed by uplift to aurora in bug 984018.