I notice the issue is with atomic gcc functions, may them be rewritten to 
access alignment issues?? (custom built forestdb may be?)

El jueves, 5 de marzo de 2015, 17:21:45 (UTC-5), Frederic Yesid Peña 
Sánchez escribió:
>
> Umm ok, i will test on a iPad Mini 2nd gen (x64) tomorrow, i will keep in 
> touch.
>
> There is some insight on how to fix it by myself (some sparing time this 
> night after college).
>
> El jueves, 5 de marzo de 2015, 17:18:58 (UTC-5), Jens Alfke escribió:
>>
>>
>> On Mar 5, 2015, at 1:50 PM, Frederic Yesid Peña Sánchez <
>> [email protected]> wrote:
>>
>> I note it tries to acess an atomic_val_add_64 but my iPad Mini is 1st gen 
>> (no x64) so it might be the issue, but no idea how to specify the platform 
>> for library.
>>
>>
>> My suspicion is that ForestDB may not work correctly in 32-bit ARM at the 
>> moment. The CPU exception is
>> EXC_BAD_ACCESS (code=EXC_ARM_DA_ALIGN, address=0x18083b34)
>> Looks like the CPU is complaining about invalid data alignment — I know 
>> ARM is much stricter about this than x86. Since it’s operating on an 8-byte 
>> value I think that means it needs to be 8-byte aligned, and that address 
>> isn’t. There’s probably something that needs to be fixed about the struct 
>> alignment in the calling code; I’ll let the ForestDB team know.
>>
>> (We haven’t gotten to formal testing of the ForestDB-based CBL yet; when 
>> we do it’ll include 32-bit as well as 64-bit. For now the only devices I’ve 
>> tested on both run 64-bit…)
>>
>> —Jens
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Couchbase Mobile" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/mobile-couchbase/fc58d1fe-af0d-4a8c-a2af-835c3dda29df%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to