ensure endian safety for non-little-endian machines
-
Nikias Bassen- State changed from new to open
- Assigned user set to Matt Colyer
I did some investigation and found out that libiphone's code except AFC should be endian safe. For AFC a patch is attached that should make it work on big endian machines too. However I have no hardware to test it on, so perhaps somebody may verify this.
-
Matt ColyerI need someone to test this on the proper machine before I can apply.
-
Martin S.- Assigned user cleared.
- State changed from open to hold
- Milestone changed from 1.0 Release to 1.2.0 Release
Following options to test:
- Some QEMU image
- An older Mac
- ARM distro packages in big-endian mode
I think we can for sure hold this back until we have properly tested it.
As whatever issues it might cause, it still won't affect the public API, only private code.If anyone takes the challenge to make it run on a big endian machine, he'll sure be able to verify this patch and libiphone functionality.
Once verified we can then ship a new release with the internals taking endianness into account which should easily be possible past 1.0.
Moving this on hold unless someone comes up with a test of course. :)
-
Matt Colyer
- State changed from hold to resolved
(from [12a44afce569c2cdf31018c259745e09b13bffc3]) added endian safety to AFC
This should make libiphone compatible with big endian machines.
[#85 ensure endian safety for non-little-endian machines state:resolved]
Signed-off-by: Matt Colyer matt@colyer.name
http://github.com/MattColyer/libiphone/commit/12a44afce569c2cdf3101... -
Matt Colyer
- State changed from resolved to hold
Applied this to resolve #82 Use correct types in AFC read/write signatures for length.
Please Sign in or create a free account to add a new ticket.
With your very own profile, you can contribute to projects, track your activity, watch tickets, receive and update tickets through your email and much more.