[Box Backup-dev] tests passing on i386

[Ben Summers]

On 20 Dec 2005, at 21:34, Martin Ebourne wrote:

> On Tue, 2005-12-20 at 18:42 +0000, Ben Summers wrote:
>> On 20 Dec 2005, at 18:36, Martin Ebourne wrote:
>>
>>> With that last commit (which took too long to find), the tests all
>>> pass on i386 FC4, both in release and debug mode.
>>
>> Even raidfile?
>
> I don't think raidfile was ever broken on i386.

That explains the lack of it failing. I'm obviously getting confused.

>
> The only test I have outstanding to fix is raidfile on x86_64 FC4. 240
> failures. All others pass whether release or debug on x86_64, and as I
> said above, all tests now pass on i386.
>
> Maybe when I fix the raidfile test on x86_64 it'll also fix ppc. I  
> think
> all other ppc tests probably now work due to my other fixes.

Maybe. I suspect endian issues for the raidfile on 64, or perhaps  
word length. It does some quite complex stuff to store files without  
using excess blocks on discs, doing stuff like ZORing the length into  
ends of files and the like. All good fun.

>
> How is OpenBSD doing?

Two failures on 3.7: backstore (server throws RaidFile/ 
RaidFileDoesntExist) and backupstorefix (fails miserably), everything  
else is fine. Looking into it...

Also, sys/mount.h isn't detected by autoconf, even though it does  
exist. Editing lib/common/config.h to #define the HAVE_SYS_MOUNT_H  
does make everything compile OK. I shall attempt to work out why  
that's happening.

Ben