Re: [Kolab-devel] Kolab on RaspberryPi

Hi Andreas.

Welcome to the list. It's awesome there's another one using
Kolab on arm.

Am Tue, 11 Mar 2014 10:57:13 +0100
schrieb Andreas Cordes <[email protected]>:
> I found lots of posts saying, not possible or ARM packages not 
> available.

Not officially supported, yes. There were several questions regarding 
arm here (devel and users list) but no useful answers so far. 
At least not as far as I know of.

But you are not the only one using Kolab on ARM. I offered my arm
packages last Saturday to this list. And I think there are more arm
users out there. 

After using Kolab 2 for a long time on MIPS (Lemote Fuloong, official
debian packages) I started  experimenting with Kolab 3.0 on ARM in Sep
2013, put those experiments on hold for some months, restarted with
Kolab 3.1 some weeks ago and updated to kolab 3.2 a couple of days
ago. I transferred my data from the old Fuloong to the new Wandboard
some weeks ago. And it's running fine since then, even if roundcube is
a bit more sluggish than it could be. (But because horde (kolab 2) was
even slower on the old MIPS I started added squirrelmail (mail only, no
groupware) some time ago and this one is fast as hell now.)

My binaries are not usable on RasPi, because I'm using Debian
armhf on ARMv7, not Raspbian. But maybe we can join forces!? I think
for me it's ok to use binaries for ARMv6 even if I have an ARMv7
(Cortex A9). 

Do you know if it's ok to mix ARMv6 and ARMv7 binaries (e.g. shared
libraries)? I downloaded a binary from the raspbian repository to check
if it runs on my ARMv7 machine and yes, it does. But maybe this was
just luck!? If it generally works, even with mixed ARMv6 and ARMv7
shared libraries, then I could mix Kolab ARMv6 and Debian armhf base
system here.

Or maybe it's ok to have both, ARMv6 and ARMv7. It's not a big deal to
compile them. It works out of the box. (While compiling things on
Loongson 2E with MIPS64 N32 ABI is no fun - many thinks don't work ootb)

regards 
hede

(this mail was transmitted via kolab on arm)