On Sat, Sep 13, 2008 at 02:10, Justin <justin@j-schmitz.net> wrote:
> Nikos Chantziaras schrieb:
>>
>> Nikos Chantziaras wrote:
>>>
>>> [...]
>>> The only real option is to get more RAM.
>>
>> Another option is to emerge on another machine, build binary packages and
>> emerge those in the embedded system.

Yeah, it looks like this is the way to go. I've never used binary
packages so this will be an interesting exercise.

> Or mount / as nfs on another machine, chroot in that and compile. This skips
> the whole extraction part on the embedded system.

Interesting. I'm not sure if I can, though. I don't have control of
the kernel. I'll investigate.

> What about temporary giving the VM more resources?

Nope, not possible unless I pay more. I had intended to simply use a
big swap file but even that doesn't work. This virtual box doesn't
support it.

Hilco Wijbenga wrote:

On Sat, Sep 13, 2008 at 02:10, Justin <justin@j-schmitz.net> wrote:

Nikos Chantziaras schrieb:

Nikos Chantziaras wrote:

[...]
The only real option is to get more RAM.

Another option is to emerge on another machine, build binary packages and
emerge those in the embedded system.


Yeah, it looks like this is the way to go. I've never used binary
packages so this will be an interesting exercise.


If the other machine has its packages compiled with the same "-march"
GCC flag as the embedded one, you can simply do:


quickpkg atom

This will simply create a binary package of the specified package atom
without compiling anything (it will use the existing installation of the
package). This will of course only work if the package is already
installed. If not, use emerge to compile and build the package without
actually installing it.


The only thing to keep in mind is the -march flag. It must be for the
embedded system, not the current system.

On Sun, Sep 14, 2008 at 3:01 AM, Alan McKinnon <alan.mckinnon@gmail.com> wrote:



Yeah, funny that. Maybe you can write the new RFC to synchronize all reply

mail to all other replies anywhere in the world.



--

alan dot mckinnon at gmail dot com




When they figure out how to use quantum entangled bits to transfer classical information ( currently deemed impossible )* you can use that. Or we could always have the equivalent of "somebody is typing a reply" notification subsystem that broadcasts to all recipients of the original email using a message id ( Hey, that actually does not sound too bad :P ) *




--
Kent
ruby -e '[1, 2, 4, 7, 0, 9, 5, 8, 3, 10, 11, 6, 12, 13].each{|x|
print "enNOSPicAMreil kdrtf@gma.com"[(2*x)..(2*x+1)]}'