Gustavo Noronha Silva writes:
> Currently the package uses the default, I tend to use binutils-gold on
> my personal builds.
That explains the memory problem. Is there some technical problem for
using gold also for the official builds?
--
To UNSUBSCRIBE, email to debian-arm-requ...@lists.debia
Em Qui, 2013-08-22 às 09:08 +0300, Timo Juhani Lindfors escreveu:
> Gustavo Noronha Silva writes:
> > In this case the problem was the OOM killer got ld killed.
>
> Are you using ld.bfd or ld.gold?
Currently the package uses the default, I tend to use binutils-gold on
my personal builds.
--
Gu
Hello,
2013/8/22 Loïc Minier :
> FWIW, the Calxeda boxes are armhf and armel capable and make a massive
> difference in buildd performance (good SATA I/O, 4G of RAM, SMP); these
> aren't exactly cheap (yet?), but I definitely wish Debian armel and
> armhf builds all eventually move to server clas
On Wed, Aug 21, 2013, Daniel Kahn Gillmor wrote:
> If there's a problem with the buildd's for armel, can we clarify what
> the specific problem is and try to get more hardware resources thrown at
> it? debian does have some funds, and this is a classic example of what
> we should be using them for
Hector Oron wrote (ao):
> The fundamental cause for the build failure seems to be memory
> limitation, while webkit requires all the available RAM in buildd
> (1.5GB), is uses all the SWAP (3GB) and final linking stage explodes
> only after 3d 14h 13m or so [1]... we could still try to make swap
>
5 matches
Mail list logo
