| 1 |
Hi Hanni, |
| 2 |
|
| 3 |
Distcc would be especially useful to test out 'out-of-the-box' |
| 4 |
successful toolchain compilations - it takes approximately 10 (failure) |
| 5 |
- 50 (success) minutes for me to test toolchain builds on x86 |
| 6 |
|
| 7 |
To take into account every permutation of libc, binutils, |
| 8 |
kernel-headers, and gcc version - it could easily approach a week or |
| 9 |
more on a uniprocessor system! |
| 10 |
|
| 11 |
It would likely also be entirely possible to integrate that into the |
| 12 |
buildroot itself if one had access to several similar architectures |
| 13 |
running an instance of the chroot. |
| 14 |
|
| 15 |
The script is really huge - I'm currently still in the middle of trying |
| 16 |
to wrap my head around all of the toolchain-specific files / directories |
| 17 |
/ subsystems that are required to mount / create / destroy upon chroot / |
| 18 |
exit. |
| 19 |
|
| 20 |
I'll send you the script, and please try it out - no license information |
| 21 |
is included (b/c at the moment it is still alpha) but will surely be |
| 22 |
available by the free-est GPL or BSD license. |
| 23 |
|
| 24 |
|
| 25 |
It's good to hear that you could see the value in this too!! |
| 26 |
|
| 27 |
|
| 28 |
~/Chris |
| 29 |
|
| 30 |
Hanni Ali wrote: |
| 31 |
> Sounds fascinating and potentially a very useful tool. |
| 32 |
> |
| 33 |
> I would be keen to integrate distcc support into it to allow faster builds, |
| 34 |
> I would be happy to work on that aspect of the script if you like. |
| 35 |
> |
| 36 |
> Hanni |
| 37 |
> |
| 38 |
-- |
| 39 |
gentoo-embedded@g.o mailing list |
Archives