| 1 |
> brainstorm... |
| 2 |
|
| 3 |
storm rolls in from the south. |
| 4 |
|
| 5 |
> |
| 6 |
> I don't know if this was intended or a side effect of me removing the |
| 7 |
> linux-headers from the SRC_URI but after the build it sure wont exist in |
| 8 |
> all cases. (bug||feature)? |
| 9 |
|
| 10 |
bug - will search out a good list of mirrors /url for it. |
| 11 |
|
| 12 |
> KERNEL_SOURCE="/var/tmp/portage/uclibc-buildroot-20040209/work/buildroot/to |
| 13 |
>olchain_build_i386/linux" |
| 14 |
> |
| 15 |
> Log of the installed files |
| 16 |
> http://dev.gentoo.org/~solar/uclibc/uclibc-buildroot-install-Manifest.txt |
| 17 |
> |
| 18 |
|
| 19 |
Thankyou - reassuring to get a similar list. If anyone constanly gets a asm |
| 20 |
operand error I have had it fairly constantly. Will post more if I work it |
| 21 |
out. |
| 22 |
|
| 23 |
> I see we have some init files etc.. Do you think we going to have |
| 24 |
> problems with the baselayout-lite stuff the other guys have been working |
| 25 |
> on? |
| 26 |
|
| 27 |
Once baselayout-lite gets going a bit more I'll get rid of most of these. |
| 28 |
There is some benefit to keeping this self contained. |
| 29 |
|
| 30 |
> Anyway I'm rebuilding now with my USE flags from before minus nls now. |
| 31 |
> USE="-nls" emerge --debug uclibc-buildroot 2> /tmp/new.x |
| 32 |
> |
| 33 |
> Then I plan on switching to the CHOST to i386-linux-uclibc-3.2.3 and see |
| 34 |
> if the uclubc-buildroot can bootstrap itself. I'll try various other |
| 35 |
> combinations so we can get this bad boy working for ~arch. |
| 36 |
|
| 37 |
I haven't tried lots of permutations so thankyou. I will look at stripflag to |
| 38 |
see how/if it can be used on configurations. |
| 39 |
|
| 40 |
> I know mutex has some stuff he's been working on from the catalyst |
| 41 |
> approach that sound very promising, hopefully we can all tie this |
| 42 |
> (uclibc, crossdev, catalyst, baselayout stuff) back in together with |
| 43 |
> each other. |
| 44 |
|
| 45 |
All for it. |
| 46 |
|
| 47 |
> Looking at the ebuild makes me wonder if we really need sys-libs/glibc |
| 48 |
> or if virtual/glibc would work in the DEPENDS. Any insights? |
| 49 |
|
| 50 |
I haven't tried compiling uclibc-buildroot with uclibc. Once its a bit more |
| 51 |
tested a virtual would be applicable. |
| 52 |
|
| 53 |
I've removed the PROVIDE virtual for the time being. A few people reported |
| 54 |
breakages because of it. |
| 55 |
|
| 56 |
Cheers, |
| 57 |
|
| 58 |
-- |
| 59 |
Daniel Black <dragonheart@g.o> |
| 60 |
Gentoo Embedded Project |
| 61 |
irc://irc.freenode.org:6667/#gentoo-embedded |
| 62 |
http://marc.theaimsgroup.com/?l=gentoo-embedded |
| 63 |
http://embedded.gentoo.org |
| 64 |
|
| 65 |
|
| 66 |
-- |
| 67 |
gentoo-embedded@g.o mailing list |
Archives