[SM-Discuss] Glibc and kernel headers(fwd from sm-commits)
Eric Sandall
eric at sandall.us
Wed Apr 22 02:56:32 EDT 2009
On Fri, 27 Mar 2009 07:41:17 +0100 (CET)
Andraž 'ruskie' Levstik <ruskie at codemages.net> wrote:
> I'm just forwarding this to discuss so that it can be better
> discussed:
>
> :2009-03-26T09:51:Ladislav Hagara:
>
> > We should ensure we all have the same headers.
> > Seems glibc doesn't use last headers by default, that should be
> > changed. I know there are developers who don't agree with it. Even
> > glibc from our iso does't use sanitized headers.
>
> Agree with the ensure the same headers. Disagree with what those
> should be...
>
> What they should be is something either UPSTREAM or WE package. If
> upstream one less thing for us. If we then it should be part of the
> glibc spell update to always update to latest headers and let people
> test it out and see if anything breaks. If nothing that should go in
> as stable and be done with it. Then when next glibc release is out
> repeat with newer headers. If someone wants to update the headers in
> between then so be it but they should be a static target not a moving
> target as defined by upstream linux and our current spell setup.
>
> Hope this will get something out of it.
I have committed a change to test for glibc that uses the a specific
version of the linux kernel for headers, much as the sanitized headers
do. This way we get the headers directly from the kernel (as intended)
and *we* do not have to do any header packaging for each release we
want to support (and for each architecture).
-sandalle
--
Eric Sandall | Source Mage GNU/Linux Developer
eric at sandall.us PGP: 0xA8EFDD61 | http://www.sourcemage.org/
http://eric.sandall.us/ | http://counter.li.org/ #196285
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://lists.ibiblio.org/pipermail/sm-discuss/attachments/20090421/a47970ac/attachment.bin
More information about the SM-Discuss
mailing list