[roland at redhat.com: Re: More info on static binary/libnss* mystery]

Greg Schafer gschafer at zip.com.au
Mon Oct 7 13:51:26 PDT 2002


On Mon, Oct 07, 2002 at 10:30:35AM -0400, Adam Trilling wrote:
> If the glibc people recommend against it though, we should make our final
> LFS be sans-patch.  We're trying to be the best we can be, rather than as
> good as redhat :)
> 
> If I understand things correctly, the problem occurs after you build glibc
> without the patch inside chroot.  So here's what I propose:
> 
> 1) build ch 5 normally
> 2) build glibc with patch
> 3) rebuild those parts of ch 5 which break in glibc 2.3, linking aginst
> the patched glibc 2.3
> 4) rebuild glibc without the patch

An important factor here is that you don't have to wait for a full glibc
rebuild.

ie: if you leave your source & build dir intact, you can (un)apply the patch,
    reissue "make", and the makefile dependencies will take care to only rebuild
    the necessary parts

> 5) continue as normal
> 
> If we do step 3 dynamically, will things we build work both with and
> without the patch?  If so, maybe we could reorder ch 6 so that you don't
> have to do any dynamic builds twice (except for glibc).
> 
> I might be able to do a test build of this tonight, so someone let me know
> if I'm completely off-base before then.

It sounds feasible. Testing is good.

Greg
-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list