Pushing forward with Onward.

Robert Connolly robert at linuxfromscratch.org
Wed Jun 17 11:09:20 PDT 2009

On Wednesday June 17 2009 12:28:01 pm robert baker wrote:
> Build Failures:
> Libcap
> This fails to build because of some conflict between capbility.h in
> libcap and types.h in the kernel headers (I am still useing
> so this may not happen with It can be worked around by
> declaring
> #include <linux/types.h> in progs/capsh.c before #include
> <sys/capability.h>

See if this patch helps:

> Bash
> Because I was following along with the SVN-20080603 instructions, and
> using the version of bash from Onward I hit a snag. Evidently there
> are some new functions in the builtin readline that comes with bash.
> These functions are not in the latest stable readline so builds fail
> unless you allow bash to build it's builtin readline. Can we omit the
> readline package or do other things depend on it?

LFS is using --with-installed-readline (system readline). There must be a way 
to make it work.

I'm also working on glibc-2.10.1 patches. The existing patches have a couple 

I simplified the arc4random patch, using netbsd's arc4random.c ported to 
linux. Made __arc4random_stir a weak alias so the internal version is always 
used, and so it never conflicts with a function in the application or another 
library. I'm figuring out a test suite for arc4random() just to see that it 

The mktemp_urandom patch uses the wrong error number when /dev/urandom doesn't 

Fixed mkstemps to use /dev/urandom. I found a test suite from INN (the news 

Made strlcat/strlcpy non-static, so they're in the shared library. This will 
break binary compatibility with other distributions. I found a test suite 
from INN.

Some of the patches were using read(), instead of __read(). Fixed that.

What's the view on upgrading GCC?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.linuxfromscratch.org/pipermail/hlfs-dev/attachments/20090617/6fde4195/attachment.sig>

More information about the hlfs-dev mailing list