On Wed, 2020-08-12 at 02:40 +0100, Ken Moffat via blfs-dev wrote:
> I think this is a genuine issue for BLFS with glibc-2.32, my only
> intended deviations from the current LFS book are newer check and
> using 5.8.1-rc1 headers and kernel - and I see that 5.8.1 is now
> released, plus (of course) my own CFLAGS.
> 
> Yhe glibc-2.32 Release annonucement at
> https://sourceware.org/pipermail/libc-announce/2020/000029.html
> said:
> 
> > The deprecated <sys/sysctl.h> header and the sysctl function have
> > been
> >  removed.  To support old binaries, the sysctl function continues
> > to
> >  exist as a compatibility symbol (on those architectures which had
> > it),
> >  but always fails with ENOSYS.  This reflects the removal of the
> > system
> >  call from all architectures, starting with Linux 5.5.
> 
> and I've now hit that with seamonkey:
> 
> In file included from /scratch/working/seamonkey-
> 2.53.3/mozilla/memory/build/mozjemalloc.cpp:224:
> /scratch/working/seamonkey-2.53.3/obj-x86_64-pc-linux-
> gnu/dist/system_wrappers/sys/sysctl.h:3:15: fatal error:
> sys/sysctl.h: No such file or directory
>     3 | #include_next <sys/sysctl.h>
>       |               ^~~~~~~~~~~~~~
> compilation terminated.
> 
> I'll note that firefox built ok, this was just an attempt to drop in
> a second browser until I can get qtwebengine built (and yes, I know
> an issue for that has already been reported).
> 
> Fedora are using glibc-32 and have a patch against seamonkey-2.53.2:
> https://src.fedoraproject.org/rpms/seamonkey/blob/master/f/seamonkey-2.53.2-sysctl.patch
> 
> looks as if a sed to delete the line from
> mozilla/memory/build/mozjemalloc.cpp should do the job, but I
> dislike using a sed to delete 3 lines (it's fine until the next
> version, then you have to work out if it still does what was
> intended), I thoght adding '&& !defined( XP_LINUX)' to the previous
> line might do it :
> 
> sed -i 's/!defined.XP_SOLARIS.*/& \&\& !defined(XP_LINUX)/' \
>     mozilla/memory/build/mozjemalloc.cpp
> 
> But that has the same problem that it is not at all obvious what is
> being changed (turned off rather than deleted).  I did a quick
> gurgle for deleting lines before and after a match with sed, the
> suggestions looked horrendous (and might have memory implications
> because sed only works on the current patternspace).

Why not just remove the line with #include? It would leave a #if ...
#endif with empty body, which shouldn't be a problem:

sed /sysctl/d -i mozilla/memory/build/mozjemalloc.cpp
> 
> For maintenance, I think I'm going to go with the patch from fedora.
> Testing now.
> 
> ĸen
> -- 
> Juliet's version of cleanliness was next to godliness, which was to
> say it was erratic, past all understanding and was seldom seen.
>                           -- Unseen Academicals


-- 
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to