a bug in the perl patch

Gilles Espinasse g.esp at free.fr
Mon Jun 27 23:16:01 PDT 2011

----- Original Message ----- 
From: "Bruce Dubbs" <bruce.dubbs at gmail.com>
To: "LFS Developers Mailinglist" <lfs-dev at linuxfromscratch.org>
Sent: Monday, June 27, 2011 9:32 PM
Subject: Re: a bug in the perl patch

> Matthew Burgess wrote:
> > On 27/06/2011 02:34, Bruce Dubbs wrote:
> >
> >> I also see a change in
> >>
> >> by Matt.
> >>
> >> I'll be glad to change this if it was an inadvertent change, but I'd
> >> like to confirm that first.
> >>
> >> Matt?
> >
> > That definitely looks like an inadvertent change to me.
> OK, I updated the patch.
I diffed the log with unfixed and fixed patch on my build system
host libc is libc-2.7.so, compiled libc is libc-2.11.3.so, still perl-5.10.1

-cc -c   -fno-strict-aliasing -pipe -fstack-protector -D_LARGEFILE_SOURCE -D
_FILE_OFFSET_BITS=64 -O2   -DVERSION=\"1.10\" -DXS_VERSION=\"1.10\"
"-I../.."  -DLIBC="/tools_i486/lib/libc-2.7.so" DynaLoader.c
+cc -c   -fno-strict-aliasing -pipe -fstack-protector -D_LARGEFILE_SOURCE -D
_FILE_OFFSET_BITS=64 -O2   -DVERSION=\"1.10\" -DXS_VERSION=\"1.10\"
"-I../.."  -DLIBC="/tools_i486/lib/libc-2.11.3.so" DynaLoader.c
 rm -rf ../../DynaLoader.o
 cp DynaLoader.o ../../DynaLoader.o

Unfixed DLIBC is really bogus while fixed DLIBC look sane.
It may have not hurt because of the limited usage on toolchain of perl


More information about the lfs-dev mailing list