Fernando de Oliveira | 2 Dec 01:22 2011
Picon

e2fsprogs-1.42

Hi,

The upgrade from e2fsprogs-1.41.14 to e2fsprogs-1.42 is successful in LFS 6.7 and 6.8, but not 6.5, with

../configure --prefix=/usr --with-root-prefix="" \
    --enable-elf-shlibs --disable-libblkid --disable-libuuid \
    --disable-uuidd --disable-fsck

giving the error message:

...

../../misc/e4defrag.c:237:2: warning: #warning Using locally defined fallocate syscall interface.
../../misc/e4defrag.c:251:12: error: static declaration of ‘fallocate64’ follows non-static declaration
/usr/include/bits/fcntl.h:251:12: note: previous declaration of ‘fallocate64’ was here
make[2]: ** [e4defrag.o] Erro 1
make[2]: Saindo do diretório `/media/dados/home/fernando/tmp/paco-build-2011.12.01/e2fsprogs-1.42/build/misc'
make[1]: ** [all-progs-recursive] Erro 1
make[1]: Saindo do diretório `/media/dados/home/fernando/tmp/paco-build-2011.12.01/e2fsprogs-1.42/build'
make: ** [all] Erro 2

The build succeeds with the option --disable-defrag added to configure.

I believe I never used defrag since the old days of the other OS. Is it safe to add this option?

In PCLinuxOS, libext2fs2 has been upgraded from 1.41.14 to 1.42, but in the same upgrade, e2fsprogs is
explicitly mentioned but not upgraded. Something like "... mantained e2fsprogs (1.41.14-1pclos2010
=> 1.42-1pclos2011)".

Trying to explicitly upgrade, with synaptics, one gets the message.
(Continue reading)

Matthew Burgess | 2 Dec 10:03 2011
Picon

Re: e2fsprogs-1.42

On Thu, 1 Dec 2011 16:22:18 -0800 (PST), Fernando de Oliveira <famobr <at> yahoo.com.br> wrote:
> Hi,
> 
> The upgrade from e2fsprogs-1.41.14 to e2fsprogs-1.42 is successful in LFS
> 6.7 and 6.8, but not 6.5, with
> 
> ../configure --prefix=/usr --with-root-prefix="" \
>     --enable-elf-shlibs --disable-libblkid --disable-libuuid \
>     --disable-uuidd --disable-fsck

I'm not quite sure what you mean here.  Are you trying to build from a LFS-6.5 host, or
are you trying to upgrade e2fsprogs on a LFS-6.5 host.  The error message you are seeing
looks like you're attempting the latter as it's complaining about the contents of
/usr/include/bits/fcntl.h which, at this stage in the build, should be coming from the
newer version of Glibc in your chroot, and should definitely not be coming from your
host.

In that case, I don't see a problem with you passing --disable-defrag at all if you don't
use that functionality, but I see no reason at the moment to add that to the book.

Regards,

Matt.

--

-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

(Continue reading)

Fernando de Oliveira | 2 Dec 23:55 2011
Picon

Re: e2fsprogs-1.42

Thanks for the reply, Matt.

On 02-12-2011 06:03, Matthew Burgess wrote:
> On Thu, 1 Dec 2011 16:22:18 -0800 (PST), Fernando de Oliveira <famobr <at> yahoo.com.br> wrote:
>> Hi,
>>
>> The upgrade from e2fsprogs-1.41.14 to e2fsprogs-1.42 is successful in LFS
>> 6.7 and 6.8, but not 6.5, with
>>
>> ../configure --prefix=/usr --with-root-prefix="" \
>>     --enable-elf-shlibs --disable-libblkid --disable-libuuid \
>>     --disable-uuidd --disable-fsck
> I'm not quite sure what you mean here.  Are you trying to build from a LFS-6.5 host, or
> are you trying to upgrade e2fsprogs on a LFS-6.5 host.
I am using cssh from a host to connect to LFS-6.5, -6.7 and -6.8. So, I assume I am inside the LFS machines,
right? These machines have e2fsprogs-1.41.14 and I wish to upgrade to e2fsprogs-1.42, in each one.
>   The error message you are seeing
> looks like you're attempting the latter as it's complaining about the contents of
> /usr/include/bits/fcntl.h which, at this stage in the build, should be coming from the
> newer version of Glibc in your chroot, and should definitely not be coming from your
> host.

I did not notice the "/" before "usr" and would never know  it is related to having an old Glibc, so thanks again.

> In that case, I don't see a problem with you passing --disable-defrag at all if you don't
> use that functionality, but I see no reason at the moment to add that to the book.
>
> Regards,
>
> Matt.
(Continue reading)


Gmane