On Wed, May 24, 2017 at 10:42 AM, Thomas Trepl <[email protected]>
wrote:

> Am 2017-05-24 17:13, schrieb Bruce Dubbs:
>
>> Thomas Trepl wrote:
>>
>>> Hi all,
>>>
>>> shouldn't the instruction (6.9. Glibc-2.25)
>>>
>>>    case $(uname -m) in
>>>        x86)    GCC_INCDIR=/usr/lib/gcc/x86-pc-linux-gnu/7.1.0/include
>>>                ln -s ld-linux.so.2 /lib/ld-lsb.so.3
>>>        ;;
>>>        x86_64) GCC_INCDIR=/usr/lib/gcc/x86_64-pc-linux-gnu/7.1.0/include
>>>                ln -s ../lib/ld-linux-x86-64.so.2 /lib64
>>>                ln -s ../lib/ld-linux-x86-64.so.2
>>> /lib64/ld-lsb-x86-64.so.3
>>>        ;;
>>>    esac
>>>
>>> read like
>>>
>>>    case $(uname -m) in
>>>        i?86)   GCC_INCDIR=/usr/lib/gcc/$(uname
>>> -m)-pc-linux-gnu/7.1.0/include
>>>    ...
>>>
>>
>> No.  The kernel does not differentiate between 32-bit x86 processors
>> in its machine name.
>>
>>   -- Bruce
>>
>
> No, too.   $(uname -m) does not output "x86" on a 32bit installation - it
> outputs "i686". So, to the case-label "x86" will never be jumped to, never
> gets executed. Same in the GCC_INCDIR in that command, "x86-pc-linux-gnu"
> does not exist, it's named "/usr/lib/gcc/i686-pc-linux-gnu/...".
>
> --
> Thomas
>


Greetings, Thomas.

My i686 LFS development build just finished. I'll update the book in a few
minutes.

You can thank Bruce for convincing me to build it :-)

Thanks again for reporting it!
-- 
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Reply via email to