[lfs-dev] [lfs-book] r11154 - in trunk/BOOK: . chapter01 chapter05 chapter06

DJ Lucas dj at linuxfromscratch.org
Sat Dec 17 09:56:48 PST 2016

On 12/17/2016 11:37 AM, Bruce Dubbs wrote:
> Pierre Labastie wrote:
>> On 17/12/2016 17:22, DJ Lucas wrote:
>>> On 12/17/2016 03:18 AM, Pierre Labastie wrote:
>>>> On 17/12/2016 07:46, dj at higgs.linuxfromscratch.org wrote:
>>>>> Author: dj
>>>>> Date: Fri Dec 16 22:46:18 2016
>>>>> New Revision: 11154
>>>>> Merge nosym branch.
>>>>> [...]
>>>> There is a note in "6.10. Adjusting the Toolchain", telling:
>>>> "On 64-bit systems, the path above will be
>>>> /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.0/../../../../lib64/. This reduces
>>>> to /usr/lib64 and /usr/lib64 is a symlink that points to /usr/lib."
>>>> Hopefully, this is no more true. I haven't built with the new layout
>>>> yet, so I am not sure how what should be changed, or whether the note
>>>> should remain, sorry for not proposing a fix.
>> Actually, what I have on a 64-bit build with jhalfs is:
>> /usr/lib/../lib/crt1.o succeeded
>> /usr/lib/../lib/crti.o succeeded
>> /usr/lib/../lib/crtn.o succeeded
>> I am not sure we should keep the note...
> I agree that the note should be removed.

Actually, we have in the book /usr/lib/crt?.o. I'm thinking since we are 
already adding the note about x86 not being regularly tested by the 
editors, we should start using the 64-bit values. Any reason not to add 
the extra path elements and showing explicit x86_64 output in the book 
with the notes applying to x86 changes?

> Also in Chapter 6, gcc we have:
> "Depending on your machine architecture, the above may differ slightly,
> the difference usually being the name of the directory after
> /usr/lib/gcc. The important thing to look for here is that gcc has found
> all three crt*.o files under the /usr/lib directory"
> What I have is:
> /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.0/../../../../lib/crt1.o succeeded
> /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.0/../../../../lib/crti.o succeeded
> /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.0/../../../../lib/crtn.o succeeded
> so that paragraph *may* need to be revised or removed.

It looks correct to me. Again x86_64 values in the book?

> Also later in the same chapter:
> "The output of the last command (allowing for a lib64 directory on
> 64-bit hosts) should be:
> attempt to open /lib/libc.so.6 succeeded"
> so the parenthetical remark needs to be removed.

Yes, it appears I had lost some of those textual changes when moving 
away from the first take (which broke spec).

> There may be other places.  grep in /jhalfs/logs for lib64.

Yes, I plan to go through it with a fine toothed comb now.


More information about the lfs-dev mailing list