Re: [lfs-support] Building Check-0.9.10 fails and fixed

[Loïc Le Ninan]

Loïc Le Ninan loic.leninan at gmail.com writes:
 
 So something probably went wrong in gcc pass 2, and it would be best to
 start over again from scratch I suppose  

Hello, 

Self-replying my previous message after further investigation on this
problem, and finally solving it. 

Both my gcc and binutils from pass 2 were built from the host toolchain, and
not from pass 1. 

And it appears that the reason for this is a copy/paste error... More
precisely : selecting the 'configure' commands from the PDF version of LFS
7.4 did not select the backslash at the end of each line on pages numbered
45 and 48. And I did not notice the problem until step 5.14 crashed.

I started over from scratch using the right 'configure' commands, and step
5.14 went fine. Problem solved. 

Conclusion : Do not patch if it is not specified in the book, but
double-check everything instead. 


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

[lfs-support] Update chap 6.13

[Frans de Boer]

Dear reader,

I do not know where else to post the next:

One of my (server)systems is an i686 with 640MB. During checking I 
always get FAIL: gas/i386/rept. I noticed that when running without 
any services, this error did not appear. So after some searching I found 
a file which cause this error to appear time after time.

While still on the root source, I applied the next line to reduce the 
memory reserved during the tests:

   sed -i 's@0x4*210@0x4*100@' gas/testsuite/i386/rept.s

After which all tests ran without errors.

Note: with *100 I am at the limit of physical memory. Around 200MB was 
pushed to swap while reserving memory. So, maybe *92 or *64 might be better.

Maybe this help others too,
regards, Frans.
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page