[lfs-support] lfs tool chain technique LFS-7.1

Eleanore Boyd cara117 at cox.net
Thu Aug 23 07:24:27 PDT 2012


On 8/23/2012 9:13 AM, ??? wrote:
>
> Hi,
>
> I am very confused about tool chain technique, especially when 
> digesting more details. In LFS-7.1 chapter 5 (constructing a temporary 
> system) the book says "Binutils is installed first because the 
> *configure *runs of both GCC and Glibc perform various feature tests 
> on the assembler and linker to determine which software features to 
> enable or disable.". But after "Binutils-2.22 - Pass 1" the binutils 
> (ld/as etc) are the host's. Or you can say that after binutils pass 1, 
> if compiling dummy.c, gcc still uses the host's binutils, not those 
> installed by binutils -- pass 1. On the other hand, when configuring 
> and compiling gcc-4.6.2-pass 2, the following print will come out:
>
> checking for ar... /tools/i686-lfs-linux-gnu/bin/ar
>
> checking for as... /tools/i686-lfs-linux-gnu/bin/as
>
> So I guess that when binutils/gcc/glibc know they are cross-compiling, 
> they will use the cross tools already installed instead if existing.
>
> Am I right? Or is there some tool chain mechanism, which is very 
> complicated?
>
> ----Hadi
>
>
>
The code doesn't necessarily know it's cross-compiling, it's reading the 
environment variables that have been set, after the shell has 
interpreted the command given. That's why the /tools folder is set in 
the PATH as the first place to check for the tools, then host folders 
are searched. Hashing is also turned off to ensure that the newly built 
tools are the tools being used once they exist. Though I'm sure that 
Bruce can explain even better... :)

Elly
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfromscratch.org/pipermail/lfs-support/attachments/20120823/8e7c8608/attachment.html>


More information about the lfs-support mailing list