On 3/1/12 4:27 PM, Jeremy Huntwork wrote: > And because of the pre-adjusting there's even less chance to bring in > something from the host system. The limits.h file is an example. The > first pass of GCC doesn't install a full-featured limits.h file because > it can't find one in the include paths we've specified.
One more thing. Because of tightly controlling where the compiler searches for headers, etc, the statement about the adjustment for the fixincludes script in pass 2 does not appear to be true any more: It says "In fact, running this script may actually pollute the build environment by installing fixed headers from the host system into GCC's private include directory." For the proposed build method, this does not appear to be the case: Fixing headers into /mnt/lfs/sources/gcc-build/gcc/include-fixed for x86_64-unknown-linux-gnu target Forbidden identifiers: linux unix Finding directories and links to directories Searching /tools/include/. Making symbolic directory links Fixing directory /tools/include into /mnt/lfs/sources/gcc-build/gcc/include-fixed So it only searches for headers in /tools/include Letting it run does 'fix' one header: Fixed: linux/a.out.h I'm unsure of the impact of letting it run, but I doubt it's very large, especially given that we allow it to run in chapter 6 under very similar circumstances. I believe this another attestation to having a cleaner separation from the host system. JH -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
