On 5/9/12 4:29 PM, Matt Burgess wrote: > On Wed, 2012-05-09 at 16:24 -0400, Jeremy Huntwork wrote: > >> I'll dig a bit and get back to you. > > Note that I'm busy digging too. Having fixincludes run in chapter 5 > looks safe; GCC only searches for headers under /mnt/lfs/tools/include > or /tools/include. > > I'll have results for chapter 6 tomorrow, hopefully!
I think we may have been too quick on the draw with removing this command entirely. The reason being that this script was originally added into gcc to "fix" non-ANSI headers in various packages. http://gcc.gnu.org/viewcvs/trunk/fixincludes/README-fixinc?view=markup Especially given the success of many recent years of LFS without 'fixed' headers, I'd say the case of broken headers is these days extremely rare. And because of the automated nature of the script there exists the possibility that it actually 'fixes' things we don't need or want 'fixed'. It's probably safer to add back the command that disables this script but just make sure that our explanation for it is accurate. JH -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
