On Fri, Aug 17, 2007 at 04:49:04PM -0600, Trent Shea wrote: > > I'm still concerned about my actual SBU time. One SBU unit for this machine > is > anywhere from 230-300 seconds, so, the glibc in chapter six with tests should > take anywhere from 4500-5800 seconds. It would be nice to see if anyone who > didn't encounter errors in the glibc-tests has SBU times that reflect those > listed in the book.
So, take a look at the unexpurgated log. Search for 'Entering' to see which subdirectories it went into. On the box I'm sitting at (ppc, glibc-2.4) I can see it entering and running tests, to create one or more .out files, in csu, iconv, iconvdata, locale, localedata, assert, ctype, intl, catgets, math, setjmp, signal, stdlib, stdio-common, libio, dlfcn, malloc, string, wcsmbs, timezone, time, dirent, grp, pwd, posix, io, resource, misc, gmon, wctype, argp, crypt, nptl, resolv, nss, rt, debug, inet, sunrpc, login, elf (which then goes through the directories again for rtld-all). If your log shows it did something similar, then all is well (but I mistrust your SBU measurement - 230-300 is a rather large variance, and anyway for a specific build of LFS it should be a constant value). Alternatively, the tests ended early and there must be a different message to show something was wrong. ĸen -- das eine Mal als Tragödie, das andere Mal als Farce -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
