> On Tuesday 12 July 2005 15:36, linux <[EMAIL PROTECTED]> > (<[EMAIL PROTECTED]>) wrote: > >> I was wondering if anybody has successfully built a 2.6.x kernel on an >> Athlon chip? > > Yes, I built one from Debian's 2.6.11-7 sources about three weeks ago, on > and for my Athlon. Since that time, a GCC upgrade has made GCC 4.0 the > default compiler -- the symlink in /usr/bin is dated July 9.
Yes, gcc 4.0.1 is what is being used here (I had other probs with gcc 3.x.x)... >> fakeroot make-kpkg --append_to_version -486 --initrd --revision=whatEver >> kernel_images > > Please: This is an invalid command line for make-kpkg. Providing the > exact > command which fails may be essential to diagnosis. fakeroot make-kpkg --append_to_version -486 --initrd \ --revision=july2005 kernel_image is what I used (okay, erronous 's' in my retyping before) >> pointer targets in passing argument 1 of 'use_config' differ in >> signedness > > Providing the exact and complete set of error messages may be essential to > diagnosis. Errors said make-kpkg are at: http://www.networkingnewsletter.org.uk/debian/make-11.log I'll play with your other suggestions (below), shortly. Thanks, Mcihael > > There are a large number of warnings which appear early in the make-kpkg > process. They're generated by GCC 4.0, but not GCC 3.3. I changed these > lines in /usr/src/kernel-source-2.6.11/Makefile > > HOSTCC = gcc > HOSTCXX = g++ > > to read > > HOSTCC = gcc-3.3 > HOSTCXX = g++-3.3 > > I configured the kernel to include "Multi-Tech multiport card support > (EXPERIMENTAL)", and then ran > > make-kpkg kernel_image > > and the initial warnings went away. > >> CC [M] drivers/char/isicom.o >> drivers/char/isicom.c:154: error: array type has incomplete element type >> drivers/char/isicom.c:155: error: array type has incomplete element type >> make[3]: *** [drivers/char/isicom.o] Error 1 >> make[2]: *** [drivers/char] Error 2 >> make[1]: *** [drivers] Error 2 >> make[1]: Leaving directory `/usr/src/kernel-source-2.6.11' >> make: *** [stamp-build] Error 2 > > The error message above don't go away. Line 154 references a type, > 'isi_board', which doesn't appear to be defined anywhere. Both GCC 3.3 > and > GCC 4.0 stop on this error; if I'm correct that isi_board isn't defined, > it's obviously an unrecoverable failure. > >> Any ideas on what I've done wrong? > > Sorry, no. Near the top of drivers/char/isicom.c, there is a message: > > * To use this driver you also need the support package. You > * can find this in RPM format on > * ftp://ftp.linux.org.uk/pub/linux/alan > > Have you tried this? (I have no clue what's in the package.) > > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
