| 
      
      
      From: Michel  <mic...@ii...> - 2001-06-17 00:03:50
      
     | 
| Ken Tyler wrote: > > On Sat, 16 Jun 2001, Geert Uytterhoeven wrote: > > > This either means bad RAM or so, or a bug in the kernel MM. > > OK, but before I start looking for 32 Meg SIMMS, I tried Glenn's config > removing PERMEDIA and PCI, putting in VIRGE, running the 2.4.5 kernel > compiled with > > Reading specs from /usr/lib/gcc-lib/ppc-redhat-linux/2.95.2/specs > gcc version 2.95.2 19991024 (release/franzo) > > under 2.2.10 has the same problems. > > make -j 10 : > > gcc: Internal compiler error: program cpp got fatal signal 4 > make[1]: *** [ppc_defs.h] Error 1 > make[1]: Leaving directory `/usr/src/linux_CVS/2.4_new/arch/ppc/kernel' > make: *** [_dir_arch/ppc/kernel] Error 2 > make[1]: *** [first_rule] Illegal instruction (core dumped) > make[1]: Leaving directory `/usr/src/linux_CVS/2.4_new/kernel' > make: *** [_dir_kernel] Error 2 > sched.c: In function `schedule': > sched.c:711: Internal compiler error: > sched.c:711: output pipe has been closed > cpp: output pipe has been closed > gcc: Internal compiler error: program as got fatal signal 4 > and etc > > (is this compiler OK ?) > > I'll now try just with AMIFB, no VIRGE. > > Couldn't be caused by not having any PCI compiled in could it ? > > Next is pulling out the mem and giving its pins a clean. > > A bit odd that make -j 10 is OK on 2.2.10 ? 2.4 demands more swap (at least twice the RAM size) than 2.2 . Have you watched it during the build? -- Earthling Michel Dänzer (MrCooper) \ Debian GNU/Linux (powerpc) developer CS student, Free Software enthusiast \ XFree86 and DRI project member |