From: Jan H. <bu...@uc...> - 2002-12-10 12:22:03
|
Hello, My uml dies of segfault. The very same kernel with the very same setup runs in tt mode. It's currently 2.4.19-36um, but 2.4.19-34um exhibited the same behaviour. It's quite hard to get sane info from gdb and I don't much understand what's going on there. This is what gdb tells me, if I set it to stop on SIGSEGV (everyting prefixed with `" '): " * arp -Ds 192.168.1.253 tap1 pub " * arp -Ds 192.168.1.253 tap0 pub " mke2fs 1.25 (20-Sep-2001) " mount: proc already mounted on /proc " /dev/rd/0 on /tmp type ext2 (rw) " " Program received signal SIGSEGV, Segmentation fault. " 0xa00a412b in new_thread (stack=Cannot access memory at address 0x8 " ) at process.c:182 " 182 } " (gdb) bt " #0 0xa00a412b in new_thread (stack=Cannot access memory at address 0x8 " ) at process.c:182 " Cannot access memory at address 0x0 " (gdb) c " Continuing. " " Program received signal SIGSEGV, Segmentation fault. " 0xa00a412b in new_thread (stack=Cannot access memory at address 0x8 " ) at process.c:182 " 182 } " (gdb) bt " #0 0xa00a412b in new_thread (stack=Cannot access memory at address 0x8 " ) at process.c:182 " Cannot access memory at address 0x0 " (gdb) c " Continuing. " " Program terminated with signal SIGSEGV, Segmentation fault. " The program no longer exists. " (gdb) The 5 lines at the begining are output of the umlinux. The first two are from uml_net as a result if ifconfig. Then mke2fs makes ext2 filesystem on a ramdisk and that is then mounted by mount -a (the last of output). The relevant part of the rc script looks like: " mount -avt nonfs " /sbin/portmap " /sbin/rpc.statd " /sbin/rpc.lockd " mount -avt nfs Where the first command probably completes correctly. I can have a closer look at what's happening, but I am not sure what to look at first. ------------------------------------------------------------------------------- Jan 'Bulb' Hudec <bu...@uc...> |