On an amd64 platform running in 64bit mode, the addresses in the System.map file look much different than the 32bit counter part. $cat /boot/System.map-2.6.5-52-smp 000000000014bfd1 A __crc_smp_call_function 00000000001aae09 A __crc_generic_commit_write 00000000003848fd A __crc_pci_unmap_single 000000000038b6b9 A __crc_highmem_start_page 0000000000470c0a A __crc_class_device_get 000000000060e766 A __crc_tty_termios_baud_rate 000000000077940e A __crc_blk_rq_prep_restart . . . 00000000ff3587a9 A __crc_delete_partition 00000000ff5480f6 A __crc_kmem_cache_destroy 00000000ff5e83c2 A __crc_xfrm_policy_register_afinfo 00000000ff963ed8 A __crc_net_srandom 00000000ff9ca065 A __crc_fb_edid_to_monspecs 00000000ffb4b0b5 A __crc_sock_map_fd 00000000ffd4e8bc A __crc_agp_alloc_bridge ffffffff80100000 A _text <--- HERE DOWN ffffffff80100000 t startup_32 ffffffff80100081 t reach_compatibility_mode ffffffff8010008e t second ffffffff80100100 t reach_long64 ffffffff8010013d T initial_code . . . . Is the above correct or is this a bug of some kind? This box has only 2 gb of ram. I thought these were physical addresses of where the variable/function or whatever will be located. On the 2.6 kernel they no longer look like addresses within the kernel? On the amd64 platform (above) they look even stranger to me. On a 32bit machine I get: cat /boot/System.map-2.6.4-54.5-smp 000f1b14 A __crc_netpoll_cleanup 0014bfd1 A __crc_smp_call_function 001f0a76 A __crc_generic_write_checks 002f0f24 A __crc_fb_prepare_logo 00444c0e A __crc_sysctl_intvec 0060e766 A __crc_tty_termios_baud_rate 00758fd7 A __crc_inode_change_ok . . . ff1f2a0e A __crc_blkdev_put ff2ca39a A __crc_tr_source_route ff963ed8 A __crc_net_srandom ff9ca065 A __crc_fb_edid_to_monspecs fff96251 A __crc_ide_setup_pci_devices ffffe000 A VSYSCALL_BASE ffffe400 A __kernel_vsyscall ffffe410 A SYSENTER_RETURN ffffe420 A __kernel_sigreturn ffffe440 A __kernel_rt_sigreturn Am I missing something or has the format and function of this file changed Thanks Mark