During boot of i386 2.6.24 kernels with CONFIG_BLK_DEV_CMD640 set on VirtualBox 1.5.2, I get the following failure: @@: buggy cmd640<3>BUG soft lockup - CPU#0 stuck for 11s! [modprobe:572] Pid: 572, comm: modprobe Not tainted (2.6.24-rc2 #1) EIP: 0060:[] EFLAGS: 00000286 CPU: 0 EIP is at _spin_lock_irqsave+0x11/0x27 EAX: d0be66b4 EBX: d0bb0b74 ECX: c039a760 EDX: 00000246 ESI: 00000000 EDI: 00000052 EBP: d0be6840 ESP: cec13df0 DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0069 CR0: 8005003b CR2: ffffffff CR3: 0ec3b000 CR4: 000006d0 DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 DR6: 00000000 DR7: 00000000 [} put_cmd640_reg+0x11/0x30 [ide_core] [] ide_probe_for_cmd640x+0x2bd/0x450 [ide_core] [] create_proc_entry+0x72/0x86 [] init_module+0x577/0x586 [ide_core] [] default_wake_function+0x0/0x8 [] sys_init_module+0x1672/0x172a [] rb_insert_color+0x4c/0xd [] disable_irq+0x0/0x25 [] syscall_call+0x7/0xb ================== After some time the "BUG soft lockup" repeats. The first 2 characters on the first line (where I put "@@") look to be unicode characters 2630 and 2665 (hex). The code that prints the message is in drivers/ide/pci/cmd640.c (762): printk("%s: buggy cmd640%c interface on %s, config=0x%02x\n", cmd_hwif0->name, 'a' + cmd640_chip_version - 1, bus_type, cfr); Note that the message is not printed completely. The BUG does not occur if I boot with 2.6.22 or 2.6.23 and disappears if I compile 2.6.24 with CONFIG_BLK_DEV_CMD640 unset. VirtualBox does not have CMD64x, but ATA_PIIX. Cheers, FJP