From Bugzilla Helper: User-Agent: Mozilla/5.0 (compatible; Konqueror/3.2; Linux) (KHTML, like Gecko) Description of problem: After rebuilding kernel, turning on preempt (no other changes to standard fedora config): Debug: sleeping function called from invalid context at mm/slab.c:1856 Feb 11 19:47:54 localhost kernel: in_atomic():1, irqs_disabled():0 Feb 11 19:47:54 localhost kernel: Call Trace: Feb 11 19:47:54 localhost kernel: [<c012776c>] __might_sleep+0xac/0xe0 Feb 11 19:47:54 localhost kernel: [<c015ac94>] kmem_cache_alloc+0x194/0x1a0 Feb 11 19:47:54 localhost kernel: [<c018b13f>] cdev_alloc+0x1f/0x60 Feb 11 19:47:54 localhost kernel: [<e091a75a>] st_probe+0x3ba/0x790 [st] Feb 11 19:47:54 localhost kernel: [<c019ce4c>] dput+0x1c/0x710 Feb 11 19:47:54 localhost kernel: [<c01d71c5>] create_dir+0xb5/0xc0 Feb 11 19:47:54 localhost kernel: [<c01d7222>] sysfs_create_dir+0x32/0x70 Feb 11 19:47:54 localhost kernel: [<c0266302>] bus_match+0x32/0x60 Feb 11 19:47:55 localhost kernel: [<c0266429>] driver_attach+0x59/0x90 Feb 11 19:47:55 localhost kernel: [<c0204e82>] kobject_register+0x22/0x60 Feb 11 19:47:55 localhost kernel: [<c02666e1>] bus_add_driver+0x91/0xb0 Feb 11 19:47:55 localhost cups: cupsd startup succeeded Feb 11 19:47:55 localhost kernel: [<c0266b00>] driver_register+0x80/0x90 Feb 11 19:47:55 localhost kernel: [<e08ad051>] init_st+0x51/0x8f [st] Feb 11 19:47:55 localhost kernel: [<c014b8c6>] sys_init_module+0x1c6/0x390 Feb 11 19:47:55 localhost kernel: [<c010c69b>] syscall_call+0x7/0xb Feb 11 19:47:56 localhost kernel: Version-Release number of selected component (if applicable): 2.6.1-1.65custom How reproducible: Didn't try Steps to Reproduce: 1.boot 2. 3. Additional info:
does it happen without preempt ? things like this are one reason we decided not to enable it.