Description of problem: see summary Version-Release number of selected component (if applicable): 2.6.19-1.2895_1.fc6 and 2.6.20-1.2922.fc7 How reproducible: Always Steps to Reproduce: 1. Suspend 2. Resume Actual results: system doesn't resume Expected results: system should resume normally Additional info: relevant part from /var/log/messages Feb 12 13:13:44 ananda NetworkManager: <information> Going to sleep. Feb 12 13:13:47 ananda hidd[2493]: Exit Feb 12 13:13:47 ananda sdpd[2401]: terminating... Feb 12 13:13:47 ananda hcid[2397]: Unregister path:/org/bluez Feb 12 13:13:47 ananda hcid[2397]: Exit Feb 12 13:13:47 ananda kernel: NET: Unregistered protocol family 31 Feb 12 13:13:49 ananda kernel: Disabling non-boot CPUs ... Feb 12 13:13:49 ananda kernel: Feb 12 13:13:49 ananda kernel: ======================================================= Feb 12 13:13:49 ananda kernel: [ INFO: possible circular locking dependency detected ] Feb 12 13:13:49 ananda kernel: 2.6.20-1.2922.fc7 #1 Feb 12 13:13:49 ananda kernel: ------------------------------------------------------- Feb 12 13:13:49 ananda kernel: pm-suspend/3141 is trying to acquire lock: Feb 12 13:13:49 ananda kernel: (cpu_bitmask_lock){--..}, at: [<c0603d59>] mutex_lock+0x21/0x24 Feb 12 13:13:49 ananda kernel: Feb 12 13:13:49 ananda kernel: but task is already holding lock: Feb 12 13:13:49 ananda kernel: (workqueue_mutex){--..}, at: [<c0603d59>] mutex_lock+0x21/0x24 Feb 12 13:13:49 ananda kernel: Feb 12 13:13:49 ananda kernel: which lock already depends on the new lock. Feb 12 13:13:49 ananda kernel: Feb 12 13:13:49 ananda kernel: Feb 12 13:13:49 ananda kernel: the existing dependency chain (in reverse order) is: Feb 12 13:13:49 ananda kernel: Feb 12 13:13:49 ananda kernel: -> #3 (workqueue_mutex){--..}: Feb 12 13:13:49 ananda kernel: [<c0440754>] __lock_acquire+0x8dd/0xa04 Feb 12 13:13:49 ananda kernel: [<c0440b64>] lock_acquire+0x56/0x6f Feb 12 13:13:49 ananda kernel: [<c0603bd3>] __mutex_lock_slowpath+0xe5/0x24a Feb 12 13:13:49 ananda kernel: [<c0603d59>] mutex_lock+0x21/0x24 Feb 12 13:13:49 ananda kernel: [<c04358ca>] __create_workqueue+0x61/0x128 Feb 12 13:13:49 ananda kernel: [<f8b3862f>] cpufreq_governor_dbs+0xa1/0x30e [cpufreq_ondemand] Feb 12 13:13:49 ananda kernel: [<c0590751>] __cpufreq_governor+0x64/0x98 Feb 12 13:13:49 ananda kernel: [<c059090c>] __cpufreq_set_policy+0x187/0x209 Feb 12 13:13:49 ananda kernel: [<c0590b68>] store_scaling_governor+0x164/0x1b1 Feb 12 13:13:49 ananda kernel: [<c059002d>] store+0x37/0x48 Feb 12 13:13:49 ananda kernel: [<c04b0b69>] sysfs_write_file+0xb8/0xe0 Feb 12 13:13:49 ananda kernel: [<c0478c68>] vfs_write+0xaf/0x163 Feb 12 13:13:49 ananda kernel: [<c04792b6>] sys_write+0x3d/0x61 Feb 12 13:13:49 ananda kernel: [<c0404058>] syscall_call+0x7/0xb Feb 12 13:13:49 ananda kernel: [<ffffffff>] 0xffffffff Feb 12 13:13:49 ananda kernel: Feb 12 13:13:49 ananda kernel: -> #2 (dbs_mutex){--..}: Feb 12 13:13:49 ananda kernel: [<c0440754>] __lock_acquire+0x8dd/0xa04 Feb 12 13:13:49 ananda kernel: [<c0440b64>] lock_acquire+0x56/0x6f Feb 12 13:13:49 ananda kernel: [<c0603bd3>] __mutex_lock_slowpath+0xe5/0x24a Feb 12 13:13:49 ananda kernel: [<c0603d59>] mutex_lock+0x21/0x24 Feb 12 13:13:49 ananda kernel: [<f8b38613>] cpufreq_governor_dbs+0x85/0x30e [cpufreq_ondemand] Feb 12 13:13:49 ananda kernel: [<c0590751>] __cpufreq_governor+0x64/0x98 Feb 12 13:13:49 ananda kernel: [<c059090c>] __cpufreq_set_policy+0x187/0x209 Feb 12 13:15:17 ananda syslogd 1.4.1: restart.
Created attachment 147891 [details] entire /var/log/messages with possible useful info of harwdware (CPU...)
If I use kernel version 2.6.18-1.2869.fc6 or lower everything works perfectly (suspend, hibernation).
I am going to mark this a duplicate of bug 223604 even though they are not exactly the same. The cpufreq_ondemand driver needs to be disabled during suspend/resume and that is not happening... Please try the workaround reported in that bug. *** This bug has been marked as a duplicate of 223604 ***
I tried the workaround of bug 223604 but the problem remains (not resuming) and now a message "lockdep: not fixing up alternatives." appears in the messages log. When I try to resume, the hard disk starts working and after 1 or 2 seconds the computer gets completely unresponsive, with CAPS LOCK led flashing and the screen completely blank. I need to power off so that the system starts again. Here are the relevant messages: Feb 14 13:02:17 ananda NetworkManager: <information> Going to sleep. Feb 14 13:02:20 ananda hidd[2465]: Exit Feb 14 13:02:20 ananda sdpd[2376]: terminating... Feb 14 13:02:20 ananda hcid[2372]: Unregister path:/org/bluez Feb 14 13:02:20 ananda hcid[2372]: Exit Feb 14 13:02:20 ananda kernel: NET: Unregistered protocol family 31 Feb 14 13:02:22 ananda cpuspeed: Disabling ondemand cpu frequency scaling governor Feb 14 13:02:22 ananda kernel: Disabling non-boot CPUs ... Feb 14 13:02:22 ananda kernel: Breaking affinity for irq 1 Feb 14 13:02:22 ananda kernel: Breaking affinity for irq 14 Feb 14 13:02:22 ananda kernel: CPU 1 is now offline Feb 14 13:02:22 ananda kernel: lockdep: not fixing up alternatives. Feb 14 13:03:45 ananda syslogd 1.4.1: restart.
This problem disappeared in new kernel: 2.6.20-1.2925.fc6