When I add kernel.hung_task_timeout_secs = 0 in my sysctl.conf, Most of the time, I have small freeze from 1/2 seconds max...
But sometimes, I have a big freeze (need a reboot). The disk is new (and there is no problem on it - checked before...)

Here a trace:
Feb 28 21:05:10 redshield kernel: [    1.760466] Hardware name: P55A-UD5                       
Feb 28 21:05:10 redshield kernel: [    1.760467] Modules linked in:                            
Feb 28 21:05:10 redshield kernel: [    1.760469] Pid: 0, comm: swapper Not tainted 2.6.31-19-generic #56-Ubuntu                                                                               
Feb 28 21:05:10 redshield kernel: [    1.760471] Call Trace:                                   
Feb 28 21:05:10 redshield kernel: [    1.760475]  [<ffffffff8105e678>] warn_slowpath_common+0x78/0xb0                                                                                         
Feb 28 21:05:10 redshield kernel: [    1.760477]  [<ffffffff8105e6bf>] warn_slowpath_null+0xf/0x20                                                                                            
Feb 28 21:05:10 redshield kernel: [    1.760479]  [<ffffffff81034335>] hpet_next_event+0x75/0x90                                                                                              
Feb 28 21:05:10 redshield kernel: [    1.760481]  [<ffffffff8103437b>] hpet_legacy_next_event+0xb/0x10                                                                                        
Feb 28 21:05:10 redshield kernel: [    1.760484]  [<ffffffff8108556f>] clockevents_program_event+0x4f/0x90                                                                                    
Feb 28 21:05:10 redshield kernel: [    1.760486]  [<ffffffff81086a60>] tick_dev_program_event+0x40/0xd0                                                                                       
Feb 28 21:05:10 redshield kernel: [    1.760489]  [<ffffffff810863f6>] tick_broadcast_oneshot_control+0x116/0x120                                                                             
Feb 28 21:05:10 redshield kernel: [    1.760491]  [<ffffffff81085ce0>] tick_notify+0x130/0x1a0 
Feb 28 21:05:10 redshield kernel: [    1.760494]  [<ffffffff8152ca37>] notifier_call_chain+0x47/0x90                                                                                          
Feb 28 21:05:10 redshield kernel: [    1.760496]  [<ffffffff8107d6d1>] raw_notifier_call_chain+0x11/0x20                                                                                      
Feb 28 21:05:10 redshield kernel: [    1.760498]  [<ffffffff81085374>] clockevents_notify+0x34/0xa0                                                                                           
Feb 28 21:05:10 redshield kernel: [    1.760500]  [<ffffffff812d7eea>] lapic_timer_state_broadcast+0x41/0x43                                                                                  
Feb 28 21:05:10 redshield kernel: [    1.760502]  [<ffffffff812d8447>] acpi_idle_enter_bm+0x185/0x2bf                                                                                         
Feb 28 21:05:10 redshield kernel: [    1.760504]  [<ffffffff812d82b4>] ? acpi_idle_enter_c1+0xae/0xbc                                                                                         
Feb 28 21:05:10 redshield kernel: [    1.760507]  [<ffffffff813feadb>] cpuidle_idle_call+0x9b/0x100                                                                                           
Feb 28 21:05:10 redshield kernel: [    1.760510]  [<ffffffff81010e12>] cpu_idle+0xb2/0x100     
Feb 28 21:05:10 redshield kernel: [    1.760512]  [<ffffffff81524296>] start_secondary+0xa9/0xab                                                                                              
Feb 28 21:05:10 redshield kernel: [    1.760516] ---[ end trace dd8604855130fcfb ]---          
Feb 28 21:05:10 redshield kernel: [    1.761189] PM: Resume from disk failed.   

I presume it's due to the same bug, maybe it can help.