Hello,

After I installed Fedora 12 updated to the latest updates available I am seeing 
a new interrupt causing wake ups in powertop:

Cn                Avg residency
C0 (cpu ejecutando)        ( 0,1%)
polling           0,0ms ( 0,0%)
C1 halt           0,0ms ( 0,0%)
C2                1,1ms ( 0,8%)
C3               53,6ms (99,1%)
P-states (frequencies)
  1,84 Ghz     0,1%
  1333 Mhz     0,0%
  1000 Mhz    99,9%
Wakeups-from-idle per second : 26,2     interval: 15,0s
las estimaciones de ACPI sobre el uso de corriente no están disponibles
Top causes for wakeups:
  33,3% (  7,5)       <interrupt> : ata_piix 
  31,6% (  7,1)         <núcleo> : hrtimer_start_range_ns (tick_sched_timer) 
  10,6% (  2,4)         <núcleo> : hrtimer_start (tick_sched_timer) 
   4,4% (  1,0)         <núcleo> : __mod_timer (br_fdb_cleanup) 
   4,4% (  1,0)         <núcleo> : __mod_timer (b44_timer) 
   2,4% (  0,5)         <núcleo> : __mod_timer (br_hello_timer_expired) 
   2,1% (  0,5)           iwl3945 : queue_delayed_work (delayed_work_timer_fn) 
   2,1% (  0,5)   devkit-disks-da : hrtimer_start_range_ns (hrtimer_wakeup) 
   1,5% (  0,3)         <núcleo> : __mod_timer (neigh_periodic_timer) 
   1,2% (  0,3)      clock-applet : hrtimer_start_range_ns (hrtimer_wakeup) 


As you can see, ata_piix is at the top all the time, in previous versions of 
Fedora it just showed up for a little while and then disappeared.

Coincidentally I started playing a little bit with sysprof (from 
updates-testing) and running for about 5 minutes on a completely idle Fedora 
system running Gnome I see the following:

[Everything]                                                0,00% 100,00%
  [pgrep]                                                   0,00%  17,32%
  [hald]                                                    0,00%  14,71%
  [scsi_eh_1]                                               0,00%  13,41%
    - - kernel - -                                          0,00%  13,41%
      kernel_thread_helper                                  0,00%  13,41%
        kthread                                             0,00%  13,41%
          scsi_error_handler                                0,00%  13,41%
            ata_scsi_error                                  0,00%  12,66%
              ata_sff_error_handler                         0,00%  11,92%
                ata_do_eh                                   0,00%  10,80%
                  ata_eh_autopsy                            0,00%  10,61%
                    ata_eh_link_autopsy                     0,00%  10,61%
                      atapi_eh_request_sense                0,00%  10,61%
                        ata_exec_internal                   0,00%  10,61%
                          ata_exec_internal_sg              0,00%  10,24%
                            perf_swcounter_hrtimer          0,00%   9,68%
                              perf_counter_overflow         0,00%   9,68%
                                perf_counter_output         0,00%   9,68%
                                  _spin_unlock_irqrestore   9,68%   9,68%
                            ata_sff_post_internal_cmd       0,00%   0,19%
                            wait_for_completion_timeout     0,00%   0,19%
                            irq_exit                        0,00%   0,19%
                          perf_swcounter_hrtimer            0,00%   0,37%
                  perf_swcounter_hrtimer                    0,00%   0,19%
                perf_swcounter_hrtimer                      0,00%   1,12%
              scsi_eh_flush_done_q                          0,00%   0,74%
            schedule                                        0,00%   0,37%
            scsi_run_host_queues                            0,00%   0,19%
            perf_swcounter_hrtimer                          0,00%   0,19%
  [ps]                                                      0,00%  11,73%
  [devkit-disks-daemon: polling /dev/sr0]                   0,00%   7,82%
    In file /usr/libexec/devkit-disks-daemon                0,00%   7,82%

Look that by itself [scsi_eh_1] is taking 13% of the execution time which I 
guess it's related to the cdrom drive and I guess this is related to the wake 
ups from powertop in the ata_piix interrupt.

I'm really worried a little bit by this because my laptop seems more power 
hungry with Fedora 12 than with Fedora 10 or 11.

Why is this? Is this a known problem? maybe someone can give me a hint about 
what package should I file a bug report to.

Thanks in advance,


-William


      Encuentra las mejores recetas en Yahoo! Cocina.                       
http://mx.mujer.yahoo.com/cocina/


_______________________________________________
Power mailing list
[email protected]
http://www.bughost.org/mailman/listinfo/power

Reply via email to