The http://www.kerneloops.org website collects kernel oops and
warning reports from various mailing lists and bugzillas as well as
with a client users can install to auto-submit oopses.
Below is a top 9 list of the oopses collected in the last 7 days.
(Reports prior to 2.6.23 have been omitted in collecting the top 9)
This week, a total of 123 oopses and warnings have been reported.
Rank 1: implement
WARN_ON in drivers/hid/hid-core.c:785
Reported 30 times (92 total reports)
Same issue as last week; the WARN_ON is in response to an invalid command
sent from a (common) userspace application. This is clearly an obnoxious
thing to do; an -EINVAL return from the ioctl() would have been nicer
More info: http://www.kerneloops.org/search.php?search=implement
Rank 2: __ieee80211_rx_handle_packet
WARN_ON in net/mac80211/rx.c:1702
Reported 14 times (20 total reports)
This is a bug in the iwl3945 driver and is already being worked on
More info: http://www.kerneloops.org/search.php?search=__ieee80211_rx_handle_packet
Rank 3: set_dentry_child_flags
WARN_ON in fs/inotify.c:172
Only reported for pre-2.6.24 kernels
Reported 9 times (15 total reports)
Is inotify related
More info: http://www.kerneloops.org/search.php?search=set_dentry_child_flags
Rank 4: uart_flush_buffer
WARN_ON in drivers/serial/serial_core.c:544
Reported 6 times (30 total reports)
This is an old known issue; Andrew has a patch for it in -mm
More info: http://www.kerneloops.org/search.php?search=uart_flush_buffer
Rank 5: __make_request
This is a spinlock lockup when using the Oracle database application
Reported 4 times (17 total reports)
Sounds like something for Jens Axboe to chase down
More info: http://www.kerneloops.org/search.php?search=__make_request
Rank 6: cache_grow
WARN_ON in mm/slab.c:2779
Reported 3 times, but only for PPC64 2.6.24-rc8
More info: http://www.kerneloops.org/search.php?search=cache_grow
Rank 7: default_idle
Reported 3 times (7 total reports)
As always, crashes in the idle loop tend to be a symptom for hardware issues
(Not only is the most wallclock time spent here, it's also a scenario for big
swings in power consumption)
More info: http://www.kerneloops.org/search.php?search=default_idle
Rank 8: down_read
Only happened with a tripple-tainted kernel; probably not worth spending time on
Reported 2 times (5 total reports)
More info: http://www.kerneloops.org/search.php?search=down_read
Rank 9: acpi_idle_enter_bm
As always, crashes in the idle loop tend to be a symptom for hardware issues
(Not only is the most wallclock time spent here, it's also a scenario for big
Reported 2 times (5 total reports)
More info: http://www.kerneloops.org/search.php?search=acpi_idle_enter_bm