sreadahead crashed with SIGSEGV in __pause_nocancel()
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
sreadahead (Ubuntu) |
New
|
Undecided
|
Unassigned |
Bug Description
Binary package hint: sreadahead
On boot with up-to-date Karmic as of a few minutes ago, sreadahead dies on boot with the following in dmesg:
Oct 1 09:56:14 severus kernel: [ 59.282811] sreadahead[651]: segfault at 7fc2ab6b6010 ip 0000000000401f3b sp 00007fff47586570 error 4 in sreadahead[
gdm is not started. I had to run 'sudo start gdm' to get it to start (curiously, /etc/init.d/gdm start did *not* work). I am not sure if gdm not starting and sreadahead are related. I also did not see the splash screens on this boot. The following are what were updated before a reboot:
$ find . -mtime -3 -type f | sort
./apparmor_
./apparmor-
./apparmor-
./gdm_2.
./ia32-
./indicator-
./language-
./language-
./language-
./language-
./libapache2-
./libapparmor1_
./libapparmor-
./libelfg0_
./libnautilus-
./libnm-
./libnm-
./libusplash0_
./mplayer_
./mplayer-
./nautilus_
./nautilus-
./network-
./software-
./tasksel_
./tasksel-
./ubuntu-
./usplash_
./xsplash_
ProblemType: Crash
Architecture: amd64
CrashCounter: 1
Date: Thu Oct 1 09:41:30 2009
DistroRelease: Ubuntu 9.10
ExecutablePath: /sbin/sreadahead
Package: sreadahead 1.0-4
ProcCmdline: /sbin/sreadahead -t 0
ProcEnviron: PATH=(custom, no user)
ProcVersionSign
SegvAnalysis:
Segfault happened at: 0x401f3b: mov (%r15,%rdi,1),%dil
PC (0x00401f3b) ok
source "(%r15,%rdi,1)" (0x7fb39b883010) not located in a known VMA region (needed readable region)!
destination "%dil" ok
SegvReason: reading unknown VMA
Signal: 11
SourcePackage: sreadahead
StacktraceTop:
__pause_nocancel () from /lib/libpthread
?? ()
__libc_start_main () from /lib/libc.so.6
?? ()
?? ()
Title: sreadahead crashed with SIGSEGV in __pause_nocancel()
Uname: Linux 2.6.31-11-generic x86_64
UserGroups:
The following bugs are likely duplicates, but apport has not unprivated them yet:
bug 422304
bug 427314
bug 429187
bug 430020
I'd be happy to subscribe a developer to the bugs.