[Bug kprobes/4531] systemtap tests fail produce kernel oops on some configs of 2.6.22-rc2 kernel
wcohen at redhat dot com
sourceware-bugzilla@sourceware.org
Tue May 22 22:10:00 GMT 2007
------- Additional Comments From wcohen at redhat dot com 2007-05-22 23:10 -------
Put im some diagnostic printk in mark_rodata_ro() for the 2.6.22-rc2. Below is
the output:
May 22 17:31:36 dhcp231-176 kernel: num_possible_cpus()= 1
May 22 17:31:36 dhcp231-176 kernel: start= ffffffff80207000 end= ffffffff80556000
May 22 17:31:36 dhcp231-176 kernel: Write protecting the kernel read-only data:
3388k
Looking up the address from the nm:
ffffffff80207000 T _stext
ffffffff80556090 R __end_rodata
Looks like all the kprobe address falls between those two.
ffffffff8043cd00 T schedule
This is uniprocessor machine so the process is marking the text as read only
too. Thus, the following start point is not used:
ffffffff804449a8 A _etext
--
http://sourceware.org/bugzilla/show_bug.cgi?id=4531
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
More information about the Systemtap
mailing list