The ptrace code is fragile. I had a test program that would make the
kernel spin in put_long while holding the kernel lock. I didn't make
the test program by looking for boundary test cases; I made it by running
mec-replay in the normal course of development, watching my kernel hang,
and boiling down a test case.
I think a good place to start is to put a loop counter in get_long
and put_long in arch/*/kernel/ptrace.c and log messages or panic
if the loop counter exceeds 1000. Also drop in some printk's at
the beginning and end of sys_ptrace catch the call sequence. Then add
another two dozen printk's, et cetera.
Michael Elizabeth Chastain
<mailto:mec@shout.net>
"love without fear"
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/