Hi,
I am trying to debug a subtle timing-dependent bug in kernel.
I found that if I could set up a break-point from within kernel
at run-time, this would help.
The condition to trigger is
"If 0 is written at virtual address 0xCEC8F004 , then stop".
The address is on kernel-stack
My SOC has a onchip JTAG-based debug block.
What I have in mind is to do as below -
signed long __sched schedule_timeout(signed long timeout)
{
struct timer_list timer;
unsigned long expire;
// some kernel code
// setup conditional break-point
// some kernel code runs here
// some kernel code
// some kernel code
// remove the break-point
// some kernel code runs here
}
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>