[Bug runtime/5194] IO problem on begin/end probes

["hunt at redhat dot com" <sourceware-bugzilla at sourceware dot org>]

------- Additional Comments From hunt at redhat dot com  2008-01-21 17:55 -------
(In reply to comment #5)
> I'm not clear about whether there actually exists a bug described
> in comment #0, in that we do not enable sleeping or preemption
> during begin/end probes, so we can't switch CPUs.

We don't?  News to me. It's easy to test.

>cat test.stp
function in_interrupt:long () %{ /* pure */
        THIS->__retvalue = in_interrupt();
%}
function irqs_disabled:long () %{ /* pure */
        THIS->__retvalue = irqs_disabled();
%}
function preempt_count:long () %{ /* pure */
        THIS->__retvalue = preempt_count();
%}
function print_info () {
	printf("%s\n", pp())
	printf("cpu: %d\n", cpu())
	printf("in_interrupt:%d, irqs_disabled:%d, preempt_count:%d\n", in_interrupt(),
irqs_disabled(), preempt_count());
	print("------------------\n\n")
}

probe begin {
	print_info()
}

>stap test.stp
begin
cpu: 2
in_interrupt:0, irqs_disabled:0, preempt_count:0



-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=5194

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.