Re: SIGSEGV when using "perf record -g" with 3.13-rc* kernel

[Peter Zijlstra]

On Fri, Jan 10, 2014 at 02:37:10PM -0500, Waiman Long wrote:
> >@@ -641,7 +641,7 @@ no_context(struct pt_regs *regs, unsigned long error_code,
> >
> >  	/* Are we prepared to handle this kernel fault? */
> >  	if (fixup_exception(regs)) {
> >-		if (current_thread_info()->sig_on_uaccess_error&&  signal) {
> >+		if (!in_nmi()&&  current_thread_info()->sig_on_uaccess_error&&  signal) {
> >  			tsk->thread.trap_nr = X86_TRAP_PF;
> >  			tsk->thread.error_code = error_code | PF_USER;
> >  			tsk->thread.cr2 = address;
> 
> Yes, this change fixed the error that I got. I no longer see SIGSEGV when I
> run the test.

Awesome, just send a more elaborate version that should have the same
effect.

> I did tried to back out your "perf: Fix arch_perf_out_copy_user default"
> patch, but it didn't fix the problem.

The culprit was: e00b12e64be9 ("perf/x86: Further optimize
copy_from_user_nmi()")


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/