[patch 4/4 v2] unwinder: The unwinder (x86* only)

Mark Wielaard mjw at redhat.com
Fri Jun 14 10:00:07 UTC 2013


On Thu, 2013-06-13 at 19:30 +0200, Jan Kratochvil wrote:
> On Thu, 13 Jun 2013 15:48:22 +0200, Mark Wielaard wrote:
> > But I don't fully understand
> > why/where jmp is actually used in the test case.
> 
> It tests the unwinder can properly unwind even if we stop at the very first
> instruction of a function.  Therefore that ISACTIVATION is true for the bottom
> frame.

O, I finally see how it works. When the child raises SIGUSR1 then the
backtrace.c testcase explicitly sets the PC to the first instruction of
jmp in backtrace-child to check this condition. How tricky and crafty :)



More information about the elfutils-devel mailing list