This is the mail archive of the
insight@sourceware.cygnus.com
mailing list for the Insight project.
Tail Calls & Insight
- To: insight at sourceware dot cygnus dot com
- Subject: Tail Calls & Insight
- From: Duane Ellis <duane_ellis at franklin dot com>
- Date: Tue, 16 May 2000 15:43:13 -0400
- Reply-to: duane_ellis at franklin dot com
I'm working on a custom port for insight/gdb for our custom RISC cpu
and I'm having a problem with something.
When you click the 'next' button, GDB really single steps till the PC
goes somewhere expected, or unexpected.
If it's unexpected (ie: a subroutine call occurs) GDB then assumes
that the return address is on the stack and pops the return address,
sets a new breakpoint and 'runs', hoping it will hit the breakpoint.
Unfortunatly, (or fortunatly depending on your view) one optmization
trick we use is to convert tail calls into jumps.
Is there any way of describing this situation to GDB?
-Duane.