This is the mail archive of the
mailing list for the GCC project.
Re: Debugging line numbers for delayed branches
- From: Andrew STUBBS <andrew dot stubbs at st dot com>
- To: gcc-patches at gcc dot gnu dot org
- Date: Thu, 18 Oct 2007 16:11:23 +0100
- Subject: Re: Debugging line numbers for delayed branches
- References: <20071018140826.GB10045@caradoc.them.org>
Daniel Jacobowitz wrote:
I have been working on GDB support for optimized code. This turns up
all sorts of interesting quirks in GCC's debug output when optimizing.
I recall that I had a problem when NOT optimising - I was getting line
numbers starting in delay slots even at -O0 (when the slot contained
nothing but a 'nop'). This was on an SH-4 target.
This is a particularly evil problem on SH where one cannot place
breakpoints in delay slots. It's made worse because there is no general
way to determine if an instruction is in a delay slot, or not, by
instruction analysis alone.
Unfortunately I do not have any example code to hand.