This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: Line number weirdness (arm-none-eabi-gdb)
- From: Tom Tromey <tromey at redhat dot com>
- To: "Queisser\, Andrew" <andrew dot queisser at hp dot com>
- Cc: "gdb\ at sourceware dot org" <gdb at sourceware dot org>
- Date: Tue, 01 Dec 2009 12:40:13 -0700
- Subject: Re: Line number weirdness (arm-none-eabi-gdb)
- References: <D0855779B9CA91439376EA595D84B6C328C29377A4@GVW0432EXB.americas.hpqcorp.net>
- Reply-to: tromey at redhat dot com
>>>>> "Andrew" == Queisser, Andrew <andrew.queisser@hp.com> writes:
Andrew> I'm confused about the behavior of my arm-none-eabi-gdb. The
Andrew> snippet below shows that when I "p" the version of a symbol I
Andrew> get one value and GDB apparently doesn't know the line number
Andrew> information for that symbol. When I set a breakpoint on that
Andrew> symbol GDB tells me that it set the breakpoint at a location a
Andrew> little farther than the nominal value for that symbol (0x2b842
Andrew> vs. 0x2b830).
I don't know why the "list" part does not work. Maybe it is a bug, it
is hard to say without more information.
I believe the discrepancy between "break" and "print" is because "break"
puts a breakpoint on the first source line of the function -- which
occurs after the function prologue. If you really want a breakpoint at
the entry point, you can use "break *function". Usually you want the
default behavior here, though.
Tom