This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [rfa] New test sigbpt.{c,exp}
- From: mec dot gnu at mindspring dot com (Michael Elizabeth Chastain)
- To: drow at false dot org, mec dot gnu at mindspring dot com
- Cc: ac131313 at redhat dot com, gdb-patches at sources dot redhat dot com
- Date: Tue, 6 Jul 2004 14:59:52 -0400 (EDT)
- Subject: Re: [rfa] New test sigbpt.{c,exp}
mec> How about:
mec>
mec> (1) before running the program, 'disassemble bowler'
mec> (2) at the breakpoint, 'x/i $pc'
mec> check that the output of (2) matches one of the lines in (1)
drow> That's not a good generic fix; it's just a peculiarity of x86 that we
drow> get this particular problem. If that hadn't been a multibyte opcode
drow> something different would have happened.
It would be a useful test because it would never give a false positive.
If the instruction in (2) is not present in (1) then that is
always a problem. I agree that it would be full of false
negatives; like, any RISC arch with fixed-width instructions
would PASS.
drow> I think just executing another instruction here after the faulting one
drow> is enough to match the kfail, whether it's the size of a breakpoint or
drow> not.
That's okay with me too.
Michael C
not-enough-sleep-today