Re: Synth virtual time tick patch

[Bart Veer <bartv at ecoscentric dot com>]

>>>>> "Bart" == Bart Veer <bartv@ecoscentric.com> writes:

    <snip>

    Bart> Rather than just switching to the itimer I think I want a
    Bart> better understanding of what is going on, and just how gdb
    Bart> is interacting with the application. In the very near future
    Bart> synthetic target applications may start receiving SIGIO
    Bart> signals as well as SIGALRM, and there will be no way of
    Bart> stopping those signals coming in at inconvenient moments.

Further investigation has revealed that the problem was in gdb: when a
signal occurred, sometimes gdb lost track of temporary breakpoints so
a "next" turned into a "continue". This problem is supposed to be
fixed in the current gdb release. I don't have a tarball of that
handy, but I do have an up to date checkout of gdb anoncvs. When I
build that for Linux, "next" seems to work just fine irrespective of
any incoming signals.

So, if the sole reason for the patch is to make gdb work better with
the synthetic target then I am afraid the patch is rejected. Instead
the correct solution is to upgrade to a more recent version of gdb.

Bart