>> You know, of course, that most debuggers allow you to catch reads
>> and writes to a certain memory location already, right?
>
> Sure, but they usually do it by inserting an illegal instruction
> at the beginning of each statement boundary ... hardware support
> for debuggers would be hard to beat.
I posted a message within the last couple of months
that asked (and then answered) a question about
watchpoint support in x86 versions of GDB. Summary:
GDB now works with the Pentium's HW support for
monitoring accesses to a given memory location; the
system runs at full speed and the kernel no longer has
to take a breakpoint exception on every instruction
boundary to activate the SW that checks the location
in question.
>People make such a huge deal about something being implemented in
>"hardware" vs "software". In most cases, the only difference
>between them is that hardware is harder to change.
In this watchpoint case I fail to see how the HW
solution can be described as anything but a huge win.
I assume we can agree that there is a continuum of
problems that can be solved with either HW or SW (and
these days the boundary between the two is blurrier
than ever) but in commodity systems I assume you'd
agree that modems, NICs, SCSI adapters, etc are all
examples where the corresponding SW solution would
more or less have to suck by comparison...
*****************************************************************
To unsubscribe from this list, send mail to [EMAIL PROTECTED]
with the text 'unsubscribe gnhlug' in the message body.
*****************************************************************
