Mail Archives: djgpp/2000/03/17/13:24:54
Damian Yerrick wrote:
>
> On Thu, 16 Mar 2000 12:01:45 +0200 (IST), Eli Zaretskii
> <eliz AT is DOT elta DOT co DOT il> wrote:
>
> >Using IRQ 5 is not a good thing anyway, since the interrupt associated
> >with it, Int 0Dh, is in conflict with the GPF exception. If your
> >program causes GPFs (e.g., if it uses signals), you could lose
> >interrupts as a result.
>
> Are you saying preinstalled sound cards (which default to IRQ 5 if
> the user does not void the computer's warranty by opening the case)
> are incompatible with signals?
This discussion was not about compatibility (the OP's interrupt handler
hooked to IRQ 5 *did* work), it is about supporting interrupts at a high
rate.
What I was saying is that, due to the overhead incurred by exceptions being
reflected as interrupts (because they share the same numbers), using IRQ 5
for hardware interrupts in a protected-mode program might begin losing
interrupts at a lower rate than it would for another IRQ.
- Raw text -