I had added that piece of code for the experiment otherwise gcc might have
optimized it. So yes the problem exists after adding those lines of code.
-Pawan
----- Original Message -----
From: "Tomasz Motylewski" <[EMAIL PROTECTED]>
To: "Bjorn Eriksson" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>; "Pawan Singh" <[EMAIL PROTECTED]>
Sent: Tuesday, June 27, 2000 3:37 AM
Subject: RE: Timer interrupt processing latency?
> > AMD SC410, which lack Floating Point Unit? (We designed a board with
that
> > processor but since it's got so poor address decoding options (very few
> > programmable chipselects) we ditched it and went with the SC520 instead
(FPU
> > included).)
> >
> > <...>I am seeing following
> > > weirdness for compute intensive pieces of code:
> > >
> > > double x = 3.4;
> > > doubly y = 5.67;
> > > for (long i=1; i<10000; i++) {
> > > double z = x+y;
> > > }
> > >
> > >
> > > The above piece of code takes 2 seconds to 3 seconds.
> > > But if I surround the above piece of code with: asm("cli") and
asm("sti")
> > > i.e. turn of interrupts, it takes only 9 milliseconds.
>
> But when you are using cli/sti is it doing anything?
>
> change it to:
>
> double x = 3.4;
> double y = 5.67;
> double z = 0.0;
> for (long i=1; i<10000; i++) {
> z += x+y;
> }
>
> printf("%lf\n",z);
>
> and test then with and without cli/sti
>
> Best regards,
> --
> Tomek
--
To unsubscribe from this list, send a message to [EMAIL PROTECTED]
with the command "unsubscribe linux-embedded" in the message body.
For more information, see <http://waste.org/mail/linux-embedded>.
