Re: IDE + SMP Lockup (no OOPS) in 2.2.12, 2.2.10

Jes Sorensen (Jes.Sorensen@cern.ch)
30 Sep 1999 12:21:39 +0200


>>>>> "Rogier" == Rogier Wolff <R.E.Wolff@BitWizard.nl> writes:

Rogier> One way to optimize a driver is to say that IF the check for
Rogier> "did you interrupt me" is just as long as "do you have data
Rogier> for me", then you can forget the "did you interrupt me" and
Rogier> simply ask if it has any data. I mean if it didn't interrupt
Rogier> but it does have data, why not handle it anyway. Saves the
Rogier> extra interrupt overhead.

Well the real way to get more speed may actually be to change the
driver to do memory mapped I/O instead of I/O ports I/O - Ingo did
some simple benches of that a few months ago, show that reading MMIO
is like 5 times faster than inb() and writing is almost 10 times
faster.

>> Well serial is a medium that is almost guaranteed to lose data
>> once in a while ;-)

Rogier> Ok. Tell me: Why is the PCI serial chip with 128 byte buffer
Rogier> experiencing overruns, while the chip soldered to my
Rogier> motherboard (probably in one of those SMC multi-io chips (*))
Rogier> with only 16 byte buffer is NOT dropping characters?

Rogier> What the hell is going on?

No idea, dodgy hardware, bad tranceivers?

Jes

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/