Re: Today Linus redesigns the networking driver interface (was Re: tulip driver in ...)

Christer Weinigel (wingel@hog.ctrl-c.liu.se)
22 Sep 1998 21:07:28 -0000

Linus Torvalds wrote:
>> The simple rule: "if you haven't registered/marked active a BH, then
>> you aren't going to trigger BH processing, so don't worry about it"
>> seems like the right thing.
>
>That's essentially the rule now. It's simple, and it works. However, it
>doesn't give us aggregation - which could improve throughput and CPU
>usage.

So why not have a few flags which tell the interrupt return path code
that it is supposed to process signals or bottom half handlers?
(Or maybe an inverted flag if that makes more sense).

So for a fast interrupt handler which can stand higher latencies:

static void interrupt(int irq, void *dev_id, struct pt_regs *regs)
{
fill_buffer();

if (plenty_of_space_left())
defer_bh = 1;
}

/Christer

-- 
If it's tourist season, why can't we shoot them?


-
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/