Re: [patch] forcedeth: fix the NAPI poll function

[Jeff Garzik <jeff@garzik.org>]

Ingo Molnar wrote:
> * Jeff Garzik <jeff@garzik.org> wrote:
> 
>>>> The compiler certainly doesn't care, and IMO it prevents bugs.
>>> agreed. Do you have an uptodate patch/git-URI for the forcedeth 
>>> rewrite you did? I can throw it into the testbed.
>> Branch 'fe-lock' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git
>>
>> It works here locally, but at this very minute I am rewriting those 
>> changesets yet again :)
> 
> FYI, looks good so far, ontop of Linus' latest -git tree.
> 
> btw., one thing i always found weird about forcedeth is that it 
> generates an extra ~100 irqs per second even when there's no network 
> traffic:
> 
>  11:      89752    XT-PIC-XT        eth0
>  11:      89854    XT-PIC-XT        eth0
>  11:      89955    XT-PIC-XT        eth0
>  11:      90056    XT-PIC-XT        eth0
>  11:      90157    XT-PIC-XT        eth0
> 
> (irq count snapshot every second). Just in case it's easy to fix ...


May I presume that behavior occurs in both mainline and my fe-lock branch?

That is likely the NIC's timer irq, which is programmed to kick the irq 
handler every so often.  DEV_NEED_TIMERIRQ is a starting point, as well 
as "Known bugs" at the top of forcedeth.c, if you're curious.

I'm _betting_ that some of that can be eliminated, but we'll see...

	Jeff