Re: Long transmission times on SMC8432T - fix (hack) included

On Sat, 5 Jul 2003, Troy Rollo wrote:

> Subject: [tulip] Long transmission times on SMC8432T - fix (hack) included
> 
> Chip: DC21041
> 
> I have just finished doing battle with this card and the tulip drivers for 
> Linux. The problem I was getting is that packets seemed to be disappearing 
> (in some versions of the driver) or pings seemed to take an inordinately 
> long time - several seconds being common, and occasionally 24 seconds.

What driver releases are you using?
I suspect that it's not one of my releases, but rather a modified
driver.  Let me know if you encounter a 21041 problem with my release.

> To cut a long story short, after substituting many variables, it turned out 
> that the problem is with the way tulip.o configures the card - the card 
> ends up thinking the link is intermittently going down.

More likely, the transceiver wasn't configured correctly and the packet
isn't transmitted until the transmit times out.

> In hunting this one down I came across a lot of people who seemed to have 
> the same problem. Many places I looked suggested the solution was to delay 
> by 2us before writing to CSR1, but this turned out to be incorrect - both 
> ineffective and unnecessary (although there may be a problem on some clone 
> CPUs that could be rectified by that).

The only reason to delay writing CSR1 is as a hack to work around
cache-PCI coherency issues.  The people suggesting a delay do not
understand what they are doing, just blindly frobbing until the problem
isn't as obvious.

-- 
Donald Becker				[email protected]
Scyld Computing Corporation		http://www.scyld.com
914 Bay Ridge Road, Suite 220		Scyld Beowulf cluster system
Annapolis MD 21403			410-990-9993

_______________________________________________
tulip mailing list, [email protected]
To change to digest mode or unsubscribe visit
http://www.scyld.com/mailman/listinfo/tulip