netdev
[Top] [All Lists]

Re: Realtek 8169 Lock-ups

To: dpollock@xxxxxxx
Subject: Re: Realtek 8169 Lock-ups
From: Francois Romieu <romieu@xxxxxxxxxxxxx>
Date: Wed, 14 Jan 2004 00:57:38 +0100
Cc: netdev@xxxxxxxxxxx
In-reply-to: <1074020596.5336.0.camel@localhost>; from pollockd@magma.ca on Tue, Jan 13, 2004 at 02:03:16PM -0500
References: <1073507006.5151.61.camel@localhost> <20040107232034.A22930@electric-eye.fr.zoreil.com> <1073596694.6378.6.camel@localhost> <4001080D.3090401@pobox.com> <20040111124957.A18068@electric-eye.fr.zoreil.com> <20040113000357.D5983@electric-eye.fr.zoreil.com> <1074020596.5336.0.camel@localhost>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mutt/1.2.5.1i
pollockd@xxxxxxxx <pollockd@xxxxxxxx> :
[...]
> With SMP enabled, I get the following, and the packet generator aborts
> quickly:
> 
> APIC error on CPU0: 40(40)
> eth0: Too much work at interrupt!
> eth0: Too much work at interrupt!

Well, it is not clear that the "Too muck work at interrupt" recovery logic
really recovers much (be it for vanilla or modified r8169 driver).

Does it behave differently if you make the "while (boguscnt > 0);" always
true in rtl8169_interrupt() and ask for a finite number of packets (say
pgset "count 1000" or more) ?

--
Ueimor

<Prev in Thread] Current Thread [Next in Thread>