Re: [PATCH] IBM Lanstreamer bugfixes

[Mike Phillips]

On Fri, Jan 18, 2002 at 05:02:57PM -0600 or sometime in the same epoch, Kent E Yoder scribbled:
>   Ok, I thought of one thing that might make things clearer here: when I 
> say "the write failed", I mean that we saw the write go out on the PCI bus 
> and then the box locked up.  We were looking at it on a PCI bus analyzer. 
> That, and it wasn't just this write, or just writes in general, it really 
> seemed random.
> 

Kent,

We had this on olympic for certain high end IBM boxen. Spent forever
trying to trap it as I couldn't emulate the behaviour on my test
boxes. We weren't getting correct values from pci reads/write and we
were running out of buffers as they weren't getting flushed. The
machine wouldn't lock but the adapter would stop tx/rx.

Turned out the pci bridge on the machine itself was causing the
problems. Tweaking the pci bus fixed the problem.  

-- 
Mike Phillips
Linux Token Ring Project
http://www.linuxtr.net
mailto: [e-mail]