Re: TG3 data corruption (TSO ?)

["Michael Chan" <mchan@broadcom.com>]

On Sat, 2006-09-09 at 07:46 +1000, Benjamin Herrenschmidt wrote:

> The PowerPC writel has a full sync _after_ the write, mostly to prevent
> it from leaking out of a spinlock, and for ordering it vs. other
> writel's or readl's. It doesn't provide any ordering guarantee vs
> cacheable storage (and was never intended to do so afaik). Such ordering
> shall
> be provided explicitely. It's possible that 2.4 used a big hammer
> approach but we've since been actively fixing drivers for that. It's to
> be noted that PowerPC might not be the only architecture affected as I
> don't think that in general, you have ordering guarantees between
> cacheable and non-cacheable stores unless you use explicit barriers.

I think 2.4 might have an additional sync before the write which will
guarantee that the buffer descriptor is written before telling the chip
to DMA it.

> 
> Thus I disagree with "fixing" the powerpc writel(). The barries shall
> definitely go into tg3.
> 

You'll have to take this up with David.