Re: [PATCH/RFC] I/O-check interface for driver's error handling

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Wed, 2005-03-02 at 12:22 -0600, Linas Vepstas wrote:
> On Tue, Mar 01, 2005 at 08:49:45AM -0800, Linus Torvalds was heard to remark:
> > 
> > The new API is what _allows_ a driver to care. It doesn't handle DMA, but
> > I think that's because nobody knows how to handle it (ie it's probably
> > hw-dependent and all existign implementations would thus be
> > driver-specific anyway).
> 
> ?  
> We could add a call 
> 
> int pci_was_there_an_error_during_dma (struct pci_dev);
> 
> right?  And it could return true/false, right?  I can certainly 
> do that today with ppc64.  I just can't tell you which dma triggered
> the problem.

That's ugly. I prefer asynchronous notification by far.

> > And yes, CLEARLY drivers will have to do all the heavy lifting. 
> 
> well .. maybe.  On ppc64, we have one hack-ish solution for
> hotplug-capable but pci-error-unaware device drivers, and that
> is to hot unplug the driver, clear the pci error condition, and 
> and replug the driver.  Works great for ethernet; haven't tested 
> USB.
> 
> I'm getting greif from the guys over here because my hack-ish code
> is hackish, and isn't arch-generic, and Paul Mackerras doesn't like it, 
> which is why Benh is threatening to re-write it, and etc. ... 
> which is why Seto is involved, and we're having this conversation ... 
> 
> 
> --linas
-- 
Benjamin Herrenschmidt <benh@kernel.crashing.org>