Re: bus_dmamap_sync() for uhci(4)

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Mon, Jun 09, 2008 at 05:15:23PM -0400, Michael Lorenz wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hello,
> 
> On Jun 9, 2008, at 16:16, Manuel Bouyer wrote:
> 
> >On Mon, Jun 09, 2008 at 03:43:05PM -0400, Michael Lorenz wrote:
> >>>I just did that. Now the kernel doesn't find /any/ USB devices
> >>>except the root hubs. It can't really be a problem with crime
> >>>interrupts - ehci finds devices and other cards that used interrupt
> >>>10 worked before so at least one of the uhcis should work.
> >>
> >>Hmm, an old, unpatched kernel does find devices at uhci.
> >
> >thanks. That's strange: if I read things properly, all the memory used
> >by uhci or ehci should be mapped uncached (use of BUS_DMA_COHERENT),
> >so the _sync operations I added should just issue wmb() instructions,
> >nothing more.
> 
> The O2 doesn't have any kind of DMA coherency support IIRC. The only  
> other card I tried in this machine is an ex which Just Worked(tm). So  
> do the onboard ahc's.

what do you mean ? bus_dmamap_sync() isn't a NOP, and for cached memory
it has to invalidate or flush the cache.

> 
> >BTW, I suspect the lack of memory read/write barrier are the cause of
> >the issues I'm seeing on x86 too (this is why I started this work).
> 
> Hmm, I need to read sgimips' bus_dma stuff again but since other  
> drivers work it can't be all that borked.

Sure it was not my point. It's most probably an issue with the
_sync() calls I added. But as the memory referred to by these calls should be
mapped uncached anyway, I don't understand how they could make a difference.

-- 
Manuel Bouyer <bouyer%antioche.eu.org@localhost>
     NetBSD: 26 ans d'experience feront toujours la difference
--