[prev in list] [next in list] [prev in thread] [next in thread] 

List:       opensolaris-driver-discuss
Subject:    Re: [driver-discuss] A few DDI API questions?
From:       "Garrett D'Amore" <garrett () damore ! org>
Date:       2010-11-25 17:17:01
Message-ID: 1290705421.475830.711.camel () thinkpad
[Download RAW message or body]


You need to look at the arguments to bus_dma_tag_create().  You'll need
to research the use of these arguments... I think on some platforms, for
example, the highaddr is not used.

	- Garrett

On Thu, 2010-11-25 at 08:50 -0500, Kyle McDonald wrote:
> HI Garrett,
> 
> On 11/25/2010 1:46 AM, Garrett D'Amore wrote:
> > On Wed, 2010-11-24 at 21:29 -0500, Kyle McDonald wrote: *Most* 
> > hardware gives free access to either the 32-bit or the full 64-bit 
> > address spaces (depending on whether it supports dual address 
> > cycles).
> > 
> Ok. I'm guessing this one probably falls into that category, given
> what I've seen in the code.
> > Any hardware reference you have will tell you what the alignment 
> > constraints are, as well as any other unusual segmentation or 
> > counter limitations. Of course without that, you're stuck hoping 
> > the open source code you're referencing is accurate. Good luck. 
> > :-)
> > 
> Do you have any suggestions for FreeBSD or OpenBSD structure fields or
> function arguments that would give the needed values (even if they are
> inaccurate?) Some of the ones I've found seem to contradict themselves
> (addr_lo == addr_hi ??) and make me think I'm not looking in the right
> place.
> 
> If I can figure out what to look for, I think the FreeBSD code should
> be accurate, it is copyrighted by the device manufacturer, and
> authored by one of their employees. However there are no guarantees.
> 
> Thanks,
> 
>   -Kyle
> 
> > - Garrett
> > 
> > 
> 


_______________________________________________
driver-discuss mailing list
driver-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/driver-discuss
[prev in list] [next in list] [prev in thread] [next in thread]