bigphysarea

bigphysarea

[Mark Hounschell]

 We have a pci reflective memory card that has no hardware scatter/gatter capabilities. Software scatter/gather is to slow for our needs. It has 16mb of mem that needs to be
contigously mapped into a user task. We have been using the bigphysarea patch and seems
to do what we need for this card. We have been using it sice the beginning to the 2.4 
series kernel. My question is, is this patch still nessessary or is there possibly a
way do do this now without the patch?

Ragards
Mark

Re: bigphysarea

[Pete Zaitcev]

> [...] We have been using the bigphysarea patch and seems
> to do what we need for this card. We have been using it sice the beginning to the 2.4 
> series kernel. My question is, is this patch still nessessary or is there possibly a
> way do do this now without the patch?

I think Pauline never bothered to submit it. In Welsh's times
it was considered too esoteric for inclusion, but nowdays we
carry so much bloat that I'd think bigphysarea can be included.
If you want it included, take over the issue and bug Linus.

-- Pete

Re: bigphysarea

[Jeff V. Merkey]

You can achieve the same thing without the bigphys patch.  I was
using the bigphys patch in the Dolphin SCI drivers, but with 
these systems having all this memory these days, just the
standard get pages calls work well enough.  To see an alternate 
implementation of bigphys that does not rely on big phys, look
at memalloc.c in the SCI code base at 

www.kernel.org/pub/linux/kernel/people/jmerkey/

and download the latest SCI drivers for Linux.  I used to use the 
bigphys patch, but found it really wasn't necessary to do what 
I needed to do.

Jeff

On Tue, Jul 23, 2002 at 01:08:00PM -0400, Pete Zaitcev wrote:
> > [...] We have been using the bigphysarea patch and seems
> > to do what we need for this card. We have been using it sice the beginning to the 2.4 
> > series kernel. My question is, is this patch still nessessary or is there possibly a
> > way do do this now without the patch?
> 
> I think Pauline never bothered to submit it. In Welsh's times
> it was considered too esoteric for inclusion, but nowdays we
> carry so much bloat that I'd think bigphysarea can be included.
> If you want it included, take over the issue and bug Linus.
> 
> -- Pete
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/