Is there a framework in place or a notion of how the Linux PCI stuff ought
to cope with this yet?
On Sat, 13 Mar 1999, Richard Henderson wrote:
> On Wed, Mar 10, 1999 at 11:57:38PM -0600, Bob McElrath wrote:
> > It seems simple to me to modify these two files to support up to 1TB memory
> > on the alpha, as opposed to the hard-coded limit of 1GB now. It appears the
> > 21172 will only decode 40 bits of address, not 64, but that's 1TB...
>
> With 32-bit PCI devices, it is difficult to get past 2GB RAM. More
> than that requires either bounce buffers or that PCI windows be
> adjusted at runtime to cope with the various outstanding DMA requests.
> Which would be cool, but i think fairly tricky to modify all the
> drivers to get right.
>
> However, there's no reason we can't support 2GB instead of 1.
>
>
> r~
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/