> Date: Sun, 23 Aug 1998 23:53:53 +0200 (MET DST)
> From: Gerard Roudier <groudier@club-internet.fr>
>
> The same scheme can be used for a 64 bit device on 64 bit PCI bus driven
> from a 32 bit CPU and AFAIK, the Pentia II allows 36 bit addressing for
> host memory. So, 64 bit PCI could be used on ia32 in order to DMA into the
> whole 36 bit address range.
>
> Keep in mind that Linux will probably never support a >2GB physical
> address space, let alone a 36-bit one... The memory management issues
> are just too hairy and hard to make work when the physical address
> space is larger than the virtual one.
I know that's the current status about 36 bit address for PPRO/II.
But things may change ... ;)
> (There have been many threads on this over the past 1 or 2 months,
> feel free to look those up for the details about why this is so)
I read them, but if I did'nt participate to these threads this does not
mean that I agreed with everything that had been debated.
> Gerard I believe I remember discussing the addition of such an
> interface over a year ago, but I could have a bad memory. :-)
There is lots of architecture dependant stuff in the ncr driver.
Most of this stuff has been added to support the ppc and sparc.
This seems to demonstrate that we donnot have enough abstraction to
deal cleanly with PCI in Linux.
Looking into the current PCI code status does not require good memory,
half an eye should be enough. :-)
> So you are half-right, half-wrong. I think the Alpha is in a somewhat
> different state about these issues, perhaps Richard can comment.
And 64 bit PCI seems to be half-ready in Linux. ;-)
Thanks for the answers, to you and Martin. Will assume, the remaining
half of 64 bit PCI stuff will be complete when it will be needed.
Gerard.
-
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.altern.org/andrebalsa/doc/lkml-faq.html