cvs commit: src/sys/i386/pci pci_bus.c

[John Baldwin]

On Wednesday 06 October 2004 03:08 pm, M. Warner Losh wrote:
> In message: <200410061054.42956.jhb at FreeBSD.org>
>
>             John Baldwin <jhb at FreeBSD.org> writes:
> : On Wednesday 06 October 2004 03:22 am, Warner Losh wrote:
> : > imp         2004-10-06 07:22:58 UTC
> : >
> : >   FreeBSD src repository
> : >
> : >   Modified files:
> : >     sys/i386/pci         pci_bus.c
> : >   Log:
> : >   For legacy PCI bridges, limit memory allocation to the top 32MB of
> : >   RAM.  Many older, legacy bridges only allow allocation from this
> : >   range.  This only appies to devices who don't have their memory
> : >   assigned by the BIOS (since we allocate the ranges so assigned
> : >   exactly), so should have minimal impact.
> : >
> : >   Hoewver, for CardBus bridges (cbb), they rarely get the resources
> : >   allocated by the BIOS, and this patch helps them greatly.  Typically
> : >   the 'bad Vcc' messages are caused by this problem.
> :
> : We really should be looking at the SMAP to find out what the real limits
> : of PCI space are I think since I think the SMAP includes an entry for PCI
> : memory mapped I/O.
>
> Can you point me at the details on how to do this?  I'd be absolutely
> thrilled to do this correctly, but so far my research has been focused
> on the actual host bridges.  Sadly, these aren't standardized.  If
> there's another area I can research, I'd be thrilled.

Ugh, it turns out that the SMAP doesn't list memio space, it just lists where 
there is memory.  This might be a start at least but it's only part of a fix.

-- 
John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org