[Ksummit-2012-discuss] PCI breakout session

H. Peter Anvin hpa at zytor.com
Mon Jun 18 18:27:44 UTC 2012


On 06/18/2012 10:39 AM, Bjorn Helgaas wrote:
>>
>> OK, maybe this is a crazy idea, but we could do an extension to the PCI
>> standard to allow I/O address space to be subdivided finer than 4K.  We
>> could even try to push it at the PCI-SIG; it might be useful for some
>> PCI-Express configurations, since PCI-Express has exactly the same issue.
> 
> You mean like "en1k" (currently badly implemented because it's a quirk
> and can't work with any reassignment)?
> 

I would suggest that we may want even finer grain; the 1K comes,
presumably, from the common practice in x86 of only using the bottom 256
bytes out of each 1K chunk (due to ISA legacy space alias conflict
avoidance), but that might be getting obsolete...

	-hpa



More information about the Ksummit-2012-discuss mailing list