* BAR0/1 & enumeration
@ 2012-08-31 19:29 Paolo
2012-08-31 19:46 ` Matthew Wilcox
0 siblings, 1 reply; 4+ messages in thread
From: Paolo @ 2012-08-31 19:29 UTC (permalink / raw)
To: linux-pci
As newbie to the pcie world, I'm seeking help for understanding this:
How does the bios during enumeration finds whether BAR0/1 is set for pointing to
the config-space of its function or to memory space for memory transactions?
Thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: BAR0/1 & enumeration
2012-08-31 19:29 BAR0/1 & enumeration Paolo
@ 2012-08-31 19:46 ` Matthew Wilcox
2012-08-31 19:56 ` Paolo
0 siblings, 1 reply; 4+ messages in thread
From: Matthew Wilcox @ 2012-08-31 19:46 UTC (permalink / raw)
To: Paolo; +Cc: linux-pci
On Fri, Aug 31, 2012 at 07:29:50PM +0000, Paolo wrote:
> As newbie to the pcie world, I'm seeking help for understanding this:
>
> How does the bios during enumeration finds whether BAR0/1 is set for pointing to
> the config-space of its function or to memory space for memory transactions?
BARs point to either memory or I/O space (discriminated by the bottom
bit of the register). They never point to config space (indeed, since
they are found in config space, that wouldn't make sense).
--
Matthew Wilcox Intel Open Source Technology Centre
"Bill, look, we understand that you're interested in selling us this
operating system, but compare it to ours. We can't possibly take such
a retrograde step."
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: BAR0/1 & enumeration
2012-08-31 19:46 ` Matthew Wilcox
@ 2012-08-31 19:56 ` Paolo
2012-09-01 3:12 ` Jiang Liu
0 siblings, 1 reply; 4+ messages in thread
From: Paolo @ 2012-08-31 19:56 UTC (permalink / raw)
To: linux-pci
Not really.
I'm working on a NTB. NTBs are Type-0 headers, like any PCIe end point.
BAR0/1 on NTB can be set as base address for the config-space of its function OR
as based address of the memory space for memory transactions.
Hence my initial question. Looking fwd to some hint.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: BAR0/1 & enumeration
2012-08-31 19:56 ` Paolo
@ 2012-09-01 3:12 ` Jiang Liu
0 siblings, 0 replies; 4+ messages in thread
From: Jiang Liu @ 2012-09-01 3:12 UTC (permalink / raw)
To: Paolo; +Cc: linux-pci
The genefic PCI/PCIe framework doesn't do any thing special for NTB device,
just treat it as a normal PCI device. The issue you mentioned must be
handled by BIOS/OS driver with device specific knowledge.
--Gerry
On 09/01/2012 03:56 AM, Paolo wrote:
> Not really.
>
> I'm working on a NTB. NTBs are Type-0 headers, like any PCIe end point.
> BAR0/1 on NTB can be set as base address for the config-space of its function OR
> as based address of the memory space for memory transactions.
>
> Hence my initial question. Looking fwd to some hint.
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pci" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2012-09-01 3:12 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-08-31 19:29 BAR0/1 & enumeration Paolo
2012-08-31 19:46 ` Matthew Wilcox
2012-08-31 19:56 ` Paolo
2012-09-01 3:12 ` Jiang Liu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).