Re: Need to arbitrary set a PCI bridge memory window

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Jan 24, 2006 at 07:16:43PM +0100, jfaslist wrote:
> Problem: Need to arbitrary set a PCI bridge memory window (PCI CFG  MEM 
> BASE and MEM LIMIT)
> 
> I am writing a linux vme driver for a PCI-VME bridge (Tundra universe2).
> The device is located behind a AMD8111 PCI bridge .
> When linux comes up, it reprograms all the PCI bridges (previously set 
> by the firmware). The problem is that it only allocates a memory window 
> large enough to accomodate the space requirements of subordinate PCI 
> devices as defined by the BAR registers. My problem is that the Tundra 
> PCI-VME bridge "consumes" a lot of extra address space (not decoded by 
> its PCI BARs) for the purpose of accessing the VME bus. Accessing VME is 
> done by programming several windows on the PCI-VME bridge. Anytime a PCI 
> address  falls in the programmed window is decoded, a VME access is done.
> I need around 1GBytes to access VME. So I need to be able forward that 
> big a range to the PCI bus that holds the PCI-VME bridge. But the 
> problem is that the AMD8111 doesn't let me do it since it only allows me 
> to acces the registers of the PCI-VME bridge, as defined by the BARs.
> Any idea, how I could force the AMD8111 PCI bridge to pass the 1GBytes 
> range to the PCI-VME controller?

Fix the device to have the proper BAR settings?  It sounds like this is
a broken device...

thanks,

greg k-h

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/


[Index of Archives]     [Newbies FAQ]     [Linux Kernel Mentors]     [Linux Kernel Development]     [IETF Annouce]     [Git]     [Networking]     [Security]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux SCSI]     [Linux ACPI]
  Powered by Linux