On 08/26/2016 12:55 AM, Scott Wood wrote:
> On 08/26/2016 12:26 AM, Tillmann Heidsieck wrote:
>> On 2016-08-24 23:39, Scott Wood wrote:
>>> BTW, for some reason your patch is not showing up in Patchwork.
>>
>> Are there some known pitfalls when sending patches to Patchwork?
>
> It's not the first
On 08/26/2016 12:26 AM, Tillmann Heidsieck wrote:
> Hello Scott,
>
> thanks for the fast reply!
>
> On 2016-08-24 23:39, Scott Wood wrote:
> [..]
>>
>> The second inbound window is at 256G, and it maps all of RAM. Note
>> that
>> for accesses in this window, the PCI address is different from
Hello Scott,
thanks for the fast reply!
On 2016-08-24 23:39, Scott Wood wrote:
[..]
The second inbound window is at 256G, and it maps all of RAM. Note
that
for accesses in this window, the PCI address is different from the host
address.
I probably really misunderstand the manual here ...
For systems with >4G of RAM, the current implementation adds a second
inbound PCIe window starting at 128G this leaves all memory from 4G to
128G inaccessible to inbound PCIe transactions. The according errors
can be observed by using the EDAC driver for MPC85XX.
This patch changes this behaviour
On 08/24/2016 02:48 PM, Tillmann Heidsieck wrote:
> For systems with >4G of RAM, the current implementation adds a second
> inbound PCIe window starting at 128G this leaves all memory from 4G to
> 128G inaccessible to inbound PCIe transactions.
The second inbound window is at 256G, and it maps