On 21.05.2018 19:42, Adam Wallis wrote:
On 5/21/2018 9:53 AM, Mathias Nyman wrote:
On 21.05.2018 15:56, Adam Wallis wrote:
Not sure if if there's any benefit in allocating the scratchpad structures from
a closer node, or any harm? xhci driver doesn't really access scratchpad that
frequently.
I
On 5/21/2018 9:53 AM, Mathias Nyman wrote:
> On 21.05.2018 15:56, Adam Wallis wrote:
> Not sure if if there's any benefit in allocating the scratchpad structures
> from
> a closer node, or any harm? xhci driver doesn't really access scratchpad that
> frequently.
I don't see how it would hurt and
On 21.05.2018 15:56, Adam Wallis wrote:
On 5/16/2018 2:02 AM, Greg Kroah-Hartman wrote:
On Tue, May 15, 2018 at 04:51:53PM -0400, Adam Wallis wrote:
Ok, fair enough, I was hoping that "modern" systems would have better
NUMA memory interconnects. I guess that isn't the case still :(
Things wi
On 5/16/2018 2:02 AM, Greg Kroah-Hartman wrote:
> On Tue, May 15, 2018 at 04:51:53PM -0400, Adam Wallis wrote:
>
> Ok, fair enough, I was hoping that "modern" systems would have better
> NUMA memory interconnects. I guess that isn't the case still :(
Things will keep improving, I'm sure.
Mathia
On Tue, May 15, 2018 at 04:51:53PM -0400, Adam Wallis wrote:
> On 5/15/2018 11:07 AM, Greg Kroah-Hartman wrote:
> > On Tue, May 15, 2018 at 09:53:57AM -0400, Adam Wallis wrote:
> > Does this really do anything? Given the speed of USB3 at the moment,
> > does fixing the memory to the node the PCI d
On 5/15/2018 11:07 AM, Greg Kroah-Hartman wrote:
> On Tue, May 15, 2018 at 09:53:57AM -0400, Adam Wallis wrote:
> Does this really do anything? Given the speed of USB3 at the moment,
> does fixing the memory to the node the PCI device is on show any
> measurable speedups? Last I remember about NU
On Tue, May 15, 2018 at 09:53:57AM -0400, Adam Wallis wrote:
> The xhci driver forces DMA memory to be node aware, however, there are
> several ring-related memory allocations that are not memory node aware.
> This patch resolves those *alloc functions to be allocated on the proper
> memory node.
The xhci driver forces DMA memory to be node aware, however, there are
several ring-related memory allocations that are not memory node aware.
This patch resolves those *alloc functions to be allocated on the proper
memory node.
Signed-off-by: Adam Wallis
---
drivers/usb/host/xhci-mem.c | 63 +++
