Dmitry/Anatoly-
Any comments on this RFC?
FWIW, I did some investigation of adding support for this in non-legacy
mode
and it seemed to me that this may be at odds with the dynmem approach.
My understanding is that with dynmem, the VA space is entirely contiguous,
but IOVA addresses are not.
> From: Don Wallwork [mailto:d...@xsightlabs.com]
> Sent: Thursday, 15 September 2022 14.14
>
> On 9/15/2022 3:40 AM, Morten Brørup wrote:
> >> From: Don Wallwork [mailto:d...@xsightlabs.com]
> >> Sent: Wednesday, 14 September 2022 23.12
> >>
> >> This patch maps hugepage memory such that address
On 9/15/2022 3:40 AM, Morten Brørup wrote:
From: Don Wallwork [mailto:d...@xsightlabs.com]
Sent: Wednesday, 14 September 2022 23.12
This patch maps hugepage memory such that address translation from
virtual to iova or vice versa can be done by simple addition/
subtraction of a constant value wit
> From: Don Wallwork [mailto:d...@xsightlabs.com]
> Sent: Wednesday, 14 September 2022 23.12
>
> This patch maps hugepage memory such that address translation from
> virtual to iova or vice versa can be done by simple addition/
> subtraction of a constant value without any page table walks.
>
> A
4 matches
Mail list logo
