[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZFv7XjRYXIsDT1uq@nvidia.com>
Date: Wed, 10 May 2023 17:15:26 -0300
From: Jason Gunthorpe <jgg@...dia.com>
To: Alex Williamson <alex.williamson@...hat.com>
Cc: "Tian, Kevin" <kevin.tian@...el.com>,
"kvm@...r.kernel.org" <kvm@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"Liu, Yi L" <yi.l.liu@...el.com>,
Nicolin Chen <nicolinc@...dia.com>,
"Lu, Baolu" <baolu.lu@...el.com>
Subject: Re: vPASID capability for VF
On Wed, May 10, 2023 at 11:24:49AM -0600, Alex Williamson wrote:
> Is this capability really intended to be mirrored by software to the
> VFs or do we need to expose the PASID capabilities of the VF in some
> other way? Thanks,
The VM definately needs this information, but not really the control
parts..
If we don't put it in the config space then we need to feed it through
ACPI which would solve both problems I suppose.
Jason
Powered by blists - more mailing lists