On 11/18/20 6:24 AM, David Gibson wrote:
> On Mon, Nov 16, 2020 at 04:34:22PM +0100, Greg Kurz wrote:
>> This series was largely built on the assumption that IPI numbers are
>> numerically equal to vCPU ids. Even if this happens to be the case
>> in practice with the default machine settings, this
On Mon, Nov 16, 2020 at 04:34:22PM +0100, Greg Kurz wrote:
> This series was largely built on the assumption that IPI numbers are
> numerically equal to vCPU ids. Even if this happens to be the case
> in practice with the default machine settings, this ceases to be true
> if VSMT is set to a differ
On Mon, 16 Nov 2020 16:54:32 +0100
Cédric Le Goater wrote:
> On 11/16/20 4:34 PM, Greg Kurz wrote:
> > This series was largely built on the assumption that IPI numbers are
> > numerically equal to vCPU ids. Even if this happens to be the case
> > in practice with the default machine settings, thi
On 11/16/20 4:34 PM, Greg Kurz wrote:
> This series was largely built on the assumption that IPI numbers are
> numerically equal to vCPU ids. Even if this happens to be the case
> in practice with the default machine settings, this ceases to be true
> if VSMT is set to a different value than the nu
This series was largely built on the assumption that IPI numbers are
numerically equal to vCPU ids. Even if this happens to be the case
in practice with the default machine settings, this ceases to be true
if VSMT is set to a different value than the number of vCPUs per core.
This causes bogus IPI