Re: [PATCH 6/7] xen/evtch: use smp barriers for user event ring

[Andrew Cooper]

On 08/02/2021 10:25, Jürgen Groß wrote:
> On 08.02.21 11:23, Andrew Cooper wrote:
>> On 08/02/2021 09:50, Jan Beulich wrote:
>>> On 08.02.2021 10:44, Andrew Cooper wrote:
>>>> On 06/02/2021 10:49, Juergen Gross wrote:
>>>>> The ring buffer for user events is used in the local system only, so
>>>>> smp barriers are fine for ensuring consistency.
>>>>>
>>>>> Reported-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
>>>>> Signed-off-by: Juergen Gross <jgross@xxxxxxxx>
>>>> These need to be virt_* to not break in UP builds (on non-x86).
>>> Initially I though so, too, but isn't the sole vCPU of such a
>>> VM getting re-scheduled to a different pCPU in the hypervisor
>>> an implied barrier anyway?
>>
>> Yes, but that isn't relevant to why UP builds break.
>>
>> smp_*() degrade to compiler barriers in UP builds, and while that's
>> mostly fine for x86 read/write, its not fine for ARM barriers.
>>
>> virt_*() exist specifically to be smp_*() which don't degrade to broken
>> in UP builds.
>
> But the barrier is really only necessary to serialize accesses within
> the guest against each other. There is no guest outside party involved.
>
> In case you are right this would mean that UP guests are all broken on
> Arm.

Oh - right.  This is a ring between the interrupt handler and a task. 
Not a ring between the guest and something else.

In which case smp_*() are correct.  Sorry for the noise.

~Andrew