Re: [Xen-devel] [PATCH v2 1/2] VMX: fix VMCS race on context-switch paths

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 09.11.17 at 15:16, <raistlin@xxxxxxxx> wrote:
> On Thu, 2017-11-09 at 06:08 -0700, Jan Beulich wrote:
>> Tasklets already take care of this by
>> calling sync_local_execstate() before calling the handler. But
>> for softirqs this isn't really an option; I'm surprised to see that
>> tasklet code does this independently of what kind of tasklet that
>> is. 
>>
> Good point. Weird indeed.
> 
>> Softirq tasklets aren't used very often, so I wonder if we have
>> a latent bug here. Otoh, if this was actually fine, adding a similar
>> call to rcu_do_batch() (or its caller) would be an option, I think.
>> 
> We can have a look.

I'm sorry for the noise here - I've once again forgotten that
sync_local_execstate() does nothing if a lazy switch hasn't
happened already. Hence leaving the potentially bad
performance effect aside, doing the same for RCU (or more
generally softirqs) would be an option.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel