[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH] xen: only clobber multicall elements without error

>>> On 27.11.18 at 08:37, <jgross@xxxxxxxx> wrote:
> On 26/11/2018 17:50, Jan Beulich wrote:
>>>>> On 26.11.18 at 17:17, <jgross@xxxxxxxx> wrote:
>>> I really fail to see why it is so bad to not clobber data in a case
>>> which normally should never occur.
>> See Andrew's original reply. You're also clobbering on potential
>> success paths.
> I think you are missing a "not" here.
> But yes, I agree. This is a downside _I_ think we can live with.
> How many cases (with productive hypervisors) are known where multicall
> parameters have been illegally re-used? I guess the number is exactly 0.

I'm not aware of any either, but this means exactly nothing.

> I know of least one case where not clobbering would be useful for
> diagnostic reasons.

Well, I'm in agreement with Julien that the behavior between ordinary
hypercalls and multicalls would better be as consistent as possible.


Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.