On 29.09.2022 13:11, Roger Pau Monné wrote:
> On Thu, Sep 29, 2022 at 11:51:03AM +0200, Jan Beulich wrote:
>> With the "inversion" of VCPUOP handling, processing arch-specific ones
>> first, the forwarding of this sub-op from the (common) compat handler to
>> (common) non-compat one did no longer have the intended effect. It now
>> needs forwarding between the arch-specific handlers.
>>
>> Fixes: 8a96c0ea7999 ("xen: move do_vcpu_op() to arch specific code")
>> Signed-off-by: Jan Beulich <jbeul...@suse.com>
>
> Reviewed-by: Roger Pau Monné <roger....@citrix.com>
Thanks.
> This seems prone to errors, I wonder if we should add a note to
> do_vcpu_op() to notice that hypercalls handled there need to be
> explicitly forwarded from compat_vcpu_op.
Perhaps. I guess if such a comment had been added by the original change,
the mistake corrected here would have been noticed right away.
> I've also noticed that compat_common_vcpu_op() forwards
> VCPUOP_set_singleshot_timer to do_vcpu_op(), but that seems to be an
> useless jump, shouldn't it forward directly to common_vcpu_op()?
Looks like another (less severe) oversight, yes. Do you want to make a
patch or should I?
Jan
