On 09/12/2019 16:19, Jan Beulich wrote:
> On 05.12.2019 23:30, Andrew Cooper wrote:
>> --- a/xen/arch/x86/domctl.c
>> +++ b/xen/arch/x86/domctl.c
>> @@ -326,9 +326,12 @@ long arch_do_domctl(
>>
>> switch ( domctl->cmd )
>> {
>> -
>> case XEN_DOMCTL_shadow_op:
>> ret = paging_domctl(d, &domctl->u.shadow_op, u_domctl, 0);
>> + /*
>> + * Continuations from paging_domctl() switch index to arch_1, and
>> + * can't use the common domctl continuation path.
>> + */
>> if ( ret == -ERESTART )
>> return hypercall_create_continuation(__HYPERVISOR_arch_1,
>> "h", u_domctl);
> There's also XEN_DOMCTL_getpageframeinfo3 down from here which
> now invokes a continuation.
Fixed.
>
>> @@ -1080,6 +1068,9 @@ long do_domctl(XEN_GUEST_HANDLE_PARAM(xen_domctl_t)
>> u_domctl)
>> if ( copyback && __copy_to_guest(u_domctl, op, 1) )
>> ret = -EFAULT;
>>
>> + if ( ret == -ERESTART )
>> + ret = hypercall_create_continuation(__HYPERVISOR_domctl,
>> + "h", u_domctl);
> You may want to mention in the description the bug you fix here:
> Previously the -EFAULT returning visible in context should have
> canceled any active continuation.
That would be presuming I'd even spotted it...
Having looked though the paths once again, I don't think there was a
path which continued and had copyback set, so this is at most a latent bug.
~Andrew
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
