On 2/15/2016 6:51 PM, Tamas K Lengyel wrote:
On Mon, Feb 15, 2016 at 9:44 AM, Jan Beulich > wrote:
>>> On 15.02.16 at 17:28, > wrote:
> On 2/15/2016 4:08 PM, Jan Beulich wrote:
>>
On Mon, Feb 15, 2016 at 9:44 AM, Jan Beulich wrote:
> >>> On 15.02.16 at 17:28, wrote:
> > On 2/15/2016 4:08 PM, Jan Beulich wrote:
> >>
> >>> After changing 1 to 1U though, I don't understand why we should also
> >>> range-check mop->event.
> >>> I'm
>>> On 15.02.16 at 17:28, wrote:
> On 2/15/2016 4:08 PM, Jan Beulich wrote:
>>
>>> After changing 1 to 1U though, I don't understand why we should also
>>> range-check mop->event.
>>> I'm imagining when (mop->event > 31):
>>> * (1U << mop->event) = 0 or >= (0x1 +
On 2/15/2016 4:08 PM, Jan Beulich wrote:
After changing 1 to 1U though, I don't understand why we should also
range-check mop->event.
I'm imagining when (mop->event > 31):
* (1U << mop->event) = 0 or >= (0x1 + 0x) (?)
No, it's plain undefined.
Weirdo C, didn't know that!
I've just
On Mon, Feb 15, 2016 at 7:08 AM, Jan Beulich wrote:
> >>> On 15.02.16 at 14:29, wrote:
> > On 2/15/2016 2:44 PM, Jan Beulich wrote:
> >>
> >>> switch ( mop->op )
> >>> {
> >>> case XEN_DOMCTL_MONITOR_OP_ENABLE:
> >>> case
>>> On 15.02.16 at 14:29, wrote:
> On 2/15/2016 2:44 PM, Jan Beulich wrote:
>>
>>> switch ( mop->op )
>>> {
>>> case XEN_DOMCTL_MONITOR_OP_ENABLE:
>>> case XEN_DOMCTL_MONITOR_OP_DISABLE:
>>> /* Check if event type is available. */
>>>
On 2/15/2016 2:44 PM, Jan Beulich wrote:
switch ( mop->op )
{
case XEN_DOMCTL_MONITOR_OP_ENABLE:
case XEN_DOMCTL_MONITOR_OP_DISABLE:
/* Check if event type is available. */
if ( unlikely(!(arch_monitor_get_capabilities(d) & (1 <<
mop->event))) )
>>> On 15.02.16 at 13:14, wrote:
> On 2/15/2016 1:41 PM, Jan Beulich wrote:
> On 15.02.16 at 07:37, wrote:
>>> default:
>>> -return -EOPNOTSUPP;
>>> +/*
>>> + * Should not be reached unless
On 2/15/2016 2:25 PM, Stefano Stabellini wrote:
On Mon, 15 Feb 2016, Jan Beulich wrote:
On 15.02.16 at 07:37, wrote:
default:
-return -EOPNOTSUPP;
+/*
+ * Should not be reached unless arch_monitor_get_capabilities() is not
+ *
On Mon, 15 Feb 2016, Jan Beulich wrote:
> >>> On 15.02.16 at 07:37, wrote:
> > default:
> > -return -EOPNOTSUPP;
> > +/*
> > + * Should not be reached unless arch_monitor_get_capabilities() is
> > not
> > + * properly implemented. In
On 2/15/2016 1:41 PM, Jan Beulich wrote:
+++ b/xen/include/xen/monitor.h
@@ -0,0 +1,30 @@
+/*
+ * include/xen/monitor.h
+ *
+ * Common monitor_op domctl handler.
+ *
+ * Copyright (c) 2015 Tamas K Lengyel (ta...@tklengyel.com)
+ * Copyright (c) 2016, Bitdefender S.R.L.
+ *
+ * This program is
On 2/15/2016 1:41 PM, Jan Beulich wrote:
On 15.02.16 at 07:37, wrote:
default:
-return -EOPNOTSUPP;
+/*
+ * Should not be reached unless arch_monitor_get_capabilities() is not
+ * properly implemented. In that case, since reaching
>>> On 15.02.16 at 07:37, wrote:
> default:
> -return -EOPNOTSUPP;
> +/*
> + * Should not be reached unless arch_monitor_get_capabilities() is
> not
> + * properly implemented. In that case, since reaching this point does
> + *
On 2/15/2016 8:37 AM, Corneliu ZUZU wrote:
diff --git a/xen/common/monitor.c b/xen/common/monitor.c
new file mode 100644
index 000..b708cab
--- /dev/null
+++ b/xen/common/monitor.c
+int rc;
+bool_t requested_status = 0;
+
+if ( unlikely(current->domain == d) ) /* no
This patch moves monitor_domctl to common-side.
Purpose: move what's common to common, prepare for implementation
of such vm-events on ARM.
* move get_capabilities to arch-side => arch_monitor_get_capabilities.
* add arch-side monitor op handling function => arch_monitor_domctl_op.
e.g.
15 matches
Mail list logo