Hi Peter,
On 8/7/2020 5:02 PM, pet...@infradead.org wrote:
On Fri, Aug 07, 2020 at 02:24:30PM +0800, Jin, Yao wrote:
Hi Peter,
On 8/6/2020 7:00 PM, pet...@infradead.org wrote:
On Thu, Aug 06, 2020 at 11:18:27AM +0200, pet...@infradead.org wrote:
Suppose we have nested virt:
L0-hv
|
G0/L1-hv
|
G1
And we're running in G0, then:
- 'exclude_hv' would exclude L0 events
- 'exclude_host' would ... exclude L1-hv events?
- 'exclude_guest' would ... exclude G1 events?
So in arch/x86/events/intel/core.c we have:
static inline void intel_set_masks(struct perf_event *event, int idx)
{
struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
if (event->attr.exclude_host)
__set_bit(idx, (unsigned long *)&cpuc->intel_ctrl_guest_mask);
if (event->attr.exclude_guest)
__set_bit(idx, (unsigned long *)&cpuc->intel_ctrl_host_mask);
if (event_is_checkpointed(event))
__set_bit(idx, (unsigned long *)&cpuc->intel_cp_status);
}
exclude_host is now set by guest (pmc_reprogram_counter,
arch/x86/kvm/pmu.c). When enabling the event, we can check exclude_host to
know if it's a guest.
Otherwise we may need more flags in event->attr to indicate the status.
which is, afaict, just plain wrong. Should that not be something like:
if (!event->attr.exclude_host)
__set_bit(idx, (unsigned long *)&cpuc->intel_ctrl_host_mask);
if (!event->attr.exclude_guest)
__set_bit(idx, (unsigned long *)&cpuc->intel_ctrl_guest_mask);
How can we know it's guest or host even if exclude_host is set in guest?
I'm not following you, consider:
xh xg h g h' g'
0 0 0 0 1 1
0 1 1 0 1 0
1 0 0 1 0 1
1 1 1 1 0 0
Thanks for the table! It clearly shows the combinations of different conditions.
My understanding is:
xh = exclude_host
xg = exclude_guest
h = intel_ctrl_host_mask (before)
g = intel_ctrl_guest_mask (before)
h' = intel_ctrl_host_mask (after)
g' = intel_ctrl_guest_mask (after)
For guest, exclude_host = 1 and exclude_guest = 0
xh xg h g h' g'
1 0 0 1 0 1
before/after values are not changed.
For host, exclude_host = 0 and exclude_guest = 1
xh xg h g h' g'
0 1 1 0 1 0
before/after values are not changed.
So the 0,0 and 1,1 cases get flipped. I have a suspicion, but this
_really_ should have fat comments all over :-(
I'm not very sure about other cases.
xh xg h g h' g'
0 0 0 0 1 1
1 1 1 1 0 0
The before/after values are just reversed. I don't know if there will be some negative impacts?
Maybe we need more reviews here.
What a sodding trainwreck..
:(
Thanks
Jin Yao
