[PATCH V2] powerpc/perf: Exclude kernel samples while counting events in user space.

[Athira Rajeev]

Perf event attritube supports exclude_kernel flag
to avoid sampling/profiling in supervisor state (kernel).
Based on this event attr flag, Monitor Mode Control Register
bit is set to freeze on supervisor state. But sometime (due
to hardware limitation), Sampled Instruction Address
Register (SIAR) locks on to kernel address even when
freeze on supervisor is set. Patch here adds a check to
drop those samples.

Signed-off-by: Athira Rajeev <atrajeev at linux.vnet.ibm.com>
---
Changes in v2:
- Initial patch was sent along with series:
  https://patchwork.ozlabs.org/project/linuxppc-dev/list/?series=209195
  Moving this patch as separate since this change is applicable
  for all PMU platforms.

 arch/powerpc/perf/core-book3s.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c
index 08643cb..40aa117 100644
--- a/arch/powerpc/perf/core-book3s.c
+++ b/arch/powerpc/perf/core-book3s.c
@@ -2122,6 +2122,17 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 	perf_event_update_userpage(event);
 
 	/*
+	 * Due to hardware limitation, sometimes SIAR could
+	 * lock on to kernel address even with freeze on
+	 * supervisor state (kernel) is set in MMCR2.
+	 * Check attr.exclude_kernel and address
+	 * to drop the sample in these cases.
+	 */
+	if (event->attr.exclude_kernel && record)
+		if (is_kernel_addr(mfspr(SPRN_SIAR)))
+			record = 0;
+
+	/*
 	 * Finally record data if requested.
 	 */
 	if (record) {
-- 
1.8.3.1