Thread (43 messages) 43 messages, 4 authors, 2025-02-18

Re: [PATCH v19 03/11] perf: arm_pmuv3: Don't disable counter in armv8pmu_enable_event()

From: Anshuman Khandual <hidden>
Date: 2025-02-03 06:38:43
Also in: kvmarm, linux-doc, linux-perf-users, lkml

On 2/3/25 06:12, Rob Herring (Arm) wrote:
From: Mark Rutland <mark.rutland@arm.com>

Currently armv8pmu_enable_event() starts by disabling the event counter
it has been asked to enable. This should not be necessary as the counter
(and the PMU as a whole) should not be active when
armv8pmu_enable_event() is called.
Makes sense.
Remove the redundant call to armv8pmu_disable_event_counter(). At the
same time, remove the comment immeditately above as everything it says
s/immeditately/immediately
is obvious from the function names below.
But should this comment drop change be folded into the next patch which
exclusively drops all obviously redundant disable/enable comments.
quoted hunk ↗ jump to hunk
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
---
 drivers/perf/arm_pmuv3.c | 5 -----
 1 file changed, 5 deletions(-)
diff --git a/drivers/perf/arm_pmuv3.c b/drivers/perf/arm_pmuv3.c
index 9ebc950559c0..5406b9ca591a 100644
--- a/drivers/perf/arm_pmuv3.c
+++ b/drivers/perf/arm_pmuv3.c
@@ -795,11 +795,6 @@ static void armv8pmu_enable_user_access(struct arm_pmu *cpu_pmu)
 
 static void armv8pmu_enable_event(struct perf_event *event)
 {
-	/*
-	 * Enable counter and interrupt, and set the counter to count
-	 * the event that we're interested in.
-	 */
-	armv8pmu_disable_event_counter(event);
 	armv8pmu_write_event_type(event);
 	armv8pmu_enable_event_irq(event);
 	armv8pmu_enable_event_counter(event);
Otherwise LGTM.

Reviewed-by: Anshuman Khandual <redacted>
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help