Although we treat the DTM counters as free-running such that we're not
too concerned about the initial DTM state, it's possible for a previous
user to have left DTM counters enabled and paired with DTC counters.
Thus if the first events are scheduled using some, but not all, DTMs,
the as-yet-unused ones could end up adding spurious increments to the
event counts at the DTC. Make sure we sync our initial DTM_PMU_CONFIG
state to all the DTMs at probe time to avoid that possibility.
Signed-off-by: Robin Murphy <[email protected]>
---
drivers/perf/arm-cmn.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/perf/arm-cmn.c b/drivers/perf/arm-cmn.c
index 0b2df72cee9f..81b0dfd511aa 100644
--- a/drivers/perf/arm-cmn.c
+++ b/drivers/perf/arm-cmn.c
@@ -1862,6 +1862,7 @@ static void arm_cmn_init_dtm(struct arm_cmn_dtm *dtm, struct arm_cmn_node *xp, i
dtm->base = xp->pmu_base + CMN_DTM_OFFSET(idx);
dtm->pmu_config_low = CMN_DTM_PMU_CONFIG_PMU_EN;
+ writeq_relaxed(dtm->pmu_config_low, dtm->base + CMN_DTM_PMU_CONFIG);
for (i = 0; i < 4; i++) {
dtm->wp_event[i] = -1;
writeq_relaxed(0, dtm->base + CMN_DTM_WPn_MASK(i));
--
2.36.1.dirty
On Mon, 5 Dec 2022 19:46:13 +0000, Robin Murphy wrote:
> Although we treat the DTM counters as free-running such that we're not
> too concerned about the initial DTM state, it's possible for a previous
> user to have left DTM counters enabled and paired with DTC counters.
> Thus if the first events are scheduled using some, but not all, DTMs,
> the as-yet-unused ones could end up adding spurious increments to the
> event counts at the DTC. Make sure we sync our initial DTM_PMU_CONFIG
> state to all the DTMs at probe time to avoid that possibility.
>
> [...]
Applied to will (for-next/perf), thanks!
[1/1] perf/arm-cmn: Reset DTM_PMU_CONFIG at probe
https://git.kernel.org/will/c/bb21ef19a3d8
Cheers,
--
Will
https://fixes.arm64.dev
https://next.arm64.dev
https://will.arm64.dev