6.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Robin Murphy <robin.murphy@xxxxxxx>
[ Upstream commit 88b63a82c84ed9bbcbdefb10cb6f75dd1dd04887 ]
Apparently pmu_event_sel is offset by 8 for all CCLA nodes, not just
the CCLA_RNI combination type.
Fixes: 23760a014417 ("perf/arm-cmn: Add CMN-700 support")
Acked-by: Mark Rutland <mark.rutland@xxxxxxx>
Reviewed-by: Ilkka Koskinen <ilkka@xxxxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Robin Murphy <robin.murphy@xxxxxxx>
Link: https://lore.kernel.org/r/6e7bb06fef6046f83e7647aad0e5be544139763f.1725296395.git.robin.murphy@xxxxxxx
Signed-off-by: Will Deacon <will@xxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/perf/arm-cmn.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/drivers/perf/arm-cmn.c b/drivers/perf/arm-cmn.c
index 6cd4fd9667aa6..f33fd110081c3 100644
--- a/drivers/perf/arm-cmn.c
+++ b/drivers/perf/arm-cmn.c
@@ -70,7 +70,8 @@
/* Technically this is 4 bits wide on DNs, but we only use 2 there anyway */
#define CMN__PMU_OCCUP1_ID GENMASK_ULL(34, 32)
-/* HN-Ps are weird... */
+/* Some types are designed to coexist with another device in the same node */
+#define CMN_CCLA_PMU_EVENT_SEL 0x008
#define CMN_HNP_PMU_EVENT_SEL 0x008
/* DTMs live in the PMU space of XP registers */
@@ -2323,10 +2324,13 @@ static int arm_cmn_discover(struct arm_cmn *cmn, unsigned int rgn_offset)
case CMN_TYPE_CXHA:
case CMN_TYPE_CCRA:
case CMN_TYPE_CCHA:
- case CMN_TYPE_CCLA:
case CMN_TYPE_HNS:
dn++;
break;
+ case CMN_TYPE_CCLA:
+ dn->pmu_base += CMN_CCLA_PMU_EVENT_SEL;
+ dn++;
+ break;
/* Nothing to see here */
case CMN_TYPE_MPAM_S:
case CMN_TYPE_MPAM_NS:
@@ -2344,7 +2348,7 @@ static int arm_cmn_discover(struct arm_cmn *cmn, unsigned int rgn_offset)
case CMN_TYPE_HNP:
case CMN_TYPE_CCLA_RNI:
dn[1] = dn[0];
- dn[0].pmu_base += CMN_HNP_PMU_EVENT_SEL;
+ dn[0].pmu_base += CMN_CCLA_PMU_EVENT_SEL;
dn[1].type = arm_cmn_subtype(dn->type);
dn += 2;
break;
--
2.43.0
