[PATCH v3 1/2] perf/arm-cmn: Move struct arm_cmn_hw_event into struct hw_perf_event

[Aviv Bakal]

In order to increase CMN_MAX_DIMENSION beyond 12 (required for meshes
larger than 12x12, such as Graviton5), the arm_cmn_hw_event struct must
grow. Since it is overlaid on the beginning of hw_perf_event via an
unsafe cast, increasing its size would violate the static_assert that
guards against overflowing into the 'target' field.

Resolve this by moving struct arm_cmn_hw_event into the hw_perf_event
union as a proper named member, eliminating the cast in to_cmn_hw() and
making the size reservation explicit. Set CMN_MAX_DIMENSION to 14 to
accommodate larger mesh topologies.

Signed-off-by: Aviv Bakal <avivb at amazon.com>
---
 drivers/perf/arm-cmn.c     | 26 +-------------------------
 include/linux/perf_event.h | 22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+), 25 deletions(-)

diff --git a/drivers/perf/arm-cmn.c b/drivers/perf/arm-cmn.c
index f5305c8fdca4..3443b819afed 100644
--- a/drivers/perf/arm-cmn.c
+++ b/drivers/perf/arm-cmn.c
@@ -31,13 +31,8 @@
 #define CMN_CHILD_NODE_ADDR		GENMASK(29, 0)
 #define CMN_CHILD_NODE_EXTERNAL		BIT(31)
 
-#define CMN_MAX_DIMENSION		12
-#define CMN_MAX_XPS			(CMN_MAX_DIMENSION * CMN_MAX_DIMENSION)
 #define CMN_MAX_DTMS			(CMN_MAX_XPS + (CMN_MAX_DIMENSION - 1) * 4)
 
-/* Currently XPs are the node type we can have most of; others top out at 128 */
-#define CMN_MAX_NODES_PER_EVENT		CMN_MAX_XPS
-
 /* The CFG node has various info besides the discovery tree */
 #define CMN_CFGM_PERIPH_ID_01		0x0008
 #define CMN_CFGM_PID0_PART_0		GENMASK_ULL(7, 0)
@@ -148,7 +143,6 @@
 #define CMN_DT_PMSRR_SS_REQ		BIT(0)
 
 #define CMN_DT_NUM_COUNTERS		8
-#define CMN_MAX_DTCS			4
 
 /*
  * Even in the worst case a DTC counter can't wrap in fewer than 2^42 cycles,
@@ -595,24 +589,6 @@ static void arm_cmn_debugfs_init(struct arm_cmn *cmn, int id)
 static void arm_cmn_debugfs_init(struct arm_cmn *cmn, int id) {}
 #endif
 
-struct arm_cmn_hw_event {
-	struct arm_cmn_node *dn;
-	u64 dtm_idx[DIV_ROUND_UP(CMN_MAX_NODES_PER_EVENT * 2, 64)];
-	s8 dtc_idx[CMN_MAX_DTCS];
-	u8 num_dns;
-	u8 dtm_offset;
-
-	/*
-	 * WP config registers are divided to UP and DOWN events. We need to
-	 * keep to track only one of them.
-	 */
-	DECLARE_BITMAP(wp_idx, CMN_MAX_XPS);
-
-	bool wide_sel;
-	enum cmn_filter_select filter_sel;
-};
-static_assert(sizeof(struct arm_cmn_hw_event) <= offsetof(struct hw_perf_event, target));
-
 #define for_each_hw_dn(hw, dn, i) \
 	for (i = 0, dn = hw->dn; i < hw->num_dns; i++, dn++)
 
@@ -622,7 +598,7 @@ static_assert(sizeof(struct arm_cmn_hw_event) <= offsetof(struct hw_perf_event,
 
 static struct arm_cmn_hw_event *to_cmn_hw(struct perf_event *event)
 {
-	return (struct arm_cmn_hw_event *)&event->hw;
+	return &event->hw.cmn;
 }
 
 static void arm_cmn_set_index(u64 x[], unsigned int pos, unsigned int val)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 48d851fbd8ea..c38576a8e338 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -119,6 +119,7 @@ struct perf_branch_stack {
 };
 
 struct task_struct;
+struct arm_cmn_node;
 
 /*
  * extra PMU register associated with an event
@@ -200,6 +201,27 @@ struct hw_perf_event {
 			u64	conf;
 			u64	conf1;
 		};
+#ifdef CONFIG_ARM_CMN
+/* Some implementations use a mesh larger than the architectural max of 12 */
+#define CMN_MAX_DIMENSION		14
+#define CMN_MAX_XPS			(CMN_MAX_DIMENSION * CMN_MAX_DIMENSION)
+#define CMN_MAX_NODES_PER_EVENT		CMN_MAX_XPS
+#define CMN_MAX_DTCS			4
+		struct arm_cmn_hw_event { /* arm_cmn */
+			/*
+			 * CMN PMU event state overlaid on hw_perf_event.
+			 * Must fit before the 'target' field.
+			 */
+			struct arm_cmn_node	*dn;
+			u64			dtm_idx[DIV_ROUND_UP(CMN_MAX_NODES_PER_EVENT * 2, 64)];
+			s8			dtc_idx[CMN_MAX_DTCS];
+			u8			num_dns;
+			u8			dtm_offset;
+			DECLARE_BITMAP(wp_idx, CMN_MAX_XPS);
+			bool			wide_sel;
+			int			filter_sel;
+		} cmn;
+#endif
 	};
 	/*
 	 * If the event is a per task event, this will point to the task in
-- 
2.47.3