tracing/rseq: Add mm_cid field to rseq_update

Add the mm_cid field to the rseq_update event, allowing tracers to
follow which mm_cid is observed by user-space, and whether negative
mm_cid values are visible in case of internal scheduler implementation
issues.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20221122203932.231377-22-mathieu.desnoyers@efficios.com
This commit is contained in:
Mathieu Desnoyers 2022-11-22 15:39:23 -05:00 committed by Peter Zijlstra
parent a3798e6ffb
commit 14a8644d4f

View File

@ -17,14 +17,17 @@ TRACE_EVENT(rseq_update,
TP_STRUCT__entry(
__field(s32, cpu_id)
__field(s32, node_id)
__field(s32, mm_cid)
),
TP_fast_assign(
__entry->cpu_id = raw_smp_processor_id();
__entry->node_id = cpu_to_node(__entry->cpu_id);
__entry->mm_cid = task_mm_cid(t);
),
TP_printk("cpu_id=%d node_id=%d", __entry->cpu_id, __entry->node_id)
TP_printk("cpu_id=%d node_id=%d mm_cid=%d", __entry->cpu_id,
__entry->node_id, __entry->mm_cid)
);
TRACE_EVENT(rseq_ip_fixup,