perf/ftrace: Optimize perf/tracepoint interaction for single events

[mirror_ubuntu-jammy-kernel.git] / kernel / trace / trace_event_perf.c

diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 81f691eb3a30a508706afa874415a6343042adfb..89b780a7c5224e514bd475c485dc261679608f58 100644 (file)
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -17,18 +17,25 @@ EXPORT_SYMBOL_GPL(perf_arch_fetch_caller_regs);
 static char *perf_trace_buf;
 static char *perf_trace_buf_nmi;
 
-typedef typeof(char [PERF_MAX_TRACE_SIZE]) perf_trace_t ;
+/*
+ * Force it to be aligned to unsigned long to avoid misaligned accesses
+ * suprises
+ */
+typedef typeof(unsigned long [PERF_MAX_TRACE_SIZE / sizeof(unsigned long)])
+       perf_trace_t;
 
 /* Count the events in use (per event id, not per instance) */
 static int     total_ref_count;
 
-static int perf_trace_event_enable(struct ftrace_event_call *event)
+static int perf_trace_event_enable(struct ftrace_event_call *event, void *data)
 {
        char *buf;
        int ret = -ENOMEM;
 
-       if (event->perf_refcount++ > 0)
+       if (event->perf_refcount++ > 0) {
+               event->perf_data = NULL;
                return 0;
+       }
 
        if (!total_ref_count) {
                buf = (char *)alloc_percpu(perf_trace_t);
@@ -46,6 +53,7 @@ static int perf_trace_event_enable(struct ftrace_event_call *event)
 
        ret = event->perf_event_enable(event);
        if (!ret) {
+               event->perf_data = data;
                total_ref_count++;
                return 0;
        }
@@ -63,7 +71,7 @@ fail_buf:
        return ret;
 }
 
-int perf_trace_enable(int event_id)
+int perf_trace_enable(int event_id, void *data)
 {
        struct ftrace_event_call *event;
        int ret = -EINVAL;
@@ -72,7 +80,7 @@ int perf_trace_enable(int event_id)
        list_for_each_entry(event, &ftrace_events, list) {
                if (event->id == event_id && event->perf_event_enable &&
                    try_module_get(event->mod)) {
-                       ret = perf_trace_event_enable(event);
+                       ret = perf_trace_event_enable(event, data);
                        break;
                }
        }
@@ -130,6 +138,8 @@ __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
        char *trace_buf, *raw_data;
        int pc, cpu;
 
+       BUILD_BUG_ON(PERF_MAX_TRACE_SIZE % sizeof(unsigned long));
+
        pc = preempt_count();
 
        /* Protect the per cpu buffer, begin the rcu read side */
@@ -152,7 +162,7 @@ __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
        raw_data = per_cpu_ptr(trace_buf, cpu);
 
        /* zero the dead bytes from align to not leak stack to user */
-       *(u64 *)(&raw_data[size - sizeof(u64)]) = 0ULL;
+       memset(&raw_data[size - sizeof(u64)], 0, sizeof(u64));
 
        entry = (struct trace_entry *)raw_data;
        tracing_generic_entry_update(entry, *irq_flags, pc);