perf bpf: Move perf_event_output() from stdio.h to bpf.h

So that we don't always carry that __bpf_output__ map, leaving that to
the scripts wanting to use that facility.

'perf trace' will be changed to look if that map is present and only
setup the bpf-output events if so.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-azwys8irxqx9053vpajr0k5h@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

tools/perf/include/bpf/bpf.h

@@ -55,4 +55,6 @@ int _version SEC("version") = LINUX_VERSION_CODE;
 static int (*probe_read)(void *dst, int size, const void *unsafe_addr) = (void *)BPF_FUNC_probe_read;
 static int (*probe_read_str)(void *dst, int size, const void *unsafe_addr) = (void *)BPF_FUNC_probe_read_str;
 
+static int (*perf_event_output)(void *, struct bpf_map *, int, void *, unsigned long) = (void *)BPF_FUNC_perf_event_output;
+
 #endif /* _PERF_BPF_H */

tools/perf/include/bpf/stdio.h

@@ -9,9 +9,6 @@ struct bpf_map SEC("maps") __bpf_stdout__ = {
        .max_entries = __NR_CPUS__,
 };
 
-static int (*perf_event_output)(void *, struct bpf_map *, int, void *, unsigned long) =
-	(void *)BPF_FUNC_perf_event_output;
-
 #define puts(from) \
 	({ const int __len = sizeof(from); \
 	   char __from[__len] = from; \