perf augmented_raw_syscalls: Postpone tmp map lookup to after pid_filter
authorArnaldo Carvalho de Melo <acme@redhat.com>
Fri, 23 Aug 2019 18:38:08 +0000 (15:38 -0300)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Mon, 26 Aug 2019 14:58:29 +0000 (11:58 -0300)
No sense in doing that lookup before figuring out if it will be used,
i.e. if the pid is being filtered that tmp space lookup will be useless.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-o74yggieorucfg4j74tb6rta@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/examples/bpf/augmented_raw_syscalls.c

index b85b177c6726e7c25f23831aede2aca653469ccb..0a8d217d65c7c6f00ff083b078c42b608e081a01 100644 (file)
@@ -250,13 +250,13 @@ int sys_enter(struct syscall_enter_args *args)
        struct syscall *syscall;
        int key = 0;
 
+       if (pid_filter__has(&pids_filtered, getpid()))
+               return 0;
+
         augmented_args = bpf_map_lookup_elem(&augmented_args_tmp, &key);
         if (augmented_args == NULL)
                 return 1;
 
-       if (pid_filter__has(&pids_filtered, getpid()))
-               return 0;
-
        probe_read(&augmented_args->args, sizeof(augmented_args->args), args);
 
        /*