perf trace: Check and discard not only 'nr' but also '__syscall_nr'
authorTaeung Song <treeze.taeung@gmail.com>
Fri, 26 Feb 2016 13:14:25 +0000 (22:14 +0900)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Mon, 29 Feb 2016 14:34:28 +0000 (11:34 -0300)
Format fields of a syscall have the first variable '__syscall_nr' or
'nr' that mean the syscall number.  But it isn't relevant here so drop
it.

'nr' among fields of syscall was renamed '__syscall_nr'.  So add
exception handling to drop '__syscall_nr' and modify the comment for
this excpetion handling.

Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Taeung Song <treeze.taeung@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1456492465-5946-1-git-send-email-treeze.taeung@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/builtin-trace.c

index 26a337f939d80e9e1d6de7bd3fdb2592ad4587ac..8dc98c598b1aed734ab6a6695c7308f3c16e0910 100644 (file)
@@ -1725,8 +1725,12 @@ static int trace__read_syscall_info(struct trace *trace, int id)
 
        sc->args = sc->tp_format->format.fields;
        sc->nr_args = sc->tp_format->format.nr_fields;
-       /* drop nr field - not relevant here; does not exist on older kernels */
-       if (sc->args && strcmp(sc->args->name, "nr") == 0) {
+       /*
+        * We need to check and discard the first variable '__syscall_nr'
+        * or 'nr' that mean the syscall number. It is needless here.
+        * So drop '__syscall_nr' or 'nr' field but does not exist on older kernels.
+        */
+       if (sc->args && (!strcmp(sc->args->name, "__syscall_nr") || !strcmp(sc->args->name, "nr"))) {
                sc->args = sc->args->next;
                --sc->nr_args;
        }