perf trace: Do not accept --no-syscalls together with -e
authorArnaldo Carvalho de Melo <acme@redhat.com>
Tue, 12 Apr 2016 19:05:02 +0000 (16:05 -0300)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Wed, 13 Apr 2016 13:11:52 +0000 (10:11 -0300)
Doesn't make sense and was causing a segfault, fix it.

  # trace -e clone --no-syscalls --event sched:*exec firefox
  The -e option can't be used with --no-syscalls.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-ccrahezikdk2uebptzr1eyyi@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/builtin-trace.c

index b842ddd3ad0cfc816086bee4859b4db1e563236d..d49c131bb5de172c228389e0a8db97f6538b99aa 100644 (file)
@@ -3344,6 +3344,8 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
                goto out;
        }
 
+       err = -1;
+
        if (trace.trace_pgfaults) {
                trace.opts.sample_address = true;
                trace.opts.sample_time = true;
@@ -3368,6 +3370,11 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
                return -1;
        }
 
+       if (!trace.trace_syscalls && ev_qualifier_str) {
+               pr_err("The -e option can't be used with --no-syscalls.\n");
+               goto out;
+       }
+
        if (output_name != NULL) {
                err = trace__open_output(&trace, output_name);
                if (err < 0) {