perf trace: Associate vfs_getname()'ed pathname with fd returned from 'openat'

When the vfs_getname() wannabe tracepoint is in place:

  # perf probe -l
    probe:vfs_getname    (on getname_flags:73@acme/git/linux/fs/namei.c with pathname)
  #

'perf trace' will use it to get the pathname when it is copied from
userspace to the kernel, right after syscalls:sys_enter_open, copied
in the 'probe:vfs_getname', stash it somewhere and then, at
syscalls:sys_exit_open time, if the 'open' return is not -1, i.e. a
successfull open syscall, associate that pathname to this return, i.e.
the fd.

We were not doing this for the 'openat' syscall, which would cause 'perf
trace' to fallback to using /proc to get the fd, change it so that we
use what we got from probe:vfs_getname, reducing the 'openat'
beautification process cost, ditching the syscalls performed to read
procfs state and avoiding some possible races in the process.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.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-xnp44ao3bkb6ejeczxfnjwsh@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 88561eed79505f737570aeacd10cd0ec0b8db0b8..2a85f5198da00c300377d542e9ce49f588bb235d 100644 (file)
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -121,7 +121,6 @@ struct trace {
        bool                    force;
        bool                    vfs_getname;
        int                     trace_pgfaults;
-       int                     open_id;
 };
 
 struct tp_field {
@@ -805,12 +804,17 @@ static struct syscall_fmt *syscall_fmt__find(const char *name)
        return bsearch(name, syscall_fmts, nmemb, sizeof(struct syscall_fmt), syscall_fmt__cmp);
 }
 
+/*
+ * is_exit: is this "exit" or "exit_group"?
+ * is_open: is this "open" or "openat"? To associate the fd returned in sys_exit with the pathname in sys_enter.
+ */
 struct syscall {
        struct event_format *tp_format;
        int                 nr_args;
+       bool                is_exit;
+       bool                is_open;
        struct format_field *args;
        const char          *name;
-       bool                is_exit;
        struct syscall_fmt  *fmt;
        struct syscall_arg_fmt *arg_fmt;
 };
@@ -1299,6 +1303,7 @@ static int trace__read_syscall_info(struct trace *trace, int id)
        }
 
        sc->is_exit = !strcmp(name, "exit_group") || !strcmp(name, "exit");
+       sc->is_open = !strcmp(name, "open") || !strcmp(name, "openat");
 
        return syscall__set_arg_fmts(sc);
 }
@@ -1722,7 +1727,7 @@ static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel,
 
        ret = perf_evsel__sc_tp_uint(evsel, ret, sample);
 
-       if (id == trace->open_id && ret >= 0 && ttrace->filename.pending_open) {
+       if (sc->is_open && ret >= 0 && ttrace->filename.pending_open) {
                trace__set_fd_pathname(thread, ret, ttrace->filename.name);
                ttrace->filename.pending_open = false;
                ++trace->stats.vfs_getname;
@@ -3205,8 +3210,6 @@ int cmd_trace(int argc, const char **argv)
                }
        }
 
-       trace.open_id = syscalltbl__id(trace.sctbl, "open");
-
        err = target__validate(&trace.opts.target);
        if (err) {
                target__strerror(&trace.opts.target, err, bf, sizeof(bf));