Minki/linux
1
0
Fork 1
mirror of https://github.com/torvalds/linux.git synced 2024-11-14 16:12:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

perf trace beauty: Disable fd->pathname when close() not enabled

Browse Source 
As we invalidate the fd->pathname table in the SCA_CLOSE_FD beautifier,
if we don't have it we may end up keeping an fd->pathname association
that then gets misprinted.

The previous behaviour continues when the close() syscall is enabled,
which may still be a a problem if we lose records (i.e. we may lose a
'close' record and then get that fd reused by socket()) but then the
tool will notify that records are being lost and the user will be warned
that some of the heuristics will fall apart.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-b7t6h8sq9lebemvfy2zh3qq1@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This commit is contained in:
Arnaldo Carvalho de Melo 2019-07-16 16:56:49 -03:00
parent 1d86275225
commit 79d725cdf2
1 changed files with 16 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
tools/perf/builtin-trace.c
View File

@ -127,6 +127,7 @@ struct trace {
unsigned int min_stack;
int raw_augmented_syscalls_args_size;
bool raw_augmented_syscalls;
bool fd_path_disabled;
bool sort_events;
bool not_ev_qualifier;
bool live;
@ -1178,7 +1179,7 @@ static const char *thread__fd_path(struct thread *thread, int fd,
{
struct thread_trace *ttrace = thread__priv(thread);
if (ttrace == NULL)
if (ttrace == NULL || trace->fd_path_disabled)
return NULL;
if (fd < 0)
@ -2097,7 +2098,7 @@ static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel,
ret = perf_evsel__sc_tp_uint(evsel, ret, sample);
if (sc->is_open && ret >= 0 && ttrace->filename.pending_open) {
if (!trace->fd_path_disabled && 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;
@ -3206,7 +3207,6 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
if (trace->syscalls.prog_array.sys_enter)
trace__init_syscalls_bpf_prog_array_maps(trace);
if (trace->ev_qualifier_ids.nr > 0) {
err = trace__set_ev_qualifier_filter(trace);
if (err < 0)
@ -3218,6 +3218,19 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
}
}
/*
* If the "close" syscall is not traced, then we will not have the
* opportunity to, in syscall_arg__scnprintf_close_fd() invalidate the
* fd->pathname table and were ending up showing the last value set by
* syscalls opening a pathname and associating it with a descriptor or
* reading it from /proc/pid/fd/ in cases where that doesn't make
* sense.
*
* So just disable this beautifier (SCA_FD, SCA_FDAT) when 'close' is
* not in use.
*/
trace->fd_path_disabled = !trace__syscall_enabled(trace, syscalltbl__id(trace->sctbl, "close"));
err = perf_evlist__apply_filters(evlist, &evsel);
if (err < 0)
goto out_error_apply_filters;