linux-stable.git/kernel/trace/trace_events_hist.c, branch v5.18.3

tracing: Fix potential double free in create_var_ref()

2022-06-09T08:30:43+00:00

commit 99696a2592bca641eb88cc9a80c90e591afebd0f upstream.

In create_var_ref(), init_var_ref() is called to initialize the fields
of variable ref_field, which is allocated in the previous function call
to create_hist_field(). Function init_var_ref() allocates the
corresponding fields such as ref_field->system, but frees these fields
when the function encounters an error. The caller later calls
destroy_hist_field() to conduct error handling, which frees the fields
and the variable itself. This results in double free of the fields which
are already freed in the previous function.

Fix this by storing NULL to the corresponding fields when they are freed
in init_var_ref().

Link: https://lkml.kernel.org/r/20220425063739.3859998-1-keitasuzuki.park@sslab.ics.keio.ac.jp

Fixes: 067fe038e70f ("tracing: Add variable reference handling to hist triggers")
CC: stable@vger.kernel.org
Reviewed-by: Masami Hiramatsu 
Reviewed-by: Tom Zanussi 
Signed-off-by: Keita Suzuki 
Signed-off-by: Steven Rostedt (Google) 
Signed-off-by: Greg Kroah-Hartman

Merge tag 'trace-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

2022-03-23T18:40:25+00:00

Pull tracing updates from Steven Rostedt:

- New user_events interface. User space can register an event with the
kernel describing the format of the event. Then it will receive a
byte in a page mapping that it can check against. A privileged task
can then enable that event like any other event, which will change
the mapped byte to true, telling the user space application to start
writing the event to the tracing buffer.

- Add new "ftrace_boot_snapshot" kernel command line parameter. When
set, the tracing buffer will be saved in the snapshot buffer at boot
up when the kernel hands things over to user space. This will keep
the traces that happened at boot up available even if user space boot
up has tracing as well.

- Have TRACE_EVENT_ENUM() also update trace event field type
descriptions. Thus if a static array defines its size with an enum,
the user space trace event parsers can still know how to parse that
array.

- Add new TRACE_CUSTOM_EVENT() macro. This acts the same as the
TRACE_EVENT() macro, but will attach to an existing tracepoint. This
will make one tracepoint be able to trace different content and not
be stuck at only what the original TRACE_EVENT() macro exports.

- Fixes to tracing error logging.

- Better saving of cmdlines to PIDs when tracing (use the wakeup events
for mapping).

* tag 'trace-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (30 commits)
tracing: Have type enum modifications copy the strings
user_events: Add trace event call as root for low permission cases
tracing/user_events: Use alloc_pages instead of kzalloc() for register pages
tracing: Add snapshot at end of kernel boot up
tracing: Have TRACE_DEFINE_ENUM affect trace event types as well
tracing: Fix strncpy warning in trace_events_synth.c
user_events: Prevent dyn_event delete racing with ioctl add/delete
tracing: Add TRACE_CUSTOM_EVENT() macro
tracing: Move the defines to create TRACE_EVENTS into their own files
tracing: Add sample code for custom trace events
tracing: Allow custom events to be added to the tracefs directory
tracing: Fix last_cmd_set() string management in histogram code
user_events: Fix potential uninitialized pointer while parsing field
tracing: Fix allocation of last_cmd in last_cmd_set()
user_events: Add documentation file
user_events: Add sample code for typical usage
user_events: Add self-test for validator boundaries
user_events: Add self-test for perf_event integration
user_events: Add self-test for dynamic_events integration
user_events: Add self-test for ftrace integration
...

tracing: Fix last_cmd_set() string management in histogram code

2022-03-10T22:41:18+00:00

Using strnlen(dest, str, n) is confusing, as the size of dest must be
strlen(dest) + n + 1. Even more confusing, using sizeof(string constant)
gives you strlen(string constant) + 1 and not just strlen().

These two together made using strncat() with a constant string a bit off
in the calculations as we have:

	len = sizeof(HIST_PREFIX) + strlen(str) + 1;
	kfree(last_cmd);
	last_cmd = kzalloc(len, GFP_KERNEL);
	strcpy(last_cmd, HIST_PREFIX);
	len -= sizeof(HIST_PREFIX) + 1;
	strncat(last_cmd, str, len);

The above works if we s/sizeof/strlen/ with HIST_PREFIX (which is defined
as "hist:", but because sizeof(HIST_PREFIX) is equal to
strlen(HIST_PREFIX) + 1, we can drop the +1 in the code. But at least
comment that we are doing so.

Link: https://lore.kernel.org/all/202203082112.Iu7tvFl4-lkp@intel.com/

Fixes: 9f8e5aee93ed2 ("tracing: Fix allocation of last_cmd in last_cmd_set()")
Reported-by: kernel test robot 
Signed-off-by: Steven Rostedt (Google)

tracing/histogram: Fix sorting on old "cpu" value

2022-03-02T03:48:30+00:00

When trying to add a histogram against an event with the "cpu" field, it
was impossible due to "cpu" being a keyword to key off of the running CPU.
So to fix this, it was changed to "common_cpu" to match the other generic
fields (like "common_pid"). But since some scripts used "cpu" for keying
off of the CPU (for events that did not have "cpu" as a field, which is
most of them), a backward compatibility trick was added such that if "cpu"
was used as a key, and the event did not have "cpu" as a field name, then
it would fallback and switch over to "common_cpu".

This fix has a couple of subtle bugs. One was that when switching over to
"common_cpu", it did not change the field name, it just set a flag. But
the code still found a "cpu" field. The "cpu" field is used for filtering
and is returned when the event does not have a "cpu" field.

This was found by:

  # cd /sys/kernel/tracing
  # echo hist:key=cpu,pid:sort=cpu > events/sched/sched_wakeup/trigger
  # cat events/sched/sched_wakeup/hist

Which showed the histogram unsorted:

{ cpu:         19, pid:       1175 } hitcount:          1
{ cpu:          6, pid:        239 } hitcount:          2
{ cpu:         23, pid:       1186 } hitcount:         14
{ cpu:         12, pid:        249 } hitcount:          2
{ cpu:          3, pid:        994 } hitcount:          5

Instead of hard coding the "cpu" checks, take advantage of the fact that
trace_event_field_field() returns a special field for "cpu" and "CPU" if
the event does not have "cpu" as a field. This special field has the
"filter_type" of "FILTER_CPU". Check that to test if the returned field is
of the CPU type instead of doing the string compare.

Also, fix the sorting bug by testing for the hist_field flag of
HIST_FIELD_FL_CPU when setting up the sort routine. Otherwise it will use
the special CPU field to know what compare routine to use, and since that
special field does not have a size, it returns tracing_map_cmp_none.

Cc: stable@vger.kernel.org
Fixes: 1e3bac71c505 ("tracing/histogram: Rename "cpu" to "common_cpu"")
Reported-by: Daniel Bristot de Oliveira 
Signed-off-by: Steven Rostedt (Google)

tracing: Fix allocation of last_cmd in last_cmd_set()

2022-02-24T04:18:54+00:00

The strncat() used in last_cmd_set() includes the nul byte of length of
the string being copied in, when it should only hold the size of the
string being copied (not the nul byte). Change it to subtract the length
of the allocated space and the nul byte to pass that into the strncat().

Also, assign "len" instead of initializing it to zero and its first update
is to do a "+=".

Link: https://lore.kernel.org/all/202202140628.fj6e4w4v-lkp@intel.com/

Reported-by: kernel test robot 
Signed-off-by: Steven Rostedt (Google)

tracing: Remove size restriction on hist trigger cmd error logging

2022-02-11T03:27:33+00:00

Currently, hist trigger command error strings are restricted to a
length of MAX_FILTER_STR_VAL (256), which is too short for some
commands already seen in the wild (with cmd strings longer than that
showing up truncated in err_log).

Remove the restriction so that no hist trigger command error string is
ever truncated.

Link: https://lkml.kernel.org/r/0f9d46407222eaf6632cd3b417bc50a11f401b71.1643399022.git.zanussi@kernel.org

Signed-off-by: Tom Zanussi 
Signed-off-by: Steven Rostedt (Google)

tracing: Propagate is_signed to expression

2022-01-28T00:15:49+00:00

During expression parsing, a new expression field is created which
should inherit the properties of the operands, such as size and
is_signed.

is_signed propagation was missing, causing spurious errors with signed
operands.  Add it in parse_expr() and parse_unary() to fix the problem.

Link: https://lkml.kernel.org/r/f4dac08742fd7a0920bf80a73c6c44042f5eaa40.1643319703.git.zanussi@kernel.org

Cc: stable@vger.kernel.org
Fixes: 100719dcef447 ("tracing: Add simple expression support to hist triggers")
Reported-by: Yordan Karadzhov 
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=215513
Signed-off-by: Tom Zanussi 
Signed-off-by: Steven Rostedt (Google)

tracing: Fix smatch warning for do while check in event_hist_trigger_parse()

2022-01-28T00:15:48+00:00

The patch ec5ce0987541: "tracing: Allow whitespace to surround hist
trigger filter" from Jan 15, 2018, leads to the following Smatch
static checker warning:

    kernel/trace/trace_events_hist.c:6199 event_hist_trigger_parse()
    warn: 'p' can't be NULL.

Since p is always checked for a NULL value at the top of loop and
nothing in the rest of the loop will set it to NULL, the warning
is correct and might as well be 1 to silence the warning.

Link: https://lkml.kernel.org/r/a1d4c79766c0cf61e20438dc35244d216633fef6.1643319703.git.zanussi@kernel.org

Fixes: ec5ce09875410 ("tracing: Allow whitespace to surround hist trigger filter")
Reported-by: kernel test robot 
Reported-by: Dan Carpenter 
Signed-off-by: Tom Zanussi 
Signed-off-by: Steven Rostedt (Google)

tracing: Fix smatch warning for null glob in event_hist_trigger_parse()

2022-01-28T00:15:47+00:00

The recent rename of event_hist_trigger_parse() caused smatch
re-evaluation of trace_events_hist.c and as a result an old warning
was found:

    kernel/trace/trace_events_hist.c:6174 event_hist_trigger_parse()
    error: we previously assumed 'glob' could be null (see line 6166)

glob should never be null (and apparently smatch can also figure that
out and skip the warning when using the cross-function DB (but which
can't be used with a 0day build as it takes too much time to
generate)).

Nonetheless for clarity, remove the test but add a WARN_ON() in case
the code ever changes.

Link: https://lkml.kernel.org/r/96925e5c1f116654ada7ea0613d930b1266b5e1c.1643319703.git.zanussi@kernel.org

Fixes: f404da6e1d46c ("tracing: Add 'last error' error facility for hist triggers")
Reported-by: kernel test robot 
Reported-by: Dan Carpenter 
Signed-off-by: Tom Zanussi 
Signed-off-by: Steven Rostedt (Google)

tracing/histogram: Fix a potential memory leak for kstrdup()

2022-01-28T00:15:44+00:00

kfree() is missing on an error path to free the memory allocated by
kstrdup():

  p = param = kstrdup(data->params[i], GFP_KERNEL);

So it is better to free it via kfree(p).

Link: https://lkml.kernel.org/r/tencent_C52895FD37802832A3E5B272D05008866F0A@qq.com

Cc: stable@vger.kernel.org
Fixes: d380dcde9a07c ("tracing: Fix now invalid var_ref_vals assumption in trace action")
Signed-off-by: Xiaoke Wang 
Signed-off-by: Steven Rostedt (Google)