perf: Reorganize scripting support
Subsystems (4)
Ian Rogers
reviewerPATCH v10 9/19 perf python: Add wrapper for perf_data file abstraction
In reply to Ian RogersThe perf_data struct is needed for session supported.
---
v2:
1. Fixed Memory & FD Leaks in pyrf_data__init : Added cleanup of old state (closing file and freeing path) if __init__ is called multiple times on the same object.
2. Fixed Invalid Free in pyrf_data__delete : Ensured pdata->data.path is always dynamically allocated via strdup() , even for the default "perf.data" . This avoids passing a static string literal to free().
3. Fixed NULL Pointer Dereference in pyrf_data__str : Added a check for NULL path to prevent segfaults if str() or repr() is called on an uninitialized object.
4. Guarded fd Argument Usage: Added a check to ensure that if an fd is provided, it corresponds to a pipe, failing gracefully with a ValueError otherwise.
v7:
- Added pyrf_data__new to zero-initialize pdata->data to fix crash on re-initialization.
--- tools/perf/util/python.c | 122 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 121 insertions(+), 1 deletion(-)
tools/perf/util/python.c +121 -1
@@ -6,6 +6,7 @@#include <linux/err.h> #include <poll.h> +#include <unistd.h> #include <internal/lib.h> #include <perf/cpumap.h>
@@ -14,6 +15,7 @@#include "callchain.h" #include "counts.h" +#include "data.h" #include "event.h" #include "evlist.h" #include "evsel.h"
@@ -2454,6 +2456,120 @@ static PyObject *pyrf__metrics(PyObject *self, PyObject *args)return list; } +struct pyrf_data { +PyObject_HEAD + + struct perf_data data; +}; + +static int pyrf_data__init (struct pyrf_data *pdata,PyObject *args,PyObject *kwargs) +{ +static char *kwlist[] = {"path" ,"fd" ,NULL }; +char *path =NULL ; +int fd =-1 ; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"|si" , kwlist, &path, &fd)) +return -1 ; + +if (pdata->data.open) +perf_data__close (&pdata->data); +free ((char *)pdata->data.path); + +if (fd !=-1 ) { +struct stat st; + +if (fstat (fd, &st) <0 || !S_ISFIFO (st.st_mode)) { +PyErr_SetString (PyExc_ValueError, +"fd argument is only supported for pipes" ); +return -1 ; + } +if (!path) + path ="-" ; +else if (strcmp (path,"-" ) !=0 ) { +PyErr_SetString (PyExc_ValueError, +"path must be '-' when fd is provided" ); +return -1 ; + } + fd =dup (fd); +if (fd <0 ) { +PyErr_SetFromErrno (PyExc_OSError); +return -1 ; + } + }else if (path &&strcmp (path,"-" ) ==0 ) { + fd =dup (0 ); +if (fd <0 ) { +PyErr_SetFromErrno (PyExc_OSError); +return -1 ; + } + } + +if (!path) + path ="perf.data" ; + + pdata->data.path =strdup (path); +if (!pdata->data.path) { +if (fd !=-1 ) +close (fd); +PyErr_NoMemory (); +return -1 ; + } + + pdata->data.mode =PERF_DATA_MODE_READ ; + pdata->data.file.fd = fd; +if (perf_data__open (&pdata->data) <0 ) { +PyErr_Format (PyExc_IOError,"Failed to open perf data: %s" , + pdata->data.path ? pdata->data.path :"perf.data" ); +return -1 ; + } +return 0 ; +} + +static void pyrf_data__delete (struct pyrf_data *pdata) +{ +perf_data__close (&pdata->data); +free ((char *)pdata->data.path); +Py_TYPE (pdata)->tp_free ((PyObject *)pdata); +} + +static PyObject *pyrf_data__str (PyObject *self) +{ +const struct pyrf_data *pdata = (const struct pyrf_data *)self; + +if (!pdata->data.path) +return PyUnicode_FromString ("[uninitialized]" ); +return PyUnicode_FromString (pdata->data.path); +} + +static PyObject *pyrf_data__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) +{ +struct pyrf_data *pdata; + + pdata = (struct pyrf_data *)PyType_GenericNew (type, args, kwargs); +if (pdata) +memset (&pdata->data,0 ,sizeof (pdata->data)); +return (PyObject *)pdata; +} + +static const char pyrf_data__doc[] =PyDoc_STR ("perf data file object." ); + +static PyTypeObject pyrf_data__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.data" , + .tp_basicsize =sizeof (struct pyrf_data ), + .tp_dealloc = (destructor )pyrf_data__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_data__doc, + .tp_init = (initproc )pyrf_data__init, + .tp_repr = pyrf_data__str, + .tp_str = pyrf_data__str, +}; + +static int pyrf_data__setup_types (void ) +{ + pyrf_data__type.tp_new = pyrf_data__new; +return PyType_Ready (&pyrf_data__type); +} +static PyMethodDef perf__methods[] = { { .ml_name ="metrics" ,
@@ -2516,7 +2632,8 @@ PyMODINIT_FUNC PyInit_perf(void)pyrf_cpu_map__setup_types () <0 ||pyrf_pmu_iterator__setup_types () <0 ||pyrf_pmu__setup_types () <0 || - pyrf_counts_values__setup_types() < 0) +pyrf_counts_values__setup_types () <0 || + pyrf_data__setup_types() < 0) return module;/* The page_size is placed in util object. */
@@ -2564,6 +2681,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_counts_values__type);PyModule_AddObject (module,"counts_values" , (PyObject *)&pyrf_counts_values__type); +Py_INCREF (&pyrf_data__type); +PyModule_AddObject (module,"data" , (PyObject *)&pyrf_data__type); + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 09/19] perf python: Add wrapper for perf_data file abstraction
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] Missing pointer nullification after freeing `pdata->data.path` in `pyrf_data__init` causes a…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Missing pointer nullification after freeing pdata->data.path in pyrf_data__init causes a Use-After-Free and Double Free vulnerability if re-initialization fails. --
commit 3bd4b2111f650c3ffdc0443e905e19b99cc0ce92
Author: Ian Rogers <[email protected]>
perf python: Add wrapper for perf_data file abstraction
This commit introduces a perf_data wrapper for the perf python extension, providing session support to interact with perf.data files.
Quoted 3 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 10 lines
tools/perf/util/python.c +10
-0+static int pyrf_data__init (struct pyrf_data *pdata,PyObject *args,PyObject *kwargs) +{ +static char *kwlist[] = {"path" ,"fd" ,NULL }; +char *path =NULL ; +int fd =-1 ; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"|si" , kwlist, &path, &fd)) +return -1 ; + +if (pdata->data.open)
tools/perf/util/python.c static PyObject *pyrf__metrics(PyObject *self, PyObject *args) · line 2484
Quoted 10 lines
tools/perf/util/python.c +10
-0+perf_data__close (&pdata->data); +free ((char *)pdata->data.path); + +if (fd !=-1 ) { +struct stat st; + +if (fstat (fd, &st) <0 || !S_ISFIFO (st.st_mode)) { +PyErr_SetString (PyExc_ValueError, +"fd argument is only supported for pipes" ); +return -1 ;
[Severity: Medium]
If this object is re-initialized by calling __init__ again, pdata->data.path is freed above. If we hit this error path (or others below it) and return -1, pdata->data.path is left as a dangling pointer.
Since the Python object remains alive even if __init__ fails on re-initialization, could a later call to pyrf_data__str() dereference this freed pointer?
Additionally, would the eventual garbage collection in pyrf_data__delete() cause a double-free?
Setting pdata->data.path = NULL immediately after the free() might prevent these issues.
[ ... ]
tools/perf/util/python.c static PyObject *pyrf__metrics(PyObject *self, PyObject *args) · line 2530
Quoted 4 lines
tools/perf/util/python.c +4
-0+static void pyrf_data__delete (struct pyrf_data *pdata) +{ +perf_data__close (&pdata->data); +free ((char *)pdata->data.path);
[Severity: Medium]
This is where the dangling pointer from the failed re-initialization would be freed a second time.
[ ... ]
tools/perf/util/python.c static PyObject *pyrf__metrics(PyObject *self, PyObject *args) · line 2540
Quoted 7 lines
tools/perf/util/python.c +7
-0+static PyObject *pyrf_data__str (PyObject *self) +{ +const struct pyrf_data *pdata = (const struct pyrf_data *)self; + +if (!pdata->data.path) +return PyUnicode_FromString ("[uninitialized]" ); +return PyUnicode_FromString (pdata->data.path);
[Severity: Medium]
And this is where the use-after-free would occur during string conversion if pdata->data.path wasn't cleared after being freed.
[ ... ]
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=9
Ian Rogers
reviewerPATCH v10 10/19 perf python: Add python session abstraction wrapping perf's session
In reply to Ian RogersSessions are necessary to be able to use perf.data files within a tool. Add a wrapper python type that incorporates the tool. Allow a sample callback to be passed when creating the session. When process_events is run this callback will be called, if supplied, for sample events.
An example use looks like:
```
$ perf record -e cycles,instructions -a sleep 3
$ PYTHONPATH=..../perf/python python3
Python 3.13.7 (main, Aug 20 2025, 22:17:40) [GCC 14.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import perf
>>> count=0
... def handle_sample(x):
... global count
... if count < 3:
... print(dir(x))
... count = count + 1
... perf.session(perf.data("perf.data"),sample=handle_sample).process_events()
...
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
```Also, add the ability to get the thread associated with a session. For threads, allow the comm string to be retrieved. This can be useful for filtering threads. Connect up some of the standard event handling in psession->tool to better support queries of the machine. Also connect up the symbols.
---
v2:
1. Fixed Potential Crash in pyrf_thread__comm : Used thread__comm_str() to safely retrieve the command name, avoiding a crash if thread__comm() returns NULL.
2. Fixed Double Free Risk: Zeroed out user_regs , intr_regs , and callchain in the shallow copy of perf_sample to prevent Python from attempting to free pointers it doesn't own.
3. Fixed Memory Leak & Exception Handling in Callback: Handled the return value of PyObject_CallFunction() to avoid leaks, and checked for failure to abort the loop and propagate Python exceptions cleanly.
4. Enforced Type Safety: Used O! with &pyrf_data__type in PyArg_ParseTupleAndKeywords to prevent bad casts from passing arbitrary objects as perf.data.
5. Added Missing Build ID Handler: Registered perf_event__process_build_id to allow correct symbol resolution.
6. Fixed Double Free Crash on Init Failure: Set session and pdata to NULL on failure to prevent tp_dealloc from double-freeing them.
7. Preserved C-level Errors: Made pyrf_session__process_events return the error code integer rather than always returning None .
v7:
- Fixed NULL comm handling.
- Avoided swallowing exceptions in module init.
- Fixed checkpatch warning for missing blank line.
v8:
- Switch from pyrf_session__init to pyrf_session__new to avoid dealing with a potentially NULL session variable.
- Added pid, tid, ppid, and cpu attributes to perf.thread.
v10:
- Added CHECK_INITIALIZED checks to pyrf_session methods to prevent NULL pointer dereferences on uninitialized objects.
--- tools/perf/util/python.c | 336 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 330 insertions(+), 6 deletions(-)
tools/perf/util/python.c +330 -6
@@ -14,8 +14,10 @@#include <structmember.h> #include "callchain.h" +#include "comm.h" #include "counts.h" #include "data.h" +#include "debug.h" #include "event.h" #include "evlist.h" #include "evsel.h"
@@ -25,8 +27,12 @@#include "pmus.h" #include "print_binary.h" #include "record.h" +#include "session.h" #include "strbuf.h" +#include "symbol.h" +#include "thread.h" #include "thread_map.h" +#include "tool.h" #include "tp_pmu.h" #include "trace-event.h" #include "util/sample.h"
@@ -2570,6 +2576,313 @@ static int pyrf_data__setup_types(void)return PyType_Ready (&pyrf_data__type); } +struct pyrf_thread { +PyObject_HEAD + + struct thread *thread; +}; + +static void pyrf_thread__delete (struct pyrf_thread *pthread) +{ +thread__put (pthread->thread); +Py_TYPE (pthread)->tp_free ((PyObject *)pthread); +} + +static PyObject *pyrf_thread__comm (PyObject *obj) +{ +struct pyrf_thread *pthread = (void *)obj; +const char *str =thread__comm_str (pthread->thread); + +if (!str) + Py_RETURN_NONE; + +return PyUnicode_FromString (str); +} + +static PyMethodDef pyrf_thread__methods[] = { + { + .ml_name ="comm" , + .ml_meth = (PyCFunction )pyrf_thread__comm, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Comm(and) associated with this thread." ) + }, + { .ml_name =NULL , } +}; + +static PyObject *pyrf_thread__get_pid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__pid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_tid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__tid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_ppid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__ppid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_cpu (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__cpu (pthread->thread)); +} + +static PyGetSetDef pyrf_thread__getset[] = { + { .name ="pid" , .get = (getter )pyrf_thread__get_pid, .doc ="process ID" }, + { .name ="tid" , .get = (getter )pyrf_thread__get_tid, .doc ="thread ID" }, + { .name ="ppid" , .get = (getter )pyrf_thread__get_ppid, .doc ="parent process ID" }, + { .name ="cpu" , .get = (getter )pyrf_thread__get_cpu, .doc ="cpu number" }, + { .name =NULL } +}; + +static const char pyrf_thread__doc[] =PyDoc_STR ("perf thread object." ); + +static PyTypeObject pyrf_thread__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.thread" , + .tp_basicsize =sizeof (struct pyrf_thread ), + .tp_dealloc = (destructor )pyrf_thread__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_methods = pyrf_thread__methods, + .tp_getset = pyrf_thread__getset, + .tp_doc = pyrf_thread__doc, +}; + +static int pyrf_thread__setup_types (void ) +{ +return PyType_Ready (&pyrf_thread__type); +} + +static PyObject *pyrf_thread__from_thread (struct thread *thread) +{ + struct pyrf_thread *pthread = PyObject_New(struct pyrf_thread, &pyrf_thread__type); + +if (!pthread) +return NULL ; + + pthread->thread =thread__get (thread); +return (PyObject *)pthread; +} + +struct pyrf_session { +PyObject_HEAD + + struct perf_session *session; +struct perf_tool tool; +struct pyrf_data *pdata; +PyObject *sample; +}; + +static int pyrf_session_tool__sample (const struct perf_tool *tool, +union perf_event *event, +struct perf_sample *sample, +struct machine *machine__maybe_unused ) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +PyObject *pyevent =pyrf_event__new (event); +struct pyrf_event *pevent = (struct pyrf_event *)pyevent; +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + +memcpy (&pevent->event, event, event->header.size); +if (evsel__parse_sample (sample->evsel, &pevent->event, &pevent->sample) <0 ) { +Py_DECREF (pyevent); +return -1 ; + } + + ret =PyObject_CallFunction (psession->sample,"O" , pyevent); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} + +static PyObject *pyrf_session__find_thread (struct pyrf_session *psession,PyObject *args) +{ +struct machine *machine; +struct thread *thread =NULL ; +PyObject *result; +int pid; + +CHECK_INITIALIZED (psession->session,"session" ); + +if (!PyArg_ParseTuple (args,"i" , &pid)) +return NULL ; + + machine = &psession->session->machines.host; + thread =machine__find_thread (machine, pid, pid); + +if (!thread) { + machine =perf_session__find_machine (psession->session, pid); +if (machine) + thread =machine__find_thread (machine, pid, pid); + } + +if (!thread) { +PyErr_Format (PyExc_TypeError,"Failed to find thread %d" , pid); +return NULL ; + } + result =pyrf_thread__from_thread (thread); +thread__put (thread); +return result; +} + +static PyObject *pyrf_session__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) +{ +struct pyrf_data *pdata; +PyObject *sample =NULL ; +static char *kwlist[] = {"data" ,"sample" ,NULL }; +struct pyrf_session *psession; +struct perf_session *session; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|O" , kwlist, &pyrf_data__type, &pdata, + &sample)) +return NULL ; + + psession = PyObject_New(struct pyrf_session, type); +if (!psession) +return NULL ; + + psession->session =NULL ; + psession->sample =NULL ; + psession->pdata =NULL ; + +Py_INCREF (pdata); + psession->pdata = pdata; + +perf_tool__init (&psession->tool,/*ordered_events=*/ true); + psession->tool.ordering_requires_timestamps = true; + +#define ADD_TOOL (name) \ + do { \ + if (name) { \ + if (!PyCallable_Check(name)) { \ + PyErr_SetString(PyExc_TypeError, #name " must be callable"); \ + goto err_out; \ + } \ + psession->tool.name = pyrf_session_tool__##name; \ + Py_INCREF(name); \ + psession->name = name; \ + } \ + } while (0) + +ADD_TOOL (sample); +#undef ADD_TOOL + + psession->tool.comm = perf_event__process_comm; + psession->tool.mmap = perf_event__process_mmap; + psession->tool.mmap2 = perf_event__process_mmap2; + psession->tool.namespaces = perf_event__process_namespaces; + psession->tool.cgroup = perf_event__process_cgroup; + psession->tool.exit = perf_event__process_exit; + psession->tool.fork = perf_event__process_fork; + psession->tool.ksymbol = perf_event__process_ksymbol; + psession->tool.text_poke = perf_event__process_text_poke; + psession->tool.build_id = perf_event__process_build_id; + session =perf_session__new (&pdata->data, &psession->tool); +if (IS_ERR (session)) { +PyErr_Format (PyExc_IOError,"failed to create session: %ld" ,PTR_ERR (session)); + goto err_out; + } + psession->session = session; + +if (symbol__init (perf_session__env (session)) <0 ) { +PyErr_SetString (PyExc_OSError,"perf: symbol__init failed" ); + goto err_out; + } + +if (perf_session__create_kernel_maps (session) <0 ) +pr_warning ("Cannot read kernel map\n" ); + +return (PyObject *)psession; +err_out: +Py_DECREF (psession); +return NULL ; +} + +static void pyrf_session__delete (struct pyrf_session *psession) +{ +perf_session__delete (psession->session); +Py_XDECREF (psession->pdata); +Py_XDECREF (psession->sample); +Py_TYPE (psession)->tp_free ((PyObject *)psession); +} + +static PyObject *pyrf_session__find_thread_events (struct pyrf_session *psession) +{ +int err; + +CHECK_INITIALIZED (psession->session,"session" ); + + err =perf_session__process_events (psession->session); + +if (PyErr_Occurred ()) +return NULL ; + +return PyLong_FromLong (err); +} + +static PyMethodDef pyrf_session__methods[] = { + { + .ml_name ="process_events" , + .ml_meth = (PyCFunction )pyrf_session__find_thread_events, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Iterate and process events." ) + }, + { + .ml_name ="find_thread" , + .ml_meth = (PyCFunction )pyrf_session__find_thread, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Returns the thread associated with a pid." ) + }, + { .ml_name =NULL , } +}; + +static const char pyrf_session__doc[] =PyDoc_STR ("perf session object." ); + +static PyObject *pyrf_session__getattro (struct pyrf_session *psession,PyObject *attr_name) +{ +if (!psession->session) { +PyErr_SetString (PyExc_ValueError,"session not initialized" ); +return NULL ; + } +return PyObject_GenericGetAttr ((PyObject *) psession, attr_name); +} + +static int pyrf_session__setattro (struct pyrf_session *psession,PyObject *attr_name,PyObject *value) +{ +if (!psession->session) { +PyErr_SetString (PyExc_ValueError,"session not initialized" ); +return -1 ; + } +return PyObject_GenericSetAttr ((PyObject *) psession, attr_name, value); +} + +static PyTypeObject pyrf_session__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.session" , + .tp_basicsize =sizeof (struct pyrf_session ), + .tp_dealloc = (destructor )pyrf_session__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_methods = pyrf_session__methods, + .tp_doc = pyrf_session__doc, + .tp_new = pyrf_session__new, + .tp_getattro = (getattrofunc ) pyrf_session__getattro, + .tp_setattro = (setattrofunc ) pyrf_session__setattro, +}; + +static int pyrf_session__setup_types (void ) +{ +return PyType_Ready (&pyrf_session__type); +} +static PyMethodDef perf__methods[] = { { .ml_name ="metrics" ,
@@ -2624,8 +2937,10 @@ PyMODINIT_FUNC PyInit_perf(void) };PyObject *module =PyModule_Create (&moduledef); - if (module == NULL || - pyrf_event__setup_types() < 0 || +if (module ==NULL ) +return NULL ; + + if (pyrf_event__setup_types() < 0 || pyrf_evlist__setup_types() < 0 || pyrf_evsel__setup_types() < 0 || pyrf_thread_map__setup_types() < 0 ||
@@ -2633,8 +2948,12 @@ PyMODINIT_FUNC PyInit_perf(void) pyrf_pmu_iterator__setup_types() < 0 || pyrf_pmu__setup_types() < 0 || pyrf_counts_values__setup_types() < 0 || - pyrf_data__setup_types() < 0) - return module; + pyrf_data__setup_types() < 0 || + pyrf_session__setup_types() < 0 || + pyrf_thread__setup_types() < 0) { +Py_DECREF (module); +return NULL ; + }/* The page_size is placed in util object. */ page_size =sysconf (_SC_PAGE_SIZE);
@@ -2684,6 +3003,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_data__type);PyModule_AddObject (module,"data" , (PyObject *)&pyrf_data__type); +Py_INCREF (&pyrf_session__type); +PyModule_AddObject (module,"session" , (PyObject *)&pyrf_session__type); + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
@@ -2697,7 +3019,9 @@ PyMODINIT_FUNC PyInit_perf(void) } error: -if (PyErr_Occurred ()) -PyErr_SetString (PyExc_ImportError,"perf: Init failed!" ); +if (PyErr_Occurred ()) { +Py_XDECREF (module); +return NULL ; + }return module; }
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 10/19] perf python: Add python session abstraction wrapping perf's session
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [High] The pyrf_session wrapper fails to register the `.attr` and `.feature` callbacks in `perf_tool`…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] The pyrf_session wrapper fails to register the .attr and .feature callbacks in perf_tool, silently dropping these events and breaking event processing in pipe mode. --
commit fa89d52ee147ebcf23e53127275a3667c97e2358
Author: Ian Rogers <[email protected]>
perf python: Add python session abstraction wrapping perf's session
This commit adds a wrapper Python type for perf's session, incorporating the tool and allowing a sample callback to be passed when creating the session. When process_events is run, this callback will be executed for sample events.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 1 line
tools/perf/util/python.c +0
-0@@ -2570,6 +2576,313 @@ static int pyrf_data__setup_types(void)
[ ... ]
Quoted 2 lines
tools/perf/util/python.c +2
-0+static PyObject *pyrf_session__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) +{
[ ... ]
tools/perf/util/python.c static int pyrf_data__setup_types(void) · line 2789
Quoted 11 lines
tools/perf/util/python.c +11
-0+ psession->tool.comm = perf_event__process_comm; + psession->tool.mmap = perf_event__process_mmap; + psession->tool.mmap2 = perf_event__process_mmap2; + psession->tool.namespaces = perf_event__process_namespaces; + psession->tool.cgroup = perf_event__process_cgroup; + psession->tool.exit = perf_event__process_exit; + psession->tool.fork = perf_event__process_fork; + psession->tool.ksymbol = perf_event__process_ksymbol; + psession->tool.text_poke = perf_event__process_text_poke; + psession->tool.build_id = perf_event__process_build_id; + session =perf_session__new (&pdata->data, &psession->tool);
[Severity: High]
Are we missing callbacks for attribute and feature events here?
If the .attr and .feature callbacks in struct perf_tool are not registered here, incoming events will be silently ignored. In pipe mode, dropping perf_event_header_attr events prevents the creation of evlists and evsels, which breaks event processing entirely.
Missing .feature callbacks would similarly prevent the population of perf_env. Could we add the appropriate attr and feature callbacks here to properly support pipe mode streams?
Quoted 4 lines
tools/perf/util/python.c +4
-0+if (IS_ERR (session)) { +PyErr_Format (PyExc_IOError,"failed to create session: %ld" ,PTR_ERR (session)); + goto err_out; + }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=10
Ian Rogers
reviewerPATCH v10 11/19 perf python: Refactor and add accessors to sample event
In reply to Ian RogersAdd common evsel field for events and move sample specific fields to only be present in sample events. Add accessors for sample events. Ensure offsets are within the bounds of the event. Allocate just enough memory for the copied event, don't make the maximum event size each time.
---
v5:
1. Fix Uninitialized Memory: Restore zero-initialization of pevent->sample in pyrf_event__new() to prevent wild free crashes on error paths.
v6:
- Refactored pyrf_event__new to take evsel and session, and use dynamic allocation based on event size. Updated callers.
v8:
- Ensure events are properly deallocated.
--- tools/perf/util/python.c | 618 +++++++++++++++++++++++++++++++++------ 1 file changed, 534 insertions(+), 84 deletions(-)
tools/perf/util/python.c +534 -84
@@ -13,21 +13,28 @@#include <perf/mmap.h> #include <structmember.h> +#include "addr_location.h" +#include "build-id.h" #include "callchain.h" #include "comm.h" #include "counts.h" #include "data.h" #include "debug.h" +#include "dso.h" #include "event.h" #include "evlist.h" #include "evsel.h" #include "expr.h" +#include "map.h" #include "metricgroup.h" #include "mmap.h" #include "pmus.h" #include "print_binary.h" #include "record.h" +#include "sample.h" #include "session.h" +#include "srccode.h" +#include "srcline.h" #include "strbuf.h" #include "symbol.h" #include "thread.h"
@@ -35,7 +42,6 @@#include "tool.h" #include "tp_pmu.h" #include "trace-event.h" -#include "util/sample.h" #ifdef HAVE_LIBTRACEEVENT #include <event-parse.h>
@@ -43,6 +49,8 @@PyMODINIT_FUNC PyInit_perf (void ); +static PyObject *pyrf_evsel__from_evsel (struct evsel *evsel); +#define member_def (type, member, ptype, help) \ { #member, ptype, \ offsetof(struct pyrf_event, event) + offsetof(struct type, member), \
@@ -71,21 +79,53 @@ PyMODINIT_FUNC PyInit_perf(void);struct pyrf_event {PyObject_HEAD +/** @sample: The parsed sample from the event. */ struct perf_sample sample; -union perf_event event; +/** @al: The address location from machine__resolve, lazily computed. */ +struct addr_location al; +/** @al_resolved: True when machine__resolve been called. */ +bool al_resolved; +/** @event: The underlying perf_event that may be in a file or ring buffer. */ +union perf_event event; };#define sample_members \ - sample_member_def(sample_ip, ip, T_ULONGLONG, "event ip"), \ sample_member_def(sample_pid, pid, T_INT, "event pid"), \ sample_member_def(sample_tid, tid, T_INT, "event tid"), \ sample_member_def(sample_time, time, T_ULONGLONG, "event timestamp"), \ - sample_member_def(sample_addr, addr, T_ULONGLONG, "event addr"), \ sample_member_def(sample_id, id, T_ULONGLONG, "event id"), \ sample_member_def(sample_stream_id, stream_id, T_ULONGLONG, "event stream id"), \ sample_member_def(sample_period, period, T_ULONGLONG, "event period"), \ sample_member_def(sample_cpu, cpu, T_UINT, "event cpu"), +static PyObject *pyrf_event__get_evsel (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->sample.evsel) + Py_RETURN_NONE; + +return pyrf_evsel__from_evsel (pevent->sample.evsel); +} + +static PyGetSetDef pyrf_event__getset[] = { + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { .name =NULL , }, +}; + +static void pyrf_event__delete (struct pyrf_event *pevent) +{ +if (pevent->al_resolved) +addr_location__exit (&pevent->al); +perf_sample__exit (&pevent->sample); +Py_TYPE (pevent)->tp_free ((PyObject *)pevent); +} +static const char pyrf_mmap_event__doc[] =PyDoc_STR ("perf mmap event object." );static PyMemberDef pyrf_mmap_event__members[] = {
@@ -124,9 +164,11 @@ static PyTypeObject pyrf_mmap_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.mmap_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_mmap_event__doc, .tp_members = pyrf_mmap_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_mmap_event__repr, };
@@ -159,9 +201,11 @@ static PyTypeObject pyrf_task_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.task_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_task_event__doc, .tp_members = pyrf_task_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_task_event__repr, };
@@ -188,9 +232,11 @@ static PyTypeObject pyrf_comm_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.comm_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_comm_event__doc, .tp_members = pyrf_comm_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_comm_event__repr, };
@@ -220,9 +266,11 @@ static PyTypeObject pyrf_throttle_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.throttle_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_throttle_event__doc, .tp_members = pyrf_throttle_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_throttle_event__repr, };
@@ -255,9 +303,11 @@ static PyTypeObject pyrf_lost_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.lost_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_lost_event__doc, .tp_members = pyrf_lost_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_lost_event__repr, };
@@ -285,9 +335,11 @@ static PyTypeObject pyrf_read_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.read_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_read_event__doc, .tp_members = pyrf_read_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_read_event__repr, };
@@ -295,16 +347,17 @@ static const char pyrf_sample_event__doc[] = PyDoc_STR("perf sample event objectstatic PyMemberDef pyrf_sample_event__members[] = { sample_members +sample_member_def (sample_ip, ip,T_ULONGLONG ,"event ip" ), +sample_member_def (sample_addr, addr,T_ULONGLONG ,"event addr" ), +sample_member_def (sample_phys_addr, phys_addr,T_ULONGLONG ,"event physical addr" ), +sample_member_def (sample_weight, weight,T_ULONGLONG ,"event weight" ), +sample_member_def (sample_data_src, data_src,T_ULONGLONG ,"event data source" ), +sample_member_def (sample_insn_count, insn_cnt,T_ULONGLONG ,"event instruction count" ), +sample_member_def (sample_cyc_count, cyc_cnt,T_ULONGLONG ,"event cycle count" ),member_def (perf_event_header, type,T_UINT ,"event type" ), { .name =NULL , }, }; -static void pyrf_sample_event__delete (struct pyrf_event *pevent) -{ -perf_sample__exit (&pevent->sample); -Py_TYPE (pevent)->tp_free ((PyObject *)pevent); -} -static PyObject *pyrf_sample_event__repr (const struct pyrf_event *pevent) {PyObject *ret;
@@ -322,6 +375,8 @@ static PyObject *pyrf_sample_event__repr(const struct pyrf_event *pevent)#ifdef HAVE_LIBTRACEEVENT static bool is_tracepoint (const struct pyrf_event *pevent) { +if (!pevent->sample.evsel) +return false;return pevent->sample.evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; }
@@ -392,6 +447,199 @@ get_tracepoint_field(struct pyrf_event *pevent, PyObject *attr_name) }#endif /* HAVE_LIBTRACEEVENT */ +static int pyrf_sample_event__resolve_al (struct pyrf_event *pevent) +{ +struct evsel *evsel = pevent->sample.evsel; +struct evlist *evlist = evsel ? evsel->evlist :NULL ; +struct perf_session *session = evlist ?evlist__session (evlist) :NULL ; + +if (pevent->al_resolved) +return 0 ; + +if (!session) +return -1 ; + +addr_location__init (&pevent->al); +if (machine__resolve (&session->machines.host, &pevent->al, &pevent->sample) <0 ) { +addr_location__exit (&pevent->al); +return -1 ; + } + + pevent->al_resolved = true; +return 0 ; +} + +static PyObject *pyrf_sample_event__get_dso (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyUnicode_FromString (dso__name (map__dso (pevent->al.map))); +} + +static PyObject *pyrf_sample_event__get_dso_long_name (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyUnicode_FromString (dso__long_name (map__dso (pevent->al.map))); +} + +static PyObject *pyrf_sample_event__get_dso_bid (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +char sbuild_id[SBUILD_ID_SIZE ]; + +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +build_id__snprintf (dso__bid (map__dso (pevent->al.map)), sbuild_id,sizeof (sbuild_id)); +return PyUnicode_FromString (sbuild_id); +} + +static PyObject *pyrf_sample_event__get_map_start (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (map__start (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_map_end (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (map__end (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_map_pgoff (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (map__pgoff (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_symbol (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyUnicode_FromString (pevent->al.sym->name); +} + +static PyObject *pyrf_sample_event__get_sym_start (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->al.sym->start); +} + +static PyObject *pyrf_sample_event__get_sym_end (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->al.sym->end); +} + +static PyObject *pyrf_sample_event__get_raw_buf (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pevent->event.header.type !=PERF_RECORD_SAMPLE ) + Py_RETURN_NONE; + +return PyBytes_FromStringAndSize ((const char *)pevent->sample.raw_data, + pevent->sample.raw_size); +} + +static PyObject *pyrf_sample_event__srccode (PyObject *self,PyObject *args) +{ +struct pyrf_event *pevent = (void *)self; +u64 addr = pevent->sample.ip; +char *srcfile =NULL ; +char *srccode =NULL ; +unsigned int line =0 ; +int len =0 ; +PyObject *result; +struct addr_location al; + +if (!PyArg_ParseTuple (args,"|K" , &addr)) +return NULL ; + +if (pyrf_sample_event__resolve_al (pevent) <0 ) + Py_RETURN_NONE; + +if (addr != pevent->sample.ip) { +addr_location__init (&al); +thread__find_symbol_fb (pevent->al.thread, pevent->sample.cpumode, addr, &al); + }else { +addr_location__init (&al); + al.thread =thread__get (pevent->al.thread); + al.map =map__get (pevent->al.map); + al.sym = pevent->al.sym; + al.addr = pevent->al.addr; + } + +if (al.map) { +struct dso *dso =map__dso (al.map); + +if (dso) { + srcfile =get_srcline_split (dso,map__rip_2objdump (al.map, addr), + &line); + } + } +addr_location__exit (&al); + +if (srcfile) { + srccode =find_sourceline (srcfile, line, &len); + result =Py_BuildValue ("(sIs#)" , srcfile, line, srccode, (Py_ssize_t )len); +free (srcfile); + }else { + result =Py_BuildValue ("(sIs#)" ,NULL ,0 ,NULL , (Py_ssize_t )0 ); + } + +return result; +} + +static PyObject *pyrf_sample_event__insn (PyObject *self,PyObject *args__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; +struct thread *thread; +struct machine *machine; + +if (pyrf_sample_event__resolve_al (pevent) <0 ) + Py_RETURN_NONE; + + thread = pevent->al.thread; + +if (!thread || !thread__maps (thread)) + Py_RETURN_NONE; + + machine =maps__machine (thread__maps (thread)); +if (!machine) + Py_RETURN_NONE; + +if (pevent->sample.ip && !pevent->sample.insn_len) +perf_sample__fetch_insn (&pevent->sample, thread, machine); + +if (!pevent->sample.insn_len) + Py_RETURN_NONE; + +return PyBytes_FromStringAndSize ((const char *)pevent->sample.insn, + pevent->sample.insn_len); +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -405,13 +653,102 @@ pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name)return obj ?:PyObject_GenericGetAttr ((PyObject *) pevent, attr_name); } +static PyGetSetDef pyrf_sample_event__getset[] = { + { + .name ="raw_buf" , + .get = (getter )pyrf_sample_event__get_raw_buf, + .set =NULL , + .doc ="event raw buffer." , + }, + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { + .name ="dso" , + .get = (getter )pyrf_sample_event__get_dso, + .set =NULL , + .doc ="event dso short name." , + }, + { + .name ="dso_long_name" , + .get = (getter )pyrf_sample_event__get_dso_long_name, + .set =NULL , + .doc ="event dso long name." , + }, + { + .name ="dso_bid" , + .get = (getter )pyrf_sample_event__get_dso_bid, + .set =NULL , + .doc ="event dso build id." , + }, + { + .name ="map_start" , + .get = (getter )pyrf_sample_event__get_map_start, + .set =NULL , + .doc ="event map start address." , + }, + { + .name ="map_end" , + .get = (getter )pyrf_sample_event__get_map_end, + .set =NULL , + .doc ="event map end address." , + }, + { + .name ="map_pgoff" , + .get = (getter )pyrf_sample_event__get_map_pgoff, + .set =NULL , + .doc ="event map page offset." , + }, + { + .name ="symbol" , + .get = (getter )pyrf_sample_event__get_symbol, + .set =NULL , + .doc ="event symbol name." , + }, + { + .name ="sym_start" , + .get = (getter )pyrf_sample_event__get_sym_start, + .set =NULL , + .doc ="event symbol start address." , + }, + { + .name ="sym_end" , + .get = (getter )pyrf_sample_event__get_sym_end, + .set =NULL , + .doc ="event symbol end address." , + }, + { .name =NULL , }, +}; + +static PyMethodDef pyrf_sample_event__methods[] = { + { + .ml_name ="srccode" , + .ml_meth = (PyCFunction )pyrf_sample_event__srccode, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Get source code for an address." ) + }, + { + .ml_name ="insn" , + .ml_meth = (PyCFunction )pyrf_sample_event__insn, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Get instruction bytes for a sample." ) + }, + { .ml_name =NULL , } +}; +static PyTypeObject pyrf_sample_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name ="perf.sample_event" , .tp_basicsize =sizeof (struct pyrf_event ), + .tp_dealloc = (destructor )pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_sample_event__doc, .tp_members = pyrf_sample_event__members, + .tp_getset = pyrf_sample_event__getset, + .tp_methods = pyrf_sample_event__methods, .tp_repr = (reprfunc )pyrf_sample_event__repr, .tp_getattro = (getattrofunc ) pyrf_sample_event__getattro, };
@@ -447,25 +784,17 @@ static PyTypeObject pyrf_context_switch_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.context_switch_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_context_switch_event__doc, .tp_members = pyrf_context_switch_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_context_switch_event__repr, };static int pyrf_event__setup_types (void ) {int err; - pyrf_mmap_event__type.tp_new = - pyrf_task_event__type.tp_new = - pyrf_comm_event__type.tp_new = - pyrf_lost_event__type.tp_new = - pyrf_read_event__type.tp_new = - pyrf_sample_event__type.tp_new = - pyrf_context_switch_event__type.tp_new = - pyrf_throttle_event__type.tp_new = PyType_GenericNew; - - pyrf_sample_event__type.tp_dealloc = (destructor )pyrf_sample_event__delete, err =PyType_Ready (&pyrf_mmap_event__type);if (err <0 )
@@ -509,33 +838,175 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, }; -static PyObject *pyrf_event__new (const union perf_event *event) +static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel, +struct perf_session *session__maybe_unused ) {struct pyrf_event *pevent; -PyTypeObject *ptype; +size_t size; +int err; +size_t min_size =sizeof (struct perf_event_header ); -if ((event->header.type <PERF_RECORD_MMAP || - event->header.type >PERF_RECORD_SAMPLE ) && - !(event->header.type ==PERF_RECORD_SWITCH || - event->header.type ==PERF_RECORD_SWITCH_CPU_WIDE )) { -PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , +if (event->header.type >=ARRAY_SIZE (pyrf_event__type) || + pyrf_event__type[event->header.type] ==NULL ) { +return PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , event->header.type); -return NULL ; } -// FIXME this better be dynamic or we need to parse everything -// before calling perf_mmap__consume(), including tracepoint fields. -if (sizeof (pevent->event) < event->header.size) { -PyErr_Format (PyExc_TypeError,"Unexpected event size: %zd < %u" , -sizeof (pevent->event), event->header.size); -return NULL ; +switch (event->header.type) { +case PERF_RECORD_MMAP : + min_size = offsetof(struct perf_record_mmap , filename) +1 ; +break ; +case PERF_RECORD_MMAP2 : + min_size = offsetof(struct perf_record_mmap2 , filename) +1 ; +break ; +case PERF_RECORD_COMM : + min_size =sizeof (struct perf_record_comm ); +break ; +case PERF_RECORD_FORK : +case PERF_RECORD_EXIT : + min_size =sizeof (struct perf_record_fork ); +break ; +case PERF_RECORD_THROTTLE : +case PERF_RECORD_UNTHROTTLE : + min_size =sizeof (struct perf_record_throttle ); +break ; +case PERF_RECORD_LOST : + min_size =sizeof (struct perf_record_lost ); +break ; +case PERF_RECORD_READ : + min_size =sizeof (struct perf_record_read ); +break ; +case PERF_RECORD_SWITCH : +case PERF_RECORD_SWITCH_CPU_WIDE : + min_size =sizeof (struct perf_record_switch ); +break ; +case PERF_RECORD_AUX : + min_size =sizeof (struct perf_record_aux ); +break ; +case PERF_RECORD_ITRACE_START : + min_size =sizeof (struct perf_record_itrace_start ); +break ; +case PERF_RECORD_LOST_SAMPLES : + min_size =sizeof (struct perf_record_lost_samples ); +break ; +case PERF_RECORD_NAMESPACES : + min_size = offsetof(struct perf_record_namespaces , link_info); +break ; +case PERF_RECORD_KSYMBOL : + min_size =sizeof (struct perf_record_ksymbol ); +break ; +case PERF_RECORD_BPF_EVENT : + min_size =sizeof (struct perf_record_bpf_event ); +break ; +case PERF_RECORD_CGROUP : + min_size =sizeof (struct perf_record_cgroup ); +break ; +case PERF_RECORD_TEXT_POKE : + min_size = offsetof(struct perf_record_text_poke_event , bytes); +break ; +case PERF_RECORD_AUX_OUTPUT_HW_ID : + min_size =sizeof (struct perf_record_aux_output_hw_id ); +break ; +case PERF_RECORD_CALLCHAIN_DEFERRED : + min_size =sizeof (struct perf_record_callchain_deferred ); +break ; +case PERF_RECORD_HEADER_ATTR : + min_size =sizeof (struct perf_record_header_attr ); +break ; +case PERF_RECORD_HEADER_TRACING_DATA : + min_size =sizeof (struct perf_record_header_tracing_data ); +break ; +case PERF_RECORD_HEADER_BUILD_ID : + min_size = offsetof(struct perf_record_header_build_id , filename) +1 ; +break ; +case PERF_RECORD_ID_INDEX : + min_size = offsetof(struct perf_record_id_index , entries); +break ; +case PERF_RECORD_AUXTRACE_INFO : + min_size = offsetof(struct perf_record_auxtrace_info , priv); +break ; +case PERF_RECORD_AUXTRACE : + min_size =sizeof (struct perf_record_auxtrace ); +break ; +case PERF_RECORD_AUXTRACE_ERROR : + min_size =sizeof (struct perf_record_auxtrace_error ); +break ; +case PERF_RECORD_THREAD_MAP : + min_size = offsetof(struct perf_record_thread_map , entries); +break ; +case PERF_RECORD_CPU_MAP : + min_size =sizeof (struct perf_record_cpu_map ); +break ; +case PERF_RECORD_STAT_CONFIG : + min_size = offsetof(struct perf_record_stat_config , data); +break ; +case PERF_RECORD_STAT : + min_size =sizeof (struct perf_record_stat ); +break ; +case PERF_RECORD_STAT_ROUND : + min_size =sizeof (struct perf_record_stat_round ); +break ; +case PERF_RECORD_EVENT_UPDATE : + min_size =sizeof (struct perf_record_event_update ); +break ; +case PERF_RECORD_TIME_CONV : + min_size =sizeof (struct perf_record_time_conv ); +break ; +case PERF_RECORD_HEADER_FEATURE : + min_size = offsetof(struct perf_record_header_feature , data); +break ; +case PERF_RECORD_COMPRESSED : + min_size = offsetof(struct perf_record_compressed , data); +break ; +case PERF_RECORD_COMPRESSED2 : + min_size = offsetof(struct perf_record_compressed2 , data); +break ; +case PERF_RECORD_BPF_METADATA : + min_size = offsetof(struct perf_record_bpf_metadata , entries); +break ; +case PERF_RECORD_SCHEDSTAT_CPU : + min_size =sizeof (struct perf_record_schedstat_cpu ); +break ; +case PERF_RECORD_SCHEDSTAT_DOMAIN : + min_size =sizeof (struct perf_record_schedstat_domain ); +break ; +default : +break ; + } +if (event->header.size < min_size) +return PyErr_Format (PyExc_ValueError,"Event size %u too small for type %u" , + event->header.size, event->header.type); + +/* Allocate just enough memory for the size of event. */ + size = offsetof(struct pyrf_event , event) + event->header.size; + pevent = (struct pyrf_event *)PyObject_Malloc (size); +if (pevent ==NULL ) +return PyErr_NoMemory (); + +/* Copy the event for memory safety and initilaize variables. */ +PyObject_Init ((PyObject *)pevent, pyrf_event__type[event->header.type]); +memcpy (&pevent->event, event, event->header.size); + +if (event->header.type ==PERF_RECORD_MMAP ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap , filename); + + pevent->event.mmap.filename[max_len -1 ] ='\0' ; } - ptype = pyrf_event__type[event->header.type]; - pevent = PyObject_New(struct pyrf_event, ptype); -if (pevent !=NULL ) { -memcpy (&pevent->event, event, event->header.size); -perf_sample__init (&pevent->sample,/*all=*/ false); +perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->al_resolved = false; +addr_location__init (&pevent->al); + +if (!evsel) +return (PyObject *)pevent; + +/* Parse the sample again so that pointers are within the copied event. */ + err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample); +if (err <0 ) { +Py_DECREF (pevent); +return PyErr_Format (PyExc_OSError, +"perf: can't parse sample, err=%d" , err); }return (PyObject *)pevent; }
@@ -1240,7 +1711,7 @@ static PyObject *pyrf_evsel__str(PyObject *self)if (!evsel)return PyUnicode_FromString ("evsel(uninitialized)" ); -return PyUnicode_FromFormat ("evsel(%s/%s/)" ,evsel__pmu_name (evsel),evsel__name (evsel)); +return PyUnicode_FromFormat ("evsel(%s)" ,evsel__name (evsel)); }static PyMethodDef pyrf_evsel__methods[] = {
@@ -1868,9 +2339,11 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist, {struct evlist *evlist;union perf_event *event; +struct evsel *evsel;int sample_id_all =1 , cpu;static char *kwlist[] = {"cpu" ,"sample_id_all" ,NULL };struct mmap *md; +PyObject *pyevent;int err;CHECK_INITIALIZED (pevlist->evlist,"evlist" );
@@ -1881,44 +2354,31 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,return NULL ; md =get_md (evlist, cpu); -if (!md) { -PyErr_Format (PyExc_TypeError,"Unknown CPU '%d'" , cpu); -return NULL ; - } +if (!md) +return PyErr_Format (PyExc_TypeError,"Unknown CPU '%d'" , cpu); -if (perf_mmap__read_init (&md->core) <0 ) - goto end; + err =perf_mmap__read_init (&md->core); +if (err <0 ) { +return PyErr_Format (PyExc_OSError, +"perf: error mmap read init, err=%d" , err); + } event =perf_mmap__read_event (&md->core); -if (event !=NULL ) { -PyObject *pyevent =pyrf_event__new (event); -struct pyrf_event *pevent = (struct pyrf_event *)pyevent; -struct evsel *evsel; - -if (pyevent ==NULL ) -return PyErr_NoMemory (); - - evsel =evlist__event2evsel (evlist, event); -if (!evsel) { -Py_DECREF (pyevent); -Py_INCREF (Py_None); -return Py_None; - } +if (event ==NULL ) + Py_RETURN_NONE; + evsel =evlist__event2evsel (evlist, event); +if (!evsel) { +/* Unknown evsel. */ perf_mmap__consume (&md->core); - - err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample); -if (err) { -Py_DECREF (pyevent); -return PyErr_Format (PyExc_OSError, -"perf: can't parse sample, err=%d" , err); - } - -return pyevent; + Py_RETURN_NONE; } -end: -Py_INCREF (Py_None); -return Py_None; + pyevent =pyrf_event__new (event, evsel,evlist__session (evlist)); +perf_mmap__consume (&md->core); +if (pyevent ==NULL ) +return PyErr_Occurred () ?NULL :PyErr_NoMemory (); + +return pyevent; } static PyObject *pyrf_evlist__open(struct pyrf_evlist *pevlist,
@@ -2130,10 +2590,7 @@ static PyObject *pyrf_evlist__str(PyObject *self)evlist__for_each_entry (pevlist->evlist, pos) {if (!first)strbuf_addch (&sb,',' ); -if (!pos->pmu) -strbuf_addstr (&sb,evsel__name (pos)); -else -strbuf_addf (&sb,"%s/%s/" , pos->pmu->name,evsel__name (pos)); +strbuf_addstr (&sb,evsel__name (pos)); first = false; }strbuf_addstr (&sb,"])" );
@@ -2681,19 +3138,12 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool, struct machine *machine __maybe_unused) { struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); -PyObject *pyevent =pyrf_event__new (event); -struct pyrf_event *pevent = (struct pyrf_event *)pyevent; +PyObject *pyevent =pyrf_event__new (event, sample->evsel, psession->session);PyObject *ret;if (pyevent ==NULL )return -ENOMEM ; -memcpy (&pevent->event, event, event->header.size); -if (evsel__parse_sample (sample->evsel, &pevent->event, &pevent->sample) <0 ) { -Py_DECREF (pyevent); -return -1 ; - } - ret =PyObject_CallFunction (psession->sample,"O" , pyevent);if (!ret) {Py_DECREF (pyevent);
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 11/19] perf python: Refactor and add accessors to sample event
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [High] Missing null termination enforcement for PERF_RECORD_COMM events combined with tight dynamic…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] Missing null termination enforcement for PERF_RECORD_COMM events combined with tight dynamic memory allocation leads to an out-of-bounds heap read.
- [High] Dynamic memory copy into statically-sized union member triggers _FORTIFY_SOURCE abort for large perf events.
--
commit 988867a783dddfdaf9aaad9f7c6cf721f3891d95
Author: Ian Rogers <[email protected]>
perf python: Refactor and add accessors to sample event
This commit refactors the Python perf event integration by adding common evsel fields for events and specific accessors for sample events. It also adjusts memory allocation to allocate just enough memory for the copied event, rather than always allocating the maximum event size.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 12 lines
struct
pyrf_event{tools/perf/util/python.c +7
-1PyObject_HEAD +/** @sample: The parsed sample from the event. */ struct perf_sample sample; -union perf_event event; +/** @al: The address location from machine__resolve, lazily computed. */ +struct addr_location al; +/** @al_resolved: True when machine__resolve been called. */ +bool al_resolved; +/** @event: The underlying perf_event that may be in a file or ring buffer. */ +union perf_event event; };
[ ... ]
Quoted 28 lines
tools/perf/util/python.c +8
-14-static PyObject *pyrf_event__new (const union perf_event *event) +static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel, +struct perf_session *session__maybe_unused ) {struct pyrf_event *pevent; -PyTypeObject *ptype; +size_t size; +int err; +size_t min_size =sizeof (struct perf_event_header ); -if ((event->header.type <PERF_RECORD_MMAP || - event->header.type >PERF_RECORD_SAMPLE ) && - !(event->header.type ==PERF_RECORD_SWITCH || - event->header.type ==PERF_RECORD_SWITCH_CPU_WIDE )) { -PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , +if (event->header.type >=ARRAY_SIZE (pyrf_event__type) || + pyrf_event__type[event->header.type] ==NULL ) { +return PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , event->header.type); -return NULL ; } -// FIXME this better be dynamic or we need to parse everything -// before calling perf_mmap__consume(), including tracepoint fields. -if (sizeof (pevent->event) < event->header.size) { -PyErr_Format (PyExc_TypeError,"Unexpected event size: %zd < %u" , -sizeof (pevent->event), event->header.size); -return NULL ;
[ ... ]
tools/perf/util/python.c static PyTypeObject *pyrf_event__type[] = { · line 988
Quoted 9 lines
tools/perf/util/python.c +9
-0+/* Allocate just enough memory for the size of event. */ + size = offsetof(struct pyrf_event , event) + event->header.size; + pevent = (struct pyrf_event *)PyObject_Malloc (size); +if (pevent ==NULL ) +return PyErr_NoMemory (); + +/* Copy the event for memory safety and initilaize variables. */ +PyObject_Init ((PyObject *)pevent, pyrf_event__type[event->header.type]); +memcpy (&pevent->event, event, event->header.size);
[Severity: High]
Does copying dynamically sized events into a fixed-size union member trigger a _FORTIFY_SOURCE abort?
Since the previous upper bounds check (sizeof(pevent->event) < event->header.size) was removed, event->header.size can now safely exceed the size of the union. However, because pevent->event remains strictly typed as a fixed-size union perf_event rather than a flexible array member, compilers building with -D_FORTIFY_SOURCE=2 will evaluate the maximum allowed destination size as sizeof(union perf_event).
When memcpy executes for large events, will the fortified memcpy fail this bounds check and abort the process?
tools/perf/util/python.c static PyTypeObject *pyrf_event__type[] = { · line 995
Quoted 7 lines
tools/perf/util/python.c +6
-0+ +if (event->header.type ==PERF_RECORD_MMAP ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap , filename); + + pevent->event.mmap.filename[max_len -1 ] ='\0' ; }
[Severity: High]
Is a similar string termination safeguard needed for PERF_RECORD_COMM events?
The patch updates memory allocation to exactly fit the record's length, leaving no safety margin or uninitialized buffer space. For PERF_RECORD_COMM events, the comm field is exported to Python using T_STRING_INPLACE, which relies on strlen() to parse the string length.
If a malicious or malformed perf.data file provides a comm string lacking a null terminator, will strlen() read past the end of the tight heap allocation, resulting in an out-of-bounds read?
Quoted 22 lines
tools/perf/util/python.c +13
-5- ptype = pyrf_event__type[event->header.type]; - pevent = PyObject_New(struct pyrf_event, ptype); -if (pevent !=NULL ) { -memcpy (&pevent->event, event, event->header.size); -perf_sample__init (&pevent->sample,/*all=*/ false); +perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->al_resolved = false; +addr_location__init (&pevent->al); + +if (!evsel) +return (PyObject *)pevent; + +/* Parse the sample again so that pointers are within the copied event. */ + err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample); +if (err <0 ) { +Py_DECREF (pevent); +return PyErr_Format (PyExc_OSError, +"perf: can't parse sample, err=%d" , err); }return (PyObject *)pevent; }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=11
Ian Rogers
reviewerPATCH v10 12/19 perf python: Add mmap2 event
In reply to Ian RogersIf mmap is handled so should mmap2 events. Add support as a distinct python event type.
--- tools/perf/util/python.c | 156 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 155 insertions(+), 1 deletion(-)
tools/perf/util/python.c +155 -1
@@ -131,7 +131,7 @@ static const char pyrf_mmap_event__doc[] = PyDoc_STR("perf mmap event object.");static PyMemberDef pyrf_mmap_event__members[] = { sample_membersmember_def (perf_event_header, type,T_UINT ,"event type" ), -member_def (perf_event_header, misc,T_UINT ,"event misc" ), +member_def (perf_event_header, misc,T_USHORT ,"event misc" ),member_def (perf_record_mmap, pid,T_UINT ,"event pid" ),member_def (perf_record_mmap, tid,T_UINT ,"event tid" ),member_def (perf_record_mmap, start,T_ULONGLONG ,"start of the map" ),
@@ -172,6 +172,148 @@ static PyTypeObject pyrf_mmap_event__type = { .tp_repr = (reprfunc)pyrf_mmap_event__repr, }; +static const char pyrf_mmap2_event__doc[] =PyDoc_STR ("perf mmap2 event object." ); + +static PyObject *pyrf_mmap2_event__get_maj (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (pevent->event.mmap2.maj); +} + +static PyObject *pyrf_mmap2_event__get_min (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (pevent->event.mmap2.min); +} + +static PyObject *pyrf_mmap2_event__get_ino (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->event.mmap2.ino); +} + +static PyObject *pyrf_mmap2_event__get_ino_generation (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->event.mmap2.ino_generation); +} + +static PyObject *pyrf_mmap2_event__get_build_id (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!(pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID )) + Py_RETURN_NONE; + +int size = pevent->event.mmap2.build_id_size; +if (size >20 ) + size =20 ; + +return PyBytes_FromStringAndSize ((const char *)pevent->event.mmap2.build_id, size); +} + +static PyGetSetDef pyrf_mmap2_event__getset[] = { + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { + .name ="maj" , + .get = pyrf_mmap2_event__get_maj, + .set =NULL , + .doc ="major number." , + }, + { + .name ="min" , + .get = pyrf_mmap2_event__get_min, + .set =NULL , + .doc ="minor number." , + }, + { + .name ="ino" , + .get = pyrf_mmap2_event__get_ino, + .set =NULL , + .doc ="inode number." , + }, + { + .name ="ino_generation" , + .get = pyrf_mmap2_event__get_ino_generation, + .set =NULL , + .doc ="inode generation." , + }, + { + .name ="build_id" , + .get = pyrf_mmap2_event__get_build_id, + .set =NULL , + .doc ="binary build ID." , + }, + { .name =NULL , }, +}; + +static PyMemberDef pyrf_mmap2_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), +member_def (perf_event_header, misc,T_USHORT ,"event misc" ), +member_def (perf_record_mmap2, pid,T_UINT ,"event pid" ), +member_def (perf_record_mmap2, tid,T_UINT ,"event tid" ), +member_def (perf_record_mmap2, start,T_ULONGLONG ,"start of the map" ), +member_def (perf_record_mmap2, len,T_ULONGLONG ,"map length" ), +member_def (perf_record_mmap2, pgoff,T_ULONGLONG ,"page offset" ), +member_def (perf_record_mmap2, prot,T_UINT ,"protection" ), +member_def (perf_record_mmap2, flags,T_UINT ,"flags" ), +member_def (perf_record_mmap2, filename,T_STRING_INPLACE ,"backing store" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_mmap2_event__repr (const struct pyrf_event *pevent) +{ +PyObject *ret; +char *s; + +if (asprintf (&s,"{ type: mmap2, pid: %u, tid: %u, start: %#" PRI_lx64", " +"length: %#" PRI_lx64", offset: %#" PRI_lx64", " +"flags: %#x, prot: %#x, filename: %s }" , + pevent->event.mmap2.pid, pevent->event.mmap2.tid, + pevent->event.mmap2.start, pevent->event.mmap2.len, + pevent->event.mmap2.pgoff, pevent->event.mmap2.flags, + pevent->event.mmap2.prot, pevent->event.mmap2.filename) <0 ) +return PyErr_NoMemory (); + + ret =PyUnicode_FromString (s); +free (s); +return ret; +} + +static PyTypeObject pyrf_mmap2_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.mmap2_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_mmap2_event__doc, + .tp_members = pyrf_mmap2_event__members, + .tp_getset = pyrf_mmap2_event__getset, + .tp_repr = (reprfunc )pyrf_mmap2_event__repr, +}; +static const char pyrf_task_event__doc[] =PyDoc_STR ("perf task (fork/exit) event object." );static PyMemberDef pyrf_task_event__members[] = {
@@ -797,6 +939,9 @@ static int pyrf_event__setup_types(void)int err; err =PyType_Ready (&pyrf_mmap_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_mmap2_event__type);if (err <0 ) goto out; err =PyType_Ready (&pyrf_lost_event__type);
@@ -826,6 +971,7 @@ static int pyrf_event__setup_types(void)static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_MMAP ] = &pyrf_mmap_event__type, + [PERF_RECORD_MMAP2 ] = &pyrf_mmap2_event__type, [PERF_RECORD_LOST ] = &pyrf_lost_event__type, [PERF_RECORD_COMM ] = &pyrf_comm_event__type, [PERF_RECORD_EXIT ] = &pyrf_task_event__type,
@@ -992,6 +1138,11 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evsize_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap , filename); pevent->event.mmap.filename[max_len -1 ] ='\0' ; + }else if (event->header.type ==PERF_RECORD_MMAP2 ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap2 , filename); + + pevent->event.mmap2.filename[max_len -1 ] ='\0' ; }perf_sample__init (&pevent->sample,/*all=*/ true);
@@ -3417,6 +3568,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_mmap_event__type);PyModule_AddObject (module,"mmap_event" , (PyObject *)&pyrf_mmap_event__type); +Py_INCREF (&pyrf_mmap2_event__type); +PyModule_AddObject (module,"mmap2_event" , (PyObject *)&pyrf_mmap2_event__type); +Py_INCREF (&pyrf_lost_event__type);PyModule_AddObject (module,"lost_event" , (PyObject *)&pyrf_lost_event__type);
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 13/19 perf python: Add callchain support
In reply to Ian RogersImplement pyrf_callchain_node and pyrf_callchain types for lazy iteration over callchain frames. Add callchain property to sample_event.
---
v2:
1. Eager Callchain Resolution: Moved the callchain resolution from deferred iteration to eager processing in pyrf_session_tool__sample() . This avoids risks of reading from unmapped memory or following dangling pointers to closed sessions.
2. Cached Callchain: Added a callchain field to struct pyrf_event to store the resolved object.
3. Simplified Access: pyrf_sample_event__get_callchain() now just returns the cached object if available.
4. Avoided Double Free: Handled lazy cleanups properly.
v6:
- Moved callchain resolution from session_tool__sample to pyrf_event__new.
--- tools/perf/util/python.c | 203 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 202 insertions(+), 1 deletion(-)
tools/perf/util/python.c +202 -1
@@ -85,6 +85,8 @@ struct pyrf_event {struct addr_location al;/** @al_resolved: True when machine__resolve been called. */ bool al_resolved; +/** @callchain: Resolved callchain, eagerly computed if requested. */ +PyObject *callchain;/** @event: The underlying perf_event that may be in a file or ring buffer. */ union perf_event event; };
@@ -122,6 +124,7 @@ static void pyrf_event__delete(struct pyrf_event *pevent) {if (pevent->al_resolved)addr_location__exit (&pevent->al); +Py_XDECREF (pevent->callchain);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -782,6 +785,142 @@ static PyObject *pyrf_sample_event__insn(PyObject *self, PyObject *args __maybe_ pevent->sample.insn_len); } +struct pyrf_callchain_node { +PyObject_HEAD + u64 ip; +struct map *map; +struct symbol *sym; +}; + +static void pyrf_callchain_node__delete (struct pyrf_callchain_node *pnode) +{ +map__put (pnode->map); +Py_TYPE (pnode)->tp_free ((PyObject *)pnode); +} + +static PyObject *pyrf_callchain_node__get_ip (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pnode->ip); +} + +static PyObject *pyrf_callchain_node__get_symbol (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +if (pnode->sym) +return PyUnicode_FromString (pnode->sym->name); +return PyUnicode_FromString ("[unknown]" ); +} + +static PyObject *pyrf_callchain_node__get_dso (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +const char *dsoname ="[unknown]" ; + +if (pnode->map) { +struct dso *dso =map__dso (pnode->map); +if (dso) { +if (symbol_conf.show_kernel_path &&dso__long_name (dso)) + dsoname =dso__long_name (dso); +else + dsoname =dso__name (dso); + } + } +return PyUnicode_FromString (dsoname); +} + +static PyGetSetDef pyrf_callchain_node__getset[] = { + { .name ="ip" , .get = (getter )pyrf_callchain_node__get_ip, }, + { .name ="symbol" , .get = (getter )pyrf_callchain_node__get_symbol, }, + { .name ="dso" , .get = (getter )pyrf_callchain_node__get_dso, }, + { .name =NULL , }, +}; + +static PyTypeObject pyrf_callchain_node__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.callchain_node" , + .tp_basicsize =sizeof (struct pyrf_callchain_node ), + .tp_dealloc = (destructor )pyrf_callchain_node__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf callchain node object." , + .tp_getset = pyrf_callchain_node__getset, +}; + +struct pyrf_callchain_frame { +u64 ip; +struct map *map; +struct symbol *sym; +}; + +struct pyrf_callchain { +PyObject_HEAD + struct pyrf_callchain_frame *frames; +u64 nr_frames; +}; + +static void pyrf_callchain__delete (struct pyrf_callchain *pchain) +{ +if (pchain->frames) { +for (u64 i =0 ; i < pchain->nr_frames; i++) +map__put (pchain->frames[i].map); +free (pchain->frames); + } +Py_TYPE (pchain)->tp_free ((PyObject *)pchain); +} + +static Py_ssize_t pyrf_callchain__length (PyObject *obj) +{ +struct pyrf_callchain *pchain = (void *)obj; +return pchain->nr_frames; +} + +static PyObject *pyrf_callchain__item (PyObject *obj,Py_ssize_t i) +{ +struct pyrf_callchain *pchain = (void *)obj; +struct pyrf_callchain_node *pnode; + +if (i <0 || i >= (Py_ssize_t )pchain->nr_frames) { +PyErr_SetString (PyExc_IndexError,"Index out of range" ); +return NULL ; + } + + pnode = PyObject_New(struct pyrf_callchain_node, &pyrf_callchain_node__type); +if (!pnode) +return NULL ; + + pnode->ip = pchain->frames[i].ip; + pnode->map =map__get (pchain->frames[i].map); + pnode->sym = pchain->frames[i].sym; + +return (PyObject *)pnode; +} + +static PySequenceMethods pyrf_callchain__sequence_methods = { + .sq_length = pyrf_callchain__length, + .sq_item = pyrf_callchain__item, +}; + +static PyTypeObject pyrf_callchain__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.callchain" , + .tp_basicsize =sizeof (struct pyrf_callchain ), + .tp_dealloc = (destructor )pyrf_callchain__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf callchain object." , + .tp_as_sequence = &pyrf_callchain__sequence_methods, +}; + +static PyObject *pyrf_sample_event__get_callchain (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->callchain) + Py_RETURN_NONE; + +Py_INCREF (pevent->callchain); +return pevent->callchain; +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -796,6 +935,12 @@ pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name) }static PyGetSetDef pyrf_sample_event__getset[] = { + { + .name ="callchain" , + .get = pyrf_sample_event__get_callchain, + .set =NULL , + .doc ="event callchain." , + }, { .name ="raw_buf" , .get = (getter )pyrf_sample_event__get_raw_buf,
@@ -965,6 +1110,12 @@ static int pyrf_event__setup_types(void) err =PyType_Ready (&pyrf_context_switch_event__type);if (err <0 ) goto out; + err =PyType_Ready (&pyrf_callchain_node__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_callchain__type); +if (err <0 ) + goto out; out:return err; }
@@ -985,9 +1136,11 @@ static PyTypeObject *pyrf_event__type[] = { };static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel, -struct perf_session *session__maybe_unused ) +struct perf_session *session) {struct pyrf_event *pevent; +struct perf_sample *sample; +struct machine *machine = session ? &session->machines.host :NULL ;size_t size;int err;size_t min_size =sizeof (struct perf_event_header );
@@ -1146,6 +1299,7 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev }perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->callchain =NULL ; pevent->al_resolved = false;addr_location__init (&pevent->al);
@@ -1159,6 +1313,50 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evreturn PyErr_Format (PyExc_OSError,"perf: can't parse sample, err=%d" , err); } + sample = &pevent->sample; +if (machine && sample->callchain) { +struct addr_location al; +struct callchain_cursor *cursor; +u64 i; +struct pyrf_callchain *pchain; + +addr_location__init (&al); +if (machine__resolve (machine, &al, sample) >=0 ) { + cursor =get_tls_callchain_cursor (); +if (thread__resolve_callchain (al.thread, cursor, sample, +NULL ,NULL ,PERF_MAX_STACK_DEPTH ) ==0 ) { +callchain_cursor_commit (cursor); + + pchain = PyObject_New(struct pyrf_callchain, &pyrf_callchain__type); +if (!pchain) { +addr_location__exit (&al); +Py_DECREF (pevent); +return NULL ; + } + pchain->nr_frames = cursor->nr; + pchain->frames =calloc (pchain->nr_frames, +sizeof (*pchain->frames)); +if (!pchain->frames) { +Py_DECREF (pchain); +addr_location__exit (&al); +Py_DECREF (pevent); +return PyErr_NoMemory (); + } +struct callchain_cursor_node *node; + +for (i =0 ; i < pchain->nr_frames; i++) { + node =callchain_cursor_current (cursor); + pchain->frames[i].ip = node->ip; + pchain->frames[i].map = +map__get (node->ms.map); + pchain->frames[i].sym = node->ms.sym; +callchain_cursor_advance (cursor); + } + pevent->callchain = (PyObject *)pchain; + } +addr_location__exit (&al); + } + }return (PyObject *)pevent; }
@@ -3394,6 +3592,9 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject } psession->session = session; + symbol_conf.use_callchain = true; + symbol_conf.show_kernel_path = true; + symbol_conf.inline_name = false;if (symbol__init (perf_session__env (session)) <0 ) {PyErr_SetString (PyExc_OSError,"perf: symbol__init failed" ); goto err_out;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 14/19 perf python: Extend API for stat events in python.c
In reply to Ian RogersAdd stat information to the session. Add call backs for stat events.
---
v5:
1. Fix Memory Corruption: Corrected the memory offset for stat_round_type in pyrf_stat_round_event__members by adding the base offset of struct pyrf_event.
2. Fix Memory Leak: Added Py_XDECREF() to free the unused return value of the Python callback in pyrf_session_tool__stat_round().
---
v7:
- Added comprehensive size checks in pyrf_event__new for all event types.
- Fixed line length warning.
v10:
- Added CHECK_INITIALIZED check to pyrf_evsel__get_ids to prevent NULL pointer dereference on uninitialized evsel objects.
--- tools/perf/util/python.c | 191 +++++++++++++++++++++++++++++++++++++-- 1 file changed, 184 insertions(+), 7 deletions(-)
tools/perf/util/python.c +184 -7
@@ -456,6 +456,77 @@ static PyTypeObject pyrf_lost_event__type = { .tp_repr = (reprfunc)pyrf_lost_event__repr, }; +static const char pyrf_stat_event__doc[] =PyDoc_STR ("perf stat event object." ); + +static PyMemberDef pyrf_stat_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), +member_def (perf_record_stat, id,T_ULONGLONG ,"event id" ), +member_def (perf_record_stat, cpu,T_UINT ,"event cpu" ), +member_def (perf_record_stat, thread,T_UINT ,"event thread" ), +member_def (perf_record_stat, val,T_ULONGLONG ,"counter value" ), +member_def (perf_record_stat, ena,T_ULONGLONG ,"enabled time" ), +member_def (perf_record_stat, run,T_ULONGLONG ,"running time" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_stat_event__repr (const struct pyrf_event *pevent) +{ +return PyUnicode_FromFormat ( +"{ type: stat, id: %llu, cpu: %u, thread: %u, val: %llu, ena: %llu, run: %llu }" , + pevent->event.stat.id, + pevent->event.stat.cpu, + pevent->event.stat.thread, + pevent->event.stat.val, + pevent->event.stat.ena, + pevent->event.stat.run); +} + +static PyTypeObject pyrf_stat_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.stat_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_new = PyType_GenericNew, + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_stat_event__doc, + .tp_members = pyrf_stat_event__members, + .tp_getset = pyrf_event__getset, + .tp_repr = (reprfunc )pyrf_stat_event__repr, +}; + +static const char pyrf_stat_round_event__doc[] =PyDoc_STR ("perf stat round event object." ); + +static PyMemberDef pyrf_stat_round_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), + { .name ="stat_round_type" , .type =T_ULONGLONG , + .offset = offsetof(struct pyrf_event , event) + offsetof(struct perf_record_stat_round , type), + .doc ="round type" }, +member_def (perf_record_stat_round, time,T_ULONGLONG ,"round time" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_stat_round_event__repr (const struct pyrf_event *pevent) +{ +return PyUnicode_FromFormat ("{ type: stat_round, type: %llu, time: %llu }" , + pevent->event.stat_round.type, + pevent->event.stat_round.time); +} + +static PyTypeObject pyrf_stat_round_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.stat_round_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_new = PyType_GenericNew, + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_stat_round_event__doc, + .tp_members = pyrf_stat_round_event__members, + .tp_getset = pyrf_event__getset, + .tp_repr = (reprfunc )pyrf_stat_round_event__repr, +}; +static const char pyrf_read_event__doc[] =PyDoc_STR ("perf read event object." );static PyMemberDef pyrf_read_event__members[] = {
@@ -1108,6 +1179,12 @@ static int pyrf_event__setup_types(void)if (err <0 ) goto out; err =PyType_Ready (&pyrf_context_switch_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_round_event__type);if (err <0 ) goto out; err =PyType_Ready (&pyrf_callchain_node__type);
@@ -1133,6 +1210,8 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SAMPLE] = &pyrf_sample_event__type, [PERF_RECORD_SWITCH] = &pyrf_context_switch_event__type, [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, + [PERF_RECORD_STAT] = &pyrf_stat_event__type, + [PERF_RECORD_STAT_ROUND] = &pyrf_stat_round_event__type, }; static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evsel,
@@ -1165,13 +1244,13 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evcase PERF_RECORD_EXIT : min_size =sizeof (struct perf_record_fork );break ; +case PERF_RECORD_LOST : + min_size =sizeof (struct perf_record_lost ); +break ;case PERF_RECORD_THROTTLE :case PERF_RECORD_UNTHROTTLE : min_size =sizeof (struct perf_record_throttle );break ; -case PERF_RECORD_LOST : - min_size =sizeof (struct perf_record_lost ); -break ;case PERF_RECORD_READ : min_size =sizeof (struct perf_record_read );break ;
@@ -2236,7 +2315,45 @@ static PyObject *pyrf_evsel__get_attr_wakeup_events(PyObject *self, void */*closreturn PyLong_FromUnsignedLong (pevsel->evsel->core.attr.wakeup_events); } +static PyObject *pyrf_evsel__get_ids (struct pyrf_evsel *pevsel,void *closure__maybe_unused ) +{ +struct evsel *evsel; +PyObject *list; + +CHECK_INITIALIZED (pevsel->evsel,"evsel" ); + + evsel = pevsel->evsel; + list =PyList_New (0 ); + +if (!list) +return NULL ; + +for (u32 i =0 ; i < evsel->core.ids; i++) { +PyObject *id =PyLong_FromUnsignedLongLong (evsel->core.id[i]); +int ret; + +if (!id) { +Py_DECREF (list); +return NULL ; + } + ret =PyList_Append (list, id); +Py_DECREF (id); +if (ret <0 ) { +Py_DECREF (list); +return NULL ; + } + } + +return list; +} +static PyGetSetDef pyrf_evsel__getset[] = { + { + .name ="ids" , + .get = (getter )pyrf_evsel__get_ids, + .set =NULL , + .doc ="event IDs." , + }, { .name ="tracking" , .get = pyrf_evsel__get_tracking,
@@ -3085,6 +3202,8 @@ static const struct perf_constant perf__constants[] = {PERF_CONST (RECORD_LOST_SAMPLES ),PERF_CONST (RECORD_SWITCH ),PERF_CONST (RECORD_SWITCH_CPU_WIDE ), +PERF_CONST (RECORD_STAT ), +PERF_CONST (RECORD_STAT_ROUND ),PERF_CONST (RECORD_MISC_SWITCH_OUT ), { .name = NULL, },
@@ -3479,6 +3598,7 @@ struct pyrf_session {struct perf_tool tool;struct pyrf_data *pdata;PyObject *sample; +PyObject *stat; }; static int pyrf_session_tool__sample(const struct perf_tool *tool,
@@ -3503,6 +3623,50 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool,return 0 ; } +static int pyrf_session_tool__stat (const struct perf_tool *tool, +struct perf_session *session, +union perf_event *event) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +struct evsel *evsel =evlist__id2evsel (session->evlist, event->stat.id); +PyObject *pyevent =pyrf_event__new (event, evsel, psession->session); +const char *name = evsel ?evsel__name (evsel) :"unknown" ; +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"Oz" , pyevent, name); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} + +static int pyrf_session_tool__stat_round (const struct perf_tool *tool, +struct perf_session *session__maybe_unused , +union perf_event *event) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +PyObject *pyevent =pyrf_event__new (event,/*evsel=*/ NULL , psession->session); +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"Oz" , pyevent,NULL ); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} +static PyObject *pyrf_session__find_thread (struct pyrf_session *psession,PyObject *args) {struct machine *machine;
@@ -3536,13 +3700,13 @@ static PyObject *pyrf_session__find_thread(struct pyrf_session *psession, PyObjestatic PyObject *pyrf_session__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) {struct pyrf_data *pdata; -PyObject *sample =NULL ; -static char *kwlist[] = {"data" ,"sample" ,NULL }; +PyObject *sample =NULL , *stat =NULL ; +static char *kwlist[] = {"data" ,"sample" ,"stat" ,NULL };struct pyrf_session *psession;struct perf_session *session; -if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|O" , kwlist, &pyrf_data__type, &pdata, - &sample)) +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|OO" , kwlist, &pyrf_data__type, &pdata, + &sample, &stat))return NULL ; psession = PyObject_New(struct pyrf_session, type);
@@ -3551,6 +3715,7 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject psession->session =NULL ; psession->sample =NULL ; + psession->stat =NULL ; psession->pdata =NULL ;Py_INCREF (pdata);
@@ -3573,8 +3738,13 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject }while (0 )ADD_TOOL (sample); +ADD_TOOL (stat);#undef ADD_TOOL +if (stat) + psession->tool.stat_round = pyrf_session_tool__stat_round; + + psession->tool.comm = perf_event__process_comm; psession->tool.mmap = perf_event__process_mmap; psession->tool.mmap2 = perf_event__process_mmap2;
@@ -3614,6 +3784,7 @@ static void pyrf_session__delete(struct pyrf_session *psession)perf_session__delete (psession->session);Py_XDECREF (psession->pdata);Py_XDECREF (psession->sample); +Py_XDECREF (psession->stat);Py_TYPE (psession)->tp_free ((PyObject *)psession); }
@@ -3796,6 +3967,12 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_context_switch_event__type);PyModule_AddObject (module,"switch_event" , (PyObject *)&pyrf_context_switch_event__type); +Py_INCREF (&pyrf_stat_event__type); +PyModule_AddObject (module,"stat_event" , (PyObject *)&pyrf_stat_event__type); + +Py_INCREF (&pyrf_stat_round_event__type); +PyModule_AddObject (module,"stat_round_event" , (PyObject *)&pyrf_stat_round_event__type); +Py_INCREF (&pyrf_thread_map__type);PyModule_AddObject (module,"thread_map" , (PyObject *)&pyrf_thread_map__type);
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 14/19] perf python: Extend API for stat events in python.c
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [High] The Python `pyrf_session` wrapper fails to register `.attr` and `.feature` callbacks for pipe…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] The Python pyrf_session wrapper fails to register .attr and .feature callbacks for pipe mode, silently dropping PERF_RECORD_HEADER_ATTR events and breaking evsel lookup. --
commit 9b2beeec32b11c94e99c583602b25a4f6100962e
Author: Ian Rogers <[email protected]>
perf python: Extend API for stat events in python.c
This commit adds stat information to the perf python session API. It provides new callbacks and Python objects to map stat events from the event stream.
Quoted 3 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 4 lines
tools/perf/util/python.c +0
-0@@ -3573,8 +3738,13 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject }while (0 )ADD_TOOL (sample);
tools/perf/util/python.c static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject · line 3750
Quoted 10 lines
tools/perf/util/python.c +5
-0+ADD_TOOL (stat);#undef ADD_TOOL +if (stat) + psession->tool.stat_round = pyrf_session_tool__stat_round; + + psession->tool.comm = perf_event__process_comm; psession->tool.mmap = perf_event__process_mmap; psession->tool.mmap2 = perf_event__process_mmap2;
[Severity: High]
Will this configuration properly handle evsel population in pipe mode?
Because psession->tool.attr and psession->tool.feature are not assigned handlers here, they appear to remain as stubs. Without processing PERF_RECORD_HEADER_ATTR and PERF_RECORD_HEADER_FEATURE events from the pipe stream, won't session->evlist remain unpopulated?
If that happens, won't evlist__id2evsel() return NULL when processing the newly added PERF_RECORD_STAT events, resulting in an "unknown" name for the stat records?
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=14
Ian Rogers
reviewerPATCH v10 15/19 perf python: Expose brstack in sample event
In reply to Ian RogersImplement pyrf_branch_entry and pyrf_branch_stack for lazy iteration over branch stack entries.
---
v2:
1. Avoided Keyword Collision: Renamed the properties "from" and "to" to "from_ip" and "to_ip" in pyrf_branch_entry__getset[] to prevent syntax errors in Python.
2. Eager Branch Stack Resolution: Updated pyrf_session_tool__sample() to allocate and copy the branch stack entries eagerly when the event is processed, instead of deferring it to iteration time. This avoids reading from potentially overwritten or unmapped mmap buffers.
3. Updated Iterators: Updated pyrf_branch_stack and pyrf_branch_stack__next() to use the copied entries rather than pointing directly to the sample's buffer.
4. Avoided Reference Leak on Init Failure: Added proper error checking for PyModule_AddObject() in the module initialization function, decrementing references on failure.
v6:
- Moved branch stack resolution from session_tool__sample to pyrf_event__new.
--- tools/perf/util/python.c | 197 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 197 insertions(+)
tools/perf/util/python.c +197 -0
@@ -22,6 +22,7 @@#include "debug.h" #include "dso.h" #include "event.h" +#include "branch.h" #include "evlist.h" #include "evsel.h" #include "expr.h"
@@ -87,6 +88,8 @@ struct pyrf_event {bool al_resolved;/** @callchain: Resolved callchain, eagerly computed if requested. */ PyObject *callchain; +/** @brstack: Resolved branch stack, eagerly computed if requested. */ +PyObject *brstack;/** @event: The underlying perf_event that may be in a file or ring buffer. */ union perf_event event; };
@@ -125,6 +128,7 @@ static void pyrf_event__delete(struct pyrf_event *pevent)if (pevent->al_resolved)addr_location__exit (&pevent->al);Py_XDECREF (pevent->callchain); +Py_XDECREF (pevent->brstack);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -992,6 +996,153 @@ static PyObject *pyrf_sample_event__get_callchain(PyObject *self, void *closurereturn pevent->callchain; } +struct pyrf_branch_entry { +PyObject_HEAD + u64 from; +u64 to; +struct branch_flags flags; +}; + +static void pyrf_branch_entry__delete (struct pyrf_branch_entry *pentry) +{ +Py_TYPE (pentry)->tp_free ((PyObject *)pentry); +} + +static PyObject *pyrf_branch_entry__get_from (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->from); +} + +static PyObject *pyrf_branch_entry__get_to (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->to); +} + +static PyObject *pyrf_branch_entry__get_mispred (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.mispred); +} + +static PyObject *pyrf_branch_entry__get_predicted (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.predicted); +} + +static PyObject *pyrf_branch_entry__get_in_tx (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.in_tx); +} + +static PyObject *pyrf_branch_entry__get_abort (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.abort); +} + +static PyObject *pyrf_branch_entry__get_cycles (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->flags.cycles); +} + +static PyObject *pyrf_branch_entry__get_type (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromLong (pentry->flags.type); +} + +static PyGetSetDef pyrf_branch_entry__getset[] = { + { .name ="from_ip" , .get = (getter )pyrf_branch_entry__get_from, }, + { .name ="to_ip" , .get = (getter )pyrf_branch_entry__get_to, }, + { .name ="mispred" , .get = (getter )pyrf_branch_entry__get_mispred, }, + { .name ="predicted" , .get = (getter )pyrf_branch_entry__get_predicted, }, + { .name ="in_tx" , .get = (getter )pyrf_branch_entry__get_in_tx, }, + { .name ="abort" , .get = (getter )pyrf_branch_entry__get_abort, }, + { .name ="cycles" , .get = (getter )pyrf_branch_entry__get_cycles, }, + { .name ="type" , .get = (getter )pyrf_branch_entry__get_type, }, + { .name =NULL , }, +}; + +static PyTypeObject pyrf_branch_entry__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.branch_entry" , + .tp_basicsize =sizeof (struct pyrf_branch_entry ), + .tp_dealloc = (destructor )pyrf_branch_entry__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf branch entry object." , + .tp_getset = pyrf_branch_entry__getset, +}; + +struct pyrf_branch_stack { +PyObject_HEAD + struct branch_entry *entries; +u64 nr; +}; + +static void pyrf_branch_stack__delete (struct pyrf_branch_stack *pstack) +{ +free (pstack->entries); +Py_TYPE (pstack)->tp_free ((PyObject *)pstack); +} + +static Py_ssize_t pyrf_branch_stack__length (PyObject *obj) +{ +struct pyrf_branch_stack *pstack = (void *)obj; +return pstack->nr; +} + +static PyObject *pyrf_branch_stack__item (PyObject *obj,Py_ssize_t i) +{ +struct pyrf_branch_stack *pstack = (void *)obj; +struct pyrf_branch_entry *pentry; + +if (i <0 || i >= (Py_ssize_t )pstack->nr) { +PyErr_SetString (PyExc_IndexError,"Index out of range" ); +return NULL ; + } + + pentry = PyObject_New(struct pyrf_branch_entry, &pyrf_branch_entry__type); +if (!pentry) +return NULL ; + + pentry->from = pstack->entries[i].from; + pentry->to = pstack->entries[i].to; + pentry->flags = pstack->entries[i].flags; + +return (PyObject *)pentry; +} + +static PySequenceMethods pyrf_branch_stack__sequence_methods = { + .sq_length = pyrf_branch_stack__length, + .sq_item = pyrf_branch_stack__item, +}; + +static PyTypeObject pyrf_branch_stack__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.branch_stack" , + .tp_basicsize =sizeof (struct pyrf_branch_stack ), + .tp_dealloc = (destructor )pyrf_branch_stack__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf branch stack object." , + .tp_as_sequence = &pyrf_branch_stack__sequence_methods, +}; + +static PyObject *pyrf_sample_event__get_brstack (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->brstack) + Py_RETURN_NONE; + +Py_INCREF (pevent->brstack); +return pevent->brstack; +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -1012,6 +1163,12 @@ static PyGetSetDef pyrf_sample_event__getset[] = { .set = NULL, .doc = "event callchain.", }, + { + .name = "brstack", + .get = pyrf_sample_event__get_brstack, + .set = NULL, + .doc = "event branch stack.", + }, { .name = "raw_buf", .get = (getter)pyrf_sample_event__get_raw_buf,
@@ -1193,6 +1350,12 @@ static int pyrf_event__setup_types(void) err =PyType_Ready (&pyrf_callchain__type);if (err <0 ) goto out; + err =PyType_Ready (&pyrf_branch_entry__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_branch_stack__type); +if (err <0 ) + goto out; out:return err; }
@@ -1379,6 +1542,7 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evperf_sample__init (&pevent->sample,/*all=*/ true); pevent->callchain =NULL ; + pevent->brstack =NULL ; pevent->al_resolved = false;addr_location__init (&pevent->al);
@@ -1436,6 +1600,27 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evaddr_location__exit (&al); } } +if (sample->branch_stack) { +struct branch_stack *bs = sample->branch_stack; +struct branch_entry *entries =perf_sample__branch_entries (sample); +struct pyrf_branch_stack *pstack; + + pstack = PyObject_New(struct pyrf_branch_stack, &pyrf_branch_stack__type); +if (!pstack) { +Py_DECREF (pevent); +return NULL ; + } + pstack->nr = bs->nr; + pstack->entries =calloc (bs->nr,sizeof (struct branch_entry )); +if (!pstack->entries) { +Py_DECREF (pstack); +Py_DECREF (pevent); +return PyErr_NoMemory (); + } +memcpy (pstack->entries, entries, + bs->nr *sizeof (struct branch_entry )); + pevent->brstack = (PyObject *)pstack; + }return (PyObject *)pevent; }
@@ -3988,6 +4173,18 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_session__type);PyModule_AddObject (module,"session" , (PyObject *)&pyrf_session__type); +Py_INCREF (&pyrf_branch_entry__type); +if (PyModule_AddObject (module,"branch_entry" , (PyObject *)&pyrf_branch_entry__type) <0 ) { +Py_DECREF (&pyrf_branch_entry__type); + goto error; + } + +Py_INCREF (&pyrf_branch_stack__type); +if (PyModule_AddObject (module,"branch_stack" , (PyObject *)&pyrf_branch_stack__type) <0 ) { +Py_DECREF (&pyrf_branch_stack__type); + goto error; + } + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 16/19 perf python: Add syscall name/id to convert syscall number and name
In reply to Ian RogersUse perf's syscalltbl support to convert syscall number to name assuming the number is for the host machine. This avoids python libaudit support as tools/perf/scripts/python/syscall-counts.py requires.
---
v2:
1. Guarded with HAVE_LIBTRACEEVENT : Wrapped the syscall functions and their entries in perf__methods with #ifdef HAVE_LIBTRACEEVENT to avoid potential linker errors if CONFIG_TRACE is disabled.
2. Changed Exception Type: Updated pyrf__syscall_id to raise a ValueError instead of a TypeError when a syscall is not found.
3. Fixed Typo: Corrected "name number" to "name" in the docstring for syscall_id.
v6:
- Added optional keyword-only elf_machine argument to syscall_name and syscall_id.
v7:
- Made syscalltbl.o unconditional in Build to fix undefined symbol when building without libtraceevent.
--- tools/perf/util/Build | 1 - tools/perf/util/python.c | 48 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 48 insertions(+), 1 deletion(-)
tools/perf/util/Build +0 -1
@@ -75,7 +75,6 @@ perf-util-y += sample.o perf-util-y += sample-raw.o perf-util-y += s390-sample-raw.o perf-util-y += amd-sample-raw.o - perf-util-y += ordered-events.o perf-util-y += namespaces.o perf-util-y += comm.o
tools/perf/util/python.c +48 -0
72 lines not on this page · raw message
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 16/19] perf python: Add syscall name/id to convert syscall number and name
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] The commit introduces `syscall_name` and `syscall_id` to the perf python module to avoid the…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The commit introduces syscall_name and syscall_id to the perf python module to avoid the dependency on python libaudit, but fails to update syscall-counts.py or its Util.py wrapper to actually use the new API. --
commit f477c88fcc2c1516b8d5945cfc3bf37bdd3b52f2
Author: Ian Rogers <[email protected]>
perf python: Add syscall name/id to convert syscall number and name
This commit adds syscall_name and syscall_id functions to the perf Python C extension module. It aims to use the internal syscalltbl support instead of relying on the external libaudit library.
Quoted 4 lines
Use perf's syscalltbl support to convert syscall number to name assuming the number is for the host machine. This avoids python libaudit support as
tools/perf/scripts/python/syscall-counts.pyrequires.
[Severity: Medium]
The commit message indicates this avoids the libaudit dependency for tools/perf/scripts/python/syscall-counts.py, but the patch doesn't seem to update that script or its wrapper in Util.py to use the new API.
Looking at tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py, the wrapper still relies entirely on the audit package:
def syscall_name(id):
try:
return audit.audit_syscall_to_name(id, machine_id)
except:
return str(id)
And tools/perf/scripts/python/syscall-counts.py still imports this legacy wrapper:
from Util import syscall_name
Should this patch also include the migration for syscall-counts.py and Util.py to actually call the newly added perf.syscall_name API?
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=16
Ian Rogers
reviewerPATCH v10 17/19 perf python: Add config file access
In reply to Ian RogersAdd perf.config_get(name) to expose the perf configuration system.
--- tools/perf/util/python.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+)
tools/perf/util/python.c +20 -0
@@ -17,6 +17,7 @@#include "build-id.h" #include "callchain.h" #include "comm.h" +#include "config.h" #include "counts.h" #include "data.h" #include "debug.h"
@@ -4077,7 +4078,26 @@ static PyObject *pyrf__syscall_id(PyObject *self, PyObject *args, PyObject *kwarreturn PyLong_FromLong (id); } +static PyObject *pyrf__config_get (PyObject *self,PyObject *args) +{ +const char *config_name, *val; + +if (!PyArg_ParseTuple (args,"s" , &config_name)) +return NULL ; + + val =perf_config_get (config_name); +if (!val) + Py_RETURN_NONE; +return PyUnicode_FromString (val); +} +static PyMethodDef perf__methods[] = { + { + .ml_name ="config_get" , + .ml_meth = (PyCFunction ) pyrf__config_get, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Get a perf config value." ) + }, { .ml_name ="metrics" , .ml_meth = (PyCFunction ) pyrf__metrics,
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v10 18/19 perf python: Add perf.pyi stubs file
In reply to Ian RogersAdd Python type stubs for the perf module to improve IDE support and static analysis. Includes docstrings for classes, methods, and constants derived from C source and JSON definitions.
---
v2:
1. Added Missing Module Functions: Added parse_metrics and pmus. Renamed metrics to parse_metrics to match python.c .
2. Added Constructors: Added __init__ methods for data , evsel , and evlist with their appropriate arguments.
3. Removed sample_comm : Removed it from sample_event since it is not exported in python.c .
4. Keyword Handling in branch_entry : I used from_ip and to_ip in the stubs to match the rename I did in python.c (in turn 145) to avoid the Python from keyword conflict.
5. Added Missing Event Classes: Added mmap_event , lost_event , comm_event , task_event , throttle_event , read_event , and switch_event .
6. Added Missing evlist Methods: Added get_pollfd and add .
7. Updated Return Types: Changed process_events to return int .
v6:
- Updated perf.pyi to use find_thread and elf_machine.
v8:
- Added mmap2_event class and new evsel attributes to perf.pyi.
- Added pid, tid, ppid, cpu attributes to class thread in perf.pyi.
--- tools/perf/Makefile.perf | 3 +- tools/perf/python/perf.pyi | 629 +++++++++++++++++++++++++++++++++++++ tools/perf/util/setup.py | 1 + 3 files changed, 632 insertions(+), 1 deletion(-) create mode 100644 tools/perf/python/perf.pyi
tools/perf/Makefile.perf +2 -1
@@ -527,7 +527,8 @@ $(OUTPUT)python/perf$(PYTHON_EXTENSION_SUFFIX): util/python.c util/setup.py $(PE CFLAGS='$(CFLAGS)' LDFLAGS='$(LDFLAGS) $(LIBS_PY)' \ $(PYTHON_WORD) util/setup.py \ --quiet build_ext; \ - cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/ + cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/; \ + cp python/perf.pyi $(OUTPUT)python/ python_perf_target: @echo "Target is: $(OUTPUT)python/perf$(PYTHON_EXTENSION_SUFFIX)"
tools/perf/python/perf.pyi new file +629 -0
629 lines not on this page · raw message
tools/perf/util/setup.py +1 -0
@@ -105,4 +105,5 @@ setup(name='perf', license='GPLv2', url='http://perf.wiki.kernel.org', ext_modules=[perf], + data_files=[('', [f'{src_perf}/python/perf.pyi'])], cmdclass={'build_ext': build_ext, 'install_lib': install_lib})
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 18/19] perf python: Add perf.pyi stubs file
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [Medium] The `setup.py` configuration incorrectly installs `perf.pyi` directly to the system's root…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] The setup.py configuration incorrectly installs perf.pyi directly to the system's root installation prefix rather than the site-packages directory alongside the extension module.
- [Low] In-tree builds fail unconditionally due to a self-copy error when $(OUTPUT) is empty.
--
commit c97699f4738dbd39b930541637409562855b081b
Author: Ian Rogers <[email protected]>
perf python: Add perf.pyi stubs file
Add Python type stubs for the perf module to improve IDE support and static analysis. Includes docstrings for classes, methods, and constants derived from C source and JSON definitions.
tools/perf/Makefile.perf $(OUTPUT)python/perf$(PYTHON_EXTENSION_SUFFIX): util/python.c util/setup.py $(PE · line 531
Quoted 11 lines
tools/perf/Makefile.perf +2
-1@@ -527,7 +527,8 @@ $(OUTPUT)python/perf$(PYTHON_EXTENSION_SUFFIX): util/python.c util/setup.py $(PE CFLAGS='$(CFLAGS)' LDFLAGS='$(LDFLAGS) $(LIBS_PY)' \ $(PYTHON_WORD) util/setup.py \ --quiet build_ext; \ - cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/ + cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/; \ + cp python/perf.pyi $(OUTPUT)python/
[Severity: Low]
Does this rule cause in-tree builds to fail?
During an in-tree build of perf, the $(OUTPUT) variable is empty. This evaluates the command to cp python/perf.pyi python/.
This causes GNU cp to exit with an error indicating the source and destination are the same file, which aborts the build process.
[ ... ]
tools/perf/util/setup.py setup(name='perf', · line 109
Quoted 10 lines
tools/perf/util/setup.py +1
-0@@ -105,4 +105,5 @@ setup(name='perf', license='GPLv2', url='http://perf.wiki.kernel.org', ext_modules=[perf], + data_files=[('', [f'{src_perf}/python/perf.pyi'])], cmdclass={'build_ext': build_ext, 'install_lib': install_lib})
[Severity: Medium]
Will this configuration install the stub file into the system root rather than the site-packages directory?
By specifying an empty string as the target directory in the data_files list, setuptools places the perf.pyi file directly into the root installation prefix, such as /usr/perf.pyi.
This avoids installing the file alongside the extension module in the Python site-packages directory. Consequently, type checkers and IDEs will fail to discover the stub.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=18
Ian Rogers
reviewerPATCH v10 19/19 perf python: Add LiveSession helper
In reply to Ian RogersAdd LiveSession class in tools/perf/python/perf_live.py to support live event collection using perf.evlist and perf.parse_events, avoiding the need to fork a separate perf record process.
---
v2:
1. Fixed File Descriptor Leak: I moved self.evlist.mmap() inside the try block so that if it raises an exception, the finally block will still be executed and call self.evlist.close() , preventing file descriptor leaks.
2. Handled InterruptedError in poll() : I wrapped the poll() call in a try-except block to catch InterruptedError and continue the loop. This prevents the live session from crashing on non-fatal signals like SIGWINCH .
3. Added evlist.config() : I added a call to self.evlist.config() in the constructor after parse_events() . This applies the default record options to the events, enabling sampling and setting up PERF_SAMPLE_* fields so that the kernel will actually generate PERF_RECORD_SAMPLE events.
4. Enable the evlist and be robust to exceptions from reading unsupported events like mmap2.
v8:
- Drain all events from a CPU before moving to the next.
--- tools/perf/python/perf_live.py | 56 ++++++++++++++++++++++ tools/perf/tests/shell/lib/setup_python.sh | 12 +++++ 2 files changed, 68 insertions(+) create mode 100755 tools/perf/python/perf_live.py
tools/perf/python/perf_live.py new file +56 -0
@@ -0,0 +1,56 @@ +# SPDX-License-Identifier: GPL-2.0 +""" +Live event session helper using perf.evlist. + +This module provides a LiveSession class that allows running a callback +for each event collected live from the system, similar to perf.session +but without requiring a perf.data file. +""" + +import perf + + +class LiveSession: + """Represents a live event collection session.""" + + def __init__(self, event_string: str, sample_callback): + self.event_string = event_string + self.sample_callback = sample_callback + # Create a cpu map for all online CPUs + self.cpus = perf.cpu_map() + # Parse events and set maps + self.evlist = perf.parse_events(self.event_string, self.cpus) + self.evlist.config() + + def run(self): + """Run the live session.""" + try: + self.evlist.open() + self.evlist.mmap() + self.evlist.enable() + + while True: + # Poll for events with 100ms timeout + try: + self.evlist.poll(100) + except InterruptedError: + continue + for cpu in self.cpus: + for _ in range(1000): # Limit to 1000 events per CPU per poll to prevent starvation + try: + event = self.evlist.read_on_cpu(cpu) + if event is None: + break + if event.type == perf.RECORD_SAMPLE: + self.sample_callback(event) + except OSError as e: + # OS/mmap errors are unrecoverable and will cause busy-looping if ignored + raise e + except Exception as e: + # Parsing or type errors are recoverable as the event was consumed + import sys + print(f"Error parsing event on CPU {cpu}: {e}", file=sys.stderr) + except KeyboardInterrupt: + pass + finally: + self.evlist.close()
tools/perf/tests/shell/lib/setup_python.sh +12 -0
@@ -14,3 +14,15 @@ then echo Skipping test, python not detected please set environment variable PYTHON. exit 2 fi + +# Set PYTHONPATH to find the in-tree built perf.so first, avoiding system-wide perf.so +if [ "x$PYTHONPATH" = "x" ]; then + # If PERF_EXEC_PATH is set (pointing to the build directory), prioritize it + if [ -n "$PERF_EXEC_PATH" ] && [ -d "$PERF_EXEC_PATH/python" ]; then + export PYTHONPATH="$PERF_EXEC_PATH/python" + elif [ -d "$(dirname "$0")/../../python" ]; then + export PYTHONPATH="$(dirname "$0")/../../python" + elif [ -d "$(dirname "$0")/../python" ]; then + export PYTHONPATH="$(dirname "$0")/../python" + fi +fi
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v10 19/19] perf python: Add LiveSession helper
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [High] LiveSession crashes unconditionally on empty ring buffers due to re-raising -EAGAIN as OSError.…
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] LiveSession crashes unconditionally on empty ring buffers due to re-raising -EAGAIN as OSError.
- [Medium] LiveSession busy-loops and spams stderr on unmapped CPUs.
- [Medium] Broad exception handler swallows user callback exceptions and misreports them as parsing errors. --
commit 4b0aea1c86d61d02f883cb514ba6d1082d3b2a72
Author: Ian Rogers <[email protected]>
perf python: Add LiveSession helper
Add LiveSession class in tools/perf/python/perf_live.py to support live event collection using perf.evlist and perf.parse_events, avoiding the need to fork a separate perf record process.
Quoted 5 lines
tools/perf/python/perf_live.py new file +0
-0
[ ... ]
Quoted 11 lines
tools/perf/python/perf_live.py new file +11
-0+ def run(self): + """Run the live session.""" + try: + self.evlist.open() + self.evlist.mmap() + self.evlist.enable() + + while True: + # Poll for events with 100ms timeout + try: + self.evlist.poll(100)
tools/perf/python/perf_live.py line 45
Quoted 10 lines
tools/perf/python/perf_live.py new file +10
-0+ except InterruptedError: + continue + for cpu in self.cpus: + for _ in range(1000): # Limit to 1000 events per CPU per poll to prevent starvation + try: + event = self.evlist.read_on_cpu(cpu) + if event is None: + break + if event.type == perf.RECORD_SAMPLE: + self.sample_callback(event)
[Severity: Medium]
Will this broad exception handler swallow exceptions raised by the user's sample_callback?
If the user's callback raises a standard Python exception like ValueError or KeyError, it looks like it gets caught below and misreported as an event parsing error, obscuring bugs in the tracing logic.
tools/perf/python/perf_live.py line 48
Quoted 3 lines
tools/perf/python/perf_live.py new file +3
-0+ except OSError as e: + # OS/mmap errors are unrecoverable and will cause busy-looping if ignored + raise e
[Severity: High]
Does this unconditionally crash the live session on an empty ring buffer?
When the CPU's ring buffer is empty, pyrf_evlist__read_on_cpu() calls perf_mmap__read_init(), which returns -EAGAIN (-11). This causes the C extension to throw an OSError.
Since it's normal for some CPUs to have no events when poll() returns, re-raising the OSError here would terminate the session the first time it checks an empty buffer.
tools/perf/python/perf_live.py line 52
Quoted 4 lines
tools/perf/python/perf_live.py new file +4
-0+ except Exception as e: + # Parsing or type errors are recoverable as the event was consumed + import sys + print(f"Error parsing event on CPU {cpu}: {e}", file=sys.stderr)
[Severity: Medium]
Can this lead to a busy-loop that spams stderr on unmapped CPUs?
If a CPU is not mapped by the event list (e.g., when tracing uncore PMU events), pyrf_evlist__read_on_cpu() returns NULL internally and throws a TypeError.
Since the exception handler here catches the TypeError but doesn't break out of the 1000-iteration loop, it will repeatedly print this error 1000 times for each unmapped CPU on every single poll cycle.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=19
Ian Rogers
reviewerv11 cover perf python: Modernize and extend Python API (Phase 1)
In reply to Ian RogersThe perf script command has long supported running Python and Perl scripts by embedding libpython and libperl. This approach has several drawbacks:
- overhead by creating Python dictionaries for every event (whether used or not),
- complex build dependencies on specific Python/Perl versions,
- complications with threading due to perf being the interpreter,
- no clear way to run standalone scripts like ilist.py.
This series takes a different approach with some initial implementation posted as an RFC last October:
https://lore.kernel.org/linux-perf-users/[email protected]/
with the motivation coming up on the mailing list earlier:
https://lore.kernel.org/lkml/CAP-5=fWDqE8SYfOLZkg_0=4Ayx6E7O+h7uUp4NDeCFkiN4b7-w@mail.gmail.com/
The ultimate goal is to remove the embedded libpython and libperl support from perf entirely, expanding the existing perf Python module to provide full access to perf data files and events, allowing scripts to be run as standalone Python applications.
To make the review process more manageable, the original 58-patch series has been split. This v11 series represents "Phase 1: API & Infrastructure" (19 patches). The first 4 patches of Phase 1 (cleanups and arch-specific header sorting) have already been merged upstream.
This remaining set contains:
1. Missed explicit dependency cleanups and header sorting for util/ and python.
2. Crucial core safety infrastructure (reference counting for evlist/evsel) to support safe lifecycle management in garbage-collected Python.
3. The core Python API extensions (session wrappers, perf_data wrappers, sample accessors, stubs, and LiveSession helper).
The subsequent "Phase 2" series will contain the actual porting of all existing Python/Perl scripts to the new API (which yields up to 35x speedups as demonstrated previously) and the final removal of embedded interpreters.
Note: The preliminary clang-format patch has been separated from this series to be sent independently.
---
v11 Changes
-----------
- Investigated Patch 8 (evsel leak) review feedback and verified no changes were necessary since perf_sample__exit(data) handles the evsel reference internally upon parsing failures.
- Patch 7: Fixed compiler errors when building with REFCNT_CHECKING=1 by replacing direct evlist member accesses with their correct accessor functions in util/bpf_counter_cgroup.c, builtin-lock.c, builtin- sched.c, builtin-annotate.c, and tests/uncore-event-sorting.c.
- Patch 9: Added pdata->data.path = NULL; after free in pyrf_data__init to prevent double-free/dangling pointers.
- Patch 10 & 14: Added missing default processors (.attr, .feature, .stat) to pyrf_session__new().
- Patch 11: Reverted PyObject_Malloc allocation back to PyObject_New and clipped memcpy size to sizeof(union perf_event) to fix _FORTIFY_SOURCE aborts without unnecessarily expanding the allocation size, and added explicit null termination to PERF_RECORD_COMM. Also avoided replicating validation logic in pyrf_event__new by making perf_event__too_small non-static and using it directly.
- Patch 18: Fixed Makefile.perf for empty $(OUTPUT) copying error, and rewrote setup.py's perf.pyi deployment to override install_lib using shutil.copy2 to fix install directories.
- Patch 19: Handled TypeError for offline CPUs, handled -EAGAIN correctly on empty buffers, and allowed user exceptions to bubble up instead of being universally caught in perf_live.py.
v10 Changes
-----------
- The preliminary clang-format patch has been separated from this series and updated independently based on review feedback (using Priority -1 for Python.h and eliminating the unused Priority 1 gap).
- Fixed Type Confusion on Python Object instantiation:
- Added strict type verification O!O! to pyrf_evlist__init for CPU and Thread maps to prevent interpreter crashes from bad inputs.
- Added .tp_getattro and .tp_setattro handlers to evlist, evsel, and session Python types to safely raise ValueError on uninitialized access (such as bypassing __init__ and invoking __new__ directly) instead of dereferencing NULL.
- Handled NULL structures safely inside string/representation callbacks for evlist and evsel types.
- Fixed Circular Reference memory leaks:
- Eliminated the pevent pointer and unused lazy resolution code from pyrf_callchain, breaking the cycle that prevented GC collection.
- Eliminated pevent pointer from pyrf_branch_stack, breaking the cycle.
- This prevents permanent leaks of all map, symbol, and event structures for every sample containing a callchain or branch stack.
- Converted callchain and branch_stack to true Python Sequence types:
- Replaced custom exhaustible iterators with .tp_as_sequence sequence methods (sq_length and sq_item).
- This allows standard Sequence operations like len(event.callchain), list index access (e.g. event.callchain[0]), and infinite re-iterations.
- Fixed potential Memory Leak on test failure:
- Fixed a leak of the newly allocated evsel structure in do_test inside sample-parsing.c on malloc failure by routing through the cleanup block.
- Fixed compiler warning for mixed declarations:
- Moved the C-statement perf_sample__init after struct and variable declarations in cs_etm__synth_branch_sample to ensure C89/compiler compatibility.
- Fixed compiler warning for discarded const qualifiers:
- Reverted custom static char * const kwlist[] to the standard static char *kwlist[] for pyrf__syscall_name and pyrf__syscall_id in python.c.
- Fixed unrecoverable file descriptor leak and busy loop in LiveSession:
- Moved self.evlist.open() inside the try block of LiveSession.run() to guarantee finally block cleanup and prevent descriptor leaks on early interrupts.
- Handled unrecoverable OSError (like mmap read init errors) by propagating/raising them to safely terminate the session instead of getting stuck in an infinite CPU-pegging poll busy loop.
- Bounded read_on_cpu() reads to at most 1000 events per CPU per poll iteration to prevent high-volume starvation of other CPU channels.
- Exported stat_event and stat_round_event to the perf module namespace to allow type checks like isinstance(event, perf.stat_event).
- Fixed Python stat callback signatures:
- Changed Python stat callbacks in C to consistently pass 2 arguments (using "Oz" and passing None for stat_round events), preventing TypeError failures in scripts defining standard 2-parameter signatures.
- Fixed data-corruption in misc fields:
- Changed misc member definition from T_UINT to T_USHORT in both mmap and mmap2 events to avoid reading adjacent size bytes from 16-bit struct headers.
- Exposed maj, min, ino, and ino_generation members to Python mmap2 events.
- Fixed mmap2 union overlay field access:
- Added custom getters to mmap2 events for maj, min, ino, ino_generation, and build_id based on the PERF_RECORD_MISC_MMAP_BUILD_ID misc header flag.
- This correctly exposes build_id (as bytes) when present, or maj/min/ino when build ID is absent, returning None for the inactive union fields.
- Enabled deployment of stubs:
- Updated setup.py to install perf.pyi alongside the extension in site-packages.
- Updated Makefile.perf to copy perf.pyi to the build directory for in-tree usage.
- Added tracepoint and missing optional parse_metrics arguments to stubs.
- Fixed a pre-existing Use-After-Free bug in iostat_prepare:
- Modified iostat_prepare to take struct evlist **evlist_ptr, allowing it to properly reassign the caller's evlist pointer, avoiding use-after-free in cmd_stat.
- Fixed in-tree python test runner stability:
- Updated setup_python.sh to automatically export PYTHONPATH pointing to the in-tree built perf.so directory (handling O= build folders), preventing Python from accidentally loading the system-wide outdated /usr/lib/perf.so module and failing with AttributeError during perf test.
v9 Changes
----------
- This series is now split, containing only the first 23 patches of the previous 58-patch series. This "Phase 1: API & Infrastructure" set focuses on modernizing and extending the Python API and adding crucial safety infrastructure (reference counting). The script porting and legacy interpreter removal will be sent in a subsequent Phase 2.
- Fixed Type Confusion in pyrf_evlist__init: Added strict type validation to CPU and Thread map arguments (using O!O!) to prevent crashes from unsafe casts.
- Fixed Infinite Loop in LiveSession.run: Added a break statement in the exception block of the event reading loop to prevent 100% CPU spinning on persistent OS errors (like mmap read init failures).
- Fixed Inconsistent Exception Handling in Session Callbacks:
- Removed the swallowing PyErr_Print() call from pyrf_session_tool__stat to preserve exceptions.
- Updated pyrf_session_tool__stat_round to check the callback return value and return -1 on failure, aborting the event loop and propagating the exception cleanly.
- Fixed Uninitialized State in pyrf_session__new: Added explicit psession->pdata = NULL initialization immediately after allocation to prevent potential crashes in tp_dealloc on early failures.
v8 Changes
----------
- Make schedstat and itrace=L fixes separate patches:
https://lore.kernel.org/lkml/[email protected]/
https://lore.kernel.org/lkml/[email protected]/
- Fixed Heap Out-Of-Bounds / Uninitialized Memory in pyrf_event__new:
Use /*all=*/true in perf_sample__init to prevent garbage memory in sample structures.
- Fixed Type Confusion in pyrf_evlist__add: Added strict O! type validation to avoid unsafe casts when adding evsels to an evlist.
- Exposed Thread Identifiers: Added pid, tid, ppid, and cpu attributes to the Python perf.thread type to allow thread identification.
- Fixed Process Resolution: Wrapped thread resolution in compaction-times.py, check-perf-trace.py, and task-analyzer.py in try-except blocks to safely handle untracked PIDs instead of raising uncaught TypeError crashes.
- Fixed Potential Data Loss in futex-contention.py: Updated process resolution in handle_start to fall back to 'unknown' on lookup errors, ensuring events are always tracked.
- Synchronized Type Stubs File: Added the mmap2_event class and new evsel and thread attributes to perf.pyi.
v7 Changes
----------
- Fixed heap out-of-bounds in pyrf_event__new by adding comprehensive size checks for all event types.
- Fixed undefined symbol syscalltbl__id when building without libtraceevent by making syscalltbl.o unconditional in Build.
- Fixed several issues in python.c:
- Handled NULL return from thread__comm_str in pyrf_thread__comm.
- Avoided swallowing exceptions in module initialization.
- Added custom tp_new methods for evlist, evsel, and data types to zero-initialize pointers and avoid crashes on re-initialization.
- Fixed lower priority review comments:
- Avoided permanent iterator exhaustion on brstack in perf_brstack_max.py by converting it to a list.
- Removed dead code (unused self.unhandled dictionary) in failed-syscalls-by-pid.py.
v6 Changes
----------
- Refactored pyrf_event__new to take evsel and session arguments, and use dynamic allocation based on the actual event size to improve memory safety and efficiency.
- Moved callchain and branch stack resolution logic from pyrf_session_tool__sample into pyrf_event__new, centralizing initialization.
- Added an optional keyword-only elf_machine argument to syscall_name and syscall_id functions to allow specifying non-host architectures, defaulting to EM_HOST.
- Renamed process method to find_thread in the Python API and C implementation for better intention-revealing naming.
- Fixed a terminal injection vulnerability in flamegraph.py by not printing unverified downloaded content in the prompt.
- Fixed CWD exposure and symlink attack risks in gecko.py by using a secure temporary directory for the HTTP server.
- Fixed a severe performance issue in event_analyzing_sample.py by removing SQLite autocommit mode and batching commits.
- Fixed AttributeError crashes in rw-by-file.py and rw-by-pid.py by correctly extracting event names.
- Fixed man page formatting issues in perf-script-python.txt by using indented code blocks.
- Updated perf.pyi stubs file to reflect all API changes.
- Verified all commit messages with checkpatch.pl and ensured lines are wrapped appropriately.
- Fixed segmentation faults in perf sched stats in diff mode.
v5 Changes
----------
Resending due to partial send of v4 due to a quota limit.
v4 Changes
----------
1. Git Fixup Cleanups
- Squashed the lingering fixup! commit remaining from the previous session back into perf check-perf-trace: Port check-perf-trace to use python module.
v3 Changes
----------
1. Memory Safety & Reference Counting Fixes
- Stored transient mmap event data inside the Python object's permanent pevent->event and invoked evsel__parse_sample() to safely point attributes into it, resolving Use-After-Free vulnerabilities.
- Nullified sample->evsel after calling evsel__put() in perf_sample__exit() to protect against potential refcount double-put crashes in error paths.
- Reordered operations inside evlist__remove() to invoke perf_evlist__remove() before reference release.
- Patched an evsel reference leak inside evlist__deliver_deferred_callchain().
2. Sashiko AI Review Cleanups
- Corrected the broken event name equality check in gecko.py to search for a substring match within the parsed event string.
- Fixed a latent AttributeError crash in task-analyzer.py by properly assigning the session instance.
- Safeguarded thread reporting in check-perf-trace.py by utilizing sample_tid instead of sample_pid, and wrapping the session thread resolution in a try-except block.
3. Omitted Minor Issues
- The minor review comments (such as permanent iterator exhaustion on brstack, or dead-code in failed-syscalls-by-pid.py) have been omitted because they do not affect correctness, lead to crashes, or require significant architectural rework.
v2 Changes
----------
1. String Match and Event Name Accuracy
- Replaced loose substring event matching across the script suite with exact matches or specific prefix constraints (syscalls:sys_exit_, evsel(skb:kfree_skb), etc.).
- Added getattr() safety checks to prevent script failures caused by unresolved attributes from older kernel traces.
2. OOM and Memory Protections
- Refactored netdev-times.py to compute and process network statistics chronologically on-the-fly, eliminating an unbounded in-memory list that caused Out-Of-Memory crashes on large files.
- Implemented threshold limits on intel-pt-events.py to cap memory allocation during event interleaving.
- Optimized export-to-sqlite.py to periodically commit database transactions (every 10,000 samples) to reduce temporary SQLite journal sizes.
3. Portability & Environment Independence
- Re-keyed internal tracking dictionaries in scripts like powerpc-hcalls.py to use thread PIDs instead of CPUs, ensuring correctness when threads migrate.
- Switched net_dropmonitor.py from host-specific /proc/kallsyms parsing to perf's built-in symbol resolution API.
- Added the --iomem parameter to mem-phys-addr.py to support offline analysis of data collected on different architectures.
4. Standalone Scripting Improvements
- Patched builtin-script.c to ensure --input parameters are successfully passed down to standalone execution pipelines via execvp().
- Guarded against string buffer overflows during .py extension path resolving.
5. Code Cleanups
- Removed stale perl subdirectories from being detected by the TUI script browser.
- Ran the entire script suite through mypy and pylint to achieve strict static type checking and resolve unreferenced variables.
Ian Rogers (19):
perf util: Sort includes and add missed explicit dependencies perf python: Add missed explicit dependencies
perf evsel/evlist: Avoid unnecessary #includes
perf data: Add open flag
perf evlist: Add reference count
perf evsel: Add reference count
perf evlist: Add reference count checking
perf python: Use evsel in sample in pyrf_event
perf python: Add wrapper for perf_data file abstraction perf python: Add python session abstraction wrapping perf's session perf python: Refactor and add accessors to sample event perf python: Add mmap2 event
perf python: Add callchain support
perf python: Extend API for stat events in python.c
perf python: Expose brstack in sample event
perf python: Add syscall name/id to convert syscall number and name perf python: Add config file access
perf python: Add perf.pyi stubs file
perf python: Add LiveSession helper
tools/perf/Makefile.perf | 5 +- tools/perf/arch/arm/util/cs-etm.c | 10 +- tools/perf/arch/arm64/util/arm-spe.c | 8 +- tools/perf/arch/arm64/util/hisi-ptt.c | 2 +- tools/perf/arch/x86/tests/hybrid.c | 22 +- tools/perf/arch/x86/tests/topdown.c | 4 +- tools/perf/arch/x86/util/auxtrace.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 6 +- tools/perf/arch/x86/util/intel-pt.c | 9 +- tools/perf/arch/x86/util/iostat.c | 14 +- tools/perf/bench/evlist-open-close.c | 29 +- tools/perf/builtin-annotate.c | 7 +- tools/perf/builtin-ftrace.c | 14 +- tools/perf/builtin-inject.c | 4 +- tools/perf/builtin-kvm.c | 14 +- tools/perf/builtin-kwork.c | 8 +- tools/perf/builtin-lock.c | 4 +- tools/perf/builtin-record.c | 95 +- tools/perf/builtin-report.c | 6 +- tools/perf/builtin-sched.c | 30 +- tools/perf/builtin-script.c | 15 +- tools/perf/builtin-stat.c | 83 +- tools/perf/builtin-top.c | 104 +- tools/perf/builtin-trace.c | 60 +- tools/perf/python/perf.pyi | 629 +++++ tools/perf/python/perf_live.py | 60 + tools/perf/tests/backward-ring-buffer.c | 26 +- tools/perf/tests/code-reading.c | 14 +- tools/perf/tests/event-times.c | 6 +- tools/perf/tests/event_update.c | 4 +- tools/perf/tests/evsel-roundtrip-name.c | 8 +- tools/perf/tests/evsel-tp-sched.c | 4 +- tools/perf/tests/expand-cgroup.c | 12 +- tools/perf/tests/hists_cumulate.c | 2 +- tools/perf/tests/hists_filter.c | 2 +- tools/perf/tests/hists_link.c | 2 +- tools/perf/tests/hists_output.c | 2 +- tools/perf/tests/hwmon_pmu.c | 7 +- tools/perf/tests/keep-tracking.c | 10 +- tools/perf/tests/mmap-basic.c | 24 +- tools/perf/tests/openat-syscall-all-cpus.c | 6 +- tools/perf/tests/openat-syscall-tp-fields.c | 26 +- tools/perf/tests/openat-syscall.c | 6 +- tools/perf/tests/parse-events.c | 139 +- tools/perf/tests/parse-metric.c | 8 +- tools/perf/tests/parse-no-sample-id-all.c | 2 +- tools/perf/tests/perf-record.c | 38 +- tools/perf/tests/perf-time-to-tsc.c | 12 +- tools/perf/tests/pfm.c | 12 +- tools/perf/tests/pmu-events.c | 11 +- tools/perf/tests/pmu.c | 4 +- tools/perf/tests/sample-parsing.c | 44 +- tools/perf/tests/shell/lib/setup_python.sh | 12 + tools/perf/tests/sw-clock.c | 20 +- tools/perf/tests/switch-tracking.c | 10 +- tools/perf/tests/task-exit.c | 20 +- tools/perf/tests/time-utils-test.c | 14 +- tools/perf/tests/tool_pmu.c | 7 +- tools/perf/tests/topology.c | 4 +- tools/perf/tests/uncore-event-sorting.c | 6 +- tools/perf/ui/browsers/annotate.c | 2 +- tools/perf/ui/browsers/hists.c | 22 +- tools/perf/util/Build | 1 - tools/perf/util/amd-sample-raw.c | 2 +- tools/perf/util/annotate-data.c | 2 +- tools/perf/util/annotate.c | 10 +- tools/perf/util/auxtrace.c | 14 +- tools/perf/util/block-info.c | 4 +- tools/perf/util/bpf_counter.c | 2 +- tools/perf/util/bpf_counter_cgroup.c | 14 +- tools/perf/util/bpf_ftrace.c | 9 +- tools/perf/util/bpf_lock_contention.c | 12 +- tools/perf/util/bpf_off_cpu.c | 44 +- tools/perf/util/bpf_trace_augment.c | 8 +- tools/perf/util/cgroup.c | 26 +- tools/perf/util/cs-etm.c | 5 +- tools/perf/util/data-convert-bt.c | 2 +- tools/perf/util/data.c | 27 +- tools/perf/util/data.h | 4 +- tools/perf/util/evlist.c | 492 ++-- tools/perf/util/evlist.h | 273 +- tools/perf/util/evsel.c | 39 +- tools/perf/util/evsel.h | 40 +- tools/perf/util/expr.c | 2 +- tools/perf/util/header.c | 69 +- tools/perf/util/header.h | 2 +- tools/perf/util/intel-tpebs.c | 7 +- tools/perf/util/iostat.c | 2 +- tools/perf/util/iostat.h | 2 +- tools/perf/util/map.h | 9 +- tools/perf/util/metricgroup.c | 12 +- tools/perf/util/parse-events.c | 10 +- tools/perf/util/parse-events.y | 2 +- tools/perf/util/perf_api_probe.c | 20 +- tools/perf/util/pfm.c | 4 +- tools/perf/util/print-events.c | 2 +- tools/perf/util/python.c | 2604 ++++++++++++++++--- tools/perf/util/record.c | 11 +- tools/perf/util/s390-sample-raw.c | 19 +- tools/perf/util/sample-raw.c | 4 +- tools/perf/util/sample.c | 17 +- tools/perf/util/session.c | 61 +- tools/perf/util/session.h | 2 + tools/perf/util/setup.py | 5 + tools/perf/util/sideband_evlist.c | 40 +- tools/perf/util/sort.c | 2 +- tools/perf/util/stat-display.c | 6 +- tools/perf/util/stat-shadow.c | 24 +- tools/perf/util/stat.c | 20 +- tools/perf/util/stream.c | 4 +- tools/perf/util/synthetic-events.c | 11 +- tools/perf/util/time-utils.c | 12 +- tools/perf/util/top.c | 4 +- 113 files changed, 4380 insertions(+), 1431 deletions(-) create mode 100644 tools/perf/python/perf.pyi create mode 100755 tools/perf/python/perf_live.py
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 1/19 perf util: Sort includes and add missed explicit dependencies
Fix missing includes found while cleaning the evsel/evlist header files. Sort the remaining header files for consistency with the rest of the code.
--- tools/perf/util/bpf_off_cpu.c | 30 +++++----- tools/perf/util/bpf_trace_augment.c | 8 +-- tools/perf/util/evlist.c | 91 +++++++++++++++-------------- tools/perf/util/map.h | 9 ++- tools/perf/util/perf_api_probe.c | 18 +++--- tools/perf/util/s390-sample-raw.c | 19 +++--- tools/perf/util/stat-shadow.c | 20 ++++--- tools/perf/util/stat.c | 16 +++-- 8 files changed, 113 insertions(+), 98 deletions(-)
tools/perf/util/bpf_off_cpu.c +16 -14
@@ -1,23 +1,25 @@// SPDX-License-Identifier: GPL-2.0 -#include "util/bpf_counter.h" -#include "util/debug.h" -#include "util/evsel.h" -#include "util/evlist.h" -#include "util/off_cpu.h" -#include "util/perf-hooks.h" -#include "util/record.h" -#include "util/session.h" -#include "util/target.h" -#include "util/cpumap.h" -#include "util/thread_map.h" -#include "util/cgroup.h" -#include "util/strlist.h" +#include <linux/time64.h> +#include <bpf/bpf.h> #include <bpf/btf.h> #include <internal/xyarray.h> -#include <linux/time64.h> +#include "bpf_counter.h" #include "bpf_skel/off_cpu.skel.h" +#include "cgroup.h" +#include "cpumap.h" +#include "debug.h" +#include "evlist.h" +#include "evsel.h" +#include "off_cpu.h" +#include "parse-events.h" +#include "perf-hooks.h" +#include "record.h" +#include "session.h" +#include "strlist.h" +#include "target.h" +#include "thread_map.h" #define MAX_STACKS 32#define MAX_PROC 4096
tools/perf/util/bpf_trace_augment.c +4 -4
@@ -1,11 +1,11 @@#include <bpf/libbpf.h> #include <internal/xyarray.h> -#include "util/debug.h" -#include "util/evlist.h" -#include "util/trace_augment.h" -#include "bpf_skel/augmented_raw_syscalls.skel.h" +#include "debug.h" +#include "evlist.h" +#include "parse-events.h" +#include "trace_augment.h" static struct augmented_raw_syscalls_bpf *skel;static struct evsel *bpf_output;
tools/perf/util/evlist.c +46 -45
113 lines not on this page · raw message
tools/perf/util/map.h +4 -5
@@ -2,14 +2,13 @@#ifndef __PERF_MAP_H#define __PERF_MAP_H -#include <linux/refcount.h> -#include <linux/compiler.h> -#include <linux/list.h> -#include <linux/rbtree.h> +#include <stdbool.h> #include <stdio.h> #include <string.h> -#include <stdbool.h> + +#include <linux/refcount.h> #include <linux/types.h> +#include <internal/rc_check.h> struct dso ;
tools/perf/util/perf_api_probe.c +11 -7
@@ -1,14 +1,18 @@/* SPDX-License-Identifier: GPL-2.0 */ +#include "perf_api_probe.h" -#include "perf-sys.h" -#include "util/cloexec.h" -#include "util/evlist.h" -#include "util/evsel.h" -#include "util/parse-events.h" -#include "util/perf_api_probe.h" -#include <perf/cpumap.h> #include <errno.h> +#include <perf/cpumap.h> + +#include "cloexec.h" +#include "evlist.h" +#include "evsel.h" +#include "parse-events.h" +#include "perf-sys.h" +#include "pmu.h" +#include "pmus.h" +typedef void (*setup_probe_fn_t )(struct evsel *evsel);static int perf_do_probe_api (setup_probe_fn_t fn,struct perf_cpu cpu,const char *str)
tools/perf/util/s390-sample-raw.c +10 -9
@@ -12,25 +12,26 @@* sample was taken from. */ -#include <unistd.h> +#include <inttypes.h> #include <stdio.h> #include <string.h> -#include <inttypes.h> -#include <sys/stat.h> +#include <asm/byteorder.h> #include <linux/compiler.h> #include <linux/err.h> -#include <asm/byteorder.h> +#include <sys/stat.h> +#include <unistd.h> +#include "color.h" #include "debug.h" -#include "session.h" #include "evlist.h" -#include "color.h" #include "hashmap.h" -#include "sample-raw.h" +#include "pmu.h" +#include "pmus.h" #include "s390-cpumcf-kernel.h" -#include "util/pmu.h" -#include "util/sample.h" +#include "sample-raw.h" +#include "sample.h" +#include "session.h" static size_t ctrset_size (struct cf_ctrset_entry *set) {
tools/perf/util/stat-shadow.c +12 -8
@@ -2,20 +2,24 @@#include <errno.h> #include <math.h> #include <stdio.h> -#include "evsel.h" -#include "stat.h" + +#include <linux/zalloc.h> + +#include "cgroup.h" #include "color.h" #include "debug.h" -#include "pmu.h" -#include "rblist.h" #include "evlist.h" +#include "evsel.h" #include "expr.h" -#include "metricgroup.h" -#include "cgroup.h" -#include "units.h" +#include "hashmap.h" #include "iostat.h" -#include "util/hashmap.h" +#include "metricgroup.h" +#include "pmu.h" +#include "pmus.h" +#include "rblist.h" +#include "stat.h" #include "tool_pmu.h" +#include "units.h" static bool tool_pmu__is_time_event (const struct perf_stat_config *config,const struct evsel *evsel,int *tool_aggr_idx)
tools/perf/util/stat.c +10 -6
@@ -1,21 +1,25 @@// SPDX-License-Identifier: GPL-2.0 +#include "stat.h" +#include <errno.h> -#include <linux/err.h> #include <inttypes.h> #include <math.h> #include <string.h> + +#include <linux/err.h> +#include <linux/zalloc.h> +#include "counts.h" #include "cpumap.h" #include "debug.h" +#include "evlist.h" +#include "evsel.h" +#include "hashmap.h" #include "header.h" -#include "stat.h" +#include "pmu.h" #include "session.h" #include "target.h" -#include "evlist.h" -#include "evsel.h" #include "thread_map.h" -#include "util/hashmap.h" -#include <linux/zalloc.h> void update_stats (struct stats *stats,u64 val) {
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 2/19 perf python: Add missed explicit dependencies
In reply to Ian RogersFix missing #include of pmus.h found while cleaning the evsel/evlist header files.
Also define PY_SSIZE_T_CLEAN before including Python.h to comply with modern Python 3 C-API requirements, and introduce CHECK_INITIALIZED and CHECK_INITIALIZED_INT safety macros to be used by subsequent patches.
--- tools/perf/util/python.c | 41 +++++++++++++++++++++++++++++++--------- 1 file changed, 32 insertions(+), 9 deletions(-)
tools/perf/util/python.c +32 -9
@@ -1,30 +1,37 @@// SPDX-License-Identifier: GPL-2.0 +#define PY_SSIZE_T_CLEAN #include <Python.h> -#include <structmember.h> +#include <inttypes.h> -#include <poll.h> +#include <linux/err.h> +#include <poll.h> + +#include <internal/lib.h> #include <perf/cpumap.h> -#ifdef HAVE_LIBTRACEEVENT -#include <event-parse.h> -#endif #include <perf/mmap.h> +#include <structmember.h> +#include "callchain.h" #include "counts.h" +#include "event.h" #include "evlist.h" #include "evsel.h" -#include "event.h" #include "expr.h" +#include "metricgroup.h" +#include "mmap.h" +#include "pmus.h" #include "print_binary.h" #include "record.h" #include "strbuf.h" #include "thread_map.h" #include "tp_pmu.h" #include "trace-event.h" -#include "metricgroup.h" -#include "mmap.h" #include "util/sample.h" -#include <internal/lib.h> + +#ifdef HAVE_LIBTRACEEVENT +#include <event-parse.h> +#endif PyMODINIT_FUNC PyInit_perf (void );
@@ -38,6 +45,22 @@ PyMODINIT_FUNC PyInit_perf(void); offsetof(struct pyrf_event , sample) + offsetof(struct perf_sample , member), \ 0, help } +#define CHECK_INITIALIZED (ptr, msg) \ + do { \ + if (!(ptr)) { \ + PyErr_SetString(PyExc_ValueError, msg " not initialized"); \ + return NULL; \ + } \ + } while (0) + +#define CHECK_INITIALIZED_INT (ptr, msg) \ + do { \ + if (!(ptr)) { \ + PyErr_SetString(PyExc_ValueError, msg " not initialized"); \ + return -1; \ + } \ + } while (0) +struct pyrf_event {PyObject_HEAD struct evsel *evsel;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 3/19 perf evsel/evlist: Avoid unnecessary #includes
In reply to Ian RogersUse forward declarations and remove unnecessary #includes in evsel.h. Sort the forward declarations in evsel.h and evlist.h. Move some PMU code into evsel.c.
--- tools/perf/util/evlist.h | 15 +++++++++------ tools/perf/util/evsel.c | 5 +++++ tools/perf/util/evsel.h | 25 ++++++++++++------------- 3 files changed, 26 insertions(+), 19 deletions(-)
tools/perf/util/evlist.h +9 -6
@@ -2,29 +2,32 @@#ifndef __PERF_EVLIST_H#define __PERF_EVLIST_H 1 +#include <signal.h> +#include <linux/compiler.h> #include <linux/kernel.h> -#include <linux/refcount.h> #include <linux/list.h> +#include <linux/refcount.h> +#include <pthread.h> +#include <unistd.h> +#include <api/fd/array.h> #include <internal/evlist.h> #include <internal/evsel.h> #include <perf/evlist.h> +#include "affinity.h" #include "events_stats.h" #include "evsel.h" #include "rblist.h" -#include <pthread.h> -#include <signal.h> -#include <unistd.h> -struct pollfd ; -struct thread_map ;struct perf_cpu_map ;struct perf_stat_config ; +struct pollfd ;struct record_opts ;struct strbuf ;struct target ; +struct thread_map ;/* * State machine of bkw_mmap_state:
tools/perf/util/evsel.c +5 -0
@@ -370,6 +370,11 @@ void evsel__set_sample_id(struct evsel *evsel, evsel->core.attr.read_format |=PERF_FORMAT_ID ; } +bool evsel__is_non_perf_event_open_pmu (const struct evsel *evsel) +{ +return evsel->pmu && evsel->pmu->type >PERF_PMU_TYPE_PE_END ; +} +/** * evsel__is_function_event - Return whether given evsel is a function * trace event
tools/perf/util/evsel.h +12 -13
@@ -2,28 +2,30 @@#ifndef __PERF_EVSEL_H#define __PERF_EVSEL_H 1 -#include <linux/list.h> #include <stdbool.h> -#include <sys/types.h> + +#include <linux/list.h> #include <linux/perf_event.h> #include <linux/types.h> +#include <sys/types.h> +#include <internal/evsel.h> #include <perf/evsel.h> +#include "symbol_conf.h" -#include "pmus.h" -#include "pmu.h" +struct bperf_follower_bpf ; +struct bperf_leader_bpf ; +struct bpf_counter_ops ;struct bpf_object ;struct cgroup ; +struct hashmap ;struct perf_counts ; +struct perf_pmu ;struct perf_stat_config ;struct perf_stat_evsel ; -union perf_event ; -struct bpf_counter_ops ;struct target ; -struct hashmap ; -struct bperf_leader_bpf ; -struct bperf_follower_bpf ; +union perf_event ;typedef int (evsel__sb_cb_t )(union perf_event *event,void *data);
@@ -356,10 +358,6 @@ void arch_evsel__apply_ratio_to_prev(struct evsel *evsel, struct perf_event_attrint evsel__set_filter (struct evsel *evsel,const char *filter);int evsel__append_tp_filter (struct evsel *evsel,const char *filter);int evsel__append_addr_filter (struct evsel *evsel,const char *filter); -static inline bool evsel__is_non_perf_event_open_pmu (const struct evsel *evsel) -{ -return evsel->pmu && evsel->pmu->type >PERF_PMU_TYPE_PE_END ; -}int evsel__enable_cpu (struct evsel *evsel,int cpu_map_idx);int evsel__enable (struct evsel *evsel);
@@ -478,6 +476,7 @@ static inline bool evsel__is_group_event(struct evsel *evsel)return evsel__is_group_leader (evsel) && evsel->core.nr_members >1 ; } +bool evsel__is_non_perf_event_open_pmu (const struct evsel *evsel);bool evsel__is_function_event (struct evsel *evsel);static inline bool evsel__is_bpf_output (struct evsel *evsel)
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 4/19 perf data: Add open flag
In reply to Ian RogersAvoid double opens and ensure only open files are closed. This addresses some issues with python integration where the data file wants to be opened before being given to a session.
---
Changes in v2:
1. Fixed File Rotation: In perf_data__switch() , I added data->open = false; after the file is closed. This ensures that the subsequent perf_data__open() call will not exit early and will successfully open the new file.
2. Fixed Memory Leak: In open_dir() , I added a call to zfree(&data->file.path) if mkdir() fails, preventing the leak of the path string.
--- tools/perf/util/data.c | 27 ++++++++++++++++++++++----- tools/perf/util/data.h | 4 +++- 2 files changed, 25 insertions(+), 6 deletions(-)
tools/perf/util/data.c +22 -5
@@ -345,9 +345,10 @@ static int open_dir(struct perf_data *data)if (asprintf (&data->file.path,"%s/data" , data->path) <0 )return -1 ; -if (perf_data__is_write (data) && -mkdir (data->path,S_IRWXU ) <0 ) +if (perf_data__is_write (data) &&mkdir (data->path,0700 ) <0 ) { +zfree (&data->file.path);return -1 ; + } ret =open_file (data);
@@ -360,9 +361,16 @@ static int open_dir(struct perf_data *data)int perf_data__open (struct perf_data *data) { -if (check_pipe (data)) +int ret; + +if (data->open)return 0 ; +if (check_pipe (data)) { + data->open = true; +return 0 ; + } +/* currently it allows stdio for pipe only */ data->file.use_stdio = false;
@@ -375,16 +383,24 @@ int perf_data__open(struct perf_data *data)if (perf_data__is_read (data)) data->is_dir =is_dir (data); -return perf_data__is_dir (data) ? -open_dir (data) :open_file_dup (data); + ret =perf_data__is_dir (data) ?open_dir (data) :open_file_dup (data); + +if (!ret) + data->open = true; + +return ret; }void perf_data__close (struct perf_data *data) { +if (!data->open) +return ; +if (perf_data__is_dir (data))perf_data__close_dir (data);perf_data_file__close (&data->file); + data->open = false; }static ssize_t perf_data_file__read (struct perf_data_file *file,void *buf,size_t size)
@@ -457,6 +473,7 @@ int perf_data__switch(struct perf_data *data,if (!at_exit) {perf_data_file__close (&data->file); + data->open = false; ret =perf_data__open (data);if (ret <0 ) goto out;
tools/perf/util/data.h +3 -1
@@ -50,6 +50,8 @@ struct perf_data {const char *path;/** @file: Underlying file to be used. */ struct perf_data_file file; +/** @open: Has the file or directory been opened. */ +bool open;/** @is_pipe: Underlying file is a pipe. */ bool is_pipe;/** @is_dir: Underlying file is a directory. */
@@ -59,7 +61,7 @@ struct perf_data {/** @in_place_update: A file opened for reading but will be written to. */ bool in_place_update;/** @mode: Read or write mode. */ -enum perf_data_mode mode; + enum perf_data_mode mode:8;struct {/** @version: perf_dir_version. */
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 04/19] perf data: Add open flag
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Pre-existing issues:
- [High] If rename() fails during log rotation in perf_data__switch(), the original data file is truncated and its contents are lost. --
commit c62907f96616fda36f9afb612d87efd788da81f6
Author: Ian Rogers <[email protected]>
perf data: Add open flag
This commit adds an open flag to struct perf_data to avoid double opens and track the open state of the file or directory. This helps resolve python integration issues where the data file is opened before being passed to a session.
Quoted 4 lines
tools/perf/util/data.c +0
-0
[ ... ]
tools/perf/util/data.c int perf_data__switch(struct perf_data *data, · line 477
Quoted 6 lines
tools/perf/util/data.c +1
-0@@ -457,6 +473,7 @@ int perf_data__switch(struct perf_data *data,if (!at_exit) {perf_data_file__close (&data->file); + data->open = false; ret =perf_data__open (data);
[Severity: High]
This is a pre-existing issue, but does reopening the file here truncate the existing data if the preceding rename() fails?
In perf_data__switch(), if rename() fails during log rotation, a warning is printed and the code falls through with the intent to "continue fill original file":
if (rename(data->path, *new_filepath))
pr_warning("Failed to rename %s to %s\n", ...);However, the file is then closed and reopened via perf_data__open(). Because the file is opened for writing, the call chain is:perf_data__open() -> open_file_dup() -> open_file() -> open_file_write(), which unconditionally opens the file with the O_TRUNC flag.
Since the original file was not successfully renamed and remains at data->path, doesn't reopening it with O_TRUNC completely wipe out all previously recorded profiling data instead of preserving it?
Quoted 2 lines
tools/perf/util/data.c +0
-0if (ret <0 ) goto out;
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=4
Ian Rogers
reviewerPATCH v11 5/19 perf evlist: Add reference count
In reply to Ian RogersThis a no-op for most of the perf tool. The reference count is set to 1 at allocation, the put will see the 1, decrement it and perform the delete. The purpose for adding the reference count is for the python code. Prior to this change the python code would clone evlists, but this has issues if events are opened, etc. This change adds a reference count for the evlists and a later change will add it to evsels. The combination is needed for the python code to operate correctly (not hit asserts in the evsel clone), but the changes are broken apart for the sake of smaller patches.
---
v2: Added evlist__put to pyrf_evlist__init in case init is called more
than once.
I double-checked trace__replay() and confirmed that trace->evlist
is not assigned to session->evlist in that function.
trace__replay creates a new session and uses its own evlist for
processing file events, leaving trace->evlist pointing to the
empty list created at startup. Therefore, the
evlist__put(trace->evlist) call in trace__exit() is safe and
correct to avoid leaking that empty list.v7:
- Added pyrf_evlist__new to zero-initialize pevlist->evlist to fix crash on re-initialization.
--- tools/perf/arch/x86/tests/hybrid.c | 2 +- tools/perf/arch/x86/tests/topdown.c | 4 +- tools/perf/arch/x86/util/iostat.c | 2 +- tools/perf/bench/evlist-open-close.c | 18 +- tools/perf/builtin-ftrace.c | 8 +- tools/perf/builtin-kvm.c | 4 +- tools/perf/builtin-lock.c | 2 +- tools/perf/builtin-record.c | 4 +- tools/perf/builtin-sched.c | 6 +- tools/perf/builtin-script.c | 2 +- tools/perf/builtin-stat.c | 10 +- tools/perf/builtin-top.c | 52 ++--- tools/perf/builtin-trace.c | 26 +-- tools/perf/tests/backward-ring-buffer.c | 18 +- tools/perf/tests/code-reading.c | 4 +- tools/perf/tests/event-times.c | 4 +- tools/perf/tests/event_update.c | 2 +- tools/perf/tests/evsel-roundtrip-name.c | 8 +- tools/perf/tests/expand-cgroup.c | 8 +- tools/perf/tests/hists_cumulate.c | 2 +- tools/perf/tests/hists_filter.c | 2 +- tools/perf/tests/hists_link.c | 2 +- tools/perf/tests/hists_output.c | 2 +- tools/perf/tests/hwmon_pmu.c | 2 +- tools/perf/tests/keep-tracking.c | 2 +- tools/perf/tests/mmap-basic.c | 18 +- tools/perf/tests/openat-syscall-tp-fields.c | 18 +- tools/perf/tests/parse-events.c | 4 +- tools/perf/tests/parse-metric.c | 4 +- tools/perf/tests/parse-no-sample-id-all.c | 2 +- tools/perf/tests/perf-record.c | 18 +- tools/perf/tests/perf-time-to-tsc.c | 2 +- tools/perf/tests/pfm.c | 4 +- tools/perf/tests/pmu-events.c | 6 +- tools/perf/tests/pmu.c | 4 +- tools/perf/tests/sw-clock.c | 14 +- tools/perf/tests/switch-tracking.c | 2 +- tools/perf/tests/task-exit.c | 14 +- tools/perf/tests/tool_pmu.c | 2 +- tools/perf/tests/topology.c | 2 +- tools/perf/tests/uncore-event-sorting.c | 2 +- tools/perf/util/cgroup.c | 4 +- tools/perf/util/data-convert-bt.c | 2 +- tools/perf/util/evlist.c | 20 +- tools/perf/util/evlist.h | 7 +- tools/perf/util/expr.c | 2 +- tools/perf/util/header.c | 22 +- tools/perf/util/metricgroup.c | 6 +- tools/perf/util/parse-events.c | 4 +- tools/perf/util/perf_api_probe.c | 2 +- tools/perf/util/python.c | 242 +++++++++----------- tools/perf/util/record.c | 2 +- tools/perf/util/session.c | 2 +- tools/perf/util/sideband_evlist.c | 16 +- 54 files changed, 315 insertions(+), 328 deletions(-)
tools/perf/arch/x86/tests/hybrid.c +1 -1
@@ -268,7 +268,7 @@ static int test_event(const struct evlist_test *e) ret = e->check (evlist); }parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/arch/x86/tests/topdown.c +2 -2
@@ -56,7 +56,7 @@ static int event_cb(void *state, struct pmu_event_info *info) *ret =TEST_FAIL ; } } -evlist__delete (evlist); +evlist__put (evlist);return 0 ; }
@@ -174,7 +174,7 @@ static int test_sort(const char *str, int expected_slots_group_size,CHECK_COND (slots_seen,"slots seen" ); ret =TEST_OK ; out_err: -evlist__delete (evlist); +evlist__put (evlist);parse_events_error__exit (&err);return ret; }
tools/perf/arch/x86/util/iostat.c +1 -1
@@ -337,7 +337,7 @@ int iostat_prepare(struct evlist *evlist, struct perf_stat_config *config)if (evlist->core.nr_entries >0 ) {pr_warning ("The -e and -M options are not supported." "All chosen events/metrics will be dropped\n" ); -evlist__delete (evlist); +evlist__put (evlist); evlist =evlist__new ();if (!evlist)return -ENOMEM ;
tools/perf/bench/evlist-open-close.c +9 -9
@@ -76,7 +76,7 @@ static struct evlist *bench__create_evlist(char *evstr, const char *uid_str)parse_events_error__exit (&err);pr_err ("Run 'perf list' for a list of valid events\n" ); ret =1 ; - goto out_delete_evlist; + goto out_put_evlist; }parse_events_error__exit (&err);if (uid_str) {
@@ -85,24 +85,24 @@ static struct evlist *bench__create_evlist(char *evstr, const char *uid_str)if (uid ==UINT_MAX ) {pr_err ("Invalid User: %s" , uid_str); ret = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; } ret =parse_uid_filter (evlist, uid);if (ret) - goto out_delete_evlist; + goto out_put_evlist; } ret =evlist__create_maps (evlist, &opts.target);if (ret <0 ) {pr_err ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }evlist__config (evlist, &opts,NULL );return evlist; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist);return NULL ; }
@@ -151,7 +151,7 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str) evlist->core.nr_entries, evlist__count_evsel_fds(evlist));printf (" Number of iterations:\t%d\n" , iterations); -evlist__delete (evlist); +evlist__put (evlist);for (i =0 ; i < iterations; i++) {pr_debug ("Started iteration %d\n" , i);
@@ -162,7 +162,7 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str)gettimeofday (&start,NULL ); err =bench__do_evlist_open_close (evlist);if (err) { -evlist__delete (evlist); +evlist__put (evlist);return err; }
@@ -171,7 +171,7 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str) runtime_us =timeval2usec (&diff);update_stats (&time_stats, runtime_us); -evlist__delete (evlist); +evlist__put (evlist);pr_debug ("Iteration %d took:\t%" PRIu64"us\n" , i, runtime_us); }
tools/perf/builtin-ftrace.c +4 -4
@@ -1999,20 +1999,20 @@ int cmd_ftrace(int argc, const char **argv) ret =evlist__create_maps (ftrace.evlist, &ftrace.target);if (ret <0 ) - goto out_delete_evlist; + goto out_put_evlist;if (argc) { ret =evlist__prepare_workload (ftrace.evlist, &ftrace.target, argv, false, ftrace__workload_exec_failed_signal);if (ret <0 ) - goto out_delete_evlist; + goto out_put_evlist; } ret =cmd_func (&ftrace); -out_delete_evlist: -evlist__delete (ftrace.evlist); +out_put_evlist: +evlist__put (ftrace.evlist); out_delete_filters:delete_filter_func (&ftrace.filters);
tools/perf/builtin-kvm.c +2 -2
@@ -1810,7 +1810,7 @@ static struct evlist *kvm_live_event_list(void) out:if (err) { -evlist__delete (evlist); +evlist__put (evlist); evlist =NULL ; }
@@ -1941,7 +1941,7 @@ static int kvm_events_live(struct perf_kvm_stat *kvm, out:perf_session__delete (kvm->session); kvm->session =NULL ; -evlist__delete (kvm->evlist); +evlist__put (kvm->evlist);return err; }
tools/perf/builtin-lock.c +1 -1
@@ -2149,7 +2149,7 @@ static int __cmd_contention(int argc, const char **argv) out_delete:lock_filter_finish (); -evlist__delete (con.evlist); +evlist__put (con.evlist);lock_contention_finish (&con);perf_session__delete (session);perf_env__exit (&host_env);
tools/perf/builtin-record.c +2 -2
@@ -4290,7 +4290,7 @@ int cmd_record(int argc, const char **argv) goto out;evlist__splice_list_tail (rec->evlist, &def_evlist->core.entries); -evlist__delete (def_evlist); +evlist__put (def_evlist); }if (rec->opts.target.tid && !rec->opts.no_inherit_set)
@@ -4400,7 +4400,7 @@ int cmd_record(int argc, const char **argv)auxtrace_record__free (rec->itr); out_opts:evlist__close_control (rec->opts.ctl_fd, rec->opts.ctl_fd_ack, &rec->opts.ctl_fd_close); -evlist__delete (rec->evlist); +evlist__put (rec->evlist);return err; }
tools/perf/builtin-sched.c +3 -3
@@ -3848,7 +3848,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched, session =perf_session__new (&data, &sched->tool);if (IS_ERR (session)) {pr_err ("Perf session creation failed.\n" ); -evlist__delete (evlist); +evlist__put (evlist);return PTR_ERR (session); }
@@ -3947,7 +3947,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched,else fprintf (stderr,"[ perf sched stats: Failed !! ]\n" ); -evlist__delete (evlist); +evlist__put (evlist);close (fd);return err; }
@@ -4851,7 +4851,7 @@ static int perf_sched__schedstat_live(struct perf_sched *sched,free_cpu_domain_info (cd_map, sv, nr); out:free_schedstat (&cpu_head); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/builtin-script.c +1 -1
@@ -2274,7 +2274,7 @@ static int script_find_metrics(const struct pmu_metric *pm, }pr_debug ("Found metric '%s' whose evsels match those of in the perf data\n" , pm->metric_name); -evlist__delete (metric_evlist); +evlist__put (metric_evlist); out:return 0 ; }
tools/perf/builtin-stat.c +5 -5
@@ -2113,7 +2113,7 @@ static int add_default_events(void) stat_config.user_requested_cpu_list, stat_config.system_wide, stat_config.hardware_aware_grouping) < 0) { -evlist__delete (metric_evlist); +evlist__put (metric_evlist); ret =-1 ;break ; }
@@ -2125,7 +2125,7 @@ static int add_default_events(void)metricgroup__copy_metric_events (evlist,/*cgrp=*/ NULL , &evlist->metric_events, &metric_evlist->metric_events); -evlist__delete (metric_evlist); +evlist__put (metric_evlist); }list_sort (/*priv=*/ NULL , &evlist->core.entries, default_evlist_evsel_cmp);
@@ -2146,7 +2146,7 @@ static int add_default_events(void)metricgroup__copy_metric_events (evsel_list,/*cgrp=*/ NULL , &evsel_list->metric_events, &evlist->metric_events); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -2381,7 +2381,7 @@ static int __cmd_report(int argc, const char **argv) perf_stat.session = session; stat_config.output = stderr; -evlist__delete (evsel_list); +evlist__put (evsel_list); evsel_list = session->evlist; ret =perf_session__process_events (session);
@@ -3060,7 +3060,7 @@ int cmd_stat(int argc, const char **argv)if (smi_cost && smi_reset)sysfs__write_int (FREEZE_ON_SMI_PATH ,0 ); -evlist__delete (evsel_list); +evlist__put (evsel_list);evlist__close_control (stat_config.ctl_fd, stat_config.ctl_fd_ack, &stat_config.ctl_fd_close);
tools/perf/builtin-top.c +26 -26
184 lines not on this page · raw message
tools/perf/builtin-trace.c +13 -13
@@ -4418,7 +4418,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (trace->summary_bpf) {if (trace_prepare_bpf_summary (trace->summary_mode) <0 ) - goto out_delete_evlist; + goto out_put_evlist;if (trace->summary_only) goto create_maps;
@@ -4486,19 +4486,19 @@ static int trace__run(struct trace *trace, int argc, const char **argv) err =evlist__create_maps (evlist, &trace->opts.target);if (err <0 ) {fprintf (trace->output,"Problems parsing the target to trace, check your options!\n" ); - goto out_delete_evlist; + goto out_put_evlist; } err =trace__symbols_init (trace, argc, argv, evlist);if (err <0 ) {fprintf (trace->output,"Problems initializing symbol libraries!\n" ); - goto out_delete_evlist; + goto out_put_evlist; }if (trace->summary_mode ==SUMMARY__BY_TOTAL && !trace->summary_bpf) { trace->syscall_stats =alloc_syscall_stats ();if (!trace->syscall_stats) - goto out_delete_evlist; + goto out_put_evlist; }evlist__config (evlist, &trace->opts, &callchain_param);
@@ -4507,7 +4507,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) err =evlist__prepare_workload (evlist, &trace->opts.target, argv, false,NULL );if (err <0 ) {fprintf (trace->output,"Couldn't run the workload!\n" ); - goto out_delete_evlist; + goto out_put_evlist; } workload_pid = evlist->workload.pid; }
@@ -4555,7 +4555,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) err =trace__expand_filters (trace, &evsel);if (err) - goto out_delete_evlist; + goto out_put_evlist; err =evlist__apply_filters (evlist, &evsel, &trace->opts.target);if (err <0 ) goto out_error_apply_filters;
@@ -4672,12 +4672,12 @@ static int trace__run(struct trace *trace, int argc, const char **argv) } } -out_delete_evlist: +out_put_evlist:trace_cleanup_bpf_summary ();delete_syscall_stats (trace->syscall_stats);trace__symbols__exit (trace);evlist__free_syscall_tp_fields (evlist); -evlist__delete (evlist); +evlist__put (evlist);cgroup__put (trace->cgroup); trace->evlist =NULL ; trace->live = false;
@@ -4702,21 +4702,21 @@ static int trace__run(struct trace *trace, int argc, const char **argv) out_error:fprintf (trace->output,"%s\n" , errbuf); - goto out_delete_evlist; + goto out_put_evlist; out_error_apply_filters:fprintf (trace->output,"Failed to set filter \"%s\" on event %s: %m\n" , evsel->filter,evsel__name (evsel)); - goto out_delete_evlist; + goto out_put_evlist; } out_error_mem:fprintf (trace->output,"Not enough memory to run!\n" ); - goto out_delete_evlist; + goto out_put_evlist; out_errno:fprintf (trace->output,"%m\n" ); - goto out_delete_evlist; + goto out_put_evlist; }static int trace__replay (struct trace *trace)
@@ -5396,7 +5396,7 @@ static void trace__exit(struct trace *trace)zfree (&trace->syscalls.table); }zfree (&trace->perfconfig_events); -evlist__delete (trace->evlist); +evlist__put (trace->evlist); trace->evlist =NULL ;ordered_events__free (&trace->oe.data);#ifdef HAVE_LIBBPF_SUPPORT
tools/perf/tests/backward-ring-buffer.c +9 -9
@@ -111,7 +111,7 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, in err =evlist__create_maps (evlist, &opts.target);if (err <0 ) {pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }parse_events_error__init (&parse_error);
@@ -124,7 +124,7 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, inif (err) {pr_debug ("Failed to parse tracepoint event, try use root\n" ); ret =TEST_SKIP ; - goto out_delete_evlist; + goto out_put_evlist; }evlist__config (evlist, &opts,NULL );
@@ -133,19 +133,19 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, inif (err <0 ) {pr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } ret =TEST_FAIL ; err =do_test (evlist, opts.mmap_pages, &sample_count, &comm_count);if (err !=TEST_OK ) - goto out_delete_evlist; + goto out_put_evlist;if ((sample_count !=NR_ITERS ) || (comm_count !=NR_ITERS )) {pr_err ("Unexpected counter: sample_count=%d, comm_count=%d\n" , sample_count, comm_count); - goto out_delete_evlist; + goto out_put_evlist; }evlist__close (evlist);
@@ -154,16 +154,16 @@ static int test__backward_ring_buffer(struct test_suite *test __maybe_unused, inif (err <0 ) {pr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } err =do_test (evlist,1 , &sample_count, &comm_count);if (err !=TEST_OK ) - goto out_delete_evlist; + goto out_put_evlist; ret =TEST_OK ; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist);return ret; }
tools/perf/tests/code-reading.c +2 -2
@@ -807,7 +807,7 @@ static int do_test_code_reading(bool try_kcore) }perf_evlist__set_maps (&evlist->core,NULL ,NULL ); -evlist__delete (evlist); +evlist__put (evlist); evlist =NULL ;continue ; }
@@ -844,7 +844,7 @@ static int do_test_code_reading(bool try_kcore) out_put:thread__put (thread); out_err: -evlist__delete (evlist); +evlist__put (evlist);perf_cpu_map__put (cpus);perf_thread_map__put (threads);machine__delete (machine);
tools/perf/tests/event-times.c +2 -2
@@ -186,7 +186,7 @@ static int test_times(int (attach)(struct evlist *), err =attach (evlist);if (err ==TEST_SKIP ) {pr_debug (" SKIP : not enough rights\n" ); -evlist__delete (evlist); +evlist__put (evlist);return err; }
@@ -205,7 +205,7 @@ static int test_times(int (attach)(struct evlist *), count.ena, count.run); out_err: -evlist__delete (evlist); +evlist__put (evlist);return !err ?TEST_OK :TEST_FAIL ; }
tools/perf/tests/event_update.c +1 -1
@@ -117,7 +117,7 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtesTEST_ASSERT_VAL ("failed to synthesize attr update cpus" , !perf_event__synthesize_event_update_cpus (&tmp.tool, evsel, process_event_cpus)); -evlist__delete (evlist); +evlist__put (evlist);return 0 ; }
tools/perf/tests/evsel-roundtrip-name.c +4 -4
@@ -33,7 +33,7 @@ static int perf_evsel__roundtrip_cache_name_test(void)if (err) {pr_debug ("Failure to parse cache event '%s' possibly as PMUs don't support it" , name); -evlist__delete (evlist); +evlist__put (evlist);continue ; }evlist__for_each_entry (evlist, evsel) {
@@ -42,7 +42,7 @@ static int perf_evsel__roundtrip_cache_name_test(void) ret =TEST_FAIL ; } } -evlist__delete (evlist); +evlist__put (evlist); } } }
@@ -66,7 +66,7 @@ static int perf_evsel__name_array_test(const char *const names[], int nr_names)if (err) {pr_debug ("failed to parse event '%s', err %d\n" , names[i], err); -evlist__delete (evlist); +evlist__put (evlist); ret =TEST_FAIL ;continue ; }
@@ -76,7 +76,7 @@ static int perf_evsel__name_array_test(const char *const names[], int nr_names) ret =TEST_FAIL ; } } -evlist__delete (evlist); +evlist__put (evlist); }return ret; }
tools/perf/tests/expand-cgroup.c +4 -4
@@ -106,7 +106,7 @@ static int expand_default_events(void)TEST_ASSERT_VAL ("failed to get evlist" , evlist); ret =test_expand_events (evlist); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -133,7 +133,7 @@ static int expand_group_events(void) ret =test_expand_events (evlist); out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -164,7 +164,7 @@ static int expand_libpfm_events(void) ret =test_expand_events (evlist); out: -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -188,7 +188,7 @@ static int expand_metric_events(void) ret =test_expand_events (evlist); out: -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/hists_cumulate.c +1 -1
@@ -743,7 +743,7 @@ static int test__hists_cumulate(struct test_suite *test __maybe_unused, int subt out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hists_filter.c +1 -1
@@ -331,7 +331,7 @@ static int test__hists_filter(struct test_suite *test __maybe_unused, int subtes out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);reset_output_field ();machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hists_link.c +1 -1
@@ -352,7 +352,7 @@ static int test__hists_link(struct test_suite *test __maybe_unused, int subtest out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);reset_output_field ();machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hists_output.c +1 -1
@@ -630,7 +630,7 @@ static int test__hists_output(struct test_suite *test __maybe_unused, int subtes out:/* tear down everything */ -evlist__delete (evlist); +evlist__put (evlist);machines__exit (&machines);put_fake_samples ();
tools/perf/tests/hwmon_pmu.c +1 -1
@@ -215,7 +215,7 @@ static int do_test(size_t i, bool with_pmu, bool with_alias) out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/keep-tracking.c +1 -1
@@ -153,7 +153,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte out_err:if (evlist) {evlist__disable (evlist); -evlist__delete (evlist); +evlist__put (evlist); }perf_cpu_map__put (cpus);perf_thread_map__put (threads);
tools/perf/tests/mmap-basic.c +9 -9
@@ -94,7 +94,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest/* Permissions failure, flag the failure as a skip. */ err =TEST_SKIP ; } - goto out_delete_evlist; + goto out_put_evlist; } evsels[i]->core.attr.wakeup_events =1 ;
@@ -106,7 +106,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestpr_debug ("failed to open counter: %s, " "tweak /proc/sys/kernel/perf_event_paranoid?\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } nr_events[i] =0 ;
@@ -116,7 +116,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (evlist__mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }for (i =0 ; i < nsyscalls; ++i)
@@ -134,7 +134,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (event->header.type !=PERF_RECORD_SAMPLE ) {pr_debug ("unexpected %s event\n" ,perf_event__name (event->header.type)); - goto out_delete_evlist; + goto out_put_evlist; }perf_sample__init (&sample,/*all=*/ false);
@@ -142,7 +142,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (err) {pr_err ("Can't parse sample, err = %d\n" , err);perf_sample__exit (&sample); - goto out_delete_evlist; + goto out_put_evlist; } err =-1 ;
@@ -153,7 +153,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtestif (evsel ==NULL ) {pr_debug ("event with id %" PRIu64" doesn't map to an evsel\n" , sample.id); - goto out_delete_evlist; + goto out_put_evlist; } nr_events[evsel->core.idx]++;perf_mmap__consume (&md->core);
@@ -168,12 +168,12 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest expected_nr_events[evsel->core.idx], evsel__name(evsel), nr_events[evsel->core.idx]); err =-1 ; - goto out_delete_evlist; + goto out_put_evlist; } } -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist); out_free_cpus:perf_cpu_map__put (cpus); out_free_threads:
tools/perf/tests/openat-syscall-tp-fields.c +9 -9
@@ -51,7 +51,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (IS_ERR (evsel)) {pr_debug ("%s: evsel__newtp\n" , __func__); ret =PTR_ERR (evsel) == -EACCES ?TEST_SKIP :TEST_FAIL ; - goto out_delete_evlist; + goto out_put_evlist; }evlist__add (evlist, evsel);
@@ -59,7 +59,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused err =evlist__create_maps (evlist, &opts.target);if (err <0 ) {pr_debug ("%s: evlist__create_maps\n" , __func__); - goto out_delete_evlist; + goto out_put_evlist; }evsel__config (evsel, &opts,NULL );
@@ -70,14 +70,14 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (err <0 ) {pr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; } err =evlist__mmap (evlist,UINT_MAX );if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }evlist__enable (evlist);
@@ -115,7 +115,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (err) {pr_debug ("Can't parse sample, err = %d\n" , err);perf_sample__exit (&sample); - goto out_delete_evlist; + goto out_put_evlist; } tp_flags =perf_sample__intval (&sample,"flags" );
@@ -126,7 +126,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused (tp_flags & flags) != flags) {pr_debug ("%s: Expected flags=%#x, got %#x\n" , __func__, flags, tp_flags); - goto out_delete_evlist; + goto out_put_evlist; } goto out_ok;
@@ -139,13 +139,13 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedif (++nr_polls >5 ) {pr_debug ("%s: no events!\n" , __func__); - goto out_delete_evlist; + goto out_put_evlist; } } out_ok: ret =TEST_OK ; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist); out:return ret; }
tools/perf/tests/parse-events.c +2 -2
@@ -2568,7 +2568,7 @@ static int test_event(const struct evlist_test *e) ret = e->check (evlist); }parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -2594,7 +2594,7 @@ static int test_event_fake_pmu(const char *str) }parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/parse-metric.c +2 -2
@@ -84,7 +84,7 @@ static int __compute_metric(const char *name, struct value *vals, cpus =perf_cpu_map__new ("0" );if (!cpus) { -evlist__delete (evlist); +evlist__put (evlist);return -ENOMEM ; }
@@ -113,7 +113,7 @@ static int __compute_metric(const char *name, struct value *vals,/* ... cleanup. */ evlist__free_stats (evlist);perf_cpu_map__put (cpus); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/parse-no-sample-id-all.c +1 -1
@@ -49,7 +49,7 @@ static int process_events(union perf_event **events, size_t count)for (i =0 ; i < count && !err; i++) err =process_event (&evlist, events[i]); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/perf-record.c +9 -9
@@ -105,7 +105,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest err =evlist__create_maps (evlist, &opts.target);if (err <0 ) {pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -117,7 +117,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest err =evlist__prepare_workload (evlist, &opts.target, argv, false,NULL );if (err <0 ) {pr_debug ("Couldn't run the workload!\n" ); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -134,7 +134,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("sched__get_first_possible_cpu: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; } cpu = err;
@@ -146,7 +146,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("sched_setaffinity: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -158,7 +158,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("perf_evlist__open: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -171,7 +171,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist); - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -209,7 +209,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestif (verbose >0 )perf_event__fprintf (event,NULL , stderr);pr_debug ("Couldn't parse sample\n" ); - goto out_delete_evlist; + goto out_put_evlist; }if (verbose >0 ) {
@@ -350,9 +350,9 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestpr_debug ("PERF_RECORD_MMAP for %s missing!\n" ,"[vdso]" ); ++errs; } -out_delete_evlist: +out_put_evlist:CPU_FREE (cpu_mask); -evlist__delete (evlist); +evlist__put (evlist); out:perf_sample__exit (&sample);if (err == -EACCES )
tools/perf/tests/perf-time-to-tsc.c +1 -1
@@ -201,7 +201,7 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su err =TEST_OK ; out_err: -evlist__delete (evlist); +evlist__put (evlist);perf_cpu_map__put (cpus);perf_thread_map__put (threads);return err;
tools/perf/tests/pfm.c +2 -2
@@ -80,7 +80,7 @@ static int test__pfm_events(struct test_suite *test __maybe_unused,evlist__nr_groups (evlist), 0); -evlist__delete (evlist); +evlist__put (evlist); }return 0 ; }
@@ -165,7 +165,7 @@ static int test__pfm_group(struct test_suite *test __maybe_unused,evlist__nr_groups (evlist), table[i].nr_groups); -evlist__delete (evlist); +evlist__put (evlist); }return 0 ; }
tools/perf/tests/pmu-events.c +3 -3
@@ -798,7 +798,7 @@ static int check_parse_id(const char *id, struct parse_events_error *error) /*warn_if_reordered=*/true, /*fake_tp=*/false);free (dup); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
@@ -865,7 +865,7 @@ static int test__parsing_callback(const struct pmu_metric *pm, cpus =perf_cpu_map__new ("0" );if (!cpus) { -evlist__delete (evlist); +evlist__put (evlist);return -ENOMEM ; }
@@ -919,7 +919,7 @@ static int test__parsing_callback(const struct pmu_metric *pm,/* ... cleanup. */ evlist__free_stats (evlist);perf_cpu_map__put (cpus); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/pmu.c +2 -2
@@ -294,7 +294,7 @@ static int test__pmu_config_helpers(struct test_suite *test __maybe_unused, ret =TEST_OK ; err_out:parse_events_terms__exit (&terms); -evlist__delete (evlist); +evlist__put (evlist);test_pmu_put (dir, pmu);return ret; }
@@ -346,7 +346,7 @@ static int test__pmu_events(struct test_suite *test __maybe_unused, int subtest ret =TEST_OK ; err_out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);test_pmu_put (dir, pmu);return ret; }
tools/perf/tests/sw-clock.c +7 -7
@@ -59,7 +59,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) evsel =evsel__new (&attr);if (evsel ==NULL ) {pr_debug ("evsel__new\n" ); - goto out_delete_evlist; + goto out_put_evlist; }evlist__add (evlist, evsel);
@@ -68,7 +68,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)if (!cpus || !threads) { err = -ENOMEM ;pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }perf_evlist__set_maps (&evlist->core, cpus, threads);
@@ -80,14 +80,14 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)pr_debug ("Couldn't open evlist: %s\nHint: check %s, using %" PRIu64" in this test.\n" ,str_error_r (errno, sbuf,sizeof (sbuf)), knob, (u64 )attr.sample_freq); - goto out_delete_evlist; + goto out_put_evlist; } err =evlist__mmap (evlist,128 );if (err <0 ) {pr_debug ("failed to mmap event: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }evlist__enable (evlist);
@@ -113,7 +113,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)if (err <0 ) {pr_debug ("Error during parse sample\n" );perf_sample__exit (&sample); - goto out_delete_evlist; + goto out_put_evlist; } total_periods += sample.period;
@@ -131,10 +131,10 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) err =-1 ; } -out_delete_evlist: +out_put_evlist:perf_cpu_map__put (cpus);perf_thread_map__put (threads); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/switch-tracking.c +1 -1
@@ -582,7 +582,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub out:if (evlist) {evlist__disable (evlist); -evlist__delete (evlist); +evlist__put (evlist); }perf_cpu_map__put (cpus);perf_thread_map__put (threads);
tools/perf/tests/task-exit.c +7 -7
@@ -74,7 +74,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _if (!cpus || !threads) { err = -ENOMEM ;pr_debug ("Not enough memory to create thread/cpu maps\n" ); - goto out_delete_evlist; + goto out_put_evlist; }perf_evlist__set_maps (&evlist->core, cpus, threads);
@@ -82,7 +82,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ err =evlist__prepare_workload (evlist, &target, argv, false, workload_exec_failed_signal);if (err <0 ) {pr_debug ("Couldn't run the workload!\n" ); - goto out_delete_evlist; + goto out_put_evlist; } evsel =evlist__first (evlist);
@@ -101,14 +101,14 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _if (err <0 ) {pr_debug ("Couldn't open the evlist: %s\n" ,str_error_r (-err, sbuf,sizeof (sbuf))); - goto out_delete_evlist; + goto out_put_evlist; }if (evlist__mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); err =-1 ; - goto out_delete_evlist; + goto out_put_evlist; }evlist__start_workload (evlist);
@@ -133,7 +133,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _if (retry_count++ >1000 ) {pr_debug ("Failed after retrying 1000 times\n" ); err =-1 ; - goto out_delete_evlist; + goto out_put_evlist; } goto retry;
@@ -144,10 +144,10 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ err =-1 ; } -out_delete_evlist: +out_put_evlist:perf_cpu_map__put (cpus);perf_thread_map__put (threads); -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/tests/tool_pmu.c +1 -1
@@ -67,7 +67,7 @@ static int do_test(enum tool_pmu_event ev, bool with_pmu) out:parse_events_error__exit (&err); -evlist__delete (evlist); +evlist__put (evlist);return ret; }
tools/perf/tests/topology.c +1 -1
@@ -58,7 +58,7 @@ static int session_write_header(char *path) !perf_session__write_header (session, session->evlist, perf_data__fd(&data), true)); -evlist__delete (session->evlist); +evlist__put (session->evlist);perf_session__delete (session);return 0 ;
tools/perf/tests/uncore-event-sorting.c +1 -1
@@ -165,7 +165,7 @@ static int test__uncore_event_sorting(struct test_suite *test __maybe_unused, ret =TEST_OK ; out_err: -evlist__delete (evlist); +evlist__put (evlist);parse_events_error__exit (&err);zfree (&pmu_prefix);zfree (&m.event1);
tools/perf/util/cgroup.c +2 -2
@@ -520,8 +520,8 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro cgrp_event_expanded = true; out_err: -evlist__delete (orig_list); -evlist__delete (tmp_list); +evlist__put (orig_list); +evlist__put (tmp_list);metricgroup__rblist_exit (&orig_metric_events);release_cgroup_list ();
tools/perf/util/data-convert-bt.c +1 -1
@@ -1362,7 +1362,7 @@ static void cleanup_events(struct perf_session *session)zfree (&evsel->priv); } -evlist__delete (evlist); +evlist__put (evlist); session->evlist =NULL ; }
tools/perf/util/evlist.c +15 -5
@@ -75,7 +75,7 @@ int sigqueue(pid_t pid, int sig, const union sigval value);#define FD (e, x, y) (*(int *)xyarray__entry(e->core.fd, x, y))#define SID (e, x, y) xyarray__entry(e->core.sample_id, x, y) -void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus, +static void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus,struct perf_thread_map *threads) {perf_evlist__init (&evlist->core);
@@ -88,6 +88,7 @@ void evlist__init(struct evlist *evlist, struct perf_cpu_map *cpus, evlist->nr_br_cntr =-1 ;metricgroup__rblist_init (&evlist->metric_events);INIT_LIST_HEAD (&evlist->deferred_samples); +refcount_set (&evlist->refcnt,1 ); }struct evlist *evlist__new (void )
@@ -139,7 +140,7 @@ struct evlist *evlist__new_default(const struct target *target, bool sample_callreturn evlist; out_err: -evlist__delete (evlist); +evlist__put (evlist);return NULL ; }
@@ -148,13 +149,19 @@ struct evlist *evlist__new_dummy(void)struct evlist *evlist =evlist__new ();if (evlist &&evlist__add_dummy (evlist)) { -evlist__delete (evlist); +evlist__put (evlist); evlist =NULL ; }return evlist; } +struct evlist *evlist__get (struct evlist *evlist) +{ +refcount_inc (&evlist->refcnt); +return evlist; +} +/** * evlist__set_id_pos - set the positions of event ids. * @evlist: selected event list
@@ -193,7 +200,7 @@ static void evlist__purge(struct evlist *evlist) evlist->core.nr_entries =0 ; } -void evlist__exit (struct evlist *evlist) +static void evlist__exit (struct evlist *evlist) {metricgroup__rblist_exit (&evlist->metric_events);event_enable_timer__exit (&evlist->eet);
@@ -202,11 +209,14 @@ void evlist__exit(struct evlist *evlist)perf_evlist__exit (&evlist->core); } -void evlist__delete (struct evlist *evlist) +void evlist__put (struct evlist *evlist) {if (evlist ==NULL )return ; +if (!refcount_dec_and_test (&evlist->refcnt)) +return ; +evlist__free_stats (evlist);evlist__munmap (evlist);evlist__close (evlist);
tools/perf/util/evlist.h +3 -4
@@ -61,6 +61,7 @@ struct event_enable_timer;struct evlist {struct perf_evlist core; +refcount_t refcnt;bool enabled;bool no_affinity;int id_pos;
@@ -109,10 +110,8 @@ struct evsel_str_handler {struct evlist *evlist__new (void );struct evlist *evlist__new_default (const struct target *target,bool sample_callchains);struct evlist *evlist__new_dummy (void ); -void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus, -struct perf_thread_map *threads); -void evlist__exit (struct evlist *evlist); -void evlist__delete (struct evlist *evlist); +struct evlist *evlist__get (struct evlist *evlist); +void evlist__put (struct evlist *evlist);void evlist__add (struct evlist *evlist,struct evsel *entry);void evlist__remove (struct evlist *evlist,struct evsel *evsel);
tools/perf/util/expr.c +1 -1
@@ -468,7 +468,7 @@ double expr__has_event(const struct expr_parse_ctx *ctx, bool compute_ids, const ret =parse_event (tmp, id) ?0 :1 ; } out: -evlist__delete (tmp); +evlist__put (tmp);return ret; }
tools/perf/util/header.c +11 -11
@@ -5187,7 +5187,7 @@ int perf_session__read_header(struct perf_session *session)pr_err ("Invalid ids section size %" PRIu64" for attr %d, not aligned to u64\n" , f_attr.ids.size, i); err = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -5200,7 +5200,7 @@ int perf_session__read_header(struct perf_session *session)pr_err ("Invalid ids section size %" PRIu64" for attr %d, too many IDs\n" , f_attr.ids.size, i); err = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; }/*
@@ -5213,19 +5213,19 @@ int perf_session__read_header(struct perf_session *session)pr_err ("Invalid ids section for attr %d: offset=%" PRIu64" size=%" PRIu64" exceeds file size %" PRIu64"\n" , i, f_attr.ids.offset, f_attr.ids.size, (u64 )input_stat.st_size); err = -EINVAL ; - goto out_delete_evlist; + goto out_put_evlist; } tmp =lseek (fd,0 ,SEEK_CUR ); evsel =evsel__new (&f_attr.attr);if (evsel ==NULL ) - goto out_delete_evlist; + goto out_put_evlist; evsel->needs_swap = header->needs_swap;/* * Do it before so that if perf_evsel__alloc_id fails, this -* entry gets purged too at evlist__delete(). +* entry gets purged too at evlist__put(). */ evlist__add (session->evlist, evsel);
@@ -5236,7 +5236,7 @@ int perf_session__read_header(struct perf_session *session)* hattr->ids threads. */ if (perf_evsel__alloc_id (&evsel->core,1 , nr_ids)) - goto out_delete_evlist; + goto out_put_evlist;lseek (fd, f_attr.ids.offset,SEEK_SET );
@@ -5266,18 +5266,18 @@ int perf_session__read_header(struct perf_session *session) err =perf_header__process_sections (header, fd, &session->tevent, perf_file_section__process);if (err <0 ) - goto out_delete_evlist; + goto out_put_evlist;if (evlist__prepare_tracepoint_events (session->evlist, session->tevent.pevent)) { err = -ENOMEM ; - goto out_delete_evlist; + goto out_put_evlist; }#else err =perf_header__process_sections (header, fd,NULL , perf_file_section__process);if (err <0 ) - goto out_delete_evlist; + goto out_put_evlist;#endif }
@@ -5303,8 +5303,8 @@ int perf_session__read_header(struct perf_session *session) out_errno:return -errno; -out_delete_evlist: -evlist__delete (session->evlist); +out_put_evlist: +evlist__put (session->evlist); session->evlist =NULL ;return err; }
tools/perf/util/metricgroup.c +3 -3
@@ -214,7 +214,7 @@ static void metric__free(struct metric *m)zfree (&m->metric_refs);expr__ctx_free (m->pctx);zfree (&m->modifier); -evlist__delete (m->evlist); +evlist__put (m->evlist);free (m); }
@@ -1331,7 +1331,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu, parsed_evlist =NULL ; err_out:parse_events_error__exit (&parse_error); -evlist__delete (parsed_evlist); +evlist__put (parsed_evlist);strbuf_release (&events);return ret; }
@@ -1542,7 +1542,7 @@ static int parse_groups(struct evlist *perf_evlist,if (combined_evlist) {evlist__splice_list_tail (perf_evlist, &combined_evlist->core.entries); -evlist__delete (combined_evlist); +evlist__put (combined_evlist); }list_for_each_entry (m, &metric_list, nd) {
tools/perf/util/parse-events.c +2 -2
@@ -2343,7 +2343,7 @@ int __parse_events(struct evlist *evlist, const char *str, const char *pmu_filte/* * There are 2 users - builtin-record and builtin-test objects. -* Both call evlist__delete in case of error, so we dont +* Both call evlist__put in case of error, so we dont * need to bother. */ return ret;
@@ -2546,7 +2546,7 @@ int parse_events_option_new_evlist(const struct option *opt, const char *str, in } ret =parse_events_option (opt, str, unset);if (ret) { -evlist__delete (*args->evlistp); +evlist__put (*args->evlistp); *args->evlistp =NULL ; }
tools/perf/util/perf_api_probe.c +1 -1
@@ -57,7 +57,7 @@ static int perf_do_probe_api(setup_probe_fn_t fn, struct perf_cpu cpu, const cha err =0 ; out_delete: -evlist__delete (evlist); +evlist__put (evlist);return err; }
tools/perf/util/python.c +110 -132
451 lines not on this page · raw message
tools/perf/util/record.c +1 -1
@@ -264,7 +264,7 @@ bool evlist__can_select_event(struct evlist *evlist, const char *str) ret = true; out_delete: -evlist__delete (temp_evlist); +evlist__put (temp_evlist);return ret; }
tools/perf/util/session.c +1 -1
@@ -264,7 +264,7 @@ void perf_session__delete(struct perf_session *session)machines__exit (&session->machines);if (session->data) {if (perf_data__is_read (session->data)) -evlist__delete (session->evlist); +evlist__put (session->evlist);perf_data__close (session->data); }#ifdef HAVE_LIBTRACEEVENT
tools/perf/util/sideband_evlist.c +8 -8
@@ -102,7 +102,7 @@ int evlist__start_sb_thread(struct evlist *evlist, struct target *target)return 0 ;if (evlist__create_maps (evlist, target)) - goto out_delete_evlist; + goto out_put_evlist;if (evlist->core.nr_entries >1 ) {bool can_sample_identifier =perf_can_sample_identifier ();
@@ -116,25 +116,25 @@ int evlist__start_sb_thread(struct evlist *evlist, struct target *target)evlist__for_each_entry (evlist, counter) {if (evsel__open (counter, evlist->core.user_requested_cpus, evlist->core.threads) <0 ) - goto out_delete_evlist; + goto out_put_evlist; }if (evlist__mmap (evlist,UINT_MAX )) - goto out_delete_evlist; + goto out_put_evlist;evlist__for_each_entry (evlist, counter) {if (evsel__enable (counter)) - goto out_delete_evlist; + goto out_put_evlist; } evlist->thread.done =0 ;if (pthread_create (&evlist->thread.th,NULL , perf_evlist__poll_thread, evlist)) - goto out_delete_evlist; + goto out_put_evlist;return 0 ; -out_delete_evlist: -evlist__delete (evlist); +out_put_evlist: +evlist__put (evlist); evlist =NULL ;return -1 ; }
@@ -145,5 +145,5 @@ void evlist__stop_sb_thread(struct evlist *evlist)return ; evlist->thread.done =1 ;pthread_join (evlist->thread.th,NULL ); -evlist__delete (evlist); +evlist__put (evlist); }
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 6/19 perf evsel: Add reference count
In reply to Ian RogersAs with evlist this a no-op for most of the perf tool. The reference count is set to 1 at allocation, the put will see the 1, decrement it and perform the delete. The purpose for adding the reference count is for the python code. Prior to this change the python code would clone evsels, but this has issues if events are opened, etc. leading to assertion failures. With a reference count the same evsel can be used and the reference count incremented for the python usage. To not change the python evsel API getset functions are added for the evsel members, no set function is provided for size as it doesn't make sense to alter this.
---
v2:
1. Fixed Potential Crash in pyrf_event__new : Initialized pevent->evsel = NULL; to avoid garbage pointer dereference if evlist__event2evsel() fails in read_on_cpu .
2. Fixed Memory Leak in pyrf_evsel__init : Added evsel__put(pevsel->evsel) before overwriting it to handle repeated __init__ calls.
3. Fixed Exception Contract: Added PyErr_NoMemory() when evsel__new() fails in pyrf_evsel__init .
4. Fixed NULL Pointer Dereference on Property Access: Added a custom tp_getattro ( pyrf_evsel__getattro ) to pyrf_evsel__type to check if pevsel->evsel is NULL and raise a ValueError if so, covering all property accesses.
5. Fixed Reference Count in pyrf_evlist__add : Added evsel__get(evsel) when adding to the evlist .
6. Fixed Reference Count in pyrf_evlist__read_on_cpu : Added evsel__get(evsel) when assigning to pevent->evsel .
v7:
- Added pyrf_evsel__new to zero-initialize pevsel->evsel to fix crash on re-initialization.
v8:
- Added O! type validation to pyrf_evlist__add to prevent type confusion.
v10:
- Added strict CHECK_INITIALIZED/CHECK_INITIALIZED_INT checks to all pyrf_evsel methods and property getters/setters to prevent NULL pointer dereferences if the object is instantiated without calling __init__.
--- tools/perf/builtin-trace.c | 12 +- tools/perf/tests/evsel-tp-sched.c | 4 +- tools/perf/tests/openat-syscall-all-cpus.c | 6 +- tools/perf/tests/openat-syscall.c | 6 +- tools/perf/util/bpf_counter_cgroup.c | 2 +- tools/perf/util/cgroup.c | 2 +- tools/perf/util/evlist.c | 2 +- tools/perf/util/evsel.c | 26 +- tools/perf/util/evsel.h | 11 +- tools/perf/util/parse-events.y | 2 +- tools/perf/util/pfm.c | 2 +- tools/perf/util/print-events.c | 2 +- tools/perf/util/python.c | 311 ++++++++++++++++++--- tools/perf/util/session.c | 3 + 14 files changed, 310 insertions(+), 81 deletions(-)
tools/perf/builtin-trace.c +6 -6
@@ -460,10 +460,10 @@ static int evsel__init_tp_ptr_field(struct evsel *evsel, struct tp_field *field, ({ struct syscall_tp *sc = __evsel__syscall_tp(evsel);\ evsel__init_tp_ptr_field(evsel, &sc->name, #name); }) -static void evsel__delete_priv(struct evsel *evsel) +static void evsel__put_and_free_priv(struct evsel *evsel) { zfree(&evsel->priv); - evsel__delete(evsel); + evsel__put(evsel); } static int evsel__init_syscall_tp(struct evsel *evsel)
@@ -543,7 +543,7 @@ static struct evsel *perf_evsel__raw_syscall_newtp(const char *direction, void *return evsel; out_delete: -evsel__delete_priv (evsel); +evsel__put_and_free_priv (evsel);return NULL ; }
@@ -3612,7 +3612,7 @@ static bool evlist__add_vfs_getname(struct evlist *evlist)list_del_init (&evsel->core.node); evsel->evlist =NULL ; -evsel__delete (evsel); +evsel__put (evsel); }return found;
@@ -3728,9 +3728,9 @@ static int trace__add_syscall_newtp(struct trace *trace)return ret; out_delete_sys_exit: -evsel__delete_priv (sys_exit); +evsel__put_and_free_priv (sys_exit); out_delete_sys_enter: -evsel__delete_priv (sys_enter); +evsel__put_and_free_priv (sys_enter); goto out; }
tools/perf/tests/evsel-tp-sched.c +2 -2
@@ -64,7 +64,7 @@ static int test__perf_evsel__tp_sched_test(struct test_suite *test __maybe_unuseif (evsel__test_field (evsel,"next_prio" ,4 , true)) ret =TEST_FAIL ; -evsel__delete (evsel); +evsel__put (evsel); evsel =evsel__newtp ("sched" ,"sched_wakeup" );
@@ -85,7 +85,7 @@ static int test__perf_evsel__tp_sched_test(struct test_suite *test __maybe_unuseif (evsel__test_field (evsel,"target_cpu" ,4 , true)) ret =TEST_FAIL ; -evsel__delete (evsel); +evsel__put (evsel);return ret; }
tools/perf/tests/openat-syscall-all-cpus.c +3 -3
@@ -59,7 +59,7 @@ static int test__openat_syscall_event_on_all_cpus(struct test_suite *test __mayb"tweak /proc/sys/kernel/perf_event_paranoid?\n" , str_error_r(errno, sbuf, sizeof(sbuf))); err =TEST_SKIP ; - goto out_evsel_delete; + goto out_evsel_put; }perf_cpu_map__for_each_cpu (cpu, idx, cpus) {
@@ -116,8 +116,8 @@ static int test__openat_syscall_event_on_all_cpus(struct test_suite *test __maybevsel__free_counts (evsel); out_close_fd:perf_evsel__close_fd (&evsel->core); -out_evsel_delete: -evsel__delete (evsel); +out_evsel_put: +evsel__put (evsel); out_cpu_map_delete:perf_cpu_map__put (cpus); out_thread_map_delete:
tools/perf/tests/openat-syscall.c +3 -3
@@ -42,7 +42,7 @@ static int test__openat_syscall_event(struct test_suite *test __maybe_unused,"tweak /proc/sys/kernel/perf_event_paranoid?\n" , str_error_r(errno, sbuf, sizeof(sbuf))); err =TEST_SKIP ; - goto out_evsel_delete; + goto out_evsel_put; }for (i =0 ; i < nr_openat_calls; ++i) {
@@ -64,8 +64,8 @@ static int test__openat_syscall_event(struct test_suite *test __maybe_unused, err =TEST_OK ; out_close_fd:perf_evsel__close_fd (&evsel->core); -out_evsel_delete: -evsel__delete (evsel); +out_evsel_put: +evsel__put (evsel); out_thread_map_delete:perf_thread_map__put (threads);return err;
tools/perf/util/bpf_counter_cgroup.c +1 -1
@@ -336,7 +336,7 @@ static int bperf_cgrp__destroy(struct evsel *evsel)return 0 ;bperf_cgroup_bpf__destroy (skel); -evsel__delete (cgrp_switch);// it'll destroy on_switch progs too +evsel__put (cgrp_switch);// it'll destroy on_switch progs too return 0 ; }
tools/perf/util/cgroup.c +1 -1
@@ -469,7 +469,7 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro/* copy the list and set to the new cgroup. */ evlist__for_each_entry (orig_list, pos) { -struct evsel *evsel =evsel__clone (/*dest=*/ NULL , pos); +struct evsel *evsel =evsel__clone (pos);if (evsel ==NULL ) goto out_err;
tools/perf/util/evlist.c +1 -1
@@ -194,7 +194,7 @@ static void evlist__purge(struct evlist *evlist)evlist__for_each_entry_safe (evlist, n, pos) {list_del_init (&pos->core.node); pos->evlist =NULL ; -evsel__delete (pos); +evsel__put (pos); } evlist->core.nr_entries =0 ;
tools/perf/util/evsel.c +16 -10
@@ -393,10 +393,11 @@ bool evsel__is_function_event(struct evsel *evsel)#undef FUNCTION_EVENT } -void evsel__init (struct evsel *evsel, +static void evsel__init (struct evsel *evsel,struct perf_event_attr *attr,int idx) {perf_evsel__init (&evsel->core, attr, idx); +refcount_set (&evsel->refcnt,1 ); evsel->tracking = !idx; evsel->unit =strdup ("" ); evsel->scale =1.0 ;
@@ -477,7 +478,7 @@ static int evsel__copy_config_terms(struct evsel *dst, struct evsel *src)* The assumption is that @orig is not configured nor opened yet. * So we only care about the attributes that can be set while it's parsed. */ -struct evsel *evsel__clone (struct evsel *dest,struct evsel *orig) +struct evsel *evsel__clone (struct evsel *orig) {struct evsel *evsel;
@@ -490,11 +491,7 @@ struct evsel *evsel__clone(struct evsel *dest, struct evsel *orig)if (orig->bpf_obj)return NULL ; -if (dest) - evsel = dest; -else - evsel =evsel__new (&orig->core.attr); - + evsel =evsel__new (&orig->core.attr);if (evsel ==NULL )return NULL ;
@@ -579,7 +576,7 @@ struct evsel *evsel__clone(struct evsel *dest, struct evsel *orig)return evsel; out_err: -evsel__delete (evsel); +evsel__put (evsel);return NULL ; }
@@ -638,6 +635,12 @@ struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx, boolreturn ERR_PTR (err); } +struct evsel *evsel__get (struct evsel *evsel) +{ +refcount_inc (&evsel->refcnt); +return evsel; +} +#ifdef HAVE_LIBTRACEEVENT struct tep_event *evsel__tp_format (struct evsel *evsel) {
@@ -1976,7 +1979,7 @@ void evsel__set_priv_destructor(void (*destructor)(void *priv)) evsel__priv_destructor = destructor; } -void evsel__exit (struct evsel *evsel) +static void evsel__exit (struct evsel *evsel) {assert (list_empty (&evsel->core.node));assert (evsel->evlist ==NULL );
@@ -2013,11 +2016,14 @@ void evsel__exit(struct evsel *evsel) } } -void evsel__delete (struct evsel *evsel) +void evsel__put (struct evsel *evsel) {if (!evsel)return ; +if (!refcount_dec_and_test (&evsel->refcnt)) +return ; +evsel__exit (evsel);free (evsel); }
tools/perf/util/evsel.h +6 -5
@@ -6,6 +6,7 @@#include <linux/list.h> #include <linux/perf_event.h> +#include <linux/refcount.h> #include <linux/types.h> #include <sys/types.h>
@@ -47,6 +48,7 @@ typedef int (evsel__sb_cb_t)(union perf_event *event, void *data);struct evsel {struct perf_evsel core;struct evlist *evlist; +refcount_t refcnt;off_t id_offset;int id_pos;int is_pos;
@@ -268,7 +270,7 @@ static inline struct evsel *evsel__new(struct perf_event_attr *attr)return evsel__new_idx (attr,0 ); } -struct evsel *evsel__clone (struct evsel *dest,struct evsel *orig); +struct evsel *evsel__clone (struct evsel *orig);int copy_config_terms (struct list_head *dst,struct list_head *src);void free_config_terms (struct list_head *config_terms);
@@ -283,14 +285,13 @@ static inline struct evsel *evsel__newtp(const char *sys, const char *name)return evsel__newtp_idx (sys, name,0 , true); } +struct evsel *evsel__get (struct evsel *evsel); +void evsel__put (struct evsel *evsel); +#ifdef HAVE_LIBTRACEEVENT struct tep_event *evsel__tp_format (struct evsel *evsel);#endif -void evsel__init (struct evsel *evsel,struct perf_event_attr *attr,int idx); -void evsel__exit (struct evsel *evsel); -void evsel__delete (struct evsel *evsel); -void evsel__set_priv_destructor (void (*destructor)(void *priv));struct callchain_param ;
tools/perf/util/parse-events.y +1 -1
@@ -47,7 +47,7 @@ static void free_list_evsel(struct list_head* list_evsel) list_for_each_entry_safe(evsel, tmp, list_evsel, core.node) { list_del_init(&evsel->core.node); - evsel__delete(evsel); + evsel__put(evsel); } free(list_evsel); }
tools/perf/util/pfm.c +1 -1
@@ -159,7 +159,7 @@ static bool is_libpfm_event_supported(const char *name, struct perf_cpu_map *cpu result = false;evsel__close (evsel); -evsel__delete (evsel); +evsel__put (evsel);return result; }
tools/perf/util/print-events.c +1 -1
@@ -174,7 +174,7 @@ bool is_event_supported(u8 type, u64 config) }evsel__close (evsel); -evsel__delete (evsel); +evsel__put (evsel); }perf_thread_map__put (tmap);
tools/perf/util/python.c +265 -46
444 lines not on this page · raw message
tools/perf/util/session.c +3 -0
@@ -1844,7 +1844,10 @@ static int evlist__deliver_deferred_callchain(struct evlist *evlist,struct evsel *saved_evsel = sample->evsel; sample->evsel =evlist__id2evsel (evlist, sample->id); +if (sample->evsel) + sample->evsel =evsel__get (sample->evsel); ret = tool->callchain_deferred (tool, event, sample, machine); +evsel__put (sample->evsel); sample->evsel = saved_evsel;return ret; }
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 06/19] perf evsel: Add reference count
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
New issues:
- [Medium] API breakage: The idx keyword argument was removed from the perf.evsel Python constructor.
Pre-existing issues:
- [High] Python pevsel objects are permanently leaked when added to a Python evlist due to an obsolete Py_INCREF.
- [High] Type confusion in multiple Python API functions due to unsafe object casting, leading to segmentation faults. --
commit 32152cc198f8b97a7c285972f68510722b3dec17
Author: Ian Rogers <[email protected]>
Subject: perf evsel: Add reference count
The commit adds a reference count to struct evsel to manage its lifecycle. This is particularly useful for Python bindings, where instead of cloning the event, the same event is reused and its reference count is incremented.
Quoted 13 lines
tools/perf/util/python.c +0
-1@@ -1028,7 +1031,6 @@ static int pyrf_evsel__init(struct pyrf_evsel *pevsel, precise_ip = 0, mmap_data = 0, sample_id_all =1 ; -int idx =0 ; if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|iKiKKiiiiiiiiiiiiiiiiiiiiiiKK", kwlist,
tools/perf/util/python.c static int pyrf_evsel__init(struct pyrf_evsel *pevsel, · line 1047
Quoted 6 lines
tools/perf/util/python.c +1
-1@@ -1041,7 +1043,7 @@ static int pyrf_evsel__init(struct pyrf_evsel *pevsel, &enable_on_exec, &task, &watermark, &precise_ip, &mmap_data, &sample_id_all, &attr.wakeup_events, &attr.bp_type, - &attr.bp_addr, &attr.bp_len, &idx)) + &attr.bp_addr, &attr.bp_len)) return -1;
[Severity: Medium]
Since the idx parameter was removed from PyArg_ParseTupleAndKeywords, won't existing Python scripts that pass idx (e.g., perf.evsel(..., idx=0)) now crash with a TypeError for an invalid keyword argument?
[ ... ]
Quoted 7 lines
tools/perf/util/python.c +1
-1@@ -1073,26 +1075,33 @@ static int pyrf_evsel__init(struct pyrf_evsel *pevsel,static PyObject *pyrf_evsel__open (struct pyrf_evsel *pevsel,PyObject *args,PyObject *kwargs) { -struct evsel *evsel = &pevsel->evsel; +struct evsel *evsel = pevsel->evsel;struct perf_cpu_map *cpus =NULL ;
tools/perf/util/python.c static int pyrf_evsel__init(struct pyrf_evsel *pevsel, · line 1089
Quoted 10 lines
tools/perf/util/python.c +2
-0struct perf_thread_map *threads =NULL ;PyObject *pcpus =NULL , *pthreads =NULL ;int group =0 , inherit =0 ;static char *kwlist[] = {"cpus" ,"threads" ,"group" ,"inherit" ,NULL }; +CHECK_INITIALIZED (evsel,"evsel" ); +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"|OOii" , kwlist, &pcpus, &pthreads, &group, &inherit))return NULL ;
[Severity: High]
This is a pre-existing issue, but pyrf_evsel__open() uses "OO" format specifiers to parse the cpus and threads arguments, allowing arbitrary Python objects to be passed in. Shortly after this code:
if (pthreads != NULL) threads = ((struct pyrf_thread_map *)pthreads)->threads; if (pcpus != NULL) cpus = ((struct pyrf_cpu_map *)pcpus)->cpus;
Could this blindly cast arbitrary Python objects to struct pyrf_thread_map and struct pyrf_cpu_map, leading to out-of-bounds reads or segmentation faults if incorrect object types are passed? Should we use the "O!" format specifier with the appropriate type objects instead?
[ ... ]
Quoted 7 lines
tools/perf/util/python.c +1
-1@@ -1584,13 +1808,13 @@ static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist,PyObject *pevsel;struct evsel *evsel; -if (!PyArg_ParseTuple (args,"O" , &pevsel)) +if (!PyArg_ParseTuple (args,"O!" , &pyrf_evsel__type, &pevsel))return NULL ;
tools/perf/util/python.c static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist, · line 1820
Quoted 10 lines
tools/perf/util/python.c +2
-2Py_INCREF (pevsel); - evsel = &((struct pyrf_evsel *)pevsel)->evsel; + evsel = ((struct pyrf_evsel *)pevsel)->evsel; evsel->core.idx = evlist->core.nr_entries; -evlist__add (evlist, evsel); +evlist__add (evlist,evsel__get (evsel));return Py_BuildValue ("i" , evlist->core.nr_entries); }
[Severity: High]
This isn't a bug introduced by this patch, but does keeping the explicit Py_INCREF(pevsel) here cause a memory leak of the Python wrapper object?
Previously, this was needed because the C evlist held a raw pointer to the evsel struct embedded within the Python object. Now that evsel is dynamically allocated and the C evlist correctly acquires its own reference to the C struct via evsel__get(evsel), the Python wrapper object reference is never decremented, which seems to result in a permanent leak.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=6
Ian Rogers
reviewerPATCH v11 7/19 perf evlist: Add reference count checking
In reply to Ian RogersNow the evlist is reference counted, add reference count checking so that gets and puts are paired and easy to debug. Reference count checking is documented here:
https://perfwiki.github.io/main/reference-count-checking/
This large patch is adding accessors to evlist functions and switching to their use. There was some minor renaming as evlist__mmap is now an accessor to the mmap variable, and the original evlist__mmap is renamed to evlist__do_mmap.
---
v2:
1. Fixed Memory Leak in evlist__new : Added free(evlist) in the else branch if ADD_RC_CHK fails, preventing a leak of the allocated raw structure.
2. Fixed Potential NULL Dereference: Added a NULL check after from_list_start(_evlist) in perf_evlist__mmap_cb_get() .
3. Fixed Use-After-Free Risk: In evlist__add() , I changed entry->evlist = evlist; to entry->evlist = evlist__get(evlist); . This ensures that the evsel holds a valid reference (wrapper) to the evlist , preventing it from becoming a dangling pointer if the original wrapper is freed.
4. Fixed Test Masking Bug: In test__perf_time__parse_for_ranges() , I replaced TEST_ASSERT_VAL with a manual check and return false; to avoid boolean evaluation of -1 inadvertently passing the test.
5. Fix reference count checker memory leaks from missed puts and due to cyclic evsel to evlist references. A leak still exists in __perf_evlist__propagate_maps due to empty CPU maps and not deleting the removed evsel.
v8:
1. Handle evsel__new failure in sample-parsing.c test and extra ref count checking integrity in evlist.c
2. Use standard per_sample initialization in cs-etm.c.
v10:
- Added CHECK_INITIALIZED checks to pyrf_evlist methods to prevent NULL pointer dereferences on uninitialized objects.
--- tools/perf/arch/arm/util/cs-etm.c | 10 +- tools/perf/arch/arm64/util/arm-spe.c | 8 +- tools/perf/arch/arm64/util/hisi-ptt.c | 2 +- tools/perf/arch/x86/tests/hybrid.c | 20 +- tools/perf/arch/x86/util/auxtrace.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 6 +- tools/perf/arch/x86/util/intel-pt.c | 9 +- tools/perf/arch/x86/util/iostat.c | 12 +- tools/perf/bench/evlist-open-close.c | 11 +- tools/perf/builtin-annotate.c | 7 +- tools/perf/builtin-ftrace.c | 6 +- tools/perf/builtin-inject.c | 4 +- tools/perf/builtin-kvm.c | 10 +- tools/perf/builtin-kwork.c | 8 +- tools/perf/builtin-lock.c | 2 +- tools/perf/builtin-record.c | 91 ++--- tools/perf/builtin-report.c | 6 +- tools/perf/builtin-sched.c | 24 +- tools/perf/builtin-script.c | 13 +- tools/perf/builtin-stat.c | 73 ++-- tools/perf/builtin-top.c | 52 +-- tools/perf/builtin-trace.c | 22 +- tools/perf/tests/backward-ring-buffer.c | 8 +- tools/perf/tests/code-reading.c | 10 +- tools/perf/tests/event-times.c | 2 +- tools/perf/tests/event_update.c | 2 +- tools/perf/tests/expand-cgroup.c | 4 +- tools/perf/tests/hwmon_pmu.c | 5 +- tools/perf/tests/keep-tracking.c | 8 +- tools/perf/tests/mmap-basic.c | 6 +- tools/perf/tests/openat-syscall-tp-fields.c | 8 +- tools/perf/tests/parse-events.c | 135 +++---- tools/perf/tests/parse-metric.c | 4 +- tools/perf/tests/perf-record.c | 20 +- tools/perf/tests/perf-time-to-tsc.c | 10 +- tools/perf/tests/pfm.c | 8 +- tools/perf/tests/pmu-events.c | 5 +- tools/perf/tests/sample-parsing.c | 44 +-- tools/perf/tests/sw-clock.c | 6 +- tools/perf/tests/switch-tracking.c | 8 +- tools/perf/tests/task-exit.c | 6 +- tools/perf/tests/time-utils-test.c | 14 +- tools/perf/tests/tool_pmu.c | 5 +- tools/perf/tests/topology.c | 2 +- tools/perf/tests/uncore-event-sorting.c | 4 +- tools/perf/ui/browsers/annotate.c | 2 +- tools/perf/ui/browsers/hists.c | 22 +- tools/perf/util/amd-sample-raw.c | 2 +- tools/perf/util/annotate-data.c | 2 +- tools/perf/util/annotate.c | 10 +- tools/perf/util/auxtrace.c | 14 +- tools/perf/util/block-info.c | 4 +- tools/perf/util/bpf_counter.c | 2 +- tools/perf/util/bpf_counter_cgroup.c | 12 +- tools/perf/util/bpf_ftrace.c | 9 +- tools/perf/util/bpf_lock_contention.c | 12 +- tools/perf/util/bpf_off_cpu.c | 14 +- tools/perf/util/cgroup.c | 20 +- tools/perf/util/cs-etm.c | 5 +- tools/perf/util/evlist.c | 391 ++++++++++++-------- tools/perf/util/evlist.h | 251 ++++++++++++- tools/perf/util/evsel.c | 6 +- tools/perf/util/evsel.h | 4 +- tools/perf/util/header.c | 47 +-- tools/perf/util/header.h | 2 +- tools/perf/util/intel-tpebs.c | 7 +- tools/perf/util/iostat.c | 2 +- tools/perf/util/iostat.h | 2 +- tools/perf/util/metricgroup.c | 6 +- tools/perf/util/parse-events.c | 6 +- tools/perf/util/pfm.c | 2 +- tools/perf/util/python.c | 92 +++-- tools/perf/util/record.c | 9 +- tools/perf/util/sample-raw.c | 4 +- tools/perf/util/session.c | 54 +-- tools/perf/util/sideband_evlist.c | 24 +- tools/perf/util/sort.c | 2 +- tools/perf/util/stat-display.c | 6 +- tools/perf/util/stat-shadow.c | 4 +- tools/perf/util/stat.c | 4 +- tools/perf/util/stream.c | 4 +- tools/perf/util/synthetic-events.c | 11 +- tools/perf/util/time-utils.c | 12 +- tools/perf/util/top.c | 4 +- 84 files changed, 1088 insertions(+), 720 deletions(-)
tools/perf/arch/arm/util/cs-etm.c +5 -5
@@ -201,7 +201,7 @@ static int cs_etm_validate_config(struct perf_pmu *cs_etm_pmu, {unsigned int idx;int err =0 ; -struct perf_cpu_map *event_cpus = evsel->evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evsel->evlist)->user_requested_cpus;struct perf_cpu_map *intersect_cpus;struct perf_cpu cpu;
@@ -325,7 +325,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, container_of(itr, struct cs_etm_recording, itr);struct perf_pmu *cs_etm_pmu = ptr->cs_etm_pmu;struct evsel *evsel, *cs_etm_evsel =NULL ; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );int err =0 ;
@@ -551,7 +551,7 @@ cs_etm_info_priv_size(struct auxtrace_record *itr, {unsigned int idx;int etmv3 =0 , etmv4 =0 , ete =0 ; -struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evlist)->user_requested_cpus;struct perf_cpu_map *intersect_cpus;struct perf_cpu cpu;struct perf_pmu *cs_etm_pmu =cs_etm_get_pmu (itr);
@@ -790,7 +790,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr,u32 offset;u64 nr_cpu, type;struct perf_cpu_map *cpu_map; -struct perf_cpu_map *event_cpus = session->evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (session->evlist)->user_requested_cpus;struct perf_cpu_map *online_cpus =perf_cpu_map__new_online_cpus ();struct cs_etm_recording *ptr = container_of(itr, struct cs_etm_recording, itr);
@@ -800,7 +800,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr,if (priv_size !=cs_etm_info_priv_size (itr, session->evlist))return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ;/* If the cpu_map has the "any" CPU all online CPUs are involved */
tools/perf/arch/arm64/util/arm-spe.c +4 -4
@@ -60,7 +60,7 @@ static bool arm_spe_is_set_freq(struct evsel *evsel)*/ static struct perf_cpu_map *arm_spe_find_cpus (struct evlist *evlist) { -struct perf_cpu_map *event_cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *event_cpus =evlist__core (evlist)->user_requested_cpus;struct perf_cpu_map *online_cpus =perf_cpu_map__new_online_cpus ();struct perf_cpu_map *intersect_cpus;
@@ -157,7 +157,7 @@ static int arm_spe_info_fill(struct auxtrace_record *itr,if (priv_size !=arm_spe_info_priv_size (itr, session->evlist))return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; cpu_map =arm_spe_find_cpus (session->evlist);
@@ -363,7 +363,7 @@ static int arm_spe_setup_tracking_event(struct evlist *evlist, {int err;struct evsel *tracking_evsel; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;/* Add dummy event to keep tracking */ err =parse_event (evlist,"dummy:u" );
@@ -396,7 +396,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr,struct arm_spe_recording *sper = container_of(itr, struct arm_spe_recording, itr);struct evsel *evsel, *tmp; -struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool discard = false;int err;u64 discard_bit;
tools/perf/arch/arm64/util/hisi-ptt.c +1 -1
@@ -53,7 +53,7 @@ static int hisi_ptt_info_fill(struct auxtrace_record *itr,if (priv_size !=HISI_PTT_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; auxtrace_info->type =PERF_AUXTRACE_HISI_PTT ;
tools/perf/arch/x86/tests/hybrid.c +10 -10
@@ -26,7 +26,7 @@ static int test__hybrid_hw_event_with_pmu(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -38,7 +38,7 @@ static int test__hybrid_hw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -57,7 +57,7 @@ static int test__hybrid_sw_hw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_SOFTWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong leader" ,evsel__has_leader (evsel, leader));
@@ -74,7 +74,7 @@ static int test__hybrid_hw_sw_group_event(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -91,7 +91,7 @@ static int test__hybrid_group_modifier1(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
@@ -113,7 +113,7 @@ static int test__hybrid_raw1(struct evlist *evlist) {struct perf_evsel *evsel; -perf_evlist__for_each_evsel (&evlist->core, evsel) { +perf_evlist__for_each_evsel (evlist__core (evlist), evsel) {struct perf_pmu *pmu =perf_pmus__find_by_type (evsel->attr.type);TEST_ASSERT_VAL ("missing pmu" , pmu);
@@ -127,7 +127,7 @@ static int test__hybrid_raw2(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_RAW == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,0x1a ));return TEST_OK ;
@@ -137,7 +137,7 @@ static int test__hybrid_cache_event(struct evlist *evlist) {struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HW_CACHE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,0x2 == (evsel->core.attr.config &0xffffffff ));return TEST_OK ;
@@ -148,7 +148,7 @@ static int test__checkevent_pmu(struct evlist *evlist)struct evsel *evsel =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,1 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,1 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_RAW == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong config" ,10 == evsel->core.attr.config);TEST_ASSERT_VAL ("wrong config1" ,1 == evsel->core.attr.config1);
@@ -168,7 +168,7 @@ static int test__hybrid_hw_group_event_2(struct evlist *evlist)struct evsel *evsel, *leader; evsel = leader =evlist__first (evlist); -TEST_ASSERT_VAL ("wrong number of entries" ,2 == evlist->core.nr_entries); +TEST_ASSERT_VAL ("wrong number of entries" ,2 ==evlist__nr_entries (evlist));TEST_ASSERT_VAL ("wrong type" ,PERF_TYPE_HARDWARE == evsel->core.attr.type);TEST_ASSERT_VAL ("wrong hybrid type" ,test_hybrid_type (evsel,PERF_TYPE_RAW ));TEST_ASSERT_VAL ("wrong config" ,test_config (evsel,PERF_COUNT_HW_CPU_CYCLES ));
tools/perf/arch/x86/util/auxtrace.c +1 -1
@@ -55,7 +55,7 @@ struct auxtrace_record *auxtrace_record__init(struct evlist *evlist, int *err) {char buffer[64 ]; -struct perf_cpu cpu =perf_cpu_map__min (evlist->core.all_cpus); +struct perf_cpu cpu =perf_cpu_map__min (evlist__core (evlist)->all_cpus);int ret; *err =0 ;
tools/perf/arch/x86/util/intel-bts.c +3 -3
@@ -79,10 +79,10 @@ static int intel_bts_info_fill(struct auxtrace_record *itr,if (priv_size !=INTEL_BTS_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].core.base; + pc =evlist__mmap (session->evlist)[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
@@ -114,7 +114,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, container_of(itr, struct intel_bts_recording, itr);struct perf_pmu *intel_bts_pmu = btsr->intel_bts_pmu;struct evsel *evsel, *intel_bts_evsel =NULL ; -const struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +const struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );if (opts->auxtrace_sample_mode) {
tools/perf/arch/x86/util/intel-pt.c +5 -4
@@ -360,10 +360,10 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, filter =intel_pt_find_filter (session->evlist, ptr->intel_pt_pmu); filter_str_len = filter ?strlen (filter) :0 ; -if (!session->evlist->core.nr_mmaps) +if (!evlist__core (session->evlist)->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].core.base; + pc =evlist__mmap (session->evlist)[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
@@ -376,7 +376,8 @@ static int intel_pt_info_fill(struct auxtrace_record *itr,ui__warning ("Intel Processor Trace: TSC not available\n" ); } - per_cpu_mmaps = !perf_cpu_map__is_any_cpu_or_is_empty (session->evlist->core.user_requested_cpus); + per_cpu_mmaps = !perf_cpu_map__is_any_cpu_or_is_empty ( +evlist__core (session->evlist)->user_requested_cpus); auxtrace_info->type =PERF_AUXTRACE_INTEL_PT ; auxtrace_info->priv[INTEL_PT_PMU_TYPE ] = intel_pt_pmu->type;
@@ -621,7 +622,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr,struct perf_pmu *intel_pt_pmu = ptr->intel_pt_pmu;bool have_timing_info, need_immediate = false;struct evsel *evsel, *intel_pt_evsel =NULL ; -const struct perf_cpu_map *cpus = evlist->core.user_requested_cpus; +const struct perf_cpu_map *cpus =evlist__core (evlist)->user_requested_cpus;bool privileged =perf_event_paranoid_check (-1 );u64 tsc_bit;int err;
tools/perf/arch/x86/util/iostat.c +7 -5
@@ -332,13 +332,15 @@ static int iostat_event_group(struct evlist *evl,return ret; } -int iostat_prepare (struct evlist *evlist,struct perf_stat_config *config) +int iostat_prepare (struct evlist **evlist_ptr,struct perf_stat_config *config) { -if (evlist->core.nr_entries >0 ) { +struct evlist *evlist = *evlist_ptr; + +if (evlist__nr_entries (evlist) >0 ) {pr_warning ("The -e and -M options are not supported." "All chosen events/metrics will be dropped\n" );evlist__put (evlist); - evlist =evlist__new (); + *evlist_ptr = evlist =evlist__new ();if (!evlist)return -ENOMEM ; }
@@ -400,7 +402,7 @@ void iostat_prefix(struct evlist *evlist,struct perf_stat_config *config, char *prefix, struct timespec *ts) { - struct iio_root_port *rp = evlist->selected->priv; + struct iio_root_port *rp = evlist__selected(evlist)->priv;if (rp) {/*
@@ -463,7 +465,7 @@ void iostat_print_counters(struct evlist *evlist,iostat_prefix (evlist, config, prefix, ts);fprintf (config->output,"%s" , prefix);evlist__for_each_entry (evlist, counter) { - perf_device = evlist->selected->priv; + perf_device =evlist__selected (evlist)->priv;if (perf_device && perf_device != counter->priv) {evlist__set_selected (evlist, counter);iostat_prefix (evlist, config, prefix, ts);
tools/perf/bench/evlist-open-close.c +6 -5
@@ -116,7 +116,7 @@ static int bench__do_evlist_open_close(struct evlist *evlist)return err; } - err =evlist__mmap (evlist, opts.mmap_pages); + err =evlist__do_mmap (evlist, opts.mmap_pages);if (err <0 ) {pr_err ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));return err;
@@ -124,7 +124,7 @@ static int bench__do_evlist_open_close(struct evlist *evlist)evlist__enable (evlist);evlist__disable (evlist); -evlist__munmap (evlist); +evlist__do_munmap (evlist);evlist__close (evlist);return 0 ;
@@ -145,10 +145,11 @@ static int bench_evlist_open_close__run(char *evstr, const char *uid_str)init_stats (&time_stats); -printf (" Number of cpus:\t%d\n" ,perf_cpu_map__nr (evlist->core.user_requested_cpus)); -printf (" Number of threads:\t%d\n" , evlist->core.threads->nr); +printf (" Number of cpus:\t%d\n" , +perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus)); +printf (" Number of threads:\t%d\n" ,evlist__core (evlist)->threads->nr);printf (" Number of events:\t%d (%d fds)\n" , - evlist->core.nr_entries,evlist__count_evsel_fds (evlist)); +evlist__nr_entries (evlist),evlist__count_evsel_fds (evlist));printf (" Number of iterations:\t%d\n" , iterations);evlist__put (evlist);
tools/perf/builtin-annotate.c +5 -2
@@ -562,7 +562,7 @@ static int __cmd_annotate(struct perf_annotate *ann) goto out;if ((use_browser ==1 || ann->use_stdio2) && ann->has_br_stack) -if (session->evlist->nr_br_cntr >0 ) +if (evlist__nr_br_cntr (session->evlist) >0 ) annotate_opts.show_br_cntr = true;if (dump_trace) {
@@ -928,8 +928,11 @@ int cmd_annotate(int argc, const char **argv)* branch counters, if the corresponding branch info is available * in the perf data in the TUI mode. */ -if ((use_browser ==1 || annotate.use_stdio2) && annotate.has_br_stack) +if ((use_browser ==1 || annotate.use_stdio2) && annotate.has_br_stack) { sort__mode =SORT_MODE__BRANCH ; +if (evlist__nr_br_cntr (annotate.session->evlist) >0 ) + annotate_opts.show_br_cntr = true; + }if (setup_sorting (/*evlist=*/ NULL ,perf_session__env (annotate.session)) <0 )usage_with_options (annotate_usage, options);
tools/perf/builtin-ftrace.c +3 -3
@@ -377,9 +377,9 @@ static int set_tracing_pid(struct perf_ftrace *ftrace)if (target__has_cpu (&ftrace->target))return 0 ; -for (i =0 ; i <perf_thread_map__nr (ftrace->evlist->core.threads); i++) { +for (i =0 ; i <perf_thread_map__nr (evlist__core (ftrace->evlist)->threads); i++) {scnprintf (buf,sizeof (buf),"%d" , -perf_thread_map__pid (ftrace->evlist->core.threads, i)); +perf_thread_map__pid (evlist__core (ftrace->evlist)->threads, i));if (append_tracing_file ("set_ftrace_pid" , buf) <0 )return -1 ; }
@@ -413,7 +413,7 @@ static int set_tracing_cpumask(struct perf_cpu_map *cpumap)static int set_tracing_cpu (struct perf_ftrace *ftrace) { -struct perf_cpu_map *cpumap = ftrace->evlist->core.user_requested_cpus; +struct perf_cpu_map *cpumap =evlist__core (ftrace->evlist)->user_requested_cpus;if (!target__has_cpu (&ftrace->target))return 0 ;
tools/perf/builtin-inject.c +2 -2
@@ -1496,7 +1496,7 @@ static int synthesize_id_index(struct perf_inject *inject, size_t new_cnt)struct perf_session *session = inject->session;struct evlist *evlist = session->evlist;struct machine *machine = &session->machines.host; -size_t from = evlist->core.nr_entries - new_cnt; +size_t from =evlist__nr_entries (evlist) - new_cnt;return __perf_event__synthesize_id_index (&inject->tool, perf_event__repipe, evlist, machine, from);
@@ -2031,7 +2031,7 @@ static int host__finished_init(const struct perf_tool *tool, struct perf_sessionif (ret)return ret; - ret =synthesize_id_index (inject, gs->session->evlist->core.nr_entries); + ret =synthesize_id_index (inject,evlist__nr_entries (gs->session->evlist));if (ret) {pr_err ("Failed to synthesize id_index\n" );return ret;
tools/perf/builtin-kvm.c +5 -5
@@ -1221,7 +1221,7 @@ static s64 perf_kvm__mmap_read_idx(struct perf_kvm_stat *kvm, int idx,int err; *mmap_time =ULLONG_MAX ; - md = &evlist->mmap[idx]; + md = &evlist__mmap (evlist)[idx]; err =perf_mmap__read_init (&md->core);if (err <0 )return (err == -EAGAIN ) ?0 :-1 ;
@@ -1266,7 +1266,7 @@ static int perf_kvm__mmap_read(struct perf_kvm_stat *kvm)s64 n, ntotal =0 ;u64 flush_time =ULLONG_MAX , mmap_time; -for (i =0 ; i < kvm->evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (kvm->evlist)->nr_mmaps; i++) { n =perf_kvm__mmap_read_idx (kvm, i, &mmap_time);if (n <0 )return -1 ;
@@ -1449,7 +1449,7 @@ static int kvm_events_live_report(struct perf_kvm_stat *kvm)evlist__enable (kvm->evlist);while (!done) { -struct fdarray *fda = &kvm->evlist->core.pollfd; +struct fdarray *fda = &evlist__core (kvm->evlist)->pollfd;int rc; rc =perf_kvm__mmap_read (kvm);
@@ -1531,7 +1531,7 @@ static int kvm_live_open_events(struct perf_kvm_stat *kvm) goto out; } -if (evlist__mmap (evlist, kvm->opts.mmap_pages) <0 ) { +if (evlist__do_mmap (evlist, kvm->opts.mmap_pages) <0 ) {ui__error ("Failed to mmap the events: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__close (evlist);
@@ -1931,7 +1931,7 @@ static int kvm_events_live(struct perf_kvm_stat *kvm,perf_session__set_id_hdr_size (kvm->session);ordered_events__set_copy_on_queue (&kvm->session->ordered_events, true);machine__synthesize_threads (&kvm->session->machines.host, &kvm->opts.target, - kvm->evlist->core.threads, true, false,1 ); +evlist__core (kvm->evlist)->threads, true, false,1 ); err =kvm_live_open_events (kvm);if (err) goto out;
tools/perf/builtin-kwork.c +4 -4
@@ -1814,7 +1814,7 @@ static int perf_kwork__check_config(struct perf_kwork *kwork, } } -list_for_each_entry (evsel, &session->evlist->core.entries, core.node) { +list_for_each_entry (evsel, &evlist__core (session->evlist)->entries, core.node) {if (kwork->show_callchain && !evsel__has_callchain (evsel)) {pr_debug ("Samples do not have callchains\n" ); kwork->show_callchain =0 ;
@@ -1864,9 +1864,9 @@ static int perf_kwork__read_events(struct perf_kwork *kwork) goto out_delete; } - kwork->nr_events = session->evlist->stats.nr_events[0 ]; - kwork->nr_lost_events = session->evlist->stats.total_lost; - kwork->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST ]; + kwork->nr_events =evlist__stats (session->evlist)->nr_events[0 ]; + kwork->nr_lost_events =evlist__stats (session->evlist)->total_lost; + kwork->nr_lost_chunks =evlist__stats (session->evlist)->nr_events[PERF_RECORD_LOST ]; out_delete:perf_session__delete (session);
tools/perf/builtin-lock.c +1 -1
@@ -2129,7 +2129,7 @@ static int __cmd_contention(int argc, const char **argv)evlist__start_workload (con.evlist);while (!done) { -if (argc &&waitpid (con.evlist->workload.pid,NULL ,WNOHANG ) >0 ) +if (argc &&waitpid (evlist__workload_pid (con.evlist),NULL ,WNOHANG ) >0 )break ;sleep (1 ); }
tools/perf/builtin-record.c +46 -45
268 lines not on this page · raw message
tools/perf/builtin-report.c +3 -3
@@ -561,7 +561,7 @@ static int evlist__tty_browse_hists(struct evlist *evlist, struct report *rep, cif (!quiet) {fprintf (stdout,"#\n# Total Lost Samples: %" PRIu64"\n#\n" , - evlist->stats.total_lost_samples); +evlist__stats (evlist)->total_lost_samples); }evlist__for_each_entry (evlist, pos) {
@@ -1156,7 +1156,7 @@ static int __cmd_report(struct report *rep) PERF_HPP_REPORT__BLOCK_AVG_CYCLES, }; -if (session->evlist->nr_br_cntr >0 ) +if (evlist__nr_br_cntr (session->evlist) >0 ) block_hpps[nr_hpps++] =PERF_HPP_REPORT__BLOCK_BRANCH_COUNTER ; block_hpps[nr_hpps++] =PERF_HPP_REPORT__BLOCK_RANGE ;
@@ -1291,7 +1291,7 @@ static int process_attr(const struct perf_tool *tool __maybe_unused,* on events sample_type. */ sample_type =evlist__combined_sample_type (*pevlist); - session = (*pevlist)->session; + session =evlist__session (*pevlist);callchain_param_setup (sample_type,perf_session__e_machine (session,/*e_flags=*/ NULL ));return 0 ; }
tools/perf/builtin-sched.c +12 -12
@@ -1972,9 +1972,9 @@ static int perf_sched__read_events(struct perf_sched *sched) goto out_delete; } - sched->nr_events = session->evlist->stats.nr_events[0 ]; - sched->nr_lost_events = session->evlist->stats.total_lost; - sched->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST ]; + sched->nr_events =evlist__stats (session->evlist)->nr_events[0 ]; + sched->nr_lost_events =evlist__stats (session->evlist)->total_lost; + sched->nr_lost_chunks =evlist__stats (session->evlist)->nr_events[PERF_RECORD_LOST ]; } rc =0 ;
@@ -3227,7 +3227,7 @@ static int timehist_check_attr(struct perf_sched *sched,struct evsel *evsel;struct evsel_runtime *er; -list_for_each_entry (evsel, &evlist->core.entries, core.node) { +list_for_each_entry (evsel, &evlist__core (evlist)->entries, core.node) { er =evsel__get_runtime (evsel);if (er ==NULL ) {pr_err ("Failed to allocate memory for evsel runtime data\n" );
@@ -3399,9 +3399,9 @@ static int perf_sched__timehist(struct perf_sched *sched) goto out; } - sched->nr_events = evlist->stats.nr_events[0 ]; - sched->nr_lost_events = evlist->stats.total_lost; - sched->nr_lost_chunks = evlist->stats.nr_events[PERF_RECORD_LOST ]; + sched->nr_events =evlist__stats (evlist)->nr_events[0 ]; + sched->nr_lost_events =evlist__stats (evlist)->total_lost; + sched->nr_lost_chunks =evlist__stats (evlist)->nr_events[PERF_RECORD_LOST ];if (sched->summary)timehist_print_summary (sched, session);
@@ -3906,7 +3906,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched,if (err <0 ) goto out; - user_requested_cpus = evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (evlist)->user_requested_cpus; err = perf_event__synthesize_schedstat(&(sched->tool), process_synthesized_schedstat_event,
@@ -3922,7 +3922,7 @@ static int perf_sched__schedstat_record(struct perf_sched *sched,evlist__start_workload (evlist);while (!done) { -if (argc &&waitpid (evlist->workload.pid,NULL ,WNOHANG ) >0 ) +if (argc &&waitpid (evlist__workload_pid (evlist),NULL ,WNOHANG ) >0 )break ;sleep (1 ); }
@@ -4623,7 +4623,7 @@ static int perf_sched__schedstat_report(struct perf_sched *sched)if (err <0 ) goto out; - user_requested_cpus = session->evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (session->evlist)->user_requested_cpus; err =perf_session__process_events (session);
@@ -4799,7 +4799,7 @@ static int perf_sched__schedstat_live(struct perf_sched *sched,if (err <0 ) goto out; - user_requested_cpus = evlist->core.user_requested_cpus; + user_requested_cpus =evlist__core (evlist)->user_requested_cpus; err = perf_event__synthesize_schedstat(&(sched->tool), process_synthesized_event_live,
@@ -4815,7 +4815,7 @@ static int perf_sched__schedstat_live(struct perf_sched *sched,evlist__start_workload (evlist);while (!done) { -if (argc &&waitpid (evlist->workload.pid,NULL ,WNOHANG ) >0 ) +if (argc &&waitpid (evlist__workload_pid (evlist),NULL ,WNOHANG ) >0 )break ;sleep (1 ); }
tools/perf/builtin-script.c +7 -6
@@ -2229,9 +2229,10 @@ static int script_find_metrics(const struct pmu_metric *pm,evlist__for_each_entry (metric_evlist, metric_evsel) {struct evsel *script_evsel =map_metric_evsel_to_script_evsel (script_evlist, metric_evsel); -struct metric_event *metric_me =metricgroup__lookup (&metric_evlist->metric_events, - metric_evsel, -/*create=*/ false); +struct metric_event *metric_me = +metricgroup__lookup (evlist__metric_events (metric_evlist), + metric_evsel, +/*create=*/ false);if (script_evsel->metric_id ==NULL ) { script_evsel->metric_id = metric_evsel->metric_id;
@@ -2251,7 +2252,7 @@ static int script_find_metrics(const struct pmu_metric *pm,if (metric_me) {struct metric_expr *expr;struct metric_event *script_me = -metricgroup__lookup (&script_evlist->metric_events, +metricgroup__lookup (evlist__metric_events (script_evlist), script_evsel,/*create=*/ true);
@@ -2321,7 +2322,7 @@ static void perf_sample__fprint_metric(struct thread *thread,assert (stat_config.aggr_mode ==AGGR_GLOBAL ); stat_config.aggr_get_id = script_aggr_cpu_id_get; stat_config.aggr_map = -cpu_aggr_map__new (evsel->evlist->core.user_requested_cpus, +cpu_aggr_map__new (evlist__core (evsel->evlist)->user_requested_cpus, aggr_cpu_id__global,/*data=*/ NULL ,/*needs_sort=*/ false); }
@@ -3909,7 +3910,7 @@ static int set_maps(struct perf_script *script)if (WARN_ONCE (script->allocated,"stats double allocation\n" ))return -EINVAL ; -perf_evlist__set_maps (&evlist->core, script->cpus, script->threads); +perf_evlist__set_maps (evlist__core (evlist), script->cpus, script->threads);if (evlist__alloc_stats (&stat_config, evlist,/*alloc_raw=*/ true))return -ENOMEM ;
tools/perf/builtin-stat.c +39 -34
206 lines not on this page · raw message
tools/perf/builtin-top.c +26 -26
175 lines not on this page · raw message
tools/perf/builtin-trace.c +11 -11
@@ -2023,7 +2023,7 @@ static int trace__symbols_init(struct trace *trace, int argc, const char **argv, goto out; err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target, - evlist->core.threads, trace__tool_process, + evlist__core(evlist)->threads, trace__tool_process, /*needs_mmap=*/callchain_param.enabled && !trace->summary_only, /*mmap_data=*/false,
@@ -4195,7 +4195,7 @@ static int trace__set_filter_pids(struct trace *trace) err =augmented_syscalls__set_filter_pids (trace->filter_pids.nr, trace->filter_pids.entries); } - }else if (perf_thread_map__pid (trace->evlist->core.threads,0 ) ==-1 ) { + }else if (perf_thread_map__pid (evlist__core (trace->evlist)->threads,0 ) ==-1 ) { err =trace__set_filter_loop_pids (trace); }
@@ -4509,7 +4509,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)fprintf (trace->output,"Couldn't run the workload!\n" ); goto out_put_evlist; } - workload_pid = evlist->workload.pid; + workload_pid =evlist__workload_pid (evlist); } err =evlist__open (evlist);
@@ -4561,7 +4561,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) goto out_error_apply_filters;if (!trace->summary_only || !trace->summary_bpf) { - err =evlist__mmap (evlist, trace->opts.mmap_pages); + err =evlist__do_mmap (evlist, trace->opts.mmap_pages);if (err <0 ) goto out_error_mmap; }
@@ -4580,8 +4580,8 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (trace->summary_bpf)trace_start_bpf_summary (); - trace->multiple_threads =perf_thread_map__pid (evlist->core.threads,0 ) ==-1 || -perf_thread_map__nr (evlist->core.threads) >1 || + trace->multiple_threads =perf_thread_map__pid (evlist__core (evlist)->threads,0 ) ==-1 || +perf_thread_map__nr (evlist__core (evlist)->threads) >1 ||evlist__first (evlist)->core.attr.inherit;/*
@@ -4598,11 +4598,11 @@ static int trace__run(struct trace *trace, int argc, const char **argv) again: before = trace->nr_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -5304,7 +5304,7 @@ static int trace__parse_cgroups(const struct option *opt, const char *str, int u {struct trace *trace = opt->value; -if (!list_empty (&trace->evlist->core.entries)) { +if (!list_empty (&evlist__core (trace->evlist)->entries)) {struct option o = { .value = &trace->evlist, };
@@ -5578,7 +5578,7 @@ int cmd_trace(int argc, const char **argv)* .perfconfig trace.add_events, and filter those out. */ if (!trace.trace_syscalls && !trace.trace_pgfaults && - trace.evlist->core.nr_entries ==0 /* Was --events used? */ ) { +evlist__nr_entries (trace.evlist) ==0 /* Was --events used? */ ) { trace.trace_syscalls = true; }/*
@@ -5664,7 +5664,7 @@ int cmd_trace(int argc, const char **argv) symbol_conf.use_callchain = true; } -if (trace.evlist->core.nr_entries >0 ) { +if (evlist__nr_entries (trace.evlist) >0 ) {bool use_btf = false;evlist__set_default_evsel_handler (trace.evlist, trace__event_handler);
tools/perf/tests/backward-ring-buffer.c +4 -4
@@ -34,8 +34,8 @@ static int count_samples(struct evlist *evlist, int *sample_count, {int i; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { -struct mmap *map = &evlist->overwrite_mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { +struct mmap *map = &evlist__overwrite_mmap (evlist)[i];union perf_event *event;perf_mmap__read_init (&map->core);
@@ -65,7 +65,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,int err;char sbuf[STRERR_BUFSIZE ]; - err =evlist__mmap (evlist, mmap_pages); + err =evlist__do_mmap (evlist, mmap_pages);if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -77,7 +77,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,evlist__disable (evlist); err =count_samples (evlist, sample_count, comm_count); -evlist__munmap (evlist); +evlist__do_munmap (evlist);return err; }
tools/perf/tests/code-reading.c +5 -5
@@ -589,8 +589,8 @@ static int process_events(struct machine *machine, struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -778,7 +778,7 @@ static int do_test_code_reading(bool try_kcore) goto out_put; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads); str = events[evidx];pr_debug ("Parsing event '%s'\n" , str);
@@ -806,7 +806,7 @@ static int do_test_code_reading(bool try_kcore)pr_debug ("perf_evlist__open() failed!\n%s\n" , errbuf); } -perf_evlist__set_maps (&evlist->core,NULL ,NULL ); +perf_evlist__set_maps (evlist__core (evlist),NULL ,NULL );evlist__put (evlist); evlist =NULL ;continue ;
@@ -817,7 +817,7 @@ static int do_test_code_reading(bool try_kcore)if (events[evidx] ==NULL ) goto out_put; - ret =evlist__mmap (evlist,UINT_MAX ); + ret =evlist__do_mmap (evlist,UINT_MAX );if (ret <0 ) {pr_debug ("evlist__mmap failed\n" ); goto out_put;
tools/perf/tests/event-times.c +1 -1
@@ -50,7 +50,7 @@ static int attach__enable_on_exec(struct evlist *evlist)static int detach__enable_on_exec (struct evlist *evlist) { -waitpid (evlist->workload.pid,NULL ,0 ); +waitpid (evlist__workload_pid (evlist),NULL ,0 );return 0 ; }
tools/perf/tests/event_update.c +1 -1
@@ -92,7 +92,7 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtesTEST_ASSERT_VAL ("failed to allocate ids" , !perf_evsel__alloc_id (&evsel->core,1 ,1 )); -perf_evlist__id_add (&evlist->core, &evsel->core,0 ,0 ,123 ); +perf_evlist__id_add (evlist__core (evlist), &evsel->core,0 ,0 ,123 );free ((char *)evsel->unit); evsel->unit =strdup ("KRAVA" );
tools/perf/tests/expand-cgroup.c +2 -2
@@ -28,7 +28,7 @@ static int test_expand_events(struct evlist *evlist)TEST_ASSERT_VAL ("evlist is empty" , !evlist__empty (evlist)); - nr_events = evlist->core.nr_entries; + nr_events =evlist__nr_entries (evlist); ev_name =calloc (nr_events,sizeof (*ev_name));if (ev_name ==NULL ) {pr_debug ("memory allocation failure\n" );
@@ -54,7 +54,7 @@ static int test_expand_events(struct evlist *evlist) } ret =TEST_FAIL ; -if (evlist->core.nr_entries != nr_events * nr_cgrps) { +if (evlist__nr_entries (evlist) != nr_events * nr_cgrps) {pr_debug ("event count doesn't match\n" ); goto out; }
tools/perf/tests/hwmon_pmu.c +3 -2
@@ -184,9 +184,10 @@ static int do_test(size_t i, bool with_pmu, bool with_alias) } ret =TEST_OK ; -if (with_pmu ? (evlist->core.nr_entries !=1 ) : (evlist->core.nr_entries <1 )) { +if (with_pmu ? (evlist__nr_entries (evlist) !=1 ) + : (evlist__nr_entries (evlist) <1 )) {pr_debug ("FAILED %s:%d Unexpected number of events for '%s' of %d\n" , - __FILE__, __LINE__, str, evlist->core.nr_entries); + __FILE__, __LINE__, str,evlist__nr_entries (evlist)); ret =TEST_FAIL ; goto out; }
tools/perf/tests/keep-tracking.c +4 -4
@@ -37,8 +37,8 @@ static int find_comm(struct evlist *evlist, const char *comm)int i, found; found =0 ; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;while ((event =perf_mmap__read_event (&md->core)) !=NULL ) {
@@ -87,7 +87,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte evlist =evlist__new ();CHECK_NOT_NULL__ (evlist); -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);CHECK__ (parse_event (evlist,"dummy:u" ));CHECK__ (parse_event (evlist,"cpu-cycles:u" ));
@@ -106,7 +106,7 @@ static int test__keep_tracking(struct test_suite *test __maybe_unused, int subte goto out_err; } -CHECK__ (evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__do_mmap (evlist,UINT_MAX ));/* * First, test that a 'comm' event can be found when the event is
tools/perf/tests/mmap-basic.c +3 -3
@@ -81,7 +81,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest goto out_free_cpus; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);for (i =0 ; i < nsyscalls; ++i) {char name[64 ];
@@ -113,7 +113,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest expected_nr_events[i] =1 +rand () %127 ; } -if (evlist__mmap (evlist,128 ) <0 ) { +if (evlist__do_mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_put_evlist;
@@ -124,7 +124,7 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest syscalls[i](); } - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/openat-syscall-tp-fields.c +4 -4
@@ -64,7 +64,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedevsel__config (evsel, &opts,NULL ); -perf_thread_map__set_pid (evlist->core.threads,0 ,getpid ()); +perf_thread_map__set_pid (evlist__core (evlist)->threads,0 ,getpid ()); err =evlist__open (evlist);if (err <0 ) {
@@ -73,7 +73,7 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unused goto out_put_evlist; } - err =evlist__mmap (evlist,UINT_MAX ); + err =evlist__do_mmap (evlist,UINT_MAX );if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -90,11 +90,11 @@ static int test__syscall_openat_tp_fields(struct test_suite *test __maybe_unusedwhile (1 ) {int before = nr_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
tools/perf/tests/parse-events.c +70 -65
519 lines not on this page · raw message
tools/perf/tests/parse-metric.c +2 -2
@@ -53,7 +53,7 @@ static double compute_single(struct evlist *evlist, const char *name)struct evsel *evsel;evlist__for_each_entry (evlist, evsel) { - me =metricgroup__lookup (&evlist->metric_events, evsel, false); + me =metricgroup__lookup (evlist__metric_events (evlist), evsel, false);if (me !=NULL ) { list_for_each_entry (mexp, &me->head, nd) {if (strcmp (mexp->metric_name, name))
@@ -88,7 +88,7 @@ static int __compute_metric(const char *name, struct value *vals,return -ENOMEM ; } -perf_evlist__set_maps (&evlist->core, cpus,NULL ); +perf_evlist__set_maps (evlist__core (evlist), cpus,NULL );/* Parse the metric into metric_events list. */ pme_test =find_core_metrics_table ("testarch" ,"testcpu" );
tools/perf/tests/perf-record.c +10 -10
@@ -129,7 +129,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestevsel__set_sample_bit (evsel,TIME );evlist__config (evlist, &opts,NULL ); - err =sched__get_first_possible_cpu (evlist->workload.pid, cpu_mask); + err =sched__get_first_possible_cpu (evlist__workload_pid (evlist), cpu_mask);if (err <0 ) {pr_debug ("sched__get_first_possible_cpu: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -142,7 +142,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest/* * So that we can check perf_sample.cpu on all the samples. */ -if (sched_setaffinity (evlist->workload.pid, cpu_mask_size, cpu_mask) <0 ) { +if (sched_setaffinity (evlist__workload_pid (evlist), cpu_mask_size, cpu_mask) <0 ) {pr_debug ("sched_setaffinity: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__cancel_workload (evlist);
@@ -166,7 +166,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest* fds in the same CPU to be injected in the same mmap ring buffer * (using ioctl(PERF_EVENT_IOC_SET_OUTPUT)). */ - err =evlist__mmap (evlist, opts.mmap_pages); + err =evlist__do_mmap (evlist, opts.mmap_pages);if (err <0 ) {pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -188,11 +188,11 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtestwhile (1 ) {int before = total_events; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) {union perf_event *event;struct mmap *md; - md = &evlist->mmap[i]; + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -231,15 +231,15 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest ++errs; } -if ((pid_t )sample.pid != evlist->workload.pid) { +if ((pid_t )sample.pid !=evlist__workload_pid (evlist)) {pr_debug ("%s with unexpected pid, expected %d, got %d\n" , - name, evlist->workload.pid, sample.pid); + name,evlist__workload_pid (evlist), sample.pid); ++errs; } -if ((pid_t )sample.tid != evlist->workload.pid) { +if ((pid_t )sample.tid !=evlist__workload_pid (evlist)) {pr_debug ("%s with unexpected tid, expected %d, got %d\n" , - name, evlist->workload.pid, sample.tid); + name,evlist__workload_pid (evlist), sample.tid); ++errs; }
@@ -248,7 +248,7 @@ static int test__PERF_RECORD(struct test_suite *test __maybe_unused, int subtest type == PERF_RECORD_MMAP2 || type == PERF_RECORD_FORK || type == PERF_RECORD_EXIT) && - (pid_t)event->comm.pid != evlist->workload.pid) { + (pid_t)event->comm.pid != evlist__workload_pid(evlist)) {pr_debug ("%s with unexpected pid/tid\n" , name); ++errs; }
tools/perf/tests/perf-time-to-tsc.c +5 -5
@@ -99,7 +99,7 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su evlist =evlist__new ();CHECK_NOT_NULL__ (evlist); -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);CHECK__ (parse_event (evlist,"cpu-cycles:u" ));
@@ -121,9 +121,9 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int su goto out_err; } -CHECK__ (evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__do_mmap (evlist,UINT_MAX )); - pc = evlist->mmap[0 ].core.base; + pc =evlist__mmap (evlist)[0 ].core.base; ret =perf_read_tsc_conversion (pc, &tc);if (ret) {if (ret == -EOPNOTSUPP ) {
@@ -145,8 +145,8 @@ static int test__perf_time_to_tsc(struct test_suite *test __maybe_unused, int suevlist__disable (evlist); -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
tools/perf/tests/pfm.c +4 -4
@@ -69,12 +69,12 @@ static int test__pfm_events(struct test_suite *test __maybe_unused,if (evlist ==NULL )return -ENOMEM ; - opt.value = evlist; + opt.value = &evlist;parse_libpfm_events_option (&opt, table[i].events,0 );TEST_ASSERT_EQUAL (table[i].events, -count_pfm_events (&evlist->core), +count_pfm_events (evlist__core (evlist)), table[i].nr_events); TEST_ASSERT_EQUAL(table[i].events, evlist__nr_groups(evlist),
@@ -154,12 +154,12 @@ static int test__pfm_group(struct test_suite *test __maybe_unused,if (evlist ==NULL )return -ENOMEM ; - opt.value = evlist; + opt.value = &evlist;parse_libpfm_events_option (&opt, table[i].events,0 );TEST_ASSERT_EQUAL (table[i].events, -count_pfm_events (&evlist->core), +count_pfm_events (evlist__core (evlist)), table[i].nr_events); TEST_ASSERT_EQUAL(table[i].events, evlist__nr_groups(evlist),
tools/perf/tests/pmu-events.c +3 -2
@@ -869,7 +869,7 @@ static int test__parsing_callback(const struct pmu_metric *pm,return -ENOMEM ; } -perf_evlist__set_maps (&evlist->core, cpus,NULL ); +perf_evlist__set_maps (evlist__core (evlist), cpus,NULL ); err =metricgroup__parse_groups_test (evlist, table, pm->metric_name);if (err) {
@@ -895,7 +895,8 @@ static int test__parsing_callback(const struct pmu_metric *pm, k++; }evlist__for_each_entry (evlist, evsel) { -struct metric_event *me =metricgroup__lookup (&evlist->metric_events, evsel, false); +struct metric_event *me =metricgroup__lookup (evlist__metric_events (evlist), + evsel, false);if (me !=NULL ) {struct metric_expr *mexp;
tools/perf/tests/sample-parsing.c +23 -21
111 lines not on this page · raw message
tools/perf/tests/sw-clock.c +3 -3
@@ -71,7 +71,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) goto out_put_evlist; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);if (evlist__open (evlist)) {const char *knob ="/proc/sys/kernel/perf_event_max_sample_rate" ;
@@ -83,7 +83,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) goto out_put_evlist; } - err =evlist__mmap (evlist,128 ); + err =evlist__do_mmap (evlist,128 );if (err <0 ) {pr_debug ("failed to mmap event: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf)));
@@ -98,7 +98,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)evlist__disable (evlist); - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/switch-tracking.c +4 -4
@@ -282,8 +282,8 @@ static int process_events(struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->core.nr_mmaps; i++) { - md = &evlist->mmap[i]; +for (i =0 ; i <evlist__core (evlist)->nr_mmaps; i++) { + md = &evlist__mmap (evlist)[i];if (perf_mmap__read_init (&md->core) <0 )continue ;
@@ -374,7 +374,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub goto out_err; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads);/* First event */ err =parse_event (evlist,"cpu-clock:u" );
@@ -471,7 +471,7 @@ static int test__switch_tracking(struct test_suite *test __maybe_unused, int sub goto out; } - err =evlist__mmap (evlist,UINT_MAX ); + err =evlist__do_mmap (evlist,UINT_MAX );if (err) {pr_debug ("evlist__mmap failed!\n" ); goto out_err;
tools/perf/tests/task-exit.c +3 -3
@@ -77,7 +77,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ goto out_put_evlist; } -perf_evlist__set_maps (&evlist->core, cpus, threads); +perf_evlist__set_maps (evlist__core (evlist), cpus, threads); err =evlist__prepare_workload (evlist, &target, argv, false, workload_exec_failed_signal);if (err <0 ) {
@@ -104,7 +104,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _ goto out_put_evlist; } -if (evlist__mmap (evlist,128 ) <0 ) { +if (evlist__do_mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); err =-1 ;
@@ -114,7 +114,7 @@ static int test__task_exit(struct test_suite *test __maybe_unused, int subtest _evlist__start_workload (evlist); retry: - md = &evlist->mmap[0 ]; + md = &evlist__mmap (evlist)[0 ];if (perf_mmap__read_init (&md->core) <0 ) goto out_init;
tools/perf/tests/time-utils-test.c +9 -5
@@ -69,16 +69,19 @@ struct test_data {static bool test__perf_time__parse_for_ranges (struct test_data *d) { -struct evlist evlist = { - .first_sample_time = d->first, - .last_sample_time = d->last, - }; -struct perf_session session = { .evlist = &evlist }; +struct evlist *evlist =evlist__new (); +struct perf_session session = { .evlist = evlist };struct perf_time_interval *ptime =NULL ;int range_size, range_num;bool pass = false;int i, err; +if (!evlist) { +pr_debug ("Missing evlist\n" ); +return false; + } +evlist__set_first_sample_time (evlist, d->first); +evlist__set_last_sample_time (evlist, d->last);pr_debug ("\nperf_time__parse_for_ranges(\"%s\")\n" , d->str);if (strchr (d->str,'%' ))
@@ -127,6 +130,7 @@ static bool test__perf_time__parse_for_ranges(struct test_data *d) pass = true; out: +evlist__put (evlist);free (ptime);return pass; }
tools/perf/tests/tool_pmu.c +3 -2
@@ -40,9 +40,10 @@ static int do_test(enum tool_pmu_event ev, bool with_pmu) } ret =TEST_OK ; -if (with_pmu ? (evlist->core.nr_entries !=1 ) : (evlist->core.nr_entries <1 )) { +if (with_pmu ? (evlist__nr_entries (evlist) !=1 ) + : (evlist__nr_entries (evlist) <1 )) {pr_debug ("FAILED %s:%d Unexpected number of events for '%s' of %d\n" , - __FILE__, __LINE__, str, evlist->core.nr_entries); + __FILE__, __LINE__, str,evlist__nr_entries (evlist)); ret =TEST_FAIL ; goto out; }
tools/perf/tests/topology.c +1 -1
@@ -46,7 +46,7 @@ static int session_write_header(char *path) session->evlist =evlist__new_default (&target,/*sample_callchains=*/ false);TEST_ASSERT_VAL ("can't get evlist" , session->evlist); - session->evlist->session = session; +evlist__set_session (session->evlist, session);perf_header__set_feat (&session->header,HEADER_CPU_TOPOLOGY );perf_header__set_feat (&session->header,HEADER_NRCPUS );
tools/perf/tests/uncore-event-sorting.c +2 -2
@@ -147,8 +147,8 @@ static int test__uncore_event_sorting(struct test_suite *test __maybe_unused, goto out_err; } -CHECK_COND (evlist->core.nr_entries >=4 ,"Number of events is >= 4" ); -CHECK_EQUAL (evlist->core.nr_entries %2 ,0 ,"Number of events is a multiple of 2" ); +CHECK_COND (evlist__nr_entries (evlist) >=4 ,"Number of events is >= 4" ); +CHECK_EQUAL (evlist__nr_entries (evlist) %2 ,0 ,"Number of events is a multiple of 2" );evlist__for_each_entry (evlist, evsel) {struct evsel *next;
tools/perf/ui/browsers/annotate.c +1 -1
@@ -597,7 +597,7 @@ static bool annotate_browser__callq(struct annotate_browser *browser, notes =symbol__annotation (dl->ops.target.sym);annotation__lock (notes); -if (!symbol__hists (dl->ops.target.sym, evsel->evlist->core.nr_entries)) { +if (!symbol__hists (dl->ops.target.sym,evlist__nr_entries (evsel->evlist))) {annotation__unlock (notes);ui__warning ("Not enough memory for annotating '%s' symbol!\n" , dl->ops.target.sym->name);
tools/perf/ui/browsers/hists.c +11 -11
@@ -688,10 +688,10 @@ static int hist_browser__handle_hotkey(struct hist_browser *browser, bool warn_lui_browser__update_nr_entries (&browser->b, nr_entries);if (warn_lost_event && - (evsel->evlist->stats.nr_lost_warned != - evsel->evlist->stats.nr_events[PERF_RECORD_LOST ])) { - evsel->evlist->stats.nr_lost_warned = - evsel->evlist->stats.nr_events[PERF_RECORD_LOST ]; + (evlist__stats (evsel->evlist)->nr_lost_warned != +evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ])) { +evlist__stats (evsel->evlist)->nr_lost_warned = +evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ];ui_browser__warn_lost_events (&browser->b); }
@@ -3321,7 +3321,7 @@ static int evsel__hists_browse(struct evsel *evsel, int nr_events, const char *h* No need to refresh, resort/decay histogram * entries if we are not collecting samples: */ -if (top->evlist->enabled) { +if (evlist__enabled (top->evlist)) { helpline ="Press 'f' to disable the events or 'h' to see other hotkeys" ; hbt->refresh = delay_secs; }else {
@@ -3493,7 +3493,7 @@ static void perf_evsel_menu__write(struct ui_browser *browser, unit, unit == ' ' ? "" : " ", ev_name);ui_browser__printf (browser,"%s" , bf); - nr_events = evsel->evlist->stats.nr_events[PERF_RECORD_LOST ]; + nr_events =evlist__stats (evsel->evlist)->nr_events[PERF_RECORD_LOST ];if (nr_events !=0 ) { menu->lost_events = true;if (!current_entry)
@@ -3559,13 +3559,13 @@ static int perf_evsel_menu__run(struct evsel_menu *menu,ui_browser__show_title (&menu->b, title);switch (key) {case K_TAB : -if (pos->core.node.next == &evlist->core.entries) +if (pos->core.node.next == &evlist__core (evlist)->entries) pos =evlist__first (evlist);else pos =evsel__next (pos); goto browse_hists;case K_UNTAB : -if (pos->core.node.prev == &evlist->core.entries) +if (pos->core.node.prev == &evlist__core (evlist)->entries) pos =evlist__last (evlist);else pos =evsel__prev (pos);
@@ -3618,7 +3618,7 @@ static int __evlist__tui_browse_hists(struct evlist *evlist, int nr_entries, construct evsel *pos;struct evsel_menu menu = { .b = { - .entries = &evlist->core.entries, + .entries = &evlist__core (evlist)->entries, .refresh = ui_browser__list_head_refresh, .seek = ui_browser__list_head_seek, .write = perf_evsel_menu__write,
@@ -3646,7 +3646,7 @@ static int __evlist__tui_browse_hists(struct evlist *evlist, int nr_entries, constatic bool evlist__single_entry (struct evlist *evlist) { -int nr_entries = evlist->core.nr_entries; +int nr_entries =evlist__nr_entries (evlist);if (nr_entries ==1 )return true;
@@ -3664,7 +3664,7 @@ static bool evlist__single_entry(struct evlist *evlist)int evlist__tui_browse_hists (struct evlist *evlist,const char *help,struct hist_browser_timer *hbt,float min_pcnt,struct perf_env *env,bool warn_lost_event) { -int nr_entries = evlist->core.nr_entries; +int nr_entries =evlist__nr_entries (evlist);if (evlist__single_entry (evlist)) { single_entry: {
tools/perf/util/amd-sample-raw.c +1 -1
@@ -421,7 +421,7 @@ static void parse_cpuid(struct perf_env *env)*/ bool evlist__has_amd_ibs (struct evlist *evlist) { -struct perf_env *env =perf_session__env (evlist->session); +struct perf_env *env =perf_session__env (evlist__session (evlist));int ret, nr_pmu_mappings =perf_env__nr_pmu_mappings (env);const char *pmu_mapping =perf_env__pmu_mappings (env);char name[sizeof ("ibs_fetch" )];
tools/perf/util/annotate-data.c +1 -1
@@ -1829,7 +1829,7 @@ int annotated_data_type__update_samples(struct annotated_data_type *adt,return 0 ;if (adt->histograms ==NULL ) { -int nr = evsel->evlist->core.nr_entries; +int nr =evlist__nr_entries (evsel->evlist);if (alloc_data_type_histograms (adt, nr) <0 )return -1 ;
tools/perf/util/annotate.c +5 -5
@@ -328,7 +328,7 @@ static int symbol__inc_addr_samples(struct map_symbol *ms,if (sym ==NULL )return 0 ; - src =symbol__hists (sym, sample->evsel->evlist->core.nr_entries); + src =symbol__hists (sym,evlist__nr_entries (sample->evsel->evlist));return src ?__symbol__inc_addr_samples (ms, src, addr, sample) :0 ; }
@@ -339,7 +339,7 @@ static int symbol__account_br_cntr(struct annotated_branch *branch, {unsigned int br_cntr_nr =evsel__leader (evsel)->br_cntr_nr;unsigned int base =evsel__leader (evsel)->br_cntr_idx; -unsigned int off = offset * evsel->evlist->nr_br_cntr; +unsigned int off = offset *evlist__nr_br_cntr (evsel->evlist);u64 *branch_br_cntr = branch->br_cntr;unsigned int i, mask, width;
@@ -369,7 +369,7 @@ static int symbol__account_cycles(u64 addr, u64 start, struct symbol *sym,if (sym ==NULL )return 0 ; - branch =symbol__find_branch_hist (sym, evsel->evlist->nr_br_cntr); + branch =symbol__find_branch_hist (sym,evlist__nr_br_cntr (evsel->evlist));if (!branch)return -ENOMEM ;if (addr < sym->start || addr >= sym->end)
@@ -511,7 +511,7 @@ static void annotation__count_and_fill(struct annotation *notes, u64 start, u64static int annotation__compute_ipc (struct annotation *notes,size_t size,struct evsel *evsel) { -unsigned int br_cntr_nr = evsel->evlist->nr_br_cntr; +unsigned int br_cntr_nr =evlist__nr_br_cntr (evsel->evlist);int err =0 ;s64 offset;
@@ -1813,7 +1813,7 @@ int annotation_br_cntr_abbr_list(char **str, struct evsel *evsel, bool header)struct evsel *pos;struct strbuf sb; -if (evsel->evlist->nr_br_cntr <=0 ) +if (evlist__nr_br_cntr (evsel->evlist) <=0 )return -ENOTSUP ;strbuf_init (&sb,/*hint=*/ 0 );
tools/perf/util/auxtrace.c +7 -7
@@ -191,7 +191,7 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp,struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu = !perf_cpu_map__has_any_cpu(evlist->core.user_requested_cpus); + bool per_cpu = !perf_cpu_map__has_any_cpu(evlist__core(evlist)->user_requested_cpus); mp->mmap_needed = evsel->needs_auxtrace_mmap;
@@ -201,11 +201,11 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, mp->idx = idx;if (per_cpu) { - mp->cpu =perf_cpu_map__cpu (evlist->core.all_cpus, idx); - mp->tid =perf_thread_map__pid (evlist->core.threads,0 ); + mp->cpu =perf_cpu_map__cpu (evlist__core (evlist)->all_cpus, idx); + mp->tid =perf_thread_map__pid (evlist__core (evlist)->threads,0 ); }else { mp->cpu.cpu =-1 ; - mp->tid =perf_thread_map__pid (evlist->core.threads, idx); + mp->tid =perf_thread_map__pid (evlist__core (evlist)->threads, idx); } }
@@ -667,10 +667,10 @@ int auxtrace_parse_snapshot_options(struct auxtrace_record *itr,static int evlist__enable_event_idx (struct evlist *evlist,struct evsel *evsel,int idx) { -bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu (evlist->core.user_requested_cpus); +bool per_cpu_mmaps = !perf_cpu_map__has_any_cpu (evlist__core (evlist)->user_requested_cpus);if (per_cpu_mmaps) { -struct perf_cpu evlist_cpu =perf_cpu_map__cpu (evlist->core.all_cpus, idx); +struct perf_cpu evlist_cpu =perf_cpu_map__cpu (evlist__core (evlist)->all_cpus, idx);int cpu_map_idx =perf_cpu_map__idx (evsel->core.cpus, evlist_cpu);if (cpu_map_idx ==-1 )
@@ -1837,7 +1837,7 @@ void perf_session__auxtrace_error_inc(struct perf_session *session,struct perf_record_auxtrace_error *e = &event->auxtrace_error;if (e->type <PERF_AUXTRACE_ERROR_MAX ) - session->evlist->stats.nr_auxtrace_errors[e->type] +=1 ; +evlist__stats (session->evlist)->nr_auxtrace_errors[e->type] +=1 ; }void events_stats__auxtrace_error_warn (const struct events_stats *stats)
tools/perf/util/block-info.c +2 -2
@@ -472,7 +472,7 @@ struct block_report *block_info__create_report(struct evlist *evlist, int *nr_reps) {struct block_report *block_reports; -int nr_hists = evlist->core.nr_entries, i =0 ; +int nr_hists =evlist__nr_entries (evlist), i =0 ;struct evsel *pos; block_reports =calloc (nr_hists,sizeof (struct block_report ));
@@ -483,7 +483,7 @@ struct block_report *block_info__create_report(struct evlist *evlist,struct hists *hists =evsel__hists (pos);process_block_report (hists, &block_reports[i], total_cycles, - block_hpps, nr_hpps, evlist->nr_br_cntr); + block_hpps, nr_hpps,evlist__nr_br_cntr (evlist)); i++; }
tools/perf/util/bpf_counter.c +1 -1
@@ -443,7 +443,7 @@ static int bperf_check_target(struct evsel *evsel, }else if (target->tid) { *filter_type =BPERF_FILTER_PID ; *filter_entry_cnt =perf_thread_map__nr (evsel->core.threads); - }else if (target->pid || evsel->evlist->workload.pid !=-1 ) { + }else if (target->pid ||evlist__workload_pid (evsel->evlist) !=-1 ) { *filter_type =BPERF_FILTER_TGID ; *filter_entry_cnt =perf_thread_map__nr (evsel->core.threads); }else {
tools/perf/util/bpf_counter_cgroup.c +6 -6
@@ -104,7 +104,7 @@ static int bperf_load_program(struct evlist *evlist)set_max_rlimit (); -if (nr_cgroups ==0 || evlist->core.nr_entries % nr_cgroups !=0 ) { +if (nr_cgroups ==0 ||evlist__nr_entries (evlist) % nr_cgroups !=0 ) {pr_err ("Invalid cgroup or event count\n" );return -EINVAL ; }
@@ -116,7 +116,7 @@ static int bperf_load_program(struct evlist *evlist)pr_err ("Failed to open cgroup skeleton\n" );return -1 ; } -setup_rodata (skel, evlist->core.nr_entries); +setup_rodata (skel,evlist__nr_entries (evlist)); err =bperf_cgroup_bpf__load (skel);if (err) {
@@ -127,12 +127,12 @@ static int bperf_load_program(struct evlist *evlist) err =-1 ; cgrp_switch =evsel__new (&cgrp_switch_attr); -if (evsel__open_per_cpu (cgrp_switch, evlist->core.all_cpus,-1 ) <0 ) { +if (evsel__open_per_cpu (cgrp_switch,evlist__core (evlist)->all_cpus,-1 ) <0 ) {pr_err ("Failed to open cgroup switches event\n" ); goto out; } -perf_cpu_map__for_each_cpu (cpu, i, evlist->core.all_cpus) { +perf_cpu_map__for_each_cpu (cpu, i,evlist__core (evlist)->all_cpus) { link =bpf_program__attach_perf_event (skel->progs.on_cgrp_switch,FD (cgrp_switch, i));if (IS_ERR (link)) {
@@ -197,7 +197,7 @@ static int bperf_load_program(struct evlist *evlist)*/ {struct evsel *leader; -int num_events = evlist->core.nr_entries / nr_cgroups; +int num_events =evlist__nr_entries (evlist) / nr_cgroups;evlist__for_each_entry (evlist, evsel) { leader =evlist__find_evsel (evlist, evsel->core.idx % num_events);
@@ -258,7 +258,7 @@ static int bperf_cgrp__sync_counters(struct evlist *evlist)unsigned int idx;int prog_fd =bpf_program__fd (skel->progs.trigger_read); -perf_cpu_map__for_each_cpu (cpu, idx, evlist->core.all_cpus) +perf_cpu_map__for_each_cpu (cpu, idx,evlist__core (evlist)->all_cpus)bperf_trigger_reading (prog_fd, cpu.cpu);return 0 ;
tools/perf/util/bpf_ftrace.c +5 -4
@@ -59,13 +59,13 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace)/* don't need to set cpu filter for system-wide mode */ if (ftrace->target.cpu_list) { - ncpus =perf_cpu_map__nr (ftrace->evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (ftrace->evlist)->user_requested_cpus);bpf_map__set_max_entries (skel->maps.cpu_filter, ncpus); skel->rodata->has_cpu =1 ; }if (target__has_task (&ftrace->target) ||target__none (&ftrace->target)) { - ntasks =perf_thread_map__nr (ftrace->evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (ftrace->evlist)->threads);bpf_map__set_max_entries (skel->maps.task_filter, ntasks); skel->rodata->has_task =1 ; }
@@ -87,7 +87,8 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (ftrace->evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu ( +evlist__core (ftrace->evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -99,7 +100,7 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (ftrace->evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (ftrace->evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } }
tools/perf/util/bpf_lock_contention.c +6 -6
@@ -223,11 +223,11 @@ int lock_contention_prepare(struct lock_contention *con)if (target__has_cpu (target)) { skel->rodata->has_cpu =1 ; - ncpus =perf_cpu_map__nr (evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus); }if (target__has_task (target)) { skel->rodata->has_task =1 ; - ntasks =perf_thread_map__nr (evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (evlist)->threads); }if (con->filters->nr_types) { skel->rodata->has_type =1 ;
@@ -334,7 +334,7 @@ int lock_contention_prepare(struct lock_contention *con) fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu (evlist__core (evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -346,13 +346,13 @@ int lock_contention_prepare(struct lock_contention *con) fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } } -if (target__none (target) && evlist->workload.pid >0 ) { -u32 pid = evlist->workload.pid; +if (target__none (target) &&evlist__workload_pid (evlist) >0 ) { +u32 pid =evlist__workload_pid (evlist);u8 val =1 ; fd =bpf_map__fd (skel->maps.task_filter);
tools/perf/util/bpf_off_cpu.c +7 -7
@@ -73,13 +73,13 @@ static void off_cpu_start(void *arg)/* update task filter for the given workload */ if (skel->rodata->has_task && skel->rodata->uses_tgid && -perf_thread_map__pid (evlist->core.threads,0 ) !=-1 ) { +perf_thread_map__pid (evlist__core (evlist)->threads,0 ) !=-1 ) {int fd;u32 pid;u8 val =1 ; fd =bpf_map__fd (skel->maps.task_filter); - pid =perf_thread_map__pid (evlist->core.threads,0 ); + pid =perf_thread_map__pid (evlist__core (evlist)->threads,0 );bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); }
@@ -168,7 +168,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target,/* don't need to set cpu filter for system-wide mode */ if (target->cpu_list) { - ncpus =perf_cpu_map__nr (evlist->core.user_requested_cpus); + ncpus =perf_cpu_map__nr (evlist__core (evlist)->user_requested_cpus);bpf_map__set_max_entries (skel->maps.cpu_filter, ncpus); skel->rodata->has_cpu =1 ; }
@@ -199,7 +199,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, skel->rodata->has_task =1 ; skel->rodata->uses_tgid =1 ; }else if (target__has_task (target)) { - ntasks =perf_thread_map__nr (evlist->core.threads); + ntasks =perf_thread_map__nr (evlist__core (evlist)->threads);bpf_map__set_max_entries (skel->maps.task_filter, ntasks); skel->rodata->has_task =1 ; }else if (target__none (target)) {
@@ -209,7 +209,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, }if (evlist__first (evlist)->cgrp) { - ncgrps = evlist->core.nr_entries -1 ;/* excluding a dummy */ + ncgrps =evlist__nr_entries (evlist) -1 ;/* excluding a dummy */ bpf_map__set_max_entries (skel->maps.cgroup_filter, ncgrps);if (!cgroup_is_v2 ("perf_event" ))
@@ -240,7 +240,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, fd =bpf_map__fd (skel->maps.cpu_filter);for (i =0 ; i < ncpus; i++) { - cpu =perf_cpu_map__cpu (evlist->core.user_requested_cpus, i).cpu; + cpu =perf_cpu_map__cpu (evlist__core (evlist)->user_requested_cpus, i).cpu;bpf_map_update_elem (fd, &cpu, &val,BPF_ANY ); } }
@@ -269,7 +269,7 @@ int off_cpu_prepare(struct evlist *evlist, struct target *target, fd =bpf_map__fd (skel->maps.task_filter);for (i =0 ; i < ntasks; i++) { - pid =perf_thread_map__pid (evlist->core.threads, i); + pid =perf_thread_map__pid (evlist__core (evlist)->threads, i);bpf_map_update_elem (fd, &pid, &val,BPF_ANY ); } }
tools/perf/util/cgroup.c +10 -10
@@ -367,7 +367,7 @@ int parse_cgroups(const struct option *opt, const char *str,char *s;int ret, i; -if (list_empty (&evlist->core.entries)) { +if (list_empty (&evlist__core (evlist)->entries)) {fprintf (stderr,"must define events before cgroups\n" );return -1 ; }
@@ -423,7 +423,7 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgroint ret =-1 ;int prefix_len; -if (evlist->core.nr_entries ==0 ) { +if (evlist__nr_entries (evlist) ==0 ) {fprintf (stderr,"must define events before cgroups\n" );return -EINVAL ; }
@@ -436,11 +436,11 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro }/* save original events and init evlist */ -evlist__splice_list_tail (orig_list, &evlist->core.entries); - evlist->core.nr_entries =0 ; +evlist__splice_list_tail (orig_list, &evlist__core (evlist)->entries); +evlist__core (evlist)->nr_entries =0 ; - orig_metric_events = evlist->metric_events; -metricgroup__rblist_init (&evlist->metric_events); + orig_metric_events = *evlist__metric_events (evlist); +metricgroup__rblist_init (evlist__metric_events (evlist));if (has_pattern_string (str)) prefix_len =match_cgroups (str);
@@ -503,15 +503,15 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgro nr_cgroups++;if (metricgroup__copy_metric_events (tmp_list, cgrp, - &evlist->metric_events, +evlist__metric_events (evlist), &orig_metric_events) <0 ) goto out_err; -evlist__splice_list_tail (evlist, &tmp_list->core.entries); - tmp_list->core.nr_entries =0 ; +evlist__splice_list_tail (evlist, &evlist__core (tmp_list)->entries); +evlist__core (tmp_list)->nr_entries =0 ; } -if (list_empty (&evlist->core.entries)) { +if (list_empty (&evlist__core (evlist)->entries)) {fprintf (stderr,"no cgroup matched: %s\n" , str); goto out_err; }
tools/perf/util/cs-etm.c +4 -1
@@ -1636,8 +1636,9 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, {int ret =0 ;struct cs_etm_auxtrace *etm = etmq->etm; -struct perf_sample sample = {.ip =0 ,}; +struct perf_sample sample;union perf_event *event = tidq->event_buf; +struct dummy_branch_stack {u64 nr;u64 hw_idx;
@@ -1645,6 +1646,7 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, } dummy_bs;u64 ip; +perf_sample__init (&sample,/*all=*/ true); ip =cs_etm__last_executed_instr (tidq->prev_packet); event->sample.header.type =PERF_RECORD_SAMPLE ;
@@ -1698,6 +1700,7 @@ static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq,"CS ETM Trace: failed to deliver instruction event, error %d\n" , ret); +perf_sample__exit (&sample);return ret; }
tools/perf/util/evlist.c +229 -162
190 lines not on this page · raw message
tools/perf/util/evlist.h +235 -16
251 lines not on this page · raw message
tools/perf/util/header.c +24 -23
47 lines not on this page · raw message
tools/perf/util/python.c +65 -27
92 lines not on this page · raw message
tools/perf/util/session.c +30 -24
54 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Part of the diff is not shown here. View the raw message.
[email protected]
AI reviewRe: [PATCH v11 07/19] perf evlist: Add reference count checking
Thank you for your contribution! Sashiko AI review found 6 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 6 potential issue(s) to consider:
New issues:
- [Critical] Double Free and TOCTOU Race in evlist__put()
- [Critical] Use-After-Free / Dangling Pointers in evlist__put cycle collection
- [High] Memory Leak of evsel via missing evsel__put() in sample delivery
- [High] Memory leak of spliced evlists due to missing re-parenting
Pre-existing issues:
- [High] Python memory leak in pyrf_evlist__read_on_cpu
- [High] Python object memory leak in pyrf_evlist__add --
commit e390129ddc82f7909e4e9e3187e55f4b0eb8b77b
Author: Ian Rogers <[email protected]>
perf evlist: Add reference count checking
This patch adds reference count checking for the perf evlist to ensure that gets and puts are correctly paired and easy to debug.
Quoted 5 lines
tools/perf/util/evlist.c +0
-0@@ -183,16 +200,21 @@ static void evlist__purge(struct evlist *evlist)
tools/perf/util/evlist.c static void evlist__purge(struct evlist *evlist) · line 210
Quoted 10 lines
tools/perf/util/evlist.c +5
-0evlist__for_each_entry_safe (evlist, n, pos) {list_del_init (&pos->core.node); +if (pos->evlist) { +/* Minimal evlist__put. */ +refcount_dec_and_test (evlist__refcnt (pos->evlist)); +RC_CHK_PUT (pos->evlist); + } pos->evlist =NULL ;evsel__put (pos); }
[Severity: High]
Does ignoring the return value of refcount_dec_and_test() cause a memory leak when splicing is used? When evlist__splice_list_tail() moves evsels from a source evlist to a destination evlist, the spliced evsels still point to the old source evlist. When this purge logic hits the source evlist, the refcount drops to 0 and refcount_dec_and_test() returns true, but because the return value is ignored, the memory for the source evlist is never actually freed.
[ ... ]
Quoted 9 lines
tools/perf/util/evlist.c +8
-0@@ -211,15 +240,43 @@ void evlist__put(struct evlist *evlist) +retry: + count =refcount_read (evlist__refcnt (evlist)); + ref_cnt = count; +evlist__for_each_entry (evlist, evsel) { +if (RC_CHK_EQUAL (evsel->evlist, evlist) && count) + count--; + } +if (refcount_read (evlist__refcnt (evlist)) != ref_cnt)
tools/perf/util/evlist.c void evlist__put(struct evlist *evlist) · line 257
Quoted 10 lines
tools/perf/util/evlist.c +9
-0+ goto retry; +if (count !=0 ) { +/* +* Not the last reference except for back references from +* evsels. +*/ +RC_CHK_PUT (evlist); +return ; + }
[Severity: Critical]
Can this TOCTOU race lead to a double free? If two threads concurrently drop an external reference to the evlist, they both might execute refcount_dec_and_test(). If the refcount drops to the number of evsels with back-references, both threads could enter the retry block simultaneously, read the count, and decrement it to 0. Since the refcount hasn't changed during the loop, the verification evaluates to false for both. Both threads would see count == 0 and fall through to out_delete, causing a double free.
[Severity: Critical]
Does this cycle collection logic prematurely free the evlist by ignoring the refcounts of the evsel objects themselves? If an external user holds a strong reference to an evsel (where its refcount > 1), it relies on the evsel's strong reference to the evlist to keep the evlist alive. By proceeding to destroy the evlist here without checking evsel refcounts, evlist__purge() will set pos->evlist = NULL, leaving the external user with a dangling or NULL evlist pointer.
Quoted 16 lines
tools/perf/util/python.c +4
-1@@ -1804,27 +1823,32 @@ static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist, PyObject *args, PyObject *kwargs __maybe_unused) { -struct evlist *evlist = pevlist->evlist; +struct evlist *evlist;PyObject *pevsel;struct evsel *evsel; +CHECK_INITIALIZED (pevlist->evlist,"evlist" ); + + evlist = pevlist->evlist;if (!PyArg_ParseTuple (args,"O!" , &pyrf_evsel__type, &pevsel))
tools/perf/util/python.c static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist, · line 1841
Quoted 10 lines
tools/perf/util/python.c +4
-2return NULL ; -Py_INCREF (pevsel); evsel = ((struct pyrf_evsel *)pevsel)->evsel; - evsel->core.idx = evlist->core.nr_entries; +CHECK_INITIALIZED (evsel,"evsel" ); + +Py_INCREF (pevsel); + evsel->core.idx =evlist__nr_entries (evlist);evlist__add (evlist,evsel__get (evsel));
[Severity: High]
This isn't a bug introduced by this patch, but does calling Py_INCREF(pevsel) here cause a memory leak? The Python wrapper object is incremented but never stored in any Python collection or decremented later. The underlying C evsel is added to the C evlist (which takes its own reference via evsel__get). Because the Python reference is incremented but never decremented, the wrapper is permanently leaked, preventing its destructor from running and leaking the initial C reference it holds.
[ ... ]
Quoted 8 lines
tools/perf/util/python.c +1
-1@@ -1836,13 +1860,16 @@ static struct mmap *get_md(struct evlist *evlist, int cpu)static PyObject *pyrf_evlist__read_on_cpu (struct pyrf_evlist *pevlist,PyObject *args,PyObject *kwargs) { -struct evlist *evlist = pevlist->evlist; +struct evlist *evlist;union perf_event *event;int sample_id_all =1 , cpu;
tools/perf/util/python.c static struct mmap *get_md(struct evlist *evlist, int cpu) · line 1875
Quoted 10 lines
tools/perf/util/python.c +3
-0static char *kwlist[] = {"cpu" ,"sample_id_all" ,NULL };struct mmap *md;int err; +CHECK_INITIALIZED (pevlist->evlist,"evlist" ); + + evlist = pevlist->evlist;if (!PyArg_ParseTupleAndKeywords (args, kwargs,"i|i" , kwlist, &cpu, &sample_id_all))return NULL ;
[Severity: High]
This is a pre-existing issue, but does this function leak the python object on failure? If evlist__parse_sample() fails, the code returns an exception via PyErr_Format() but fails to Py_DECREF() the allocated pyevent object, permanently leaking it. Additionally, if it succeeds, it calls Py_INCREF() before returning it, bumping the refcount to 2 while the caller only expects 1, preventing the object from ever being garbage collected.
Quoted 8 lines
tools/perf/util/session.c +0
-1@@ -1959,21 +1963,23 @@ static int machines__deliver_event(struct machines *machines,dump_event (evlist, event, file_offset, sample, file_path); -if (!sample->evsel)
tools/perf/util/session.c static int machines__deliver_event(struct machines *machines, · line 1973
Quoted 10 lines
tools/perf/util/session.c +5
-2+if (!sample->evsel) { sample->evsel =evlist__id2evsel (evlist, sample->id); - else +if (sample->evsel) + sample->evsel =evsel__get (sample->evsel); + }else {assert (sample->evsel ==evlist__id2evsel (evlist, sample->id)); - + } machine =machines__find_for_cpumode (machines, event, sample);
[Severity: High]
Does adding evsel__get() here introduce a memory leak? The sample delivery path populates sample->evsel and takes a reference, but the cleanup paths like perf_sample__exit() do not appear to have a matching evsel__put(). This permanently leaks the evsel object and its reference wrapper for events processed this way.
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=7
Ian Rogers
reviewerPATCH v11 8/19 perf python: Use evsel in sample in pyrf_event
In reply to Ian RogersAvoid a duplicated evsel by using the one in sample. Add evsel__get/put to the evsel in perf_sample.
--- tools/perf/util/evsel.c | 2 +- tools/perf/util/python.c | 10 +++------- tools/perf/util/sample.c | 17 ++++++++++++----- 3 files changed, 16 insertions(+), 13 deletions(-)
tools/perf/util/evsel.c +1 -1
@@ -3365,7 +3365,7 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event,union u64_swap u;perf_sample__init (data,/*all=*/ true); - data->evsel = evsel; + data->evsel =evsel__get (evsel); data->cpu = data->pid = data->tid =-1 ; data->stream_id = data->id = data->time =-1ULL ; data->period = evsel->core.attr.sample_period;
tools/perf/util/python.c +3 -7
@@ -63,7 +63,6 @@ PyMODINIT_FUNC PyInit_perf(void);struct pyrf_event {PyObject_HEAD - struct evsel *evsel; struct perf_sample sample;union perf_event event; };
@@ -294,7 +293,6 @@ static PyMemberDef pyrf_sample_event__members[] = {static void pyrf_sample_event__delete (struct pyrf_event *pevent) { -evsel__put (pevent->evsel);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -316,7 +314,7 @@ static PyObject *pyrf_sample_event__repr(const struct pyrf_event *pevent)#ifdef HAVE_LIBTRACEEVENT static bool is_tracepoint (const struct pyrf_event *pevent) { -return pevent->evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; +return pevent->sample.evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; } static PyObject*
@@ -363,7 +361,7 @@ tracepoint_field(const struct pyrf_event *pe, struct tep_format_field *field)static PyObject *get_tracepoint_field (struct pyrf_event *pevent,PyObject *attr_name) { -struct evsel *evsel = pevent->evsel; +struct evsel *evsel = pevent->sample.evsel;struct tep_event *tp_format =evsel__tp_format (evsel);struct tep_format_field *field;
@@ -529,7 +527,7 @@ static PyObject *pyrf_event__new(const union perf_event *event) pevent = PyObject_New(struct pyrf_event, ptype);if (pevent !=NULL ) {memcpy (&pevent->event, event, event->header.size); - pevent->evsel =NULL ; +perf_sample__init (&pevent->sample,/*all=*/ false); }return (PyObject *)pevent; }
@@ -1899,8 +1897,6 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,return Py_None; } - pevent->evsel =evsel__get (evsel); -perf_mmap__consume (&md->core); err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample);
tools/perf/util/sample.c +12 -5
@@ -1,18 +1,23 @@/* SPDX-License-Identifier: GPL-2.0 */ #include "sample.h" + +#include <stdlib.h> +#include <string.h> + +#include <elf.h> +#include <linux/zalloc.h> + +#include "../../arch/x86/include/asm/insn.h" #include "debug.h" +#include "evsel.h" #include "thread.h" -#include <elf.h> +#ifndef EM_CSKY #define EM_CSKY 252#endif #ifndef EM_LOONGARCH #define EM_LOONGARCH 258#endif -#include <linux/zalloc.h> -#include <stdlib.h> -#include <string.h> -#include "../../arch/x86/include/asm/insn.h" void perf_sample__init (struct perf_sample *sample,bool all) {
@@ -29,6 +34,8 @@ void perf_sample__init(struct perf_sample *sample, bool all)void perf_sample__exit (struct perf_sample *sample) { +evsel__put (sample->evsel); + sample->evsel =NULL ;zfree (&sample->user_regs);zfree (&sample->intr_regs);if (sample->merged_callchain) {
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 08/19] perf python: Use evsel in sample in pyrf_event
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [High] Overwriting `sample.evsel` without dropping the existing reference causes an `evsel` memory leak.…
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] Overwriting sample.evsel without dropping the existing reference causes an evsel memory leak.
- [High] Leaked evsel reference when evsel__parse_sample or evlist__parse_sample fails on error paths that bypass cleanup. --
commit f3d12718541e9a7c485a3e3991c159212f6a7942
Author: Ian Rogers <[email protected]>
perf python: Use evsel in sample in pyrf_event
Avoid a duplicated evsel by using the one in sample. Add evsel__get/put to the evsel in perf_sample.
tools/perf/util/evsel.c int evsel__parse_sample(struct evsel *evsel, union perf_event *event, · line 3368
Quoted 10 lines
tools/perf/util/evsel.c +1
-1@@ -3365,7 +3365,7 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event,union u64_swap u;perf_sample__init (data,/*all=*/ true); - data->evsel = evsel; + data->evsel =evsel__get (evsel);
[Severity: High]
Since data->evsel now holds a reference count acquired here, do existing callers that manually overwrite sample.evsel need to be updated?
For example, in tools/perf/util/session.c:session__flush_deferred_samples():
sample.evsel = evlist__id2evsel(evlist, sample.id);
if (sample.evsel)
sample.evsel = evsel__get(sample.evsel);If evlist__parse_sample() already populated sample.evsel with an evsel__get() reference, does this manual overwrite permanently leak the original reference since perf_sample__exit() will only drop the new one?
[Severity: High]
If parsing fails later in evsel__parse_sample() (for example, by jumping to out_efault), the function returns an error, but data->evsel already holds the reference.
Since several callers abort on parsing failure without calling perf_sample__exit(), does this error path permanently leak the evsel reference?
For example, in tools/perf/builtin-inject.c:guest_session__fetch():
ret = evlist__parse_sample(gs->session->evlist, gs->ev.event,
&gs->ev.sample);
if (ret) {
pr_err("Parse failed fetching guest event");
return ret;
}Quoted 3 lines
tools/perf/util/evsel.c +0
-0data->cpu = data->pid = data->tid =-1 ; data->stream_id = data->id = data->time =-1ULL ; data->period = evsel->core.attr.sample_period;
[ ... ]
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=8
Ian Rogers
reviewerPATCH v11 9/19 perf python: Add wrapper for perf_data file abstraction
In reply to Ian RogersThe perf_data struct is needed for session supported.
---
v2:
1. Fixed Memory & FD Leaks in pyrf_data__init : Added cleanup of old state (closing file and freeing path) if __init__ is called multiple times on the same object.
2. Fixed Invalid Free in pyrf_data__delete : Ensured pdata->data.path is always dynamically allocated via strdup() , even for the default "perf.data" . This avoids passing a static string literal to free().
3. Fixed NULL Pointer Dereference in pyrf_data__str : Added a check for NULL path to prevent segfaults if str() or repr() is called on an uninitialized object.
4. Guarded fd Argument Usage: Added a check to ensure that if an fd is provided, it corresponds to a pipe, failing gracefully with a ValueError otherwise.
v7:
- Added pyrf_data__new to zero-initialize pdata->data to fix crash on re-initialization.
--- tools/perf/util/python.c | 123 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 122 insertions(+), 1 deletion(-)
tools/perf/util/python.c +122 -1
@@ -6,6 +6,7 @@#include <linux/err.h> #include <poll.h> +#include <unistd.h> #include <internal/lib.h> #include <perf/cpumap.h>
@@ -14,6 +15,7 @@#include "callchain.h" #include "counts.h" +#include "data.h" #include "event.h" #include "evlist.h" #include "evsel.h"
@@ -2454,6 +2456,121 @@ static PyObject *pyrf__metrics(PyObject *self, PyObject *args)return list; } +struct pyrf_data { +PyObject_HEAD + + struct perf_data data; +}; + +static int pyrf_data__init (struct pyrf_data *pdata,PyObject *args,PyObject *kwargs) +{ +static char *kwlist[] = {"path" ,"fd" ,NULL }; +char *path =NULL ; +int fd =-1 ; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"|si" , kwlist, &path, &fd)) +return -1 ; + +if (pdata->data.open) +perf_data__close (&pdata->data); +free ((char *)pdata->data.path); + pdata->data.path =NULL ; + +if (fd !=-1 ) { +struct stat st; + +if (fstat (fd, &st) <0 || !S_ISFIFO (st.st_mode)) { +PyErr_SetString (PyExc_ValueError, +"fd argument is only supported for pipes" ); +return -1 ; + } +if (!path) + path ="-" ; +else if (strcmp (path,"-" ) !=0 ) { +PyErr_SetString (PyExc_ValueError, +"path must be '-' when fd is provided" ); +return -1 ; + } + fd =dup (fd); +if (fd <0 ) { +PyErr_SetFromErrno (PyExc_OSError); +return -1 ; + } + }else if (path &&strcmp (path,"-" ) ==0 ) { + fd =dup (0 ); +if (fd <0 ) { +PyErr_SetFromErrno (PyExc_OSError); +return -1 ; + } + } + +if (!path) + path ="perf.data" ; + + pdata->data.path =strdup (path); +if (!pdata->data.path) { +if (fd !=-1 ) +close (fd); +PyErr_NoMemory (); +return -1 ; + } + + pdata->data.mode =PERF_DATA_MODE_READ ; + pdata->data.file.fd = fd; +if (perf_data__open (&pdata->data) <0 ) { +PyErr_Format (PyExc_IOError,"Failed to open perf data: %s" , + pdata->data.path ? pdata->data.path :"perf.data" ); +return -1 ; + } +return 0 ; +} + +static void pyrf_data__delete (struct pyrf_data *pdata) +{ +perf_data__close (&pdata->data); +free ((char *)pdata->data.path); +Py_TYPE (pdata)->tp_free ((PyObject *)pdata); +} + +static PyObject *pyrf_data__str (PyObject *self) +{ +const struct pyrf_data *pdata = (const struct pyrf_data *)self; + +if (!pdata->data.path) +return PyUnicode_FromString ("[uninitialized]" ); +return PyUnicode_FromString (pdata->data.path); +} + +static PyObject *pyrf_data__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) +{ +struct pyrf_data *pdata; + + pdata = (struct pyrf_data *)PyType_GenericNew (type, args, kwargs); +if (pdata) +memset (&pdata->data,0 ,sizeof (pdata->data)); +return (PyObject *)pdata; +} + +static const char pyrf_data__doc[] =PyDoc_STR ("perf data file object." ); + +static PyTypeObject pyrf_data__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.data" , + .tp_basicsize =sizeof (struct pyrf_data ), + .tp_dealloc = (destructor )pyrf_data__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_data__doc, + .tp_init = (initproc )pyrf_data__init, + .tp_repr = pyrf_data__str, + .tp_str = pyrf_data__str, +}; + +static int pyrf_data__setup_types (void ) +{ + pyrf_data__type.tp_new = pyrf_data__new; +return PyType_Ready (&pyrf_data__type); +} +static PyMethodDef perf__methods[] = { { .ml_name ="metrics" ,
@@ -2516,7 +2633,8 @@ PyMODINIT_FUNC PyInit_perf(void)pyrf_cpu_map__setup_types () <0 ||pyrf_pmu_iterator__setup_types () <0 ||pyrf_pmu__setup_types () <0 || - pyrf_counts_values__setup_types() < 0) +pyrf_counts_values__setup_types () <0 || + pyrf_data__setup_types() < 0) return module;/* The page_size is placed in util object. */
@@ -2564,6 +2682,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_counts_values__type);PyModule_AddObject (module,"counts_values" , (PyObject *)&pyrf_counts_values__type); +Py_INCREF (&pyrf_data__type); +PyModule_AddObject (module,"data" , (PyObject *)&pyrf_data__type); + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 10/19 perf python: Add python session abstraction wrapping perf's session
In reply to Ian RogersSessions are necessary to be able to use perf.data files within a tool. Add a wrapper python type that incorporates the tool. Allow a sample callback to be passed when creating the session. When process_events is run this callback will be called, if supplied, for sample events.
An example use looks like:
```
$ perf record -e cycles,instructions -a sleep 3
$ PYTHONPATH=..../perf/python python3
Python 3.13.7 (main, Aug 20 2025, 22:17:40) [GCC 14.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import perf
>>> count=0
... def handle_sample(x):
... global count
... if count < 3:
... print(dir(x))
... count = count + 1
... perf.session(perf.data("perf.data"),sample=handle_sample).process_events()
...
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'sample_addr', 'sample_cpu', 'sample_id', 'sample_ip', 'sample_period', 'sample_pid', 'sample_stream_id', 'sample_tid', 'sample_time', 'type']
```Also, add the ability to get the thread associated with a session. For threads, allow the comm string to be retrieved. This can be useful for filtering threads. Connect up some of the standard event handling in psession->tool to better support queries of the machine. Also connect up the symbols.
---
v2:
1. Fixed Potential Crash in pyrf_thread__comm : Used thread__comm_str() to safely retrieve the command name, avoiding a crash if thread__comm() returns NULL.
2. Fixed Double Free Risk: Zeroed out user_regs , intr_regs , and callchain in the shallow copy of perf_sample to prevent Python from attempting to free pointers it doesn't own.
3. Fixed Memory Leak & Exception Handling in Callback: Handled the return value of PyObject_CallFunction() to avoid leaks, and checked for failure to abort the loop and propagate Python exceptions cleanly.
4. Enforced Type Safety: Used O! with &pyrf_data__type in PyArg_ParseTupleAndKeywords to prevent bad casts from passing arbitrary objects as perf.data.
5. Added Missing Build ID Handler: Registered perf_event__process_build_id to allow correct symbol resolution.
6. Fixed Double Free Crash on Init Failure: Set session and pdata to NULL on failure to prevent tp_dealloc from double-freeing them.
7. Preserved C-level Errors: Made pyrf_session__process_events return the error code integer rather than always returning None .
v7:
- Fixed NULL comm handling.
- Avoided swallowing exceptions in module init.
- Fixed checkpatch warning for missing blank line.
v8:
- Switch from pyrf_session__init to pyrf_session__new to avoid dealing with a potentially NULL session variable.
- Added pid, tid, ppid, and cpu attributes to perf.thread.
v10:
- Added CHECK_INITIALIZED checks to pyrf_session methods to prevent NULL pointer dereferences on uninitialized objects.
--- tools/perf/util/python.c | 341 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 335 insertions(+), 6 deletions(-)
tools/perf/util/python.c +335 -6
@@ -14,8 +14,10 @@#include <structmember.h> #include "callchain.h" +#include "comm.h" #include "counts.h" #include "data.h" +#include "debug.h" #include "event.h" #include "evlist.h" #include "evsel.h"
@@ -25,8 +27,14 @@#include "pmus.h" #include "print_binary.h" #include "record.h" +#include "session.h" #include "strbuf.h" +#include "symbol.h" +#include "stat.h" +#include "header.h" +#include "thread.h" #include "thread_map.h" +#include "tool.h" #include "tp_pmu.h" #include "trace-event.h" #include "util/sample.h"
@@ -2571,6 +2579,316 @@ static int pyrf_data__setup_types(void)return PyType_Ready (&pyrf_data__type); } +struct pyrf_thread { +PyObject_HEAD + + struct thread *thread; +}; + +static void pyrf_thread__delete (struct pyrf_thread *pthread) +{ +thread__put (pthread->thread); +Py_TYPE (pthread)->tp_free ((PyObject *)pthread); +} + +static PyObject *pyrf_thread__comm (PyObject *obj) +{ +struct pyrf_thread *pthread = (void *)obj; +const char *str =thread__comm_str (pthread->thread); + +if (!str) + Py_RETURN_NONE; + +return PyUnicode_FromString (str); +} + +static PyMethodDef pyrf_thread__methods[] = { + { + .ml_name ="comm" , + .ml_meth = (PyCFunction )pyrf_thread__comm, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Comm(and) associated with this thread." ) + }, + { .ml_name =NULL , } +}; + +static PyObject *pyrf_thread__get_pid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__pid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_tid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__tid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_ppid (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__ppid (pthread->thread)); +} + +static PyObject *pyrf_thread__get_cpu (struct pyrf_thread *pthread,void *closure__maybe_unused ) +{ +return PyLong_FromLong (thread__cpu (pthread->thread)); +} + +static PyGetSetDef pyrf_thread__getset[] = { + { .name ="pid" , .get = (getter )pyrf_thread__get_pid, .doc ="process ID" }, + { .name ="tid" , .get = (getter )pyrf_thread__get_tid, .doc ="thread ID" }, + { .name ="ppid" , .get = (getter )pyrf_thread__get_ppid, .doc ="parent process ID" }, + { .name ="cpu" , .get = (getter )pyrf_thread__get_cpu, .doc ="cpu number" }, + { .name =NULL } +}; + +static const char pyrf_thread__doc[] =PyDoc_STR ("perf thread object." ); + +static PyTypeObject pyrf_thread__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.thread" , + .tp_basicsize =sizeof (struct pyrf_thread ), + .tp_dealloc = (destructor )pyrf_thread__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_methods = pyrf_thread__methods, + .tp_getset = pyrf_thread__getset, + .tp_doc = pyrf_thread__doc, +}; + +static int pyrf_thread__setup_types (void ) +{ +return PyType_Ready (&pyrf_thread__type); +} + +static PyObject *pyrf_thread__from_thread (struct thread *thread) +{ + struct pyrf_thread *pthread = PyObject_New(struct pyrf_thread, &pyrf_thread__type); + +if (!pthread) +return NULL ; + + pthread->thread =thread__get (thread); +return (PyObject *)pthread; +} + +struct pyrf_session { +PyObject_HEAD + + struct perf_session *session; +struct perf_tool tool; +struct pyrf_data *pdata; +PyObject *sample; +}; + +static int pyrf_session_tool__sample (const struct perf_tool *tool, +union perf_event *event, +struct perf_sample *sample, +struct machine *machine__maybe_unused ) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +PyObject *pyevent =pyrf_event__new (event); +struct pyrf_event *pevent = (struct pyrf_event *)pyevent; +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + +memcpy (&pevent->event, event, event->header.size); +if (evsel__parse_sample (sample->evsel, &pevent->event, &pevent->sample) <0 ) { +Py_DECREF (pyevent); +return -1 ; + } + + ret =PyObject_CallFunction (psession->sample,"O" , pyevent); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} + +static PyObject *pyrf_session__find_thread (struct pyrf_session *psession,PyObject *args) +{ +struct machine *machine; +struct thread *thread =NULL ; +PyObject *result; +int pid; + +CHECK_INITIALIZED (psession->session,"session" ); + +if (!PyArg_ParseTuple (args,"i" , &pid)) +return NULL ; + + machine = &psession->session->machines.host; + thread =machine__find_thread (machine, pid, pid); + +if (!thread) { + machine =perf_session__find_machine (psession->session, pid); +if (machine) + thread =machine__find_thread (machine, pid, pid); + } + +if (!thread) { +PyErr_Format (PyExc_TypeError,"Failed to find thread %d" , pid); +return NULL ; + } + result =pyrf_thread__from_thread (thread); +thread__put (thread); +return result; +} + +static PyObject *pyrf_session__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) +{ +struct pyrf_data *pdata; +PyObject *sample =NULL ; +static char *kwlist[] = {"data" ,"sample" ,NULL }; +struct pyrf_session *psession; +struct perf_session *session; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|O" , kwlist, &pyrf_data__type, &pdata, + &sample)) +return NULL ; + + psession = PyObject_New(struct pyrf_session, type); +if (!psession) +return NULL ; + + psession->session =NULL ; + psession->sample =NULL ; + psession->pdata =NULL ; + +Py_INCREF (pdata); + psession->pdata = pdata; + +perf_tool__init (&psession->tool,/*ordered_events=*/ true); + psession->tool.ordering_requires_timestamps = true; + +#define ADD_TOOL (name) \ + do { \ + if (name) { \ + if (!PyCallable_Check(name)) { \ + PyErr_SetString(PyExc_TypeError, #name " must be callable"); \ + goto err_out; \ + } \ + psession->tool.name = pyrf_session_tool__##name; \ + Py_INCREF(name); \ + psession->name = name; \ + } \ + } while (0) + +ADD_TOOL (sample); +#undef ADD_TOOL + + psession->tool.comm = perf_event__process_comm; + psession->tool.mmap = perf_event__process_mmap; + psession->tool.mmap2 = perf_event__process_mmap2; + psession->tool.namespaces = perf_event__process_namespaces; + psession->tool.cgroup = perf_event__process_cgroup; + psession->tool.exit = perf_event__process_exit; + psession->tool.fork = perf_event__process_fork; + psession->tool.ksymbol = perf_event__process_ksymbol; + psession->tool.text_poke = perf_event__process_text_poke; + psession->tool.build_id = perf_event__process_build_id; + psession->tool.attr = perf_event__process_attr; + psession->tool.feature = perf_event__process_feature; + psession->tool.stat = perf_event__process_stat_event; + session =perf_session__new (&pdata->data, &psession->tool); +if (IS_ERR (session)) { +PyErr_Format (PyExc_IOError,"failed to create session: %ld" ,PTR_ERR (session)); + goto err_out; + } + psession->session = session; + +if (symbol__init (perf_session__env (session)) <0 ) { +PyErr_SetString (PyExc_OSError,"perf: symbol__init failed" ); + goto err_out; + } + +if (perf_session__create_kernel_maps (session) <0 ) +pr_warning ("Cannot read kernel map\n" ); + +return (PyObject *)psession; +err_out: +Py_DECREF (psession); +return NULL ; +} + +static void pyrf_session__delete (struct pyrf_session *psession) +{ +perf_session__delete (psession->session); +Py_XDECREF (psession->pdata); +Py_XDECREF (psession->sample); +Py_TYPE (psession)->tp_free ((PyObject *)psession); +} + +static PyObject *pyrf_session__find_thread_events (struct pyrf_session *psession) +{ +int err; + +CHECK_INITIALIZED (psession->session,"session" ); + + err =perf_session__process_events (psession->session); + +if (PyErr_Occurred ()) +return NULL ; + +return PyLong_FromLong (err); +} + +static PyMethodDef pyrf_session__methods[] = { + { + .ml_name ="process_events" , + .ml_meth = (PyCFunction )pyrf_session__find_thread_events, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Iterate and process events." ) + }, + { + .ml_name ="find_thread" , + .ml_meth = (PyCFunction )pyrf_session__find_thread, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Returns the thread associated with a pid." ) + }, + { .ml_name =NULL , } +}; + +static const char pyrf_session__doc[] =PyDoc_STR ("perf session object." ); + +static PyObject *pyrf_session__getattro (struct pyrf_session *psession,PyObject *attr_name) +{ +if (!psession->session) { +PyErr_SetString (PyExc_ValueError,"session not initialized" ); +return NULL ; + } +return PyObject_GenericGetAttr ((PyObject *) psession, attr_name); +} + +static int pyrf_session__setattro (struct pyrf_session *psession,PyObject *attr_name,PyObject *value) +{ +if (!psession->session) { +PyErr_SetString (PyExc_ValueError,"session not initialized" ); +return -1 ; + } +return PyObject_GenericSetAttr ((PyObject *) psession, attr_name, value); +} + +static PyTypeObject pyrf_session__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.session" , + .tp_basicsize =sizeof (struct pyrf_session ), + .tp_dealloc = (destructor )pyrf_session__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_methods = pyrf_session__methods, + .tp_doc = pyrf_session__doc, + .tp_new = pyrf_session__new, + .tp_getattro = (getattrofunc ) pyrf_session__getattro, + .tp_setattro = (setattrofunc ) pyrf_session__setattro, +}; + +static int pyrf_session__setup_types (void ) +{ +return PyType_Ready (&pyrf_session__type); +} +static PyMethodDef perf__methods[] = { { .ml_name ="metrics" ,
@@ -2625,8 +2943,10 @@ PyMODINIT_FUNC PyInit_perf(void) };PyObject *module =PyModule_Create (&moduledef); - if (module == NULL || - pyrf_event__setup_types() < 0 || +if (module ==NULL ) +return NULL ; + + if (pyrf_event__setup_types() < 0 || pyrf_evlist__setup_types() < 0 || pyrf_evsel__setup_types() < 0 || pyrf_thread_map__setup_types() < 0 ||
@@ -2634,8 +2954,12 @@ PyMODINIT_FUNC PyInit_perf(void) pyrf_pmu_iterator__setup_types() < 0 || pyrf_pmu__setup_types() < 0 || pyrf_counts_values__setup_types() < 0 || - pyrf_data__setup_types() < 0) - return module; + pyrf_data__setup_types() < 0 || + pyrf_session__setup_types() < 0 || + pyrf_thread__setup_types() < 0) { +Py_DECREF (module); +return NULL ; + }/* The page_size is placed in util object. */ page_size =sysconf (_SC_PAGE_SIZE);
@@ -2685,6 +3009,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_data__type);PyModule_AddObject (module,"data" , (PyObject *)&pyrf_data__type); +Py_INCREF (&pyrf_session__type); +PyModule_AddObject (module,"session" , (PyObject *)&pyrf_session__type); + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
@@ -2698,7 +3025,9 @@ PyMODINIT_FUNC PyInit_perf(void) } error: -if (PyErr_Occurred ()) -PyErr_SetString (PyExc_ImportError,"perf: Init failed!" ); +if (PyErr_Occurred ()) { +Py_XDECREF (module); +return NULL ; + }return module; }
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 11/19 perf python: Refactor and add accessors to sample event
In reply to Ian RogersAdd common evsel field for events and move sample specific fields to only be present in sample events. Add accessors for sample events. Ensure offsets are within the bounds of the event. Allocate just enough memory for the copied event, don't make the maximum event size each time.
---
v5:
1. Fix Uninitialized Memory: Restore zero-initialization of pevent->sample in pyrf_event__new() to prevent wild free crashes on error paths.
v6:
- Refactored pyrf_event__new to take evsel and session, and use dynamic allocation based on event size. Updated callers.
v8:
- Ensure events are properly deallocated.
--- tools/perf/util/python.c | 499 +++++++++++++++++++++++++++++++------- tools/perf/util/session.c | 2 +- tools/perf/util/session.h | 2 + 3 files changed, 418 insertions(+), 85 deletions(-)
tools/perf/util/python.c +415 -84
@@ -13,21 +13,28 @@#include <perf/mmap.h> #include <structmember.h> +#include "addr_location.h" +#include "build-id.h" #include "callchain.h" #include "comm.h" #include "counts.h" #include "data.h" #include "debug.h" +#include "dso.h" #include "event.h" #include "evlist.h" #include "evsel.h" #include "expr.h" +#include "map.h" #include "metricgroup.h" #include "mmap.h" #include "pmus.h" #include "print_binary.h" #include "record.h" +#include "sample.h" #include "session.h" +#include "srccode.h" +#include "srcline.h" #include "strbuf.h" #include "symbol.h" #include "stat.h"
@@ -37,7 +44,6 @@#include "tool.h" #include "tp_pmu.h" #include "trace-event.h" -#include "util/sample.h" #ifdef HAVE_LIBTRACEEVENT #include <event-parse.h>
@@ -45,6 +51,8 @@PyMODINIT_FUNC PyInit_perf (void ); +static PyObject *pyrf_evsel__from_evsel (struct evsel *evsel); +#define member_def (type, member, ptype, help) \ { #member, ptype, \ offsetof(struct pyrf_event, event) + offsetof(struct type, member), \
@@ -73,21 +81,53 @@ PyMODINIT_FUNC PyInit_perf(void);struct pyrf_event {PyObject_HEAD +/** @sample: The parsed sample from the event. */ struct perf_sample sample; -union perf_event event; +/** @al: The address location from machine__resolve, lazily computed. */ +struct addr_location al; +/** @al_resolved: True when machine__resolve been called. */ +bool al_resolved; +/** @event: The underlying perf_event that may be in a file or ring buffer. */ +union perf_event event; };#define sample_members \ - sample_member_def(sample_ip, ip, T_ULONGLONG, "event ip"), \ sample_member_def(sample_pid, pid, T_INT, "event pid"), \ sample_member_def(sample_tid, tid, T_INT, "event tid"), \ sample_member_def(sample_time, time, T_ULONGLONG, "event timestamp"), \ - sample_member_def(sample_addr, addr, T_ULONGLONG, "event addr"), \ sample_member_def(sample_id, id, T_ULONGLONG, "event id"), \ sample_member_def(sample_stream_id, stream_id, T_ULONGLONG, "event stream id"), \ sample_member_def(sample_period, period, T_ULONGLONG, "event period"), \ sample_member_def(sample_cpu, cpu, T_UINT, "event cpu"), +static PyObject *pyrf_event__get_evsel (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->sample.evsel) + Py_RETURN_NONE; + +return pyrf_evsel__from_evsel (pevent->sample.evsel); +} + +static PyGetSetDef pyrf_event__getset[] = { + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { .name =NULL , }, +}; + +static void pyrf_event__delete (struct pyrf_event *pevent) +{ +if (pevent->al_resolved) +addr_location__exit (&pevent->al); +perf_sample__exit (&pevent->sample); +Py_TYPE (pevent)->tp_free ((PyObject *)pevent); +} +static const char pyrf_mmap_event__doc[] =PyDoc_STR ("perf mmap event object." );static PyMemberDef pyrf_mmap_event__members[] = {
@@ -126,9 +166,11 @@ static PyTypeObject pyrf_mmap_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.mmap_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_mmap_event__doc, .tp_members = pyrf_mmap_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_mmap_event__repr, };
@@ -161,9 +203,11 @@ static PyTypeObject pyrf_task_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.task_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_task_event__doc, .tp_members = pyrf_task_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_task_event__repr, };
@@ -190,9 +234,11 @@ static PyTypeObject pyrf_comm_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.comm_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_comm_event__doc, .tp_members = pyrf_comm_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_comm_event__repr, };
@@ -222,9 +268,11 @@ static PyTypeObject pyrf_throttle_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.throttle_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_throttle_event__doc, .tp_members = pyrf_throttle_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_throttle_event__repr, };
@@ -257,9 +305,11 @@ static PyTypeObject pyrf_lost_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.lost_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_lost_event__doc, .tp_members = pyrf_lost_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_lost_event__repr, };
@@ -287,9 +337,11 @@ static PyTypeObject pyrf_read_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.read_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_read_event__doc, .tp_members = pyrf_read_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_read_event__repr, };
@@ -297,16 +349,17 @@ static const char pyrf_sample_event__doc[] = PyDoc_STR("perf sample event objectstatic PyMemberDef pyrf_sample_event__members[] = { sample_members +sample_member_def (sample_ip, ip,T_ULONGLONG ,"event ip" ), +sample_member_def (sample_addr, addr,T_ULONGLONG ,"event addr" ), +sample_member_def (sample_phys_addr, phys_addr,T_ULONGLONG ,"event physical addr" ), +sample_member_def (sample_weight, weight,T_ULONGLONG ,"event weight" ), +sample_member_def (sample_data_src, data_src,T_ULONGLONG ,"event data source" ), +sample_member_def (sample_insn_count, insn_cnt,T_ULONGLONG ,"event instruction count" ), +sample_member_def (sample_cyc_count, cyc_cnt,T_ULONGLONG ,"event cycle count" ),member_def (perf_event_header, type,T_UINT ,"event type" ), { .name =NULL , }, }; -static void pyrf_sample_event__delete (struct pyrf_event *pevent) -{ -perf_sample__exit (&pevent->sample); -Py_TYPE (pevent)->tp_free ((PyObject *)pevent); -} -static PyObject *pyrf_sample_event__repr (const struct pyrf_event *pevent) {PyObject *ret;
@@ -324,6 +377,8 @@ static PyObject *pyrf_sample_event__repr(const struct pyrf_event *pevent)#ifdef HAVE_LIBTRACEEVENT static bool is_tracepoint (const struct pyrf_event *pevent) { +if (!pevent->sample.evsel) +return false;return pevent->sample.evsel->core.attr.type ==PERF_TYPE_TRACEPOINT ; }
@@ -394,6 +449,199 @@ get_tracepoint_field(struct pyrf_event *pevent, PyObject *attr_name) }#endif /* HAVE_LIBTRACEEVENT */ +static int pyrf_sample_event__resolve_al (struct pyrf_event *pevent) +{ +struct evsel *evsel = pevent->sample.evsel; +struct evlist *evlist = evsel ? evsel->evlist :NULL ; +struct perf_session *session = evlist ?evlist__session (evlist) :NULL ; + +if (pevent->al_resolved) +return 0 ; + +if (!session) +return -1 ; + +addr_location__init (&pevent->al); +if (machine__resolve (&session->machines.host, &pevent->al, &pevent->sample) <0 ) { +addr_location__exit (&pevent->al); +return -1 ; + } + + pevent->al_resolved = true; +return 0 ; +} + +static PyObject *pyrf_sample_event__get_dso (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyUnicode_FromString (dso__name (map__dso (pevent->al.map))); +} + +static PyObject *pyrf_sample_event__get_dso_long_name (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyUnicode_FromString (dso__long_name (map__dso (pevent->al.map))); +} + +static PyObject *pyrf_sample_event__get_dso_bid (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +char sbuild_id[SBUILD_ID_SIZE ]; + +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +build_id__snprintf (dso__bid (map__dso (pevent->al.map)), sbuild_id,sizeof (sbuild_id)); +return PyUnicode_FromString (sbuild_id); +} + +static PyObject *pyrf_sample_event__get_map_start (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (map__start (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_map_end (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (map__end (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_map_pgoff (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.map) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (map__pgoff (pevent->al.map)); +} + +static PyObject *pyrf_sample_event__get_symbol (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyUnicode_FromString (pevent->al.sym->name); +} + +static PyObject *pyrf_sample_event__get_sym_start (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->al.sym->start); +} + +static PyObject *pyrf_sample_event__get_sym_end (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pyrf_sample_event__resolve_al (pevent) <0 || !pevent->al.sym) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->al.sym->end); +} + +static PyObject *pyrf_sample_event__get_raw_buf (struct pyrf_event *pevent, +void *closure__maybe_unused ) +{ +if (pevent->event.header.type !=PERF_RECORD_SAMPLE ) + Py_RETURN_NONE; + +return PyBytes_FromStringAndSize ((const char *)pevent->sample.raw_data, + pevent->sample.raw_size); +} + +static PyObject *pyrf_sample_event__srccode (PyObject *self,PyObject *args) +{ +struct pyrf_event *pevent = (void *)self; +u64 addr = pevent->sample.ip; +char *srcfile =NULL ; +char *srccode =NULL ; +unsigned int line =0 ; +int len =0 ; +PyObject *result; +struct addr_location al; + +if (!PyArg_ParseTuple (args,"|K" , &addr)) +return NULL ; + +if (pyrf_sample_event__resolve_al (pevent) <0 ) + Py_RETURN_NONE; + +if (addr != pevent->sample.ip) { +addr_location__init (&al); +thread__find_symbol_fb (pevent->al.thread, pevent->sample.cpumode, addr, &al); + }else { +addr_location__init (&al); + al.thread =thread__get (pevent->al.thread); + al.map =map__get (pevent->al.map); + al.sym = pevent->al.sym; + al.addr = pevent->al.addr; + } + +if (al.map) { +struct dso *dso =map__dso (al.map); + +if (dso) { + srcfile =get_srcline_split (dso,map__rip_2objdump (al.map, addr), + &line); + } + } +addr_location__exit (&al); + +if (srcfile) { + srccode =find_sourceline (srcfile, line, &len); + result =Py_BuildValue ("(sIs#)" , srcfile, line, srccode, (Py_ssize_t )len); +free (srcfile); + }else { + result =Py_BuildValue ("(sIs#)" ,NULL ,0 ,NULL , (Py_ssize_t )0 ); + } + +return result; +} + +static PyObject *pyrf_sample_event__insn (PyObject *self,PyObject *args__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; +struct thread *thread; +struct machine *machine; + +if (pyrf_sample_event__resolve_al (pevent) <0 ) + Py_RETURN_NONE; + + thread = pevent->al.thread; + +if (!thread || !thread__maps (thread)) + Py_RETURN_NONE; + + machine =maps__machine (thread__maps (thread)); +if (!machine) + Py_RETURN_NONE; + +if (pevent->sample.ip && !pevent->sample.insn_len) +perf_sample__fetch_insn (&pevent->sample, thread, machine); + +if (!pevent->sample.insn_len) + Py_RETURN_NONE; + +return PyBytes_FromStringAndSize ((const char *)pevent->sample.insn, + pevent->sample.insn_len); +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -407,13 +655,102 @@ pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name)return obj ?:PyObject_GenericGetAttr ((PyObject *) pevent, attr_name); } +static PyGetSetDef pyrf_sample_event__getset[] = { + { + .name ="raw_buf" , + .get = (getter )pyrf_sample_event__get_raw_buf, + .set =NULL , + .doc ="event raw buffer." , + }, + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { + .name ="dso" , + .get = (getter )pyrf_sample_event__get_dso, + .set =NULL , + .doc ="event dso short name." , + }, + { + .name ="dso_long_name" , + .get = (getter )pyrf_sample_event__get_dso_long_name, + .set =NULL , + .doc ="event dso long name." , + }, + { + .name ="dso_bid" , + .get = (getter )pyrf_sample_event__get_dso_bid, + .set =NULL , + .doc ="event dso build id." , + }, + { + .name ="map_start" , + .get = (getter )pyrf_sample_event__get_map_start, + .set =NULL , + .doc ="event map start address." , + }, + { + .name ="map_end" , + .get = (getter )pyrf_sample_event__get_map_end, + .set =NULL , + .doc ="event map end address." , + }, + { + .name ="map_pgoff" , + .get = (getter )pyrf_sample_event__get_map_pgoff, + .set =NULL , + .doc ="event map page offset." , + }, + { + .name ="symbol" , + .get = (getter )pyrf_sample_event__get_symbol, + .set =NULL , + .doc ="event symbol name." , + }, + { + .name ="sym_start" , + .get = (getter )pyrf_sample_event__get_sym_start, + .set =NULL , + .doc ="event symbol start address." , + }, + { + .name ="sym_end" , + .get = (getter )pyrf_sample_event__get_sym_end, + .set =NULL , + .doc ="event symbol end address." , + }, + { .name =NULL , }, +}; + +static PyMethodDef pyrf_sample_event__methods[] = { + { + .ml_name ="srccode" , + .ml_meth = (PyCFunction )pyrf_sample_event__srccode, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Get source code for an address." ) + }, + { + .ml_name ="insn" , + .ml_meth = (PyCFunction )pyrf_sample_event__insn, + .ml_flags =METH_NOARGS , + .ml_doc =PyDoc_STR ("Get instruction bytes for a sample." ) + }, + { .ml_name =NULL , } +}; +static PyTypeObject pyrf_sample_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name ="perf.sample_event" , .tp_basicsize =sizeof (struct pyrf_event ), + .tp_dealloc = (destructor )pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_sample_event__doc, .tp_members = pyrf_sample_event__members, + .tp_getset = pyrf_sample_event__getset, + .tp_methods = pyrf_sample_event__methods, .tp_repr = (reprfunc )pyrf_sample_event__repr, .tp_getattro = (getattrofunc ) pyrf_sample_event__getattro, };
@@ -449,25 +786,17 @@ static PyTypeObject pyrf_context_switch_event__type = {PyVarObject_HEAD_INIT (NULL ,0 ) .tp_name = "perf.context_switch_event", .tp_basicsize = sizeof(struct pyrf_event), + .tp_dealloc = (destructor)pyrf_event__delete, .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, .tp_doc = pyrf_context_switch_event__doc, .tp_members = pyrf_context_switch_event__members, + .tp_getset = pyrf_event__getset, .tp_repr = (reprfunc)pyrf_context_switch_event__repr, };static int pyrf_event__setup_types (void ) {int err; - pyrf_mmap_event__type.tp_new = - pyrf_task_event__type.tp_new = - pyrf_comm_event__type.tp_new = - pyrf_lost_event__type.tp_new = - pyrf_read_event__type.tp_new = - pyrf_sample_event__type.tp_new = - pyrf_context_switch_event__type.tp_new = - pyrf_throttle_event__type.tp_new = PyType_GenericNew; - - pyrf_sample_event__type.tp_dealloc = (destructor )pyrf_sample_event__delete, err =PyType_Ready (&pyrf_mmap_event__type);if (err <0 )
@@ -511,33 +840,56 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, }; -static PyObject *pyrf_event__new (const union perf_event *event) +static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel) {struct pyrf_event *pevent; -PyTypeObject *ptype; +int err; +u32 min_size; -if ((event->header.type <PERF_RECORD_MMAP || - event->header.type >PERF_RECORD_SAMPLE ) && - !(event->header.type ==PERF_RECORD_SWITCH || - event->header.type ==PERF_RECORD_SWITCH_CPU_WIDE )) { -PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , +if (event->header.type >=ARRAY_SIZE (pyrf_event__type) || + pyrf_event__type[event->header.type] ==NULL ) { +return PyErr_Format (PyExc_TypeError,"Unexpected header type %u" , event->header.type); -return NULL ; } -// FIXME this better be dynamic or we need to parse everything -// before calling perf_mmap__consume(), including tracepoint fields. -if (sizeof (pevent->event) < event->header.size) { -PyErr_Format (PyExc_TypeError,"Unexpected event size: %zd < %u" , -sizeof (pevent->event), event->header.size); -return NULL ; +if (perf_event__too_small (event, &min_size)) { +return PyErr_Format (PyExc_ValueError,"Event size %u too small for type %u" , + event->header.size, event->header.type); } - ptype = pyrf_event__type[event->header.type]; - pevent = PyObject_New(struct pyrf_event, ptype); -if (pevent !=NULL ) { -memcpy (&pevent->event, event, event->header.size); -perf_sample__init (&pevent->sample,/*all=*/ false); +size_t copy_size = event->header.size; + + pevent = PyObject_New(struct pyrf_event, pyrf_event__type[event->header.type]); +if (pevent ==NULL ) +return PyErr_NoMemory (); + +if (copy_size >sizeof (pevent->event)) + copy_size =sizeof (pevent->event); + +/* Copy the event for memory safety and initilaize variables. */ +memcpy (&pevent->event, event, copy_size); + +if (event->header.type ==PERF_RECORD_MMAP ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - + offsetof(struct perf_record_mmap , filename); + + pevent->event.mmap.filename[max_len -1 ] ='\0' ; + } + +perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->al_resolved = false; +addr_location__init (&pevent->al); + +if (!evsel) +return (PyObject *)pevent; + +/* Parse the sample again so that pointers are within the copied event. */ + err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample); +if (err <0 ) { +Py_DECREF (pevent); +return PyErr_Format (PyExc_OSError, +"perf: can't parse sample, err=%d" , err); }return (PyObject *)pevent; }
@@ -1242,7 +1594,7 @@ static PyObject *pyrf_evsel__str(PyObject *self)if (!evsel)return PyUnicode_FromString ("evsel(uninitialized)" ); -return PyUnicode_FromFormat ("evsel(%s/%s/)" ,evsel__pmu_name (evsel),evsel__name (evsel)); +return PyUnicode_FromFormat ("evsel(%s)" ,evsel__name (evsel)); }static PyMethodDef pyrf_evsel__methods[] = {
@@ -1870,9 +2222,11 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist, {struct evlist *evlist;union perf_event *event; +struct evsel *evsel;int sample_id_all =1 , cpu;static char *kwlist[] = {"cpu" ,"sample_id_all" ,NULL };struct mmap *md; +PyObject *pyevent;int err;CHECK_INITIALIZED (pevlist->evlist,"evlist" );
@@ -1883,44 +2237,31 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,return NULL ; md =get_md (evlist, cpu); -if (!md) { -PyErr_Format (PyExc_TypeError,"Unknown CPU '%d'" , cpu); -return NULL ; - } +if (!md) +return PyErr_Format (PyExc_TypeError,"Unknown CPU '%d'" , cpu); -if (perf_mmap__read_init (&md->core) <0 ) - goto end; + err =perf_mmap__read_init (&md->core); +if (err <0 ) { +return PyErr_Format (PyExc_OSError, +"perf: error mmap read init, err=%d" , err); + } event =perf_mmap__read_event (&md->core); -if (event !=NULL ) { -PyObject *pyevent =pyrf_event__new (event); -struct pyrf_event *pevent = (struct pyrf_event *)pyevent; -struct evsel *evsel; - -if (pyevent ==NULL ) -return PyErr_NoMemory (); - - evsel =evlist__event2evsel (evlist, event); -if (!evsel) { -Py_DECREF (pyevent); -Py_INCREF (Py_None); -return Py_None; - } +if (event ==NULL ) + Py_RETURN_NONE; + evsel =evlist__event2evsel (evlist, event); +if (!evsel) { +/* Unknown evsel. */ perf_mmap__consume (&md->core); - - err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample); -if (err) { -Py_DECREF (pyevent); -return PyErr_Format (PyExc_OSError, -"perf: can't parse sample, err=%d" , err); - } - -return pyevent; + Py_RETURN_NONE; } -end: -Py_INCREF (Py_None); -return Py_None; + pyevent =pyrf_event__new (event, evsel); +perf_mmap__consume (&md->core); +if (pyevent ==NULL ) +return PyErr_Occurred () ?NULL :PyErr_NoMemory (); + +return pyevent; } static PyObject *pyrf_evlist__open(struct pyrf_evlist *pevlist,
@@ -2132,10 +2473,7 @@ static PyObject *pyrf_evlist__str(PyObject *self)evlist__for_each_entry (pevlist->evlist, pos) {if (!first)strbuf_addch (&sb,',' ); -if (!pos->pmu) -strbuf_addstr (&sb,evsel__name (pos)); -else -strbuf_addf (&sb,"%s/%s/" , pos->pmu->name,evsel__name (pos)); +strbuf_addstr (&sb,evsel__name (pos)); first = false; }strbuf_addstr (&sb,"])" );
@@ -2684,19 +3022,12 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool, struct machine *machine __maybe_unused) { struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); -PyObject *pyevent =pyrf_event__new (event); -struct pyrf_event *pevent = (struct pyrf_event *)pyevent; +PyObject *pyevent =pyrf_event__new (event, sample->evsel);PyObject *ret;if (pyevent ==NULL )return -ENOMEM ; -memcpy (&pevent->event, event, event->header.size); -if (evsel__parse_sample (sample->evsel, &pevent->event, &pevent->sample) <0 ) { -Py_DECREF (pyevent); -return -1 ; - } - ret =PyObject_CallFunction (psession->sample,"O" , pyevent);if (!ret) {Py_DECREF (pyevent);
tools/perf/util/session.c +1 -1
@@ -2666,7 +2666,7 @@ static const u32 perf_event__min_size[PERF_RECORD_HEADER_MAX] = {* Caller must ensure event->header.type < PERF_RECORD_HEADER_MAX. * If min is non-NULL, stores the required minimum on failure. */ -static bool perf_event__too_small (const union perf_event *event,u32 *min) +bool perf_event__too_small (const union perf_event *event,u32 *min) {u32 min_sz = perf_event__min_size[event->header.type];
tools/perf/util/session.h +2 -0
@@ -122,6 +122,8 @@ void perf_session__delete(struct perf_session *session);void perf_event_header__bswap (struct perf_event_header *hdr); +bool perf_event__too_small (const union perf_event *event,u32 *min); + int perf_session__peek_event(struct perf_session *session, off_t file_offset, void *buf, size_t buf_sz, union perf_event **event_ptr,
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 11/19] perf python: Refactor and add accessors to sample event
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [High] Truncating the event payload during copy while preserving the original large `header.size` leads…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Truncating the event payload during copy while preserving the original large header.size leads to heap out-of-bounds reads and writes. --
commit 929f25f8c9aa5f28716697be8db6159cf17bcacf
Author: Ian Rogers <[email protected]>
perf python: Refactor and add accessors to sample event
This commit adds a common evsel field for events and moves sample specific fields. It adds accessors for sample events and changes memory allocation to only allocate enough memory for the copied event instead of the maximum event size.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 7 lines
tools/perf/util/python.c +1
-1@@ -511,33 +840,56 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, }; -static PyObject *pyrf_event__new (const union perf_event *event) +static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel) {
[ ... ]
Quoted 8 lines
tools/perf/util/python.c +8
-0+size_t copy_size = event->header.size; + + pevent = PyObject_New(struct pyrf_event, pyrf_event__type[event->header.type]); +if (pevent ==NULL ) +return PyErr_NoMemory (); + +if (copy_size >sizeof (pevent->event)) + copy_size =sizeof (pevent->event);
tools/perf/util/python.c static PyTypeObject *pyrf_event__type[] = { · line 877
Quoted 10 lines
tools/perf/util/python.c +10
-0+ +/* Copy the event for memory safety and initilaize variables. */ +memcpy (&pevent->event, event, copy_size); + +if (event->header.type ==PERF_RECORD_MMAP ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - + offsetof(struct perf_record_mmap , filename); + + pevent->event.mmap.filename[max_len -1 ] ='\0' ;
[Severity: High]
In tools/perf/util/python.c:pyrf_event__new(), if event->header.size is greater than sizeof(pevent->event), the copy is clamped but pevent->event.header.size retains the original large value.
Could this cause an out-of-bounds heap write when calculating max_len for the NUL byte termination?
tools/perf/util/python.c static PyTypeObject *pyrf_event__type[] = { · line 888
Quoted 11 lines
tools/perf/util/python.c +11
-0+ } + +perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->al_resolved = false; +addr_location__init (&pevent->al); + +if (!evsel) +return (PyObject *)pevent; + +/* Parse the sample again so that pointers are within the copied event. */ + err =evsel__parse_sample (evsel, &pevent->event, &pevent->sample);
[Severity: High]
Also in tools/perf/util/python.c:pyrf_event__new(), could passing pevent->event to evsel__parse_sample() with the original large header.size cause it to read past the end of the truncated pevent allocation into adjacent Python heap memory?
Quoted 7 lines
tools/perf/util/python.c +4
-0+if (err <0 ) { +Py_DECREF (pevent); +return PyErr_Format (PyExc_OSError, +"perf: can't parse sample, err=%d" , err); }return (PyObject *)pevent; }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=11
Ian Rogers
reviewerPATCH v11 12/19 perf python: Add mmap2 event
In reply to Ian RogersIf mmap is handled so should mmap2 events. Add support as a distinct python event type.
--- tools/perf/util/python.c | 162 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 161 insertions(+), 1 deletion(-)
tools/perf/util/python.c +161 -1
@@ -133,7 +133,7 @@ static const char pyrf_mmap_event__doc[] = PyDoc_STR("perf mmap event object.");static PyMemberDef pyrf_mmap_event__members[] = { sample_membersmember_def (perf_event_header, type,T_UINT ,"event type" ), -member_def (perf_event_header, misc,T_UINT ,"event misc" ), +member_def (perf_event_header, misc,T_USHORT ,"event misc" ),member_def (perf_record_mmap, pid,T_UINT ,"event pid" ),member_def (perf_record_mmap, tid,T_UINT ,"event tid" ),member_def (perf_record_mmap, start,T_ULONGLONG ,"start of the map" ),
@@ -174,6 +174,149 @@ static PyTypeObject pyrf_mmap_event__type = { .tp_repr = (reprfunc)pyrf_mmap_event__repr, }; +static const char pyrf_mmap2_event__doc[] =PyDoc_STR ("perf mmap2 event object." ); + +static PyObject *pyrf_mmap2_event__get_maj (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (pevent->event.mmap2.maj); +} + +static PyObject *pyrf_mmap2_event__get_min (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLong (pevent->event.mmap2.min); +} + +static PyObject *pyrf_mmap2_event__get_ino (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->event.mmap2.ino); +} + +static PyObject *pyrf_mmap2_event__get_ino_generation (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID ) + Py_RETURN_NONE; + +return PyLong_FromUnsignedLongLong (pevent->event.mmap2.ino_generation); +} + +static PyObject *pyrf_mmap2_event__get_build_id (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!(pevent->event.header.misc &PERF_RECORD_MISC_MMAP_BUILD_ID )) + Py_RETURN_NONE; + +int size = pevent->event.mmap2.build_id_size; + +if (size >20 ) + size =20 ; + +return PyBytes_FromStringAndSize ((const char *)pevent->event.mmap2.build_id, size); +} + +static PyGetSetDef pyrf_mmap2_event__getset[] = { + { + .name ="evsel" , + .get = pyrf_event__get_evsel, + .set =NULL , + .doc ="tracking event." , + }, + { + .name ="maj" , + .get = pyrf_mmap2_event__get_maj, + .set =NULL , + .doc ="major number." , + }, + { + .name ="min" , + .get = pyrf_mmap2_event__get_min, + .set =NULL , + .doc ="minor number." , + }, + { + .name ="ino" , + .get = pyrf_mmap2_event__get_ino, + .set =NULL , + .doc ="inode number." , + }, + { + .name ="ino_generation" , + .get = pyrf_mmap2_event__get_ino_generation, + .set =NULL , + .doc ="inode generation." , + }, + { + .name ="build_id" , + .get = pyrf_mmap2_event__get_build_id, + .set =NULL , + .doc ="binary build ID." , + }, + { .name =NULL , }, +}; + +static PyMemberDef pyrf_mmap2_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), +member_def (perf_event_header, misc,T_USHORT ,"event misc" ), +member_def (perf_record_mmap2, pid,T_UINT ,"event pid" ), +member_def (perf_record_mmap2, tid,T_UINT ,"event tid" ), +member_def (perf_record_mmap2, start,T_ULONGLONG ,"start of the map" ), +member_def (perf_record_mmap2, len,T_ULONGLONG ,"map length" ), +member_def (perf_record_mmap2, pgoff,T_ULONGLONG ,"page offset" ), +member_def (perf_record_mmap2, prot,T_UINT ,"protection" ), +member_def (perf_record_mmap2, flags,T_UINT ,"flags" ), +member_def (perf_record_mmap2, filename,T_STRING_INPLACE ,"backing store" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_mmap2_event__repr (const struct pyrf_event *pevent) +{ +PyObject *ret; +char *s; + +if (asprintf (&s,"{ type: mmap2, pid: %u, tid: %u, start: %#" PRI_lx64", " +"length: %#" PRI_lx64", offset: %#" PRI_lx64", " +"flags: %#x, prot: %#x, filename: %s }" , + pevent->event.mmap2.pid, pevent->event.mmap2.tid, + pevent->event.mmap2.start, pevent->event.mmap2.len, + pevent->event.mmap2.pgoff, pevent->event.mmap2.flags, + pevent->event.mmap2.prot, pevent->event.mmap2.filename) <0 ) +return PyErr_NoMemory (); + + ret =PyUnicode_FromString (s); +free (s); +return ret; +} + +static PyTypeObject pyrf_mmap2_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.mmap2_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_mmap2_event__doc, + .tp_members = pyrf_mmap2_event__members, + .tp_getset = pyrf_mmap2_event__getset, + .tp_repr = (reprfunc )pyrf_mmap2_event__repr, +}; +static const char pyrf_task_event__doc[] =PyDoc_STR ("perf task (fork/exit) event object." );static PyMemberDef pyrf_task_event__members[] = {
@@ -799,6 +942,9 @@ static int pyrf_event__setup_types(void)int err; err =PyType_Ready (&pyrf_mmap_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_mmap2_event__type);if (err <0 ) goto out; err =PyType_Ready (&pyrf_lost_event__type);
@@ -828,6 +974,7 @@ static int pyrf_event__setup_types(void)static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_MMAP ] = &pyrf_mmap_event__type, + [PERF_RECORD_MMAP2 ] = &pyrf_mmap2_event__type, [PERF_RECORD_LOST ] = &pyrf_lost_event__type, [PERF_RECORD_COMM ] = &pyrf_comm_event__type, [PERF_RECORD_EXIT ] = &pyrf_task_event__type,
@@ -875,6 +1022,16 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev offsetof(struct perf_record_mmap , filename); pevent->event.mmap.filename[max_len -1 ] ='\0' ; + }else if (event->header.type ==PERF_RECORD_MMAP2 ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap2 , filename); + + pevent->event.mmap2.filename[max_len -1 ] ='\0' ; + }else if (event->header.type ==PERF_RECORD_COMM ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_comm , comm); + + pevent->event.comm.comm[max_len -1 ] ='\0' ; }perf_sample__init (&pevent->sample,/*all=*/ true);
@@ -3304,6 +3461,9 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_mmap_event__type);PyModule_AddObject (module,"mmap_event" , (PyObject *)&pyrf_mmap_event__type); +Py_INCREF (&pyrf_mmap2_event__type); +PyModule_AddObject (module,"mmap2_event" , (PyObject *)&pyrf_mmap2_event__type); +Py_INCREF (&pyrf_lost_event__type);PyModule_AddObject (module,"lost_event" , (PyObject *)&pyrf_lost_event__type);
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 12/19] perf python: Add mmap2 event
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [High] Heap buffer overflow (out-of-bounds write) in pyrf_event__new. --
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Heap buffer overflow (out-of-bounds write) in pyrf_event__new.
--
commit 543923057ce17404acb2ef9bc546347f87be105e
Author: Ian Rogers <[email protected]>
perf python: Add mmap2 event
This commit adds support for mmap2 events as a distinct Python event type in the perf tools, parsing the necessary major, minor, inode, and build ID information alongside standard mmap properties.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
tools/perf/util/python.c static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev · line 1029
Quoted 9 lines
tools/perf/util/python.c +5
-0@@ -875,6 +1022,16 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev offsetof(struct perf_record_mmap , filename); pevent->event.mmap.filename[max_len -1 ] ='\0' ; + }else if (event->header.type ==PERF_RECORD_MMAP2 ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_mmap2 , filename); + + pevent->event.mmap2.filename[max_len -1 ] ='\0' ;
[Severity: High]
Can this write out of bounds?
If a perf.data file contains an event->header.size larger than the actual allocated size of pevent, wouldn't this calculation make max_len point past the end of the heap allocation?
While the initial copy might limit the copy size to prevent overflowing the buffer, this null-termination logic uses the unvalidated pevent->event.header.size. If the header size is maliciously set to a large value (e.g., up to 0xFFFF), could this result in writing a null byte past the end of the structure and corrupting allocator metadata?
tools/perf/util/python.c static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev · line 1034
Quoted 5 lines
tools/perf/util/python.c +5
-0+ }else if (event->header.type ==PERF_RECORD_COMM ) { +/* Ensure '\0' string termination. */ +size_t max_len = pevent->event.header.size - offsetof(struct perf_record_comm , comm); + + pevent->event.comm.comm[max_len -1 ] ='\0' ;
[Severity: High]
Does this have the same out-of-bounds write risk as the PERF_RECORD_MMAP2 case above?
Using the unvalidated pevent->event.header.size to calculate max_len could similarly write past the end of the heap-allocated struct pyrf_event.
Quoted 3 lines
tools/perf/util/python.c +0
-0}perf_sample__init (&pevent->sample,/*all=*/ true);
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=12
Ian Rogers
reviewerPATCH v11 13/19 perf python: Add callchain support
In reply to Ian RogersImplement pyrf_callchain_node and pyrf_callchain types for lazy iteration over callchain frames. Add callchain property to sample_event.
---
v2:
1. Eager Callchain Resolution: Moved the callchain resolution from deferred iteration to eager processing in pyrf_session_tool__sample() . This avoids risks of reading from unmapped memory or following dangling pointers to closed sessions.
2. Cached Callchain: Added a callchain field to struct pyrf_event to store the resolved object.
3. Simplified Access: pyrf_sample_event__get_callchain() now just returns the cached object if available.
4. Avoided Double Free: Handled lazy cleanups properly.
v6:
- Moved callchain resolution from session_tool__sample to pyrf_event__new.
--- tools/perf/util/python.c | 210 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 207 insertions(+), 3 deletions(-)
tools/perf/util/python.c +207 -3
@@ -87,6 +87,8 @@ struct pyrf_event {struct addr_location al;/** @al_resolved: True when machine__resolve been called. */ bool al_resolved; +/** @callchain: Resolved callchain, eagerly computed if requested. */ +PyObject *callchain;/** @event: The underlying perf_event that may be in a file or ring buffer. */ union perf_event event; };
@@ -124,6 +126,7 @@ static void pyrf_event__delete(struct pyrf_event *pevent) {if (pevent->al_resolved)addr_location__exit (&pevent->al); +Py_XDECREF (pevent->callchain);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -785,6 +788,144 @@ static PyObject *pyrf_sample_event__insn(PyObject *self, PyObject *args __maybe_ pevent->sample.insn_len); } +struct pyrf_callchain_node { +PyObject_HEAD + u64 ip; +struct map *map; +struct symbol *sym; +}; + +static void pyrf_callchain_node__delete (struct pyrf_callchain_node *pnode) +{ +map__put (pnode->map); +Py_TYPE (pnode)->tp_free ((PyObject *)pnode); +} + +static PyObject *pyrf_callchain_node__get_ip (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pnode->ip); +} + +static PyObject *pyrf_callchain_node__get_symbol (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +if (pnode->sym) +return PyUnicode_FromString (pnode->sym->name); +return PyUnicode_FromString ("[unknown]" ); +} + +static PyObject *pyrf_callchain_node__get_dso (struct pyrf_callchain_node *pnode, +void *closure__maybe_unused ) +{ +const char *dsoname ="[unknown]" ; + +if (pnode->map) { +struct dso *dso =map__dso (pnode->map); + +if (dso) { +if (symbol_conf.show_kernel_path &&dso__long_name (dso)) + dsoname =dso__long_name (dso); +else + dsoname =dso__name (dso); + } + } +return PyUnicode_FromString (dsoname); +} + +static PyGetSetDef pyrf_callchain_node__getset[] = { + { .name ="ip" , .get = (getter )pyrf_callchain_node__get_ip, }, + { .name ="symbol" , .get = (getter )pyrf_callchain_node__get_symbol, }, + { .name ="dso" , .get = (getter )pyrf_callchain_node__get_dso, }, + { .name =NULL , }, +}; + +static PyTypeObject pyrf_callchain_node__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.callchain_node" , + .tp_basicsize =sizeof (struct pyrf_callchain_node ), + .tp_dealloc = (destructor )pyrf_callchain_node__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf callchain node object." , + .tp_getset = pyrf_callchain_node__getset, +}; + +struct pyrf_callchain_frame { +u64 ip; +struct map *map; +struct symbol *sym; +}; + +struct pyrf_callchain { +PyObject_HEAD + struct pyrf_callchain_frame *frames; +u64 nr_frames; +}; + +static void pyrf_callchain__delete (struct pyrf_callchain *pchain) +{ +if (pchain->frames) { +for (u64 i =0 ; i < pchain->nr_frames; i++) +map__put (pchain->frames[i].map); +free (pchain->frames); + } +Py_TYPE (pchain)->tp_free ((PyObject *)pchain); +} + +static Py_ssize_t pyrf_callchain__length (PyObject *obj) +{ +struct pyrf_callchain *pchain = (void *)obj; + +return pchain->nr_frames; +} + +static PyObject *pyrf_callchain__item (PyObject *obj,Py_ssize_t i) +{ +struct pyrf_callchain *pchain = (void *)obj; +struct pyrf_callchain_node *pnode; + +if (i <0 || i >= (Py_ssize_t )pchain->nr_frames) { +PyErr_SetString (PyExc_IndexError,"Index out of range" ); +return NULL ; + } + + pnode = PyObject_New(struct pyrf_callchain_node, &pyrf_callchain_node__type); +if (!pnode) +return NULL ; + + pnode->ip = pchain->frames[i].ip; + pnode->map =map__get (pchain->frames[i].map); + pnode->sym = pchain->frames[i].sym; + +return (PyObject *)pnode; +} + +static PySequenceMethods pyrf_callchain__sequence_methods = { + .sq_length = pyrf_callchain__length, + .sq_item = pyrf_callchain__item, +}; + +static PyTypeObject pyrf_callchain__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.callchain" , + .tp_basicsize =sizeof (struct pyrf_callchain ), + .tp_dealloc = (destructor )pyrf_callchain__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf callchain object." , + .tp_as_sequence = &pyrf_callchain__sequence_methods, +}; + +static PyObject *pyrf_sample_event__get_callchain (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->callchain) + Py_RETURN_NONE; + +Py_INCREF (pevent->callchain); +return pevent->callchain; +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -799,6 +940,12 @@ pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name) }static PyGetSetDef pyrf_sample_event__getset[] = { + { + .name ="callchain" , + .get = pyrf_sample_event__get_callchain, + .set =NULL , + .doc ="event callchain." , + }, { .name ="raw_buf" , .get = (getter )pyrf_sample_event__get_raw_buf,
@@ -968,6 +1115,12 @@ static int pyrf_event__setup_types(void) err =PyType_Ready (&pyrf_context_switch_event__type);if (err <0 ) goto out; + err =PyType_Ready (&pyrf_callchain_node__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_callchain__type); +if (err <0 ) + goto out; out:return err; }
@@ -987,9 +1140,12 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, }; -static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel) +static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel, +struct perf_session *session) {struct pyrf_event *pevent; +struct perf_sample *sample; +struct machine *machine = session ? &session->machines.host :NULL ;int err;u32 min_size;
@@ -1035,6 +1191,7 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev }perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->callchain =NULL ; pevent->al_resolved = false;addr_location__init (&pevent->al);
@@ -1048,6 +1205,50 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evreturn PyErr_Format (PyExc_OSError,"perf: can't parse sample, err=%d" , err); } + sample = &pevent->sample; +if (machine && sample->callchain) { +struct addr_location al; +struct callchain_cursor *cursor; +u64 i; +struct pyrf_callchain *pchain; + +addr_location__init (&al); +if (machine__resolve (machine, &al, sample) >=0 ) { + cursor =get_tls_callchain_cursor (); +if (thread__resolve_callchain (al.thread, cursor, sample, +NULL ,NULL ,PERF_MAX_STACK_DEPTH ) ==0 ) { +callchain_cursor_commit (cursor); + + pchain = PyObject_New(struct pyrf_callchain, &pyrf_callchain__type); +if (!pchain) { +addr_location__exit (&al); +Py_DECREF (pevent); +return NULL ; + } + pchain->nr_frames = cursor->nr; + pchain->frames =calloc (pchain->nr_frames, +sizeof (*pchain->frames)); +if (!pchain->frames) { +Py_DECREF (pchain); +addr_location__exit (&al); +Py_DECREF (pevent); +return PyErr_NoMemory (); + } +struct callchain_cursor_node *node; + +for (i =0 ; i < pchain->nr_frames; i++) { + node =callchain_cursor_current (cursor); + pchain->frames[i].ip = node->ip; + pchain->frames[i].map = +map__get (node->ms.map); + pchain->frames[i].sym = node->ms.sym; +callchain_cursor_advance (cursor); + } + pevent->callchain = (PyObject *)pchain; + } +addr_location__exit (&al); + } + }return (PyObject *)pevent; }
@@ -2413,7 +2614,7 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,perf_mmap__consume (&md->core); Py_RETURN_NONE; } - pyevent =pyrf_event__new (event, evsel); + pyevent =pyrf_event__new (event, evsel,evlist__session (evlist));perf_mmap__consume (&md->core);if (pyevent ==NULL )return PyErr_Occurred () ?NULL :PyErr_NoMemory ();
@@ -3179,7 +3380,7 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool, struct machine *machine __maybe_unused) { struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); -PyObject *pyevent =pyrf_event__new (event, sample->evsel); +PyObject *pyevent =pyrf_event__new (event, sample->evsel, psession->session);PyObject *ret;if (pyevent ==NULL )
@@ -3287,6 +3488,9 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject } psession->session = session; + symbol_conf.use_callchain = true; + symbol_conf.show_kernel_path = true; + symbol_conf.inline_name = false;if (symbol__init (perf_session__env (session)) <0 ) {PyErr_SetString (PyExc_OSError,"perf: symbol__init failed" ); goto err_out;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 13/19] perf python: Add callchain support
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
New issues:
- [High] Hardcoding the host machine for callchain resolution breaks analysis of guest VM events.
Pre-existing issues:
- [Critical] Truncating oversized perf_event structures leads to out-of-bounds heap reads and writes. --
commit eb725c58d220336416770572b756cd5d942572bc
Author: Ian Rogers <[email protected]>
perf python: Add callchain support
Implement pyrf_callchain_node and pyrf_callchain types for lazy iteration over callchain frames. Add callchain property to sample_event.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
tools/perf/util/python.c static PyTypeObject *pyrf_event__type[] = { · line 1148
Quoted 11 lines
tools/perf/util/python.c +4
-1@@ -987,9 +1140,12 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, }; -static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel) +static PyObject *pyrf_event__new (const union perf_event *event,struct evsel *evsel, +struct perf_session *session) {struct pyrf_event *pevent; +struct perf_sample *sample; +struct machine *machine = session ? &session->machines.host :NULL ;
[Severity: High]
Does hardcoding the host machine here break the analysis of guest VM events?
Since this forces the host machine context into machine__resolve(), events originating from guest VMs might be erroneously resolved against the host kernel's maps and symbols, breaking callchain decoding.
[ ... ]
tools/perf/util/python.c static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev · line 1207
Quoted 12 lines
tools/perf/util/python.c +1
-0@@ -1035,6 +1191,7 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev }perf_sample__init (&pevent->sample,/*all=*/ true); + pevent->callchain =NULL ; pevent->al_resolved = false;addr_location__init (&pevent->al);@@ -1048,6 +1205,50 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evreturn PyErr_Format (PyExc_OSError,"perf: can't parse sample, err=%d" , err); }
[Severity: Critical]
This is a pre-existing issue introduced earlier in this series, but when evsel__parse_sample() parses the sample here, can it read out of bounds?
Because earlier in pyrf_event__new() the copy of the event is truncated:
memcpy(&pevent->event, event, copy_size);
but pevent->event.header.size retains its original large size (which can reach up to 65535 bytes). Since evsel__parse_sample() relies on the untruncated header.size, it appears it can read trailing variable-length arrays out-of-bounds, leaking adjacent heap memory or causing crashes.
[ ... ]
tools/perf/util/python.c static int pyrf_session_tool__sample(const struct perf_tool *tool, · line 3383
Quoted 6 lines
tools/perf/util/python.c +1
-1@@ -3179,7 +3380,7 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool, struct machine *machine __maybe_unused) { struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); -PyObject *pyevent =pyrf_event__new (event, sample->evsel); +PyObject *pyevent =pyrf_event__new (event, sample->evsel, psession->session);
[Severity: High]
Is it possible to pass the dynamically provided machine argument directly to pyrf_event__new() instead of ignoring it here?
By ignoring the machine argument and passing the overarching session instead, we sever the context needed for guest VM evaluation, leading to the hardcoded host machine fallback inside pyrf_event__new().
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=13
Ian Rogers
reviewerPATCH v11 14/19 perf python: Extend API for stat events in python.c
In reply to Ian RogersAdd stat information to the session. Add call backs for stat events.
---
v5:
1. Fix Memory Corruption: Corrected the memory offset for stat_round_type in pyrf_stat_round_event__members by adding the base offset of struct pyrf_event.
2. Fix Memory Leak: Added Py_XDECREF() to free the unused return value of the Python callback in pyrf_session_tool__stat_round().
---
v7:
- Added comprehensive size checks in pyrf_event__new for all event types.
- Fixed line length warning.
v10:
- Added CHECK_INITIALIZED check to pyrf_evsel__get_ids to prevent NULL pointer dereference on uninitialized evsel objects.
--- tools/perf/util/python.c | 185 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 181 insertions(+), 4 deletions(-)
tools/perf/util/python.c +181 -4
@@ -459,6 +459,77 @@ static PyTypeObject pyrf_lost_event__type = { .tp_repr = (reprfunc)pyrf_lost_event__repr, }; +static const char pyrf_stat_event__doc[] =PyDoc_STR ("perf stat event object." ); + +static PyMemberDef pyrf_stat_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), +member_def (perf_record_stat, id,T_ULONGLONG ,"event id" ), +member_def (perf_record_stat, cpu,T_UINT ,"event cpu" ), +member_def (perf_record_stat, thread,T_UINT ,"event thread" ), +member_def (perf_record_stat, val,T_ULONGLONG ,"counter value" ), +member_def (perf_record_stat, ena,T_ULONGLONG ,"enabled time" ), +member_def (perf_record_stat, run,T_ULONGLONG ,"running time" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_stat_event__repr (const struct pyrf_event *pevent) +{ +return PyUnicode_FromFormat ( +"{ type: stat, id: %llu, cpu: %u, thread: %u, val: %llu, ena: %llu, run: %llu }" , + pevent->event.stat.id, + pevent->event.stat.cpu, + pevent->event.stat.thread, + pevent->event.stat.val, + pevent->event.stat.ena, + pevent->event.stat.run); +} + +static PyTypeObject pyrf_stat_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.stat_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_new = PyType_GenericNew, + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_stat_event__doc, + .tp_members = pyrf_stat_event__members, + .tp_getset = pyrf_event__getset, + .tp_repr = (reprfunc )pyrf_stat_event__repr, +}; + +static const char pyrf_stat_round_event__doc[] =PyDoc_STR ("perf stat round event object." ); + +static PyMemberDef pyrf_stat_round_event__members[] = { + sample_members +member_def (perf_event_header, type,T_UINT ,"event type" ), + { .name ="stat_round_type" , .type =T_ULONGLONG , + .offset = offsetof(struct pyrf_event , event) + offsetof(struct perf_record_stat_round , type), + .doc ="round type" }, +member_def (perf_record_stat_round, time,T_ULONGLONG ,"round time" ), + { .name =NULL , }, +}; + +static PyObject *pyrf_stat_round_event__repr (const struct pyrf_event *pevent) +{ +return PyUnicode_FromFormat ("{ type: stat_round, type: %llu, time: %llu }" , + pevent->event.stat_round.type, + pevent->event.stat_round.time); +} + +static PyTypeObject pyrf_stat_round_event__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.stat_round_event" , + .tp_basicsize =sizeof (struct pyrf_event ), + .tp_new = PyType_GenericNew, + .tp_dealloc = (destructor )pyrf_event__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_stat_round_event__doc, + .tp_members = pyrf_stat_round_event__members, + .tp_getset = pyrf_event__getset, + .tp_repr = (reprfunc )pyrf_stat_round_event__repr, +}; +static const char pyrf_read_event__doc[] =PyDoc_STR ("perf read event object." );static PyMemberDef pyrf_read_event__members[] = {
@@ -1113,6 +1184,12 @@ static int pyrf_event__setup_types(void)if (err <0 ) goto out; err =PyType_Ready (&pyrf_context_switch_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_event__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_stat_round_event__type);if (err <0 ) goto out; err =PyType_Ready (&pyrf_callchain_node__type);
@@ -1138,6 +1215,8 @@ static PyTypeObject *pyrf_event__type[] = { [PERF_RECORD_SAMPLE] = &pyrf_sample_event__type, [PERF_RECORD_SWITCH] = &pyrf_context_switch_event__type, [PERF_RECORD_SWITCH_CPU_WIDE] = &pyrf_context_switch_event__type, + [PERF_RECORD_STAT] = &pyrf_stat_event__type, + [PERF_RECORD_STAT_ROUND] = &pyrf_stat_round_event__type, }; static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evsel,
@@ -2128,7 +2207,45 @@ static PyObject *pyrf_evsel__get_attr_wakeup_events(PyObject *self, void */*closreturn PyLong_FromUnsignedLong (pevsel->evsel->core.attr.wakeup_events); } +static PyObject *pyrf_evsel__get_ids (struct pyrf_evsel *pevsel,void *closure__maybe_unused ) +{ +struct evsel *evsel; +PyObject *list; + +CHECK_INITIALIZED (pevsel->evsel,"evsel" ); + + evsel = pevsel->evsel; + list =PyList_New (0 ); + +if (!list) +return NULL ; + +for (u32 i =0 ; i < evsel->core.ids; i++) { +PyObject *id =PyLong_FromUnsignedLongLong (evsel->core.id[i]); +int ret; + +if (!id) { +Py_DECREF (list); +return NULL ; + } + ret =PyList_Append (list, id); +Py_DECREF (id); +if (ret <0 ) { +Py_DECREF (list); +return NULL ; + } + } + +return list; +} +static PyGetSetDef pyrf_evsel__getset[] = { + { + .name ="ids" , + .get = (getter )pyrf_evsel__get_ids, + .set =NULL , + .doc ="event IDs." , + }, { .name ="tracking" , .get = pyrf_evsel__get_tracking,
@@ -2977,6 +3094,8 @@ static const struct perf_constant perf__constants[] = {PERF_CONST (RECORD_LOST_SAMPLES ),PERF_CONST (RECORD_SWITCH ),PERF_CONST (RECORD_SWITCH_CPU_WIDE ), +PERF_CONST (RECORD_STAT ), +PERF_CONST (RECORD_STAT_ROUND ),PERF_CONST (RECORD_MISC_SWITCH_OUT ), { .name = NULL, },
@@ -3372,6 +3491,7 @@ struct pyrf_session {struct perf_tool tool;struct pyrf_data *pdata;PyObject *sample; +PyObject *stat; }; static int pyrf_session_tool__sample(const struct perf_tool *tool,
@@ -3396,6 +3516,50 @@ static int pyrf_session_tool__sample(const struct perf_tool *tool,return 0 ; } +static int pyrf_session_tool__stat (const struct perf_tool *tool, +struct perf_session *session, +union perf_event *event) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +struct evsel *evsel =evlist__id2evsel (session->evlist, event->stat.id); +PyObject *pyevent =pyrf_event__new (event, evsel, psession->session); +const char *name = evsel ?evsel__name (evsel) :"unknown" ; +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"Oz" , pyevent, name); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} + +static int pyrf_session_tool__stat_round (const struct perf_tool *tool, +struct perf_session *session__maybe_unused , +union perf_event *event) +{ + struct pyrf_session *psession = container_of(tool, struct pyrf_session, tool); +PyObject *pyevent =pyrf_event__new (event,/*evsel=*/ NULL , psession->session); +PyObject *ret; + +if (pyevent ==NULL ) +return -ENOMEM ; + + ret =PyObject_CallFunction (psession->stat,"Oz" , pyevent,NULL ); +if (!ret) { +Py_DECREF (pyevent); +return -1 ; + } +Py_DECREF (ret); +Py_DECREF (pyevent); +return 0 ; +} +static PyObject *pyrf_session__find_thread (struct pyrf_session *psession,PyObject *args) {struct machine *machine;
@@ -3429,13 +3593,13 @@ static PyObject *pyrf_session__find_thread(struct pyrf_session *psession, PyObjestatic PyObject *pyrf_session__new (PyTypeObject *type,PyObject *args,PyObject *kwargs) {struct pyrf_data *pdata; -PyObject *sample =NULL ; -static char *kwlist[] = {"data" ,"sample" ,NULL }; +PyObject *sample =NULL , *stat =NULL ; +static char *kwlist[] = {"data" ,"sample" ,"stat" ,NULL };struct pyrf_session *psession;struct perf_session *session; -if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|O" , kwlist, &pyrf_data__type, &pdata, - &sample)) +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"O!|OO" , kwlist, &pyrf_data__type, &pdata, + &sample, &stat))return NULL ; psession = PyObject_New(struct pyrf_session, type);
@@ -3444,6 +3608,7 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject psession->session =NULL ; psession->sample =NULL ; + psession->stat =NULL ; psession->pdata =NULL ;Py_INCREF (pdata);
@@ -3466,8 +3631,13 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject }while (0 )ADD_TOOL (sample); +ADD_TOOL (stat);#undef ADD_TOOL +if (stat) + psession->tool.stat_round = pyrf_session_tool__stat_round; + + psession->tool.comm = perf_event__process_comm; psession->tool.mmap = perf_event__process_mmap; psession->tool.mmap2 = perf_event__process_mmap2;
@@ -3510,6 +3680,7 @@ static void pyrf_session__delete(struct pyrf_session *psession)perf_session__delete (psession->session);Py_XDECREF (psession->pdata);Py_XDECREF (psession->sample); +Py_XDECREF (psession->stat);Py_TYPE (psession)->tp_free ((PyObject *)psession); }
@@ -3692,6 +3863,12 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_context_switch_event__type);PyModule_AddObject (module,"switch_event" , (PyObject *)&pyrf_context_switch_event__type); +Py_INCREF (&pyrf_stat_event__type); +PyModule_AddObject (module,"stat_event" , (PyObject *)&pyrf_stat_event__type); + +Py_INCREF (&pyrf_stat_round_event__type); +PyModule_AddObject (module,"stat_round_event" , (PyObject *)&pyrf_stat_round_event__type); +Py_INCREF (&pyrf_thread_map__type);PyModule_AddObject (module,"thread_map" , (PyObject *)&pyrf_thread_map__type);
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 14/19] perf python: Extend API for stat events in python.c
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] The Python callback for `stat` events is unconditionally overwritten by the default C handler…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The Python callback for stat events is unconditionally overwritten by the default C handler, rendering the custom Python callback completely inactive. --
commit cbc98fa452fc586a169f4af327209522a8d370d7
Author: Ian Rogers <[email protected]>
perf python: Extend API for stat events in python.c
This commit extends the perf Python API by adding stat information to the session and providing callbacks for stat events.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
tools/perf/util/python.c static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject · line 3634
Quoted 5 lines
tools/perf/util/python.c +1
-0@@ -3466,8 +3631,13 @@ static PyObject *pyrf_session__new(PyTypeObject *type, PyObject *args, PyObject }while (0 )ADD_TOOL (sample); +ADD_TOOL (stat);
[Severity: Medium]
Does this ADD_TOOL(stat) assignment get overwritten later in this function?
Looking further down in pyrf_session__new(), the tool configuration unconditionally assigns the default C handler:
psession->tool.stat = perf_event__process_stat_event;
Will this overwrite the pyrf_session_tool__stat callback set up by ADD_TOOL(), preventing the custom Python stat event handler from ever being executed?
Quoted 9 lines
tools/perf/util/python.c +4
-0#undef ADD_TOOL +if (stat) + psession->tool.stat_round = pyrf_session_tool__stat_round; + + psession->tool.comm = perf_event__process_comm; psession->tool.mmap = perf_event__process_mmap; psession->tool.mmap2 = perf_event__process_mmap2;
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=14
Ian Rogers
reviewerPATCH v11 15/19 perf python: Expose brstack in sample event
In reply to Ian RogersImplement pyrf_branch_entry and pyrf_branch_stack for lazy iteration over branch stack entries.
---
v2:
1. Avoided Keyword Collision: Renamed the properties "from" and "to" to "from_ip" and "to_ip" in pyrf_branch_entry__getset[] to prevent syntax errors in Python.
2. Eager Branch Stack Resolution: Updated pyrf_session_tool__sample() to allocate and copy the branch stack entries eagerly when the event is processed, instead of deferring it to iteration time. This avoids reading from potentially overwritten or unmapped mmap buffers.
3. Updated Iterators: Updated pyrf_branch_stack and pyrf_branch_stack__next() to use the copied entries rather than pointing directly to the sample's buffer.
4. Avoided Reference Leak on Init Failure: Added proper error checking for PyModule_AddObject() in the module initialization function, decrementing references on failure.
v6:
- Moved branch stack resolution from session_tool__sample to pyrf_event__new.
--- tools/perf/util/python.c | 198 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 198 insertions(+)
tools/perf/util/python.c +198 -0
@@ -22,6 +22,7 @@#include "debug.h" #include "dso.h" #include "event.h" +#include "branch.h" #include "evlist.h" #include "evsel.h" #include "expr.h"
@@ -89,6 +90,8 @@ struct pyrf_event {bool al_resolved;/** @callchain: Resolved callchain, eagerly computed if requested. */ PyObject *callchain; +/** @brstack: Resolved branch stack, eagerly computed if requested. */ +PyObject *brstack;/** @event: The underlying perf_event that may be in a file or ring buffer. */ union perf_event event; };
@@ -127,6 +130,7 @@ static void pyrf_event__delete(struct pyrf_event *pevent)if (pevent->al_resolved)addr_location__exit (&pevent->al);Py_XDECREF (pevent->callchain); +Py_XDECREF (pevent->brstack);perf_sample__exit (&pevent->sample);Py_TYPE (pevent)->tp_free ((PyObject *)pevent); }
@@ -997,6 +1001,154 @@ static PyObject *pyrf_sample_event__get_callchain(PyObject *self, void *closurereturn pevent->callchain; } +struct pyrf_branch_entry { +PyObject_HEAD + u64 from; +u64 to; +struct branch_flags flags; +}; + +static void pyrf_branch_entry__delete (struct pyrf_branch_entry *pentry) +{ +Py_TYPE (pentry)->tp_free ((PyObject *)pentry); +} + +static PyObject *pyrf_branch_entry__get_from (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->from); +} + +static PyObject *pyrf_branch_entry__get_to (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->to); +} + +static PyObject *pyrf_branch_entry__get_mispred (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.mispred); +} + +static PyObject *pyrf_branch_entry__get_predicted (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.predicted); +} + +static PyObject *pyrf_branch_entry__get_in_tx (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.in_tx); +} + +static PyObject *pyrf_branch_entry__get_abort (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyBool_FromLong (pentry->flags.abort); +} + +static PyObject *pyrf_branch_entry__get_cycles (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromUnsignedLongLong (pentry->flags.cycles); +} + +static PyObject *pyrf_branch_entry__get_type (struct pyrf_branch_entry *pentry, +void *closure__maybe_unused ) +{ +return PyLong_FromLong (pentry->flags.type); +} + +static PyGetSetDef pyrf_branch_entry__getset[] = { + { .name ="from_ip" , .get = (getter )pyrf_branch_entry__get_from, }, + { .name ="to_ip" , .get = (getter )pyrf_branch_entry__get_to, }, + { .name ="mispred" , .get = (getter )pyrf_branch_entry__get_mispred, }, + { .name ="predicted" , .get = (getter )pyrf_branch_entry__get_predicted, }, + { .name ="in_tx" , .get = (getter )pyrf_branch_entry__get_in_tx, }, + { .name ="abort" , .get = (getter )pyrf_branch_entry__get_abort, }, + { .name ="cycles" , .get = (getter )pyrf_branch_entry__get_cycles, }, + { .name ="type" , .get = (getter )pyrf_branch_entry__get_type, }, + { .name =NULL , }, +}; + +static PyTypeObject pyrf_branch_entry__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.branch_entry" , + .tp_basicsize =sizeof (struct pyrf_branch_entry ), + .tp_dealloc = (destructor )pyrf_branch_entry__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf branch entry object." , + .tp_getset = pyrf_branch_entry__getset, +}; + +struct pyrf_branch_stack { +PyObject_HEAD + struct branch_entry *entries; +u64 nr; +}; + +static void pyrf_branch_stack__delete (struct pyrf_branch_stack *pstack) +{ +free (pstack->entries); +Py_TYPE (pstack)->tp_free ((PyObject *)pstack); +} + +static Py_ssize_t pyrf_branch_stack__length (PyObject *obj) +{ +struct pyrf_branch_stack *pstack = (void *)obj; + +return pstack->nr; +} + +static PyObject *pyrf_branch_stack__item (PyObject *obj,Py_ssize_t i) +{ +struct pyrf_branch_stack *pstack = (void *)obj; +struct pyrf_branch_entry *pentry; + +if (i <0 || i >= (Py_ssize_t )pstack->nr) { +PyErr_SetString (PyExc_IndexError,"Index out of range" ); +return NULL ; + } + + pentry = PyObject_New(struct pyrf_branch_entry, &pyrf_branch_entry__type); +if (!pentry) +return NULL ; + + pentry->from = pstack->entries[i].from; + pentry->to = pstack->entries[i].to; + pentry->flags = pstack->entries[i].flags; + +return (PyObject *)pentry; +} + +static PySequenceMethods pyrf_branch_stack__sequence_methods = { + .sq_length = pyrf_branch_stack__length, + .sq_item = pyrf_branch_stack__item, +}; + +static PyTypeObject pyrf_branch_stack__type = { +PyVarObject_HEAD_INIT (NULL ,0 ) + .tp_name ="perf.branch_stack" , + .tp_basicsize =sizeof (struct pyrf_branch_stack ), + .tp_dealloc = (destructor )pyrf_branch_stack__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc ="perf branch stack object." , + .tp_as_sequence = &pyrf_branch_stack__sequence_methods, +}; + +static PyObject *pyrf_sample_event__get_brstack (PyObject *self,void *closure__maybe_unused ) +{ +struct pyrf_event *pevent = (void *)self; + +if (!pevent->brstack) + Py_RETURN_NONE; + +Py_INCREF (pevent->brstack); +return pevent->brstack; +} +static PyObject *pyrf_sample_event__getattro (struct pyrf_event *pevent,PyObject *attr_name) {
@@ -1017,6 +1169,12 @@ static PyGetSetDef pyrf_sample_event__getset[] = { .set = NULL, .doc = "event callchain.", }, + { + .name = "brstack", + .get = pyrf_sample_event__get_brstack, + .set = NULL, + .doc = "event branch stack.", + }, { .name = "raw_buf", .get = (getter)pyrf_sample_event__get_raw_buf,
@@ -1198,6 +1356,12 @@ static int pyrf_event__setup_types(void) err =PyType_Ready (&pyrf_callchain__type);if (err <0 ) goto out; + err =PyType_Ready (&pyrf_branch_entry__type); +if (err <0 ) + goto out; + err =PyType_Ready (&pyrf_branch_stack__type); +if (err <0 ) + goto out; out:return err; }
@@ -1271,6 +1435,7 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evperf_sample__init (&pevent->sample,/*all=*/ true); pevent->callchain =NULL ; + pevent->brstack =NULL ; pevent->al_resolved = false;addr_location__init (&pevent->al);
@@ -1328,6 +1493,27 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evaddr_location__exit (&al); } } +if (sample->branch_stack) { +struct branch_stack *bs = sample->branch_stack; +struct branch_entry *entries =perf_sample__branch_entries (sample); +struct pyrf_branch_stack *pstack; + + pstack = PyObject_New(struct pyrf_branch_stack, &pyrf_branch_stack__type); +if (!pstack) { +Py_DECREF (pevent); +return NULL ; + } + pstack->nr = bs->nr; + pstack->entries =calloc (bs->nr,sizeof (struct branch_entry )); +if (!pstack->entries) { +Py_DECREF (pstack); +Py_DECREF (pevent); +return PyErr_NoMemory (); + } +memcpy (pstack->entries, entries, + bs->nr *sizeof (struct branch_entry )); + pevent->brstack = (PyObject *)pstack; + }return (PyObject *)pevent; }
@@ -3884,6 +4070,18 @@ PyMODINIT_FUNC PyInit_perf(void)Py_INCREF (&pyrf_session__type);PyModule_AddObject (module,"session" , (PyObject *)&pyrf_session__type); +Py_INCREF (&pyrf_branch_entry__type); +if (PyModule_AddObject (module,"branch_entry" , (PyObject *)&pyrf_branch_entry__type) <0 ) { +Py_DECREF (&pyrf_branch_entry__type); + goto error; + } + +Py_INCREF (&pyrf_branch_stack__type); +if (PyModule_AddObject (module,"branch_stack" , (PyObject *)&pyrf_branch_stack__type) <0 ) { +Py_DECREF (&pyrf_branch_stack__type); + goto error; + } + dict =PyModule_GetDict (module);if (dict ==NULL ) goto error;
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 15/19] perf python: Expose brstack in sample event
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
Pre-existing issues:
- [High] Heap out-of-bounds read and information leak due to truncated perf event copy. --
commit 62a7a108ea094d1b9fdf0011865d427da46d2e31
Author: Ian Rogers <[email protected]>
perf python: Expose brstack in sample event
Implement pyrf_branch_entry and pyrf_branch_stack for lazy iteration over branch stack entries.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
Quoted 13 lines
tools/perf/util/python.c +9
-0@@ -1328,6 +1493,27 @@ static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *evaddr_location__exit (&al); } } +if (sample->branch_stack) { +struct branch_stack *bs = sample->branch_stack; +struct branch_entry *entries =perf_sample__branch_entries (sample); +struct pyrf_branch_stack *pstack; + + pstack = PyObject_New(struct pyrf_branch_stack, &pyrf_branch_stack__type); +if (!pstack) { +Py_DECREF (pevent); +return NULL ;
tools/perf/util/python.c static PyObject *pyrf_event__new(const union perf_event *event, struct evsel *ev · line 1514
Quoted 10 lines
tools/perf/util/python.c +10
-0+ } + pstack->nr = bs->nr; + pstack->entries =calloc (bs->nr,sizeof (struct branch_entry )); +if (!pstack->entries) { +Py_DECREF (pstack); +Py_DECREF (pevent); +return PyErr_NoMemory (); + } +memcpy (pstack->entries, entries, + bs->nr *sizeof (struct branch_entry ));
[Severity: High]
This is a pre-existing issue, but does this out-of-bounds read pose a risk of an information leak or crash?
Earlier in pyrf_event__new(), the copy size of the event is clamped to sizeof(pevent->event), but the event header size is not adjusted:
if (copy_size > sizeof(pevent->event)) copy_size = sizeof(pevent->event); memcpy(&pevent->event, event, copy_size);
Then, evsel__parse_sample() parses this truncated buffer using the original larger header size. If the branch stack was beyond the clamped limit in the original event, the sample->branch_stack pointer ends up pointing to memory past the end of the pevent allocation.
Consequently, reading bs->nr and blindly copying that amount of adjacent heap memory into the Python object can expose arbitrary heap contents to the Python script or cause a crash.
Quoted 4 lines
tools/perf/util/python.c +2
-0+ pevent->brstack = (PyObject *)pstack; + }return (PyObject *)pevent; }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=15
Ian Rogers
reviewerPATCH v11 16/19 perf python: Add syscall name/id to convert syscall number and name
In reply to Ian RogersUse perf's syscalltbl support to convert syscall number to name assuming the number is for the host machine. This avoids python libaudit support as tools/perf/scripts/python/syscall-counts.py requires.
---
v2:
1. Guarded with HAVE_LIBTRACEEVENT : Wrapped the syscall functions and their entries in perf__methods with #ifdef HAVE_LIBTRACEEVENT to avoid potential linker errors if CONFIG_TRACE is disabled.
2. Changed Exception Type: Updated pyrf__syscall_id to raise a ValueError instead of a TypeError when a syscall is not found.
3. Fixed Typo: Corrected "name number" to "name" in the docstring for syscall_id.
v6:
- Added optional keyword-only elf_machine argument to syscall_name and syscall_id.
v7:
- Made syscalltbl.o unconditional in Build to fix undefined symbol when building without libtraceevent.
--- tools/perf/util/Build | 1 - tools/perf/util/python.c | 48 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 48 insertions(+), 1 deletion(-)
tools/perf/util/Build +0 -1
@@ -75,7 +75,6 @@ perf-util-y += sample.o perf-util-y += sample-raw.o perf-util-y += s390-sample-raw.o perf-util-y += amd-sample-raw.o - perf-util-y += ordered-events.o perf-util-y += namespaces.o perf-util-y += comm.o
tools/perf/util/python.c +48 -0
72 lines not on this page · raw message
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
[email protected]
AI reviewRe: [PATCH v11 16/19] perf python: Add syscall name/id to convert syscall number and name
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] Incomplete implementation and cross-platform analysis violation in the Python API. The patch…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Incomplete implementation and cross-platform analysis violation in the Python API. The patch adds perf.syscall_name() to avoid the libaudit dependency for syscall-counts.py, but fails to update Util.py to use it. Additionally, the new API cannot be used for cross-platform trace analysis because it defaults to EM_HOST and the Python bindings do not expose the trace's e_machine. --
commit 86a217d5e729a1a66f5768f14d13ddebb0e96ae8
Author: Ian Rogers <[email protected]>
perf python: Add syscall name/id to convert syscall number and name
This commit adds perf.syscall_name() and perf.syscall_id() to the Python bindings, utilizing perf's syscalltbl support. This is intended to replace the python-audit dependency previously required by syscall-counts.py.
Quoted 4 lines
Use perf's syscalltbl support to convert syscall number to name assuming the number is for the host machine. This avoids python libaudit support as
tools/perf/scripts/python/syscall-counts.pyrequires.
[Severity: Medium]
Does this patch update the actual python scripts to use the new API?
Looking at tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py, it seems the syscall_name() wrapper still relies on python-audit, leaving the libaudit dependency intact for syscall-counts.py.
Quoted 4 lines
tools/perf/util/python.c +0
-0
[ ... ]
tools/perf/util/python.c static int pyrf_session__setup_types(void) · line 3951
Quoted 13 lines
tools/perf/util/python.c +9
-0@@ -3938,6 +3940,40 @@ static int pyrf_session__setup_types(void)return PyType_Ready (&pyrf_session__type); } +static PyObject *pyrf__syscall_name (PyObject *self,PyObject *args,PyObject *kwargs) +{ +const char *name; +int id; +int elf_machine =EM_HOST ; +static char *kwlist[] = {"id" ,"elf_machine" ,NULL }; + +if (!PyArg_ParseTupleAndKeywords (args, kwargs,"i|$i" , kwlist, &id, &elf_machine)) +return NULL ;
[Severity: Medium]
Does defaulting to EM_HOST restrict cross-platform analysis?
Since python scripts currently have no way to access the e_machine value dynamically from the trace (such as from perf_env or evsel), it appears this will always default to the host architecture.
Will this break analysis when processing a perf.data file recorded on a different architecture, such as analyzing an ARM trace on an x86 host?
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=16
Ian Rogers
reviewerPATCH v11 17/19 perf python: Add config file access
In reply to Ian RogersAdd perf.config_get(name) to expose the perf configuration system.
--- tools/perf/util/python.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+)
tools/perf/util/python.c +20 -0
@@ -17,6 +17,7 @@#include "build-id.h" #include "callchain.h" #include "comm.h" +#include "config.h" #include "counts.h" #include "data.h" #include "debug.h"
@@ -3974,7 +3975,26 @@ static PyObject *pyrf__syscall_id(PyObject *self, PyObject *args, PyObject *kwarreturn PyLong_FromLong (id); } +static PyObject *pyrf__config_get (PyObject *self,PyObject *args) +{ +const char *config_name, *val; + +if (!PyArg_ParseTuple (args,"s" , &config_name)) +return NULL ; + + val =perf_config_get (config_name); +if (!val) + Py_RETURN_NONE; +return PyUnicode_FromString (val); +} +static PyMethodDef perf__methods[] = { + { + .ml_name ="config_get" , + .ml_meth = (PyCFunction ) pyrf__config_get, + .ml_flags =METH_VARARGS , + .ml_doc =PyDoc_STR ("Get a perf config value." ) + }, { .ml_name ="metrics" , .ml_meth = (PyCFunction ) pyrf__metrics,
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog
Ian Rogers
reviewerPATCH v11 18/19 perf python: Add perf.pyi stubs file
In reply to Ian RogersAdd Python type stubs for the perf module to improve IDE support and static analysis. Includes docstrings for classes, methods, and constants derived from C source and JSON definitions.
---
v2:
1. Added Missing Module Functions: Added parse_metrics and pmus. Renamed metrics to parse_metrics to match python.c .
2. Added Constructors: Added __init__ methods for data , evsel , and evlist with their appropriate arguments.
3. Removed sample_comm : Removed it from sample_event since it is not exported in python.c .
4. Keyword Handling in branch_entry : I used from_ip and to_ip in the stubs to match the rename I did in python.c (in turn 145) to avoid the Python from keyword conflict.
5. Added Missing Event Classes: Added mmap_event , lost_event , comm_event , task_event , throttle_event , read_event , and switch_event .
6. Added Missing evlist Methods: Added get_pollfd and add .
7. Updated Return Types: Changed process_events to return int .
v6:
- Updated perf.pyi to use find_thread and elf_machine.
v8:
- Added mmap2_event class and new evsel attributes to perf.pyi.
- Added pid, tid, ppid, cpu attributes to class thread in perf.pyi.
--- tools/perf/Makefile.perf | 5 +- tools/perf/python/perf.pyi | 629 +++++++++++++++++++++++++++++++++++++ tools/perf/util/setup.py | 5 + 3 files changed, 638 insertions(+), 1 deletion(-) create mode 100644 tools/perf/python/perf.pyi
tools/perf/Makefile.perf +4 -1
@@ -527,7 +527,10 @@ $(OUTPUT)python/perf$(PYTHON_EXTENSION_SUFFIX): util/python.c util/setup.py $(PE CFLAGS='$(CFLAGS)' LDFLAGS='$(LDFLAGS) $(LIBS_PY)' \ $(PYTHON_WORD) util/setup.py \ --quiet build_ext; \ - cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/ + cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/; \ + if [ "$(abspath python/perf.pyi)" != "$(abspath $(OUTPUT)python/perf.pyi)" ]; then \ + cp python/perf.pyi $(OUTPUT)python/; \ + fi python_perf_target: @echo "Target is: $(OUTPUT)python/perf$(PYTHON_EXTENSION_SUFFIX)"
tools/perf/python/perf.pyi new file +629 -0
629 lines not on this page · raw message
tools/perf/util/setup.py +5 -0
@@ -2,6 +2,7 @@ from os import getenv, path from subprocess import Popen, PIPE from re import sub import shlex +import shutil cc = getenv("CC") assert cc, "Environment variable CC not set"
@@ -73,6 +74,10 @@ class install_lib(_install_lib): _install_lib.finalize_options(self) self.build_dir = build_lib + def run(self): + _install_lib.run(self) + shutil.copy2(f'{src_perf}/python/perf.pyi', self.install_dir) + cflags = getenv('CFLAGS', '').split() # switch off several checks (need to be at the end of cflags list)
Signature · 1 line
2.54.0.1032.g2f8565e1d1-goog