public-mirrors/linux
1
0
Fork 0
mirror of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2025-08-05 16:54:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
linux/include/rv/da_monitor.h
Gabriele Monaco 9d475d80c9 rv: Retry when da monitor detects race conditions 
DA monitor can be accessed from multiple cores simultaneously, this is
likely, for instance when dealing with per-task monitors reacting on
events that do not always occur on the CPU where the task is running.
This can cause race conditions where two events change the next state
and we see inconsistent values. E.g.:

  [62] event_srs: 27: sleepable x sched_wakeup -> running (final)
  [63] event_srs: 27: sleepable x sched_set_state_sleepable -> sleepable
  [63] error_srs: 27: event sched_switch_suspend not expected in the state running

In this case the monitor fails because the event on CPU 62 wins against
the one on CPU 63, although the correct state should have been
sleepable, since the task get suspended.

Detect if the current state was modified by using try_cmpxchg while
storing the next value. If it was, try again reading the current state.
After a maximum number of failed retries, react by calling a special
tracepoint, print on the console and reset the monitor.

Remove the functions da_monitor_curr_state() and da_monitor_set_state()
as they only hide the underlying implementation in this case.

Monitors where this type of condition can occur must be able to account
for racing events in any possible order, as we cannot know the winner.

Cc: Ingo Molnar <mingo@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Cc: Juri Lelli <jlelli@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/20250728135022.255578-6-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Reviewed-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
2025-07-28 16:47:34 -04:00

548 lines
18 KiB
C
Raw Permalink Blame History

/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright (C) 2019-2022 Red Hat, Inc. Daniel Bristot de Oliveira <bristot@kernel.org>
*
* Deterministic automata (DA) monitor functions, to be used together
* with automata models in C generated by the dot2k tool.
*
* The dot2k tool is available at tools/verification/dot2k/
*
* For further information, see:
* Documentation/trace/rv/da_monitor_synthesis.rst
*/
#include <rv/automata.h>
#include <linux/rv.h>
#include <linux/bug.h>
#include <linux/sched.h>
#ifdef CONFIG_RV_REACTORS
#define DECLARE_RV_REACTING_HELPERS(name, type) \
static void cond_react_##name(type curr_state, type event) \
{ \
if (!rv_reacting_on() || !rv_##name.react) \
return; \
rv_##name.react("rv: monitor %s does not allow event %s on state %s\n", \
#name, \
model_get_event_name_##name(event), \
model_get_state_name_##name(curr_state)); \
}
#else /* CONFIG_RV_REACTOR */
#define DECLARE_RV_REACTING_HELPERS(name, type) \
static void cond_react_##name(type curr_state, type event) \
{ \
return; \
}
#endif
/*
* Generic helpers for all types of deterministic automata monitors.
*/
#define DECLARE_DA_MON_GENERIC_HELPERS(name, type) \
\
DECLARE_RV_REACTING_HELPERS(name, type) \
\
/* \
* da_monitor_reset_##name - reset a monitor and setting it to init state \
*/ \
static inline void da_monitor_reset_##name(struct da_monitor *da_mon) \
{ \
da_mon->monitoring = 0; \
da_mon->curr_state = model_get_initial_state_##name(); \
} \
\
/* \
* da_monitor_start_##name - start monitoring \
* \
* The monitor will ignore all events until monitoring is set to true. This \
* function needs to be called to tell the monitor to start monitoring. \
*/ \
static inline void da_monitor_start_##name(struct da_monitor *da_mon) \
{ \
da_mon->curr_state = model_get_initial_state_##name(); \
da_mon->monitoring = 1; \
} \
\
/* \
* da_monitoring_##name - returns true if the monitor is processing events \
*/ \
static inline bool da_monitoring_##name(struct da_monitor *da_mon) \
{ \
return da_mon->monitoring; \
} \
\
/* \
* da_monitor_enabled_##name - checks if the monitor is enabled \
*/ \
static inline bool da_monitor_enabled_##name(void) \
{ \
/* global switch */ \
if (unlikely(!rv_monitoring_on())) \
return 0; \
\
/* monitor enabled */ \
if (unlikely(!rv_##name.enabled)) \
return 0; \
\
return 1; \
} \
\
/* \
* da_monitor_handling_event_##name - checks if the monitor is ready to handle events \
*/ \
static inline bool da_monitor_handling_event_##name(struct da_monitor *da_mon) \
{ \
\
if (!da_monitor_enabled_##name()) \
return 0; \
\
/* monitor is actually monitoring */ \
if (unlikely(!da_monitoring_##name(da_mon))) \
return 0; \
\
return 1; \
}
/*
* Event handler for implicit monitors. Implicit monitor is the one which the
* handler does not need to specify which da_monitor to manipulate. Examples
* of implicit monitor are the per_cpu or the global ones.
*
* Retry in case there is a race between getting and setting the next state,
* warn and reset the monitor if it runs out of retries. The monitor should be
* able to handle various orders.
*/
#define DECLARE_DA_MON_MODEL_HANDLER_IMPLICIT(name, type) \
\
static inline bool \
da_event_##name(struct da_monitor *da_mon, enum events_##name event) \
{ \
enum states_##name curr_state, next_state; \
\
curr_state = READ_ONCE(da_mon->curr_state); \
for (int i = 0; i < MAX_DA_RETRY_RACING_EVENTS; i++) { \
next_state = model_get_next_state_##name(curr_state, event); \
if (next_state == INVALID_STATE) { \
cond_react_##name(curr_state, event); \
trace_error_##name(model_get_state_name_##name(curr_state), \
model_get_event_name_##name(event)); \
return false; \
} \
if (likely(try_cmpxchg(&da_mon->curr_state, &curr_state, next_state))) { \
trace_event_##name(model_get_state_name_##name(curr_state), \
model_get_event_name_##name(event), \
model_get_state_name_##name(next_state), \
model_is_final_state_##name(next_state)); \
return true; \
} \
} \
\
trace_rv_retries_error(#name, model_get_event_name_##name(event)); \
pr_warn("rv: " __stringify(MAX_DA_RETRY_RACING_EVENTS) \
" retries reached for event %s, resetting monitor %s", \
model_get_event_name_##name(event), #name); \
return false; \
} \
/*
* Event handler for per_task monitors.
*
* Retry in case there is a race between getting and setting the next state,
* warn and reset the monitor if it runs out of retries. The monitor should be
* able to handle various orders.
*/
#define DECLARE_DA_MON_MODEL_HANDLER_PER_TASK(name, type) \
\
static inline bool da_event_##name(struct da_monitor *da_mon, struct task_struct *tsk, \
enum events_##name event) \
{ \
enum states_##name curr_state, next_state; \
\
curr_state = READ_ONCE(da_mon->curr_state); \
for (int i = 0; i < MAX_DA_RETRY_RACING_EVENTS; i++) { \
next_state = model_get_next_state_##name(curr_state, event); \
if (next_state == INVALID_STATE) { \
cond_react_##name(curr_state, event); \
trace_error_##name(tsk->pid, \
model_get_state_name_##name(curr_state), \
model_get_event_name_##name(event)); \
return false; \
} \
if (likely(try_cmpxchg(&da_mon->curr_state, &curr_state, next_state))) { \
trace_event_##name(tsk->pid, \
model_get_state_name_##name(curr_state), \
model_get_event_name_##name(event), \
model_get_state_name_##name(next_state), \
model_is_final_state_##name(next_state)); \
return true; \
} \
} \
\
trace_rv_retries_error(#name, model_get_event_name_##name(event)); \
pr_warn("rv: " __stringify(MAX_DA_RETRY_RACING_EVENTS) \
" retries reached for event %s, resetting monitor %s", \
model_get_event_name_##name(event), #name); \
return false; \
}
/*
* Functions to define, init and get a global monitor.
*/
#define DECLARE_DA_MON_INIT_GLOBAL(name, type) \
\
/* \
* global monitor (a single variable) \
*/ \
static struct da_monitor da_mon_##name; \
\
/* \
* da_get_monitor_##name - return the global monitor address \
*/ \
static struct da_monitor *da_get_monitor_##name(void) \
{ \
return &da_mon_##name; \
} \
\
/* \
* da_monitor_reset_all_##name - reset the single monitor \
*/ \
static void da_monitor_reset_all_##name(void) \
{ \
da_monitor_reset_##name(da_get_monitor_##name()); \
} \
\
/* \
* da_monitor_init_##name - initialize a monitor \
*/ \
static inline int da_monitor_init_##name(void) \
{ \
da_monitor_reset_all_##name(); \
return 0; \
} \
\
/* \
* da_monitor_destroy_##name - destroy the monitor \
*/ \
static inline void da_monitor_destroy_##name(void) \
{ \
return; \
}
/*
* Functions to define, init and get a per-cpu monitor.
*/
#define DECLARE_DA_MON_INIT_PER_CPU(name, type) \
\
/* \
* per-cpu monitor variables \
*/ \
static DEFINE_PER_CPU(struct da_monitor, da_mon_##name); \
\
/* \
* da_get_monitor_##name - return current CPU monitor address \
*/ \
static struct da_monitor *da_get_monitor_##name(void) \
{ \
return this_cpu_ptr(&da_mon_##name); \
} \
\
/* \
* da_monitor_reset_all_##name - reset all CPUs' monitor \
*/ \
static void da_monitor_reset_all_##name(void) \
{ \
struct da_monitor *da_mon; \
int cpu; \
for_each_cpu(cpu, cpu_online_mask) { \
da_mon = per_cpu_ptr(&da_mon_##name, cpu); \
da_monitor_reset_##name(da_mon); \
} \
} \
\
/* \
* da_monitor_init_##name - initialize all CPUs' monitor \
*/ \
static inline int da_monitor_init_##name(void) \
{ \
da_monitor_reset_all_##name(); \
return 0; \
} \
\
/* \
* da_monitor_destroy_##name - destroy the monitor \
*/ \
static inline void da_monitor_destroy_##name(void) \
{ \
return; \
}
/*
* Functions to define, init and get a per-task monitor.
*/
#define DECLARE_DA_MON_INIT_PER_TASK(name, type) \
\
/* \
* The per-task monitor is stored a vector in the task struct. This variable \
* stores the position on the vector reserved for this monitor. \
*/ \
static int task_mon_slot_##name = RV_PER_TASK_MONITOR_INIT; \
\
/* \
* da_get_monitor_##name - return the monitor in the allocated slot for tsk \
*/ \
static inline struct da_monitor *da_get_monitor_##name(struct task_struct *tsk) \
{ \
return &tsk->rv[task_mon_slot_##name].da_mon; \
} \
\
static void da_monitor_reset_all_##name(void) \
{ \
struct task_struct *g, *p; \
int cpu; \
\
read_lock(&tasklist_lock); \
for_each_process_thread(g, p) \
da_monitor_reset_##name(da_get_monitor_##name(p)); \
for_each_present_cpu(cpu) \
da_monitor_reset_##name(da_get_monitor_##name(idle_task(cpu))); \
read_unlock(&tasklist_lock); \
} \
\
/* \
* da_monitor_init_##name - initialize the per-task monitor \
* \
* Try to allocate a slot in the task's vector of monitors. If there \
* is an available slot, use it and reset all task's monitor. \
*/ \
static int da_monitor_init_##name(void) \
{ \
int slot; \
\
slot = rv_get_task_monitor_slot(); \
if (slot < 0 || slot >= RV_PER_TASK_MONITOR_INIT) \
return slot; \
\
task_mon_slot_##name = slot; \
\
da_monitor_reset_all_##name(); \
return 0; \
} \
\
/* \
* da_monitor_destroy_##name - return the allocated slot \
*/ \
static inline void da_monitor_destroy_##name(void) \
{ \
if (task_mon_slot_##name == RV_PER_TASK_MONITOR_INIT) { \
WARN_ONCE(1, "Disabling a disabled monitor: " #name); \
return; \
} \
rv_put_task_monitor_slot(task_mon_slot_##name); \
task_mon_slot_##name = RV_PER_TASK_MONITOR_INIT; \
return; \
}
/*
* Handle event for implicit monitor: da_get_monitor_##name() will figure out
* the monitor.
*/
#define DECLARE_DA_MON_MONITOR_HANDLER_IMPLICIT(name, type) \
\
static inline void __da_handle_event_##name(struct da_monitor *da_mon, \
enum events_##name event) \
{ \
bool retval; \
\
retval = da_event_##name(da_mon, event); \
if (!retval) \
da_monitor_reset_##name(da_mon); \
} \
\
/* \
* da_handle_event_##name - handle an event \
*/ \
static inline void da_handle_event_##name(enum events_##name event) \
{ \
struct da_monitor *da_mon = da_get_monitor_##name(); \
bool retval; \
\
retval = da_monitor_handling_event_##name(da_mon); \
if (!retval) \
return; \
\
__da_handle_event_##name(da_mon, event); \
} \
\
/* \
* da_handle_start_event_##name - start monitoring or handle event \
* \
* This function is used to notify the monitor that the system is returning \
* to the initial state, so the monitor can start monitoring in the next event. \
* Thus: \
* \
* If the monitor already started, handle the event. \
* If the monitor did not start yet, start the monitor but skip the event. \
*/ \
static inline bool da_handle_start_event_##name(enum events_##name event) \
{ \
struct da_monitor *da_mon; \
\
if (!da_monitor_enabled_##name()) \
return 0; \
\
da_mon = da_get_monitor_##name(); \
\
if (unlikely(!da_monitoring_##name(da_mon))) { \
da_monitor_start_##name(da_mon); \
return 0; \
} \
\
__da_handle_event_##name(da_mon, event); \
\
return 1; \
} \
\
/* \
* da_handle_start_run_event_##name - start monitoring and handle event \
* \
* This function is used to notify the monitor that the system is in the \
* initial state, so the monitor can start monitoring and handling event. \
*/ \
static inline bool da_handle_start_run_event_##name(enum events_##name event) \
{ \
struct da_monitor *da_mon; \
\
if (!da_monitor_enabled_##name()) \
return 0; \
\
da_mon = da_get_monitor_##name(); \
\
if (unlikely(!da_monitoring_##name(da_mon))) \
da_monitor_start_##name(da_mon); \
\
__da_handle_event_##name(da_mon, event); \
\
return 1; \
}
/*
* Handle event for per task.
*/
#define DECLARE_DA_MON_MONITOR_HANDLER_PER_TASK(name, type) \
\
static inline void \
__da_handle_event_##name(struct da_monitor *da_mon, struct task_struct *tsk, \
enum events_##name event) \
{ \
bool retval; \
\
retval = da_event_##name(da_mon, tsk, event); \
if (!retval) \
da_monitor_reset_##name(da_mon); \
} \
\
/* \
* da_handle_event_##name - handle an event \
*/ \
static inline void \
da_handle_event_##name(struct task_struct *tsk, enum events_##name event) \
{ \
struct da_monitor *da_mon = da_get_monitor_##name(tsk); \
bool retval; \
\
retval = da_monitor_handling_event_##name(da_mon); \
if (!retval) \
return; \
\
__da_handle_event_##name(da_mon, tsk, event); \
} \
\
/* \
* da_handle_start_event_##name - start monitoring or handle event \
* \
* This function is used to notify the monitor that the system is returning \
* to the initial state, so the monitor can start monitoring in the next event. \
* Thus: \
* \
* If the monitor already started, handle the event. \
* If the monitor did not start yet, start the monitor but skip the event. \
*/ \
static inline bool \
da_handle_start_event_##name(struct task_struct *tsk, enum events_##name event) \
{ \
struct da_monitor *da_mon; \
\
if (!da_monitor_enabled_##name()) \
return 0; \
\
da_mon = da_get_monitor_##name(tsk); \
\
if (unlikely(!da_monitoring_##name(da_mon))) { \
da_monitor_start_##name(da_mon); \
return 0; \
} \
\
__da_handle_event_##name(da_mon, tsk, event); \
\
return 1; \
} \
\
/* \
* da_handle_start_run_event_##name - start monitoring and handle event \
* \
* This function is used to notify the monitor that the system is in the \
* initial state, so the monitor can start monitoring and handling event. \
*/ \
static inline bool \
da_handle_start_run_event_##name(struct task_struct *tsk, enum events_##name event) \
{ \
struct da_monitor *da_mon; \
\
if (!da_monitor_enabled_##name()) \
return 0; \
\
da_mon = da_get_monitor_##name(tsk); \
\
if (unlikely(!da_monitoring_##name(da_mon))) \
da_monitor_start_##name(da_mon); \
\
__da_handle_event_##name(da_mon, tsk, event); \
\
return 1; \
}
/*
* Entry point for the global monitor.
*/
#define DECLARE_DA_MON_GLOBAL(name, type) \
\
DECLARE_AUTOMATA_HELPERS(name, type) \
DECLARE_DA_MON_GENERIC_HELPERS(name, type) \
DECLARE_DA_MON_MODEL_HANDLER_IMPLICIT(name, type) \
DECLARE_DA_MON_INIT_GLOBAL(name, type) \
DECLARE_DA_MON_MONITOR_HANDLER_IMPLICIT(name, type)
/*
* Entry point for the per-cpu monitor.
*/
#define DECLARE_DA_MON_PER_CPU(name, type) \
\
DECLARE_AUTOMATA_HELPERS(name, type) \
DECLARE_DA_MON_GENERIC_HELPERS(name, type) \
DECLARE_DA_MON_MODEL_HANDLER_IMPLICIT(name, type) \
DECLARE_DA_MON_INIT_PER_CPU(name, type) \
DECLARE_DA_MON_MONITOR_HANDLER_IMPLICIT(name, type)
/*
* Entry point for the per-task monitor.
*/
#define DECLARE_DA_MON_PER_TASK(name, type) \
\
DECLARE_AUTOMATA_HELPERS(name, type) \
DECLARE_DA_MON_GENERIC_HELPERS(name, type) \
DECLARE_DA_MON_MODEL_HANDLER_PER_TASK(name, type) \
DECLARE_DA_MON_INIT_PER_TASK(name, type) \
DECLARE_DA_MON_MONITOR_HANDLER_PER_TASK(name, type)
Reference in a new issue View git blame Copy permalink