linux/tools/perf/util/lock-contention.h
Namhyung Kim 4fd06bd2dc perf lock contention: Add -G/--cgroup-filter option
The -G/--cgroup-filter is to limit lock contention collection on the
tasks in the specific cgroups only.

  $ sudo ./perf lock con -abt -G /user.slice/.../vte-spawn-52221fb8-b33f-4a52-b5c3-e35d1e6fc0e0.scope \
    ./perf bench sched messaging
  # Running 'sched/messaging' benchmark:
  # 20 sender and receiver processes per group
  # 10 groups == 400 processes run

       Total time: 0.174 [sec]
   contended   total wait     max wait     avg wait          pid   comm

           4    114.45 us     60.06 us     28.61 us       214847   sched-messaging
           2    111.40 us     60.84 us     55.70 us       214848   sched-messaging
           2    106.09 us     59.42 us     53.04 us       214837   sched-messaging
           1     81.70 us     81.70 us     81.70 us       214709   sched-messaging
          68     78.44 us      6.83 us      1.15 us       214633   sched-messaging
          69     73.71 us      2.69 us      1.07 us       214632   sched-messaging
           4     72.62 us     60.83 us     18.15 us       214850   sched-messaging
           2     71.75 us     67.60 us     35.88 us       214840   sched-messaging
           2     69.29 us     67.53 us     34.65 us       214804   sched-messaging
           2     69.00 us     68.23 us     34.50 us       214826   sched-messaging
  ...

Export cgroup__new() function as it's needed from outside.

Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Hao Luo <haoluo@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <song@kernel.org>
Cc: bpf@vger.kernel.org
Link: https://lore.kernel.org/r/20230906174903.346486-5-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2023-09-12 17:32:00 -03:00

180 lines
4 KiB
C

// SPDX-License-Identifier: GPL-2.0
#ifndef PERF_LOCK_CONTENTION_H
#define PERF_LOCK_CONTENTION_H
#include <linux/list.h>
#include <linux/rbtree.h>
struct lock_filter {
int nr_types;
int nr_addrs;
int nr_syms;
int nr_cgrps;
unsigned int *types;
unsigned long *addrs;
char **syms;
u64 *cgrps;
};
struct lock_stat {
struct hlist_node hash_entry;
struct rb_node rb; /* used for sorting */
u64 addr; /* address of lockdep_map, used as ID */
char *name; /* for strcpy(), we cannot use const */
u64 *callstack;
unsigned int nr_acquire;
unsigned int nr_acquired;
unsigned int nr_contended;
unsigned int nr_release;
union {
unsigned int nr_readlock;
unsigned int flags;
};
unsigned int nr_trylock;
/* these times are in nano sec. */
u64 avg_wait_time;
u64 wait_time_total;
u64 wait_time_min;
u64 wait_time_max;
int broken; /* flag of blacklist */
int combined;
};
/*
* States of lock_seq_stat
*
* UNINITIALIZED is required for detecting first event of acquire.
* As the nature of lock events, there is no guarantee
* that the first event for the locks are acquire,
* it can be acquired, contended or release.
*/
#define SEQ_STATE_UNINITIALIZED 0 /* initial state */
#define SEQ_STATE_RELEASED 1
#define SEQ_STATE_ACQUIRING 2
#define SEQ_STATE_ACQUIRED 3
#define SEQ_STATE_READ_ACQUIRED 4
#define SEQ_STATE_CONTENDED 5
/*
* MAX_LOCK_DEPTH
* Imported from include/linux/sched.h.
* Should this be synchronized?
*/
#define MAX_LOCK_DEPTH 48
struct lock_stat *lock_stat_find(u64 addr);
struct lock_stat *lock_stat_findnew(u64 addr, const char *name, int flags);
bool match_callstack_filter(struct machine *machine, u64 *callstack);
/*
* struct lock_seq_stat:
* Place to put on state of one lock sequence
* 1) acquire -> acquired -> release
* 2) acquire -> contended -> acquired -> release
* 3) acquire (with read or try) -> release
* 4) Are there other patterns?
*/
struct lock_seq_stat {
struct list_head list;
int state;
u64 prev_event_time;
u64 addr;
int read_count;
};
struct thread_stat {
struct rb_node rb;
u32 tid;
struct list_head seq_list;
};
/*
* CONTENTION_STACK_DEPTH
* Number of stack trace entries to find callers
*/
#define CONTENTION_STACK_DEPTH 8
/*
* CONTENTION_STACK_SKIP
* Number of stack trace entries to skip when finding callers.
* The first few entries belong to the locking implementation itself.
*/
#define CONTENTION_STACK_SKIP 4
/*
* flags for lock:contention_begin
* Imported from include/trace/events/lock.h.
*/
#define LCB_F_SPIN (1U << 0)
#define LCB_F_READ (1U << 1)
#define LCB_F_WRITE (1U << 2)
#define LCB_F_RT (1U << 3)
#define LCB_F_PERCPU (1U << 4)
#define LCB_F_MUTEX (1U << 5)
struct evlist;
struct machine;
struct target;
struct lock_contention_fails {
int task;
int stack;
int time;
int data;
};
struct lock_contention {
struct evlist *evlist;
struct target *target;
struct machine *machine;
struct hlist_head *result;
struct lock_filter *filters;
struct lock_contention_fails fails;
struct rb_root cgroups;
unsigned long map_nr_entries;
int max_stack;
int stack_skip;
int aggr_mode;
int owner;
int nr_filtered;
bool save_callstack;
};
#ifdef HAVE_BPF_SKEL
int lock_contention_prepare(struct lock_contention *con);
int lock_contention_start(void);
int lock_contention_stop(void);
int lock_contention_read(struct lock_contention *con);
int lock_contention_finish(struct lock_contention *con);
#else /* !HAVE_BPF_SKEL */
static inline int lock_contention_prepare(struct lock_contention *con __maybe_unused)
{
return 0;
}
static inline int lock_contention_start(void) { return 0; }
static inline int lock_contention_stop(void) { return 0; }
static inline int lock_contention_finish(struct lock_contention *con __maybe_unused)
{
return 0;
}
static inline int lock_contention_read(struct lock_contention *con __maybe_unused)
{
return 0;
}
#endif /* HAVE_BPF_SKEL */
#endif /* PERF_LOCK_CONTENTION_H */