#include <linux/bitops.h>
#include <linux/hash.h>
+static void perf_evlist__mmap_put(struct perf_evlist *evlist, int idx);
+static void __perf_evlist__munmap(struct perf_evlist *evlist, int idx);
+
#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
#define SID(e, x, y) xyarray__entry(e->sample_id, x, y)
INIT_HLIST_HEAD(&evlist->heads[i]);
INIT_LIST_HEAD(&evlist->entries);
perf_evlist__set_maps(evlist, cpus, threads);
+ fdarray__init(&evlist->pollfd, 64);
evlist->workload.pid = -1;
}
void perf_evlist__exit(struct perf_evlist *evlist)
{
zfree(&evlist->mmap);
- zfree(&evlist->pollfd);
+ fdarray__exit(&evlist->pollfd);
}
void perf_evlist__delete(struct perf_evlist *evlist)
return perf_evlist__enable_event_thread(evlist, evsel, idx);
}
-static int perf_evlist__grow_pollfd(struct perf_evlist *evlist, int hint)
-{
- int nr_fds_alloc = evlist->nr_fds_alloc + hint;
- size_t size = sizeof(struct pollfd) * nr_fds_alloc;
- struct pollfd *pollfd = realloc(evlist->pollfd, size);
-
- if (pollfd == NULL)
- return -ENOMEM;
-
- evlist->nr_fds_alloc = nr_fds_alloc;
- evlist->pollfd = pollfd;
- return 0;
-}
-
int perf_evlist__alloc_pollfd(struct perf_evlist *evlist)
{
int nr_cpus = cpu_map__nr(evlist->cpus);
nfds += nr_cpus * nr_threads;
}
- if (evlist->nr_fds_alloc - evlist->nr_fds < nfds &&
- perf_evlist__grow_pollfd(evlist, nfds) < 0)
+ if (fdarray__available_entries(&evlist->pollfd) < nfds &&
+ fdarray__grow(&evlist->pollfd, nfds) < 0)
return -ENOMEM;
return 0;
}
-int perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd)
+static int __perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd, int idx)
{
+ int pos = fdarray__add(&evlist->pollfd, fd, POLLIN | POLLERR | POLLHUP);
/*
- * XXX: 64 is arbitrary, just not to call realloc at each fd.
- * Find a better autogrowing heuristic
+ * Save the idx so that when we filter out fds POLLHUP'ed we can
+ * close the associated evlist->mmap[] entry.
*/
- if (evlist->nr_fds == evlist->nr_fds_alloc &&
- perf_evlist__grow_pollfd(evlist, 64) < 0)
- return -ENOMEM;
+ if (pos >= 0) {
+ evlist->pollfd.priv[pos].idx = idx;
- fcntl(fd, F_SETFL, O_NONBLOCK);
- evlist->pollfd[evlist->nr_fds].fd = fd;
- evlist->pollfd[evlist->nr_fds].events = POLLIN | POLLERR | POLLHUP;
- evlist->nr_fds++;
- return 0;
+ fcntl(fd, F_SETFL, O_NONBLOCK);
+ }
+
+ return pos;
}
-int perf_evlist__filter_pollfd(struct perf_evlist *evlist, short revents_and_mask)
+int perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd)
{
- int fd, nr_fds = 0;
-
- if (evlist->nr_fds == 0)
- return 0;
-
- for (fd = 0; fd < evlist->nr_fds; ++fd) {
- if (evlist->pollfd[fd].revents & revents_and_mask)
- continue;
+ return __perf_evlist__add_pollfd(evlist, fd, -1);
+}
- if (fd != nr_fds)
- evlist->pollfd[nr_fds] = evlist->pollfd[fd];
+static void perf_evlist__munmap_filtered(struct fdarray *fda, int fd)
+{
+ struct perf_evlist *evlist = container_of(fda, struct perf_evlist, pollfd);
- ++nr_fds;
- }
+ perf_evlist__mmap_put(evlist, fda->priv[fd].idx);
+}
- evlist->nr_fds = nr_fds;
- return nr_fds;
+int perf_evlist__filter_pollfd(struct perf_evlist *evlist, short revents_and_mask)
+{
+ return fdarray__filter(&evlist->pollfd, revents_and_mask,
+ perf_evlist__munmap_filtered);
}
int perf_evlist__poll(struct perf_evlist *evlist, int timeout)
{
- return poll(evlist->pollfd, evlist->nr_fds, timeout);
+ return fdarray__poll(&evlist->pollfd, timeout);
}
static void perf_evlist__id_hash(struct perf_evlist *evlist,
return event;
}
+static bool perf_mmap__empty(struct perf_mmap *md)
+{
+ return perf_mmap__read_head(md) != md->prev;
+}
+
+static void perf_evlist__mmap_get(struct perf_evlist *evlist, int idx)
+{
+ ++evlist->mmap[idx].refcnt;
+}
+
+static void perf_evlist__mmap_put(struct perf_evlist *evlist, int idx)
+{
+ BUG_ON(evlist->mmap[idx].refcnt == 0);
+
+ if (--evlist->mmap[idx].refcnt == 0)
+ __perf_evlist__munmap(evlist, idx);
+}
+
void perf_evlist__mmap_consume(struct perf_evlist *evlist, int idx)
{
+ struct perf_mmap *md = &evlist->mmap[idx];
+
if (!evlist->overwrite) {
- struct perf_mmap *md = &evlist->mmap[idx];
unsigned int old = md->prev;
perf_mmap__write_tail(md, old);
}
+
+ if (md->refcnt == 1 && perf_mmap__empty(md))
+ perf_evlist__mmap_put(evlist, idx);
}
static void __perf_evlist__munmap(struct perf_evlist *evlist, int idx)
if (evlist->mmap[idx].base != NULL) {
munmap(evlist->mmap[idx].base, evlist->mmap_len);
evlist->mmap[idx].base = NULL;
+ evlist->mmap[idx].refcnt = 0;
}
}
static int __perf_evlist__mmap(struct perf_evlist *evlist, int idx,
struct mmap_params *mp, int fd)
{
+ /*
+ * The last one will be done at perf_evlist__mmap_consume(), so that we
+ * make sure we don't prevent tools from consuming every last event in
+ * the ring buffer.
+ *
+ * I.e. we can get the POLLHUP meaning that the fd doesn't exist
+ * anymore, but the last events for it are still in the ring buffer,
+ * waiting to be consumed.
+ *
+ * Tools can chose to ignore this at their own discretion, but the
+ * evlist layer can't just drop it when filtering events in
+ * perf_evlist__filter_pollfd().
+ */
+ evlist->mmap[idx].refcnt = 2;
evlist->mmap[idx].prev = 0;
evlist->mmap[idx].mask = mp->mask;
evlist->mmap[idx].base = mmap(NULL, evlist->mmap_len, mp->prot,
} else {
if (ioctl(fd, PERF_EVENT_IOC_SET_OUTPUT, *output) != 0)
return -1;
+
+ perf_evlist__mmap_get(evlist, idx);
}
- if (perf_evlist__add_pollfd(evlist, fd) < 0)
+ if (__perf_evlist__add_pollfd(evlist, fd, idx) < 0) {
+ perf_evlist__mmap_put(evlist, idx);
return -1;
+ }
if ((evsel->attr.read_format & PERF_FORMAT_ID) &&
perf_evlist__id_add_fd(evlist, evsel, cpu, thread, fd) < 0)
if (evlist->mmap == NULL && perf_evlist__alloc_mmap(evlist) < 0)
return -ENOMEM;
- if (evlist->pollfd == NULL && perf_evlist__alloc_pollfd(evlist) < 0)
+ if (evlist->pollfd.entries == NULL && perf_evlist__alloc_pollfd(evlist) < 0)
return -ENOMEM;
evlist->overwrite = overwrite;
out_delete_threads:
thread_map__delete(evlist->threads);
+ evlist->threads = NULL;
return -1;
}
}
}
+static int perf_evlist__create_syswide_maps(struct perf_evlist *evlist)
+{
+ int err = -ENOMEM;
+
+ /*
+ * Try reading /sys/devices/system/cpu/online to get
+ * an all cpus map.
+ *
+ * FIXME: -ENOMEM is the best we can do here, the cpu_map
+ * code needs an overhaul to properly forward the
+ * error, and we may not want to do that fallback to a
+ * default cpu identity map :-\
+ */
+ evlist->cpus = cpu_map__new(NULL);
+ if (evlist->cpus == NULL)
+ goto out;
+
+ evlist->threads = thread_map__new_dummy();
+ if (evlist->threads == NULL)
+ goto out_free_cpus;
+
+ err = 0;
+out:
+ return err;
+out_free_cpus:
+ cpu_map__delete(evlist->cpus);
+ evlist->cpus = NULL;
+ goto out;
+}
+
int perf_evlist__open(struct perf_evlist *evlist)
{
struct perf_evsel *evsel;
int err;
+ /*
+ * Default: one fd per CPU, all threads, aka systemwide
+ * as sys_perf_event_open(cpu = -1, thread = -1) is EINVAL
+ */
+ if (evlist->threads == NULL && evlist->cpus == NULL) {
+ err = perf_evlist__create_syswide_maps(evlist);
+ if (err < 0)
+ goto out_err;
+ }
+
perf_evlist__update_id_pos(evlist);
evlist__for_each(evlist, evsel) {
sigaction(SIGUSR1, &act, NULL);
}
- if (target__none(target))
+ if (target__none(target)) {
+ if (evlist->threads == NULL) {
+ fprintf(stderr, "FATAL: evlist->threads need to be set at this point (%s:%d).\n",
+ __func__, __LINE__);
+ goto out_close_pipes;
+ }
evlist->threads->map[0] = evlist->workload.pid;
+ }
close(child_ready_pipe[1]);
close(go_pipe[0]);