diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2019-11-26 15:04:47 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2019-11-26 15:04:47 -0800 |
commit | 3f59dbcace56fae7e4ed303bab90f1bedadcfdf4 (patch) | |
tree | c425529202b9dbe3e3b3dde072c1edf51b1b9e93 /tools/perf/lib/evlist.c | |
parent | df28204bb0f29cc475c0a8893c99b46a11a4903f (diff) | |
parent | ceb9e77324fa661b1001a0ae66f061b5fcb4e4e6 (diff) |
Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf updates from Ingo Molnar:
"The main kernel side changes in this cycle were:
- Various Intel-PT updates and optimizations (Alexander Shishkin)
- Prohibit kprobes on Xen/KVM emulate prefixes (Masami Hiramatsu)
- Add support for LSM and SELinux checks to control access to the
perf syscall (Joel Fernandes)
- Misc other changes, optimizations, fixes and cleanups - see the
shortlog for details.
There were numerous tooling changes as well - 254 non-merge commits.
Here are the main changes - too many to list in detail:
- Enhancements to core tooling infrastructure, perf.data, libperf,
libtraceevent, event parsing, vendor events, Intel PT, callchains,
BPF support and instruction decoding.
- There were updates to the following tools:
perf annotate
perf diff
perf inject
perf kvm
perf list
perf maps
perf parse
perf probe
perf record
perf report
perf script
perf stat
perf test
perf trace
- And a lot of other changes: please see the shortlog and Git log for
more details"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (279 commits)
perf parse: Fix potential memory leak when handling tracepoint errors
perf probe: Fix spelling mistake "addrees" -> "address"
libtraceevent: Fix memory leakage in copy_filter_type
libtraceevent: Fix header installation
perf intel-bts: Does not support AUX area sampling
perf intel-pt: Add support for decoding AUX area samples
perf intel-pt: Add support for recording AUX area samples
perf pmu: When using default config, record which bits of config were changed by the user
perf auxtrace: Add support for queuing AUX area samples
perf session: Add facility to peek at all events
perf auxtrace: Add support for dumping AUX area samples
perf inject: Cut AUX area samples
perf record: Add aux-sample-size config term
perf record: Add support for AUX area sampling
perf auxtrace: Add support for AUX area sample recording
perf auxtrace: Move perf_evsel__find_pmu()
perf record: Add a function to test for kernel support for AUX area sampling
perf tools: Add kernel AUX area sampling definitions
perf/core: Make the mlock accounting simple again
perf report: Jump to symbol source view from total cycles view
...
Diffstat (limited to 'tools/perf/lib/evlist.c')
-rw-r--r-- | tools/perf/lib/evlist.c | 357 |
1 files changed, 357 insertions, 0 deletions
diff --git a/tools/perf/lib/evlist.c b/tools/perf/lib/evlist.c index d1496fee810c..205ddbb80bc1 100644 --- a/tools/perf/lib/evlist.c +++ b/tools/perf/lib/evlist.c @@ -8,13 +8,20 @@ #include <internal/evlist.h> #include <internal/evsel.h> #include <internal/xyarray.h> +#include <internal/mmap.h> +#include <internal/cpumap.h> +#include <internal/threadmap.h> +#include <internal/xyarray.h> +#include <internal/lib.h> #include <linux/zalloc.h> +#include <sys/ioctl.h> #include <stdlib.h> #include <errno.h> #include <unistd.h> #include <fcntl.h> #include <signal.h> #include <poll.h> +#include <sys/mman.h> #include <perf/cpumap.h> #include <perf/threadmap.h> #include <api/fd/array.h> @@ -27,6 +34,7 @@ void perf_evlist__init(struct perf_evlist *evlist) INIT_HLIST_HEAD(&evlist->heads[i]); INIT_LIST_HEAD(&evlist->entries); evlist->nr_entries = 0; + fdarray__init(&evlist->pollfd, 64); } static void __perf_evlist__propagate_maps(struct perf_evlist *evlist, @@ -101,8 +109,36 @@ perf_evlist__next(struct perf_evlist *evlist, struct perf_evsel *prev) return next; } +static void perf_evlist__purge(struct perf_evlist *evlist) +{ + struct perf_evsel *pos, *n; + + perf_evlist__for_each_entry_safe(evlist, n, pos) { + list_del_init(&pos->node); + perf_evsel__delete(pos); + } + + evlist->nr_entries = 0; +} + +void perf_evlist__exit(struct perf_evlist *evlist) +{ + perf_cpu_map__put(evlist->cpus); + perf_thread_map__put(evlist->threads); + evlist->cpus = NULL; + evlist->threads = NULL; + fdarray__exit(&evlist->pollfd); +} + void perf_evlist__delete(struct perf_evlist *evlist) { + if (evlist == NULL) + return; + + perf_evlist__munmap(evlist); + perf_evlist__close(evlist); + perf_evlist__purge(evlist); + perf_evlist__exit(evlist); free(evlist); } @@ -277,7 +313,328 @@ int perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd, return pos; } +static void perf_evlist__munmap_filtered(struct fdarray *fda, int fd, + void *arg __maybe_unused) +{ + struct perf_mmap *map = fda->priv[fd].ptr; + + if (map) + perf_mmap__put(map); +} + +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, NULL); +} + int perf_evlist__poll(struct perf_evlist *evlist, int timeout) { return fdarray__poll(&evlist->pollfd, timeout); } + +static struct perf_mmap* perf_evlist__alloc_mmap(struct perf_evlist *evlist, bool overwrite) +{ + int i; + struct perf_mmap *map; + + map = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); + if (!map) + return NULL; + + for (i = 0; i < evlist->nr_mmaps; i++) { + struct perf_mmap *prev = i ? &map[i - 1] : NULL; + + /* + * When the perf_mmap() call is made we grab one refcount, plus + * one extra to let perf_mmap__consume() get the last + * events after all real references (perf_mmap__get()) are + * dropped. + * + * Each PERF_EVENT_IOC_SET_OUTPUT points to this mmap and + * thus does perf_mmap__get() on it. + */ + perf_mmap__init(&map[i], prev, overwrite, NULL); + } + + return map; +} + +static void perf_evlist__set_sid_idx(struct perf_evlist *evlist, + struct perf_evsel *evsel, int idx, int cpu, + int thread) +{ + struct perf_sample_id *sid = SID(evsel, cpu, thread); + + sid->idx = idx; + if (evlist->cpus && cpu >= 0) + sid->cpu = evlist->cpus->map[cpu]; + else + sid->cpu = -1; + if (!evsel->system_wide && evlist->threads && thread >= 0) + sid->tid = perf_thread_map__pid(evlist->threads, thread); + else + sid->tid = -1; +} + +static struct perf_mmap* +perf_evlist__mmap_cb_get(struct perf_evlist *evlist, bool overwrite, int idx) +{ + struct perf_mmap *maps; + + maps = overwrite ? evlist->mmap_ovw : evlist->mmap; + + if (!maps) { + maps = perf_evlist__alloc_mmap(evlist, overwrite); + if (!maps) + return NULL; + + if (overwrite) + evlist->mmap_ovw = maps; + else + evlist->mmap = maps; + } + + return &maps[idx]; +} + +#define FD(e, x, y) (*(int *) xyarray__entry(e->fd, x, y)) + +static int +perf_evlist__mmap_cb_mmap(struct perf_mmap *map, struct perf_mmap_param *mp, + int output, int cpu) +{ + return perf_mmap__mmap(map, mp, output, cpu); +} + +static void perf_evlist__set_mmap_first(struct perf_evlist *evlist, struct perf_mmap *map, + bool overwrite) +{ + if (overwrite) + evlist->mmap_ovw_first = map; + else + evlist->mmap_first = map; +} + +static int +mmap_per_evsel(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops, + int idx, struct perf_mmap_param *mp, int cpu_idx, + int thread, int *_output, int *_output_overwrite) +{ + int evlist_cpu = perf_cpu_map__cpu(evlist->cpus, cpu_idx); + struct perf_evsel *evsel; + int revent; + + perf_evlist__for_each_entry(evlist, evsel) { + bool overwrite = evsel->attr.write_backward; + struct perf_mmap *map; + int *output, fd, cpu; + + if (evsel->system_wide && thread) + continue; + + cpu = perf_cpu_map__idx(evsel->cpus, evlist_cpu); + if (cpu == -1) + continue; + + map = ops->get(evlist, overwrite, idx); + if (map == NULL) + return -ENOMEM; + + if (overwrite) { + mp->prot = PROT_READ; + output = _output_overwrite; + } else { + mp->prot = PROT_READ | PROT_WRITE; + output = _output; + } + + fd = FD(evsel, cpu, thread); + + if (*output == -1) { + *output = fd; + + /* + * The last one will be done at perf_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(). + */ + refcount_set(&map->refcnt, 2); + + if (ops->mmap(map, mp, *output, evlist_cpu) < 0) + return -1; + + if (!idx) + perf_evlist__set_mmap_first(evlist, map, overwrite); + } else { + if (ioctl(fd, PERF_EVENT_IOC_SET_OUTPUT, *output) != 0) + return -1; + + perf_mmap__get(map); + } + + revent = !overwrite ? POLLIN : 0; + + if (!evsel->system_wide && + perf_evlist__add_pollfd(evlist, fd, map, revent) < 0) { + perf_mmap__put(map); + return -1; + } + + if (evsel->attr.read_format & PERF_FORMAT_ID) { + if (perf_evlist__id_add_fd(evlist, evsel, cpu, thread, + fd) < 0) + return -1; + perf_evlist__set_sid_idx(evlist, evsel, idx, cpu, + thread); + } + } + + return 0; +} + +static int +mmap_per_thread(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops, + struct perf_mmap_param *mp) +{ + int thread; + int nr_threads = perf_thread_map__nr(evlist->threads); + + for (thread = 0; thread < nr_threads; thread++) { + int output = -1; + int output_overwrite = -1; + + if (ops->idx) + ops->idx(evlist, mp, thread, false); + + if (mmap_per_evsel(evlist, ops, thread, mp, 0, thread, + &output, &output_overwrite)) + goto out_unmap; + } + + return 0; + +out_unmap: + perf_evlist__munmap(evlist); + return -1; +} + +static int +mmap_per_cpu(struct perf_evlist *evlist, struct perf_evlist_mmap_ops *ops, + struct perf_mmap_param *mp) +{ + int nr_threads = perf_thread_map__nr(evlist->threads); + int nr_cpus = perf_cpu_map__nr(evlist->cpus); + int cpu, thread; + + for (cpu = 0; cpu < nr_cpus; cpu++) { + int output = -1; + int output_overwrite = -1; + + if (ops->idx) + ops->idx(evlist, mp, cpu, true); + + for (thread = 0; thread < nr_threads; thread++) { + if (mmap_per_evsel(evlist, ops, cpu, mp, cpu, + thread, &output, &output_overwrite)) + goto out_unmap; + } + } + + return 0; + +out_unmap: + perf_evlist__munmap(evlist); + return -1; +} + +static int perf_evlist__nr_mmaps(struct perf_evlist *evlist) +{ + int nr_mmaps; + + nr_mmaps = perf_cpu_map__nr(evlist->cpus); + if (perf_cpu_map__empty(evlist->cpus)) + nr_mmaps = perf_thread_map__nr(evlist->threads); + + return nr_mmaps; +} + +int perf_evlist__mmap_ops(struct perf_evlist *evlist, + struct perf_evlist_mmap_ops *ops, + struct perf_mmap_param *mp) +{ + struct perf_evsel *evsel; + const struct perf_cpu_map *cpus = evlist->cpus; + const struct perf_thread_map *threads = evlist->threads; + + if (!ops || !ops->get || !ops->mmap) + return -EINVAL; + + mp->mask = evlist->mmap_len - page_size - 1; + + evlist->nr_mmaps = perf_evlist__nr_mmaps(evlist); + + perf_evlist__for_each_entry(evlist, evsel) { + if ((evsel->attr.read_format & PERF_FORMAT_ID) && + evsel->sample_id == NULL && + perf_evsel__alloc_id(evsel, perf_cpu_map__nr(cpus), threads->nr) < 0) + return -ENOMEM; + } + + if (evlist->pollfd.entries == NULL && perf_evlist__alloc_pollfd(evlist) < 0) + return -ENOMEM; + + if (perf_cpu_map__empty(cpus)) + return mmap_per_thread(evlist, ops, mp); + + return mmap_per_cpu(evlist, ops, mp); +} + +int perf_evlist__mmap(struct perf_evlist *evlist, int pages) +{ + struct perf_mmap_param mp; + struct perf_evlist_mmap_ops ops = { + .get = perf_evlist__mmap_cb_get, + .mmap = perf_evlist__mmap_cb_mmap, + }; + + evlist->mmap_len = (pages + 1) * page_size; + + return perf_evlist__mmap_ops(evlist, &ops, &mp); +} + +void perf_evlist__munmap(struct perf_evlist *evlist) +{ + int i; + + if (evlist->mmap) { + for (i = 0; i < evlist->nr_mmaps; i++) + perf_mmap__munmap(&evlist->mmap[i]); + } + + if (evlist->mmap_ovw) { + for (i = 0; i < evlist->nr_mmaps; i++) + perf_mmap__munmap(&evlist->mmap_ovw[i]); + } + + zfree(&evlist->mmap); + zfree(&evlist->mmap_ovw); +} + +struct perf_mmap* +perf_evlist__next_mmap(struct perf_evlist *evlist, struct perf_mmap *map, + bool overwrite) +{ + if (map) + return map->next; + + return overwrite ? evlist->mmap_ovw_first : evlist->mmap_first; +} |