linux/tools/lib/perf/mmap.c
Rob Herring 47d01e7b99 libperf: Add support for user space counter access
x86 and arm64 can both support direct access of event counters in
userspace. The access sequence is less than trivial and currently exists
in perf test code (tools/perf/arch/x86/tests/rdpmc.c) with copies in
projects such as PAPI and libpfm4.

In order to support userspace access, an event must be mmapped first
with perf_evsel__mmap(). Then subsequent calls to perf_evsel__read()
will use the fast path (assuming the arch supports it).

Committer notes:

Added a '__maybe_unused' attribute to the read_perf_counter() argument
to fix the build on arches other than x86_64 and arm.

Committer testing:

  Building and running the libperf tests in verbose mode (V=1) now shows
  those "loop = N, count = N" extra lines, testing user space counter
  access.

  # make V=1 -C tools/lib/perf tests
  make: Entering directory '/home/acme/git/perf/tools/lib/perf'
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=. obj=libperf
  make -C /home/acme/git/perf/tools/lib/api/ O= libapi.a
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=./fd obj=libapi
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=./fs obj=libapi
  make -C tests
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-cpumap-a test-cpumap.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-threadmap-a test-threadmap.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-evlist-a test-evlist.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-evsel-a test-evsel.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-cpumap-so test-cpumap.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-threadmap-so test-threadmap.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-evlist-so test-evlist.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-evsel-so test-evsel.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  make -C tests run
  running static:
  - running test-cpumap.c...OK
  - running test-threadmap.c...OK
  - running test-evlist.c...OK
  - running test-evsel.c...
  	loop = 65536, count = 333926
  	loop = 131072, count = 655781
  	loop = 262144, count = 1311141
  	loop = 524288, count = 2630126
  	loop = 1048576, count = 5256955
  	loop = 65536, count = 524594
  	loop = 131072, count = 1058916
  	loop = 262144, count = 2097458
  	loop = 524288, count = 4205429
  	loop = 1048576, count = 8406606
  OK
  running dynamic:
  - running test-cpumap.c...OK
  - running test-threadmap.c...OK
  - running test-evlist.c...OK
  - running test-evsel.c...
  	loop = 65536, count = 328102
  	loop = 131072, count = 655782
  	loop = 262144, count = 1317494
  	loop = 524288, count = 2627851
  	loop = 1048576, count = 5255187
  	loop = 65536, count = 524601
  	loop = 131072, count = 1048923
  	loop = 262144, count = 2107917
  	loop = 524288, count = 4194606
  	loop = 1048576, count = 8409322
  OK
  make: Leaving directory '/home/acme/git/perf/tools/lib/perf'
  #

Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Itaru Kitayama <itaru.kitayama@gmail.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Link: http://lore.kernel.org/lkml/20210414155412.3697605-4-robh@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2021-04-20 08:10:45 -03:00

364 lines
8.1 KiB
C

// SPDX-License-Identifier: GPL-2.0
#include <sys/mman.h>
#include <inttypes.h>
#include <asm/bug.h>
#include <errno.h>
#include <string.h>
#include <linux/ring_buffer.h>
#include <linux/perf_event.h>
#include <perf/mmap.h>
#include <perf/event.h>
#include <perf/evsel.h>
#include <internal/mmap.h>
#include <internal/lib.h>
#include <linux/kernel.h>
#include <linux/math64.h>
#include "internal.h"
void perf_mmap__init(struct perf_mmap *map, struct perf_mmap *prev,
bool overwrite, libperf_unmap_cb_t unmap_cb)
{
map->fd = -1;
map->overwrite = overwrite;
map->unmap_cb = unmap_cb;
refcount_set(&map->refcnt, 0);
if (prev)
prev->next = map;
}
size_t perf_mmap__mmap_len(struct perf_mmap *map)
{
return map->mask + 1 + page_size;
}
int perf_mmap__mmap(struct perf_mmap *map, struct perf_mmap_param *mp,
int fd, int cpu)
{
map->prev = 0;
map->mask = mp->mask;
map->base = mmap(NULL, perf_mmap__mmap_len(map), mp->prot,
MAP_SHARED, fd, 0);
if (map->base == MAP_FAILED) {
map->base = NULL;
return -1;
}
map->fd = fd;
map->cpu = cpu;
return 0;
}
void perf_mmap__munmap(struct perf_mmap *map)
{
if (map && map->base != NULL) {
munmap(map->base, perf_mmap__mmap_len(map));
map->base = NULL;
map->fd = -1;
refcount_set(&map->refcnt, 0);
}
if (map && map->unmap_cb)
map->unmap_cb(map);
}
void perf_mmap__get(struct perf_mmap *map)
{
refcount_inc(&map->refcnt);
}
void perf_mmap__put(struct perf_mmap *map)
{
BUG_ON(map->base && refcount_read(&map->refcnt) == 0);
if (refcount_dec_and_test(&map->refcnt))
perf_mmap__munmap(map);
}
static inline void perf_mmap__write_tail(struct perf_mmap *md, u64 tail)
{
ring_buffer_write_tail(md->base, tail);
}
u64 perf_mmap__read_head(struct perf_mmap *map)
{
return ring_buffer_read_head(map->base);
}
static bool perf_mmap__empty(struct perf_mmap *map)
{
struct perf_event_mmap_page *pc = map->base;
return perf_mmap__read_head(map) == map->prev && !pc->aux_size;
}
void perf_mmap__consume(struct perf_mmap *map)
{
if (!map->overwrite) {
u64 old = map->prev;
perf_mmap__write_tail(map, old);
}
if (refcount_read(&map->refcnt) == 1 && perf_mmap__empty(map))
perf_mmap__put(map);
}
static int overwrite_rb_find_range(void *buf, int mask, u64 *start, u64 *end)
{
struct perf_event_header *pheader;
u64 evt_head = *start;
int size = mask + 1;
pr_debug2("%s: buf=%p, start=%"PRIx64"\n", __func__, buf, *start);
pheader = (struct perf_event_header *)(buf + (*start & mask));
while (true) {
if (evt_head - *start >= (unsigned int)size) {
pr_debug("Finished reading overwrite ring buffer: rewind\n");
if (evt_head - *start > (unsigned int)size)
evt_head -= pheader->size;
*end = evt_head;
return 0;
}
pheader = (struct perf_event_header *)(buf + (evt_head & mask));
if (pheader->size == 0) {
pr_debug("Finished reading overwrite ring buffer: get start\n");
*end = evt_head;
return 0;
}
evt_head += pheader->size;
pr_debug3("move evt_head: %"PRIx64"\n", evt_head);
}
WARN_ONCE(1, "Shouldn't get here\n");
return -1;
}
/*
* Report the start and end of the available data in ringbuffer
*/
static int __perf_mmap__read_init(struct perf_mmap *md)
{
u64 head = perf_mmap__read_head(md);
u64 old = md->prev;
unsigned char *data = md->base + page_size;
unsigned long size;
md->start = md->overwrite ? head : old;
md->end = md->overwrite ? old : head;
if ((md->end - md->start) < md->flush)
return -EAGAIN;
size = md->end - md->start;
if (size > (unsigned long)(md->mask) + 1) {
if (!md->overwrite) {
WARN_ONCE(1, "failed to keep up with mmap data. (warn only once)\n");
md->prev = head;
perf_mmap__consume(md);
return -EAGAIN;
}
/*
* Backward ring buffer is full. We still have a chance to read
* most of data from it.
*/
if (overwrite_rb_find_range(data, md->mask, &md->start, &md->end))
return -EINVAL;
}
return 0;
}
int perf_mmap__read_init(struct perf_mmap *map)
{
/*
* Check if event was unmapped due to a POLLHUP/POLLERR.
*/
if (!refcount_read(&map->refcnt))
return -ENOENT;
return __perf_mmap__read_init(map);
}
/*
* Mandatory for overwrite mode
* The direction of overwrite mode is backward.
* The last perf_mmap__read() will set tail to map->core.prev.
* Need to correct the map->core.prev to head which is the end of next read.
*/
void perf_mmap__read_done(struct perf_mmap *map)
{
/*
* Check if event was unmapped due to a POLLHUP/POLLERR.
*/
if (!refcount_read(&map->refcnt))
return;
map->prev = perf_mmap__read_head(map);
}
/* When check_messup is true, 'end' must points to a good entry */
static union perf_event *perf_mmap__read(struct perf_mmap *map,
u64 *startp, u64 end)
{
unsigned char *data = map->base + page_size;
union perf_event *event = NULL;
int diff = end - *startp;
if (diff >= (int)sizeof(event->header)) {
size_t size;
event = (union perf_event *)&data[*startp & map->mask];
size = event->header.size;
if (size < sizeof(event->header) || diff < (int)size)
return NULL;
/*
* Event straddles the mmap boundary -- header should always
* be inside due to u64 alignment of output.
*/
if ((*startp & map->mask) + size != ((*startp + size) & map->mask)) {
unsigned int offset = *startp;
unsigned int len = min(sizeof(*event), size), cpy;
void *dst = map->event_copy;
do {
cpy = min(map->mask + 1 - (offset & map->mask), len);
memcpy(dst, &data[offset & map->mask], cpy);
offset += cpy;
dst += cpy;
len -= cpy;
} while (len);
event = (union perf_event *)map->event_copy;
}
*startp += size;
}
return event;
}
/*
* Read event from ring buffer one by one.
* Return one event for each call.
*
* Usage:
* perf_mmap__read_init()
* while(event = perf_mmap__read_event()) {
* //process the event
* perf_mmap__consume()
* }
* perf_mmap__read_done()
*/
union perf_event *perf_mmap__read_event(struct perf_mmap *map)
{
union perf_event *event;
/*
* Check if event was unmapped due to a POLLHUP/POLLERR.
*/
if (!refcount_read(&map->refcnt))
return NULL;
/* non-overwirte doesn't pause the ringbuffer */
if (!map->overwrite)
map->end = perf_mmap__read_head(map);
event = perf_mmap__read(map, &map->start, map->end);
if (!map->overwrite)
map->prev = map->start;
return event;
}
#if defined(__i386__) || defined(__x86_64__)
static u64 read_perf_counter(unsigned int counter)
{
unsigned int low, high;
asm volatile("rdpmc" : "=a" (low), "=d" (high) : "c" (counter));
return low | ((u64)high) << 32;
}
static u64 read_timestamp(void)
{
unsigned int low, high;
asm volatile("rdtsc" : "=a" (low), "=d" (high));
return low | ((u64)high) << 32;
}
#else
static u64 read_perf_counter(unsigned int counter __maybe_unused) { return 0; }
static u64 read_timestamp(void) { return 0; }
#endif
int perf_mmap__read_self(struct perf_mmap *map, struct perf_counts_values *count)
{
struct perf_event_mmap_page *pc = map->base;
u32 seq, idx, time_mult = 0, time_shift = 0;
u64 cnt, cyc = 0, time_offset = 0, time_cycles = 0, time_mask = ~0ULL;
if (!pc || !pc->cap_user_rdpmc)
return -1;
do {
seq = READ_ONCE(pc->lock);
barrier();
count->ena = READ_ONCE(pc->time_enabled);
count->run = READ_ONCE(pc->time_running);
if (pc->cap_user_time && count->ena != count->run) {
cyc = read_timestamp();
time_mult = READ_ONCE(pc->time_mult);
time_shift = READ_ONCE(pc->time_shift);
time_offset = READ_ONCE(pc->time_offset);
if (pc->cap_user_time_short) {
time_cycles = READ_ONCE(pc->time_cycles);
time_mask = READ_ONCE(pc->time_mask);
}
}
idx = READ_ONCE(pc->index);
cnt = READ_ONCE(pc->offset);
if (pc->cap_user_rdpmc && idx) {
s64 evcnt = read_perf_counter(idx - 1);
u16 width = READ_ONCE(pc->pmc_width);
evcnt <<= 64 - width;
evcnt >>= 64 - width;
cnt += evcnt;
} else
return -1;
barrier();
} while (READ_ONCE(pc->lock) != seq);
if (count->ena != count->run) {
u64 delta;
/* Adjust for cap_usr_time_short, a nop if not */
cyc = time_cycles + ((cyc - time_cycles) & time_mask);
delta = time_offset + mul_u64_u32_shr(cyc, time_mult, time_shift);
count->ena += delta;
if (idx)
count->run += delta;
cnt = mul_u64_u64_div64(cnt, count->ena, count->run);
}
count->val = cnt;
return 0;
}