* [jcmvbkbc-xtensa:xtensa-6.6-rc5-esp32 33/38] kernel/events/core.c:6374:15: warning: variable 'flags' set but not used
@ 2023-10-14 3:52 kernel test robot
0 siblings, 0 replies; only message in thread
From: kernel test robot @ 2023-10-14 3:52 UTC (permalink / raw)
To: Max Filippov; +Cc: llvm, oe-kbuild-all
tree: https://github.com/jcmvbkbc/linux-xtensa xtensa-6.6-rc5-esp32
head: c2191d1bd425905e3a6439c28449222848e9ecde
commit: 028231d5f2c4db12a43a4c54c45622bb547975be [33/38] WIP: perf: support mmapping event on noMMU
config: x86_64-rhel-8.3-rust (https://download.01.org/0day-ci/archive/20231014/202310141148.ZLbvbBSG-lkp@intel.com/config)
compiler: clang version 16.0.4 (https://github.com/llvm/llvm-project.git ae42196bc493ffe877a7e3dff8be32035dea4d07)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20231014/202310141148.ZLbvbBSG-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202310141148.ZLbvbBSG-lkp@intel.com/
All warnings (new ones prefixed by >>):
>> kernel/events/core.c:6374:15: warning: variable 'flags' set but not used [-Wunused-but-set-variable]
int ret = 0, flags = 0;
^
>> kernel/events/core.c:6583:15: warning: no previous prototype for function 'perf_get_unmapped_area' [-Wmissing-prototypes]
unsigned long perf_get_unmapped_area(struct file *file,
^
kernel/events/core.c:6583:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
unsigned long perf_get_unmapped_area(struct file *file,
^
static
kernel/events/core.c:6639:17: warning: unused function 'perf_mmap_capabilities' [-Wunused-function]
static unsigned perf_mmap_capabilities(struct file *file)
^
3 warnings generated.
vim +/flags +6374 kernel/events/core.c
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6363
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6364 static int perf_mmap(struct file *file, struct vm_area_struct *vma)
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6365 {
cdd6c482c9ff9c kernel/perf_event.c Ingo Molnar 2009-09-21 6366 struct perf_event *event = file->private_data;
22a4f650d686ee kernel/perf_counter.c Ingo Molnar 2009-06-01 6367 unsigned long user_locked, user_lock_limit;
789f90fcf6b0b5 kernel/perf_counter.c Peter Zijlstra 2009-05-15 6368 struct user_struct *user = current_user();
56de4e8f914668 kernel/events/core.c Steven Rostedt (VMware 2019-12-13 6369) struct perf_buffer *rb = NULL;
22a4f650d686ee kernel/perf_counter.c Ingo Molnar 2009-06-01 6370 unsigned long locked, lock_limit;
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6371 unsigned long vma_size;
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6372 unsigned long nr_pages;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6373 long user_extra = 0, extra = 0;
d57e34fdd60be7 kernel/perf_event.c Peter Zijlstra 2010-05-28 @6374 int ret = 0, flags = 0;
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6375
c7920614cebbf2 kernel/perf_event.c Peter Zijlstra 2010-05-18 6376 /*
c7920614cebbf2 kernel/perf_event.c Peter Zijlstra 2010-05-18 6377 * Don't allow mmap() of inherited per-task counters. This would
c7920614cebbf2 kernel/perf_event.c Peter Zijlstra 2010-05-18 6378 * create a performance issue due to all children writing to the
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 6379 * same rb.
c7920614cebbf2 kernel/perf_event.c Peter Zijlstra 2010-05-18 6380 */
c7920614cebbf2 kernel/perf_event.c Peter Zijlstra 2010-05-18 6381 if (event->cpu == -1 && event->attr.inherit)
c7920614cebbf2 kernel/perf_event.c Peter Zijlstra 2010-05-18 6382 return -EINVAL;
c7920614cebbf2 kernel/perf_event.c Peter Zijlstra 2010-05-18 6383
43a21ea81a2400 kernel/perf_counter.c Peter Zijlstra 2009-03-25 6384 if (!(vma->vm_flags & VM_SHARED))
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6385 return -EINVAL;
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6386
da97e18458fb42 kernel/events/core.c Joel Fernandes (Google 2019-10-14 6387) ret = security_perf_event_read(event);
da97e18458fb42 kernel/events/core.c Joel Fernandes (Google 2019-10-14 6388) if (ret)
da97e18458fb42 kernel/events/core.c Joel Fernandes (Google 2019-10-14 6389) return ret;
da97e18458fb42 kernel/events/core.c Joel Fernandes (Google 2019-10-14 6390)
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6391 vma_size = vma->vm_end - vma->vm_start;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6392
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6393 if (vma->vm_pgoff == 0) {
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6394 nr_pages = (vma_size / PAGE_SIZE) - 1;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6395 } else {
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6396 /*
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6397 * AUX area mapping: if rb->aux_nr_pages != 0, it's already
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6398 * mapped, all subsequent mappings should have the same size
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6399 * and offset. Must be above the normal perf buffer.
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6400 */
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6401 u64 aux_offset, aux_size;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6402
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6403 if (!event->rb)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6404 return -EINVAL;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6405
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6406 nr_pages = vma_size / PAGE_SIZE;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6407
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6408 mutex_lock(&event->mmap_mutex);
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6409 ret = -EINVAL;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6410
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6411 rb = event->rb;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6412 if (!rb)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6413 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6414
6aa7de059173a9 kernel/events/core.c Mark Rutland 2017-10-23 6415 aux_offset = READ_ONCE(rb->user_page->aux_offset);
6aa7de059173a9 kernel/events/core.c Mark Rutland 2017-10-23 6416 aux_size = READ_ONCE(rb->user_page->aux_size);
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6417
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6418 if (aux_offset < perf_data_size(rb) + PAGE_SIZE)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6419 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6420
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6421 if (aux_offset != vma->vm_pgoff << PAGE_SHIFT)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6422 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6423
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6424 /* already mapped with a different offset */
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6425 if (rb_has_aux(rb) && rb->aux_pgoff != vma->vm_pgoff)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6426 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6427
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6428 if (aux_size != vma_size || aux_size != nr_pages * PAGE_SIZE)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6429 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6430
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6431 /* already mapped with a different size */
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6432 if (rb_has_aux(rb) && rb->aux_nr_pages != nr_pages)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6433 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6434
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6435 if (!is_power_of_2(nr_pages))
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6436 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6437
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6438 if (!atomic_inc_not_zero(&rb->mmap_count))
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6439 goto aux_unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6440
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6441 if (rb_has_aux(rb)) {
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6442 atomic_inc(&rb->aux_mmap_count);
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6443 ret = 0;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6444 goto unlock;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6445 }
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6446
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6447 atomic_set(&rb->aux_mmap_count, 1);
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6448 user_extra = nr_pages;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6449
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6450 goto accounting;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6451 }
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6452
7730d8655880f4 kernel/perf_counter.c Peter Zijlstra 2009-03-25 6453 /*
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 6454 * If we have rb pages ensure they're a power-of-two number, so we
7730d8655880f4 kernel/perf_counter.c Peter Zijlstra 2009-03-25 6455 * can do bitmasks instead of modulo.
7730d8655880f4 kernel/perf_counter.c Peter Zijlstra 2009-03-25 6456 */
2ed11312eb1950 kernel/events/core.c Kan Liang 2015-03-02 6457 if (nr_pages != 0 && !is_power_of_2(nr_pages))
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6458 return -EINVAL;
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6459
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6460 if (vma_size != PAGE_SIZE * (1 + nr_pages))
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6461 return -EINVAL;
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6462
cdd6c482c9ff9c kernel/perf_event.c Ingo Molnar 2009-09-21 6463 WARN_ON_ONCE(event->ctx->parent_ctx);
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6464 again:
cdd6c482c9ff9c kernel/perf_event.c Ingo Molnar 2009-09-21 6465 mutex_lock(&event->mmap_mutex);
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 6466 if (event->rb) {
60490e7966659b kernel/events/core.c Zhipeng Xie 2022-02-09 6467 if (data_page_nr(event->rb) != nr_pages) {
ebb3c4c4cb81d6 kernel/perf_counter.c Peter Zijlstra 2009-04-06 6468 ret = -EINVAL;
ebb3c4c4cb81d6 kernel/perf_counter.c Peter Zijlstra 2009-04-06 6469 goto unlock;
ebb3c4c4cb81d6 kernel/perf_counter.c Peter Zijlstra 2009-04-06 6470 }
ebb3c4c4cb81d6 kernel/perf_counter.c Peter Zijlstra 2009-04-06 6471
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6472 if (!atomic_inc_not_zero(&event->rb->mmap_count)) {
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6473 /*
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 6474 * Raced against perf_mmap_close(); remove the
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 6475 * event and try again.
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6476 */
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 6477 ring_buffer_attach(event, NULL);
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6478 mutex_unlock(&event->mmap_mutex);
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6479 goto again;
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6480 }
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6481
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6482 goto unlock;
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6483 }
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 6484
789f90fcf6b0b5 kernel/perf_counter.c Peter Zijlstra 2009-05-15 6485 user_extra = nr_pages + 1;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6486
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6487 accounting:
cdd6c482c9ff9c kernel/perf_event.c Ingo Molnar 2009-09-21 6488 user_lock_limit = sysctl_perf_event_mlock >> (PAGE_SHIFT - 10);
a3862d3f814ce7 kernel/perf_counter.c Ingo Molnar 2009-05-24 6489
a3862d3f814ce7 kernel/perf_counter.c Ingo Molnar 2009-05-24 6490 /*
a3862d3f814ce7 kernel/perf_counter.c Ingo Molnar 2009-05-24 6491 * Increase the limit linearly with more CPUs:
a3862d3f814ce7 kernel/perf_counter.c Ingo Molnar 2009-05-24 6492 */
a3862d3f814ce7 kernel/perf_counter.c Ingo Molnar 2009-05-24 6493 user_lock_limit *= num_online_cpus();
a3862d3f814ce7 kernel/perf_counter.c Ingo Molnar 2009-05-24 6494
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6495 user_locked = atomic_long_read(&user->locked_vm);
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6496
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6497 /*
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6498 * sysctl_perf_event_mlock may have changed, so that
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6499 * user->locked_vm > user_lock_limit
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6500 */
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6501 if (user_locked > user_lock_limit)
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6502 user_locked = user_lock_limit;
003461559ef7a9 kernel/events/core.c Song Liu 2020-01-23 6503 user_locked += user_extra;
c5078f78b455fb kernel/perf_counter.c Peter Zijlstra 2009-05-05 6504
c4b75479741c9c kernel/events/core.c Alexander Shishkin 2019-11-20 6505 if (user_locked > user_lock_limit) {
d44248a4133773 kernel/events/core.c Song Liu 2019-09-04 6506 /*
d44248a4133773 kernel/events/core.c Song Liu 2019-09-04 6507 * charge locked_vm until it hits user_lock_limit;
d44248a4133773 kernel/events/core.c Song Liu 2019-09-04 6508 * charge the rest from pinned_vm
d44248a4133773 kernel/events/core.c Song Liu 2019-09-04 6509 */
789f90fcf6b0b5 kernel/perf_counter.c Peter Zijlstra 2009-05-15 6510 extra = user_locked - user_lock_limit;
d44248a4133773 kernel/events/core.c Song Liu 2019-09-04 6511 user_extra -= extra;
d44248a4133773 kernel/events/core.c Song Liu 2019-09-04 6512 }
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6513
78d7d407b62a02 kernel/perf_event.c Jiri Slaby 2010-03-05 6514 lock_limit = rlimit(RLIMIT_MEMLOCK);
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6515 lock_limit >>= PAGE_SHIFT;
70f8a3ca68d3e1 kernel/events/core.c Davidlohr Bueso 2019-02-06 6516 locked = atomic64_read(&vma->vm_mm->pinned_vm) + extra;
37d81828385f8f kernel/perf_counter.c Paul Mackerras 2009-03-23 6517
da97e18458fb42 kernel/events/core.c Joel Fernandes (Google 2019-10-14 6518) if ((locked > lock_limit) && perf_is_paranoid() &&
459ec28ab404d7 kernel/perf_counter.c Ingo Molnar 2009-09-13 6519 !capable(CAP_IPC_LOCK)) {
ebb3c4c4cb81d6 kernel/perf_counter.c Peter Zijlstra 2009-04-06 6520 ret = -EPERM;
ebb3c4c4cb81d6 kernel/perf_counter.c Peter Zijlstra 2009-04-06 6521 goto unlock;
ebb3c4c4cb81d6 kernel/perf_counter.c Peter Zijlstra 2009-04-06 6522 }
7b732a75047738 kernel/perf_counter.c Peter Zijlstra 2009-03-23 6523
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 6524 WARN_ON(!rb && event->rb);
906010b2134e14 kernel/perf_event.c Peter Zijlstra 2009-09-21 6525
d57e34fdd60be7 kernel/perf_event.c Peter Zijlstra 2010-05-28 6526 if (vma->vm_flags & VM_WRITE)
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 6527 flags |= RING_BUFFER_WRITABLE;
d57e34fdd60be7 kernel/perf_event.c Peter Zijlstra 2010-05-28 6528
:::::: The code at line 6374 was first introduced by commit
:::::: d57e34fdd60be7ffd0b1d86bfa1a553df86b7172 perf: Simplify the ring-buffer logic: make perf_buffer_alloc() do everything needed
:::::: TO: Peter Zijlstra <a.p.zijlstra@chello.nl>
:::::: CC: Ingo Molnar <mingo@elte.hu>
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2023-10-14 3:53 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-10-14 3:52 [jcmvbkbc-xtensa:xtensa-6.6-rc5-esp32 33/38] kernel/events/core.c:6374:15: warning: variable 'flags' set but not used kernel test robot
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox