| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <202208230309.KzpqxoH0-lkp@intel.com>
Date: Tue, 23 Aug 2022 03:40:29 +0800
From: kernel test robot <lkp@...el.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: kbuild-all@...ts.01.org, linux-kernel@...r.kernel.org
Subject: [peterz-queue:perf/wip.rewrite 1/5] kernel/events/core.c:12047:9:
error: implicit declaration of function 'mutex_lock_double'; did you mean
'mutex_lock_killable'?
tree: https://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git perf/wip.rewrite
head: 596b8cc92a97fa8defbff896dc9baedef441b983
commit: 6a606214738273b4c6bd00c500bab6cb53a5b9d7 [1/5] perf: Rewrite core context handling
config: mips-allyesconfig (https://download.01.org/0day-ci/archive/20220823/202208230309.KzpqxoH0-lkp@intel.com/config)
compiler: mips-linux-gcc (GCC) 12.1.0
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git/commit/?id=6a606214738273b4c6bd00c500bab6cb53a5b9d7
git remote add peterz-queue https://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git
git fetch --no-tags peterz-queue perf/wip.rewrite
git checkout 6a606214738273b4c6bd00c500bab6cb53a5b9d7
# save the config file
mkdir build_dir && cp config build_dir/.config
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=mips SHELL=/bin/bash kernel/
If you fix the issue, kindly add following tag where applicable
Reported-by: kernel test robot <lkp@...el.com>
All errors (new ones prefixed by >>):
kernel/events/core.c: In function 'perf_rotate_context':
kernel/events/core.c:4236:36: warning: variable 'task_ctx' set but not used [-Wunused-but-set-variable]
4236 | struct perf_event_context *task_ctx = NULL;
| ^~~~~~~~
kernel/events/core.c: At top level:
kernel/events/core.c:4810:1: warning: no previous prototype for 'find_get_pmu_context' [-Wmissing-prototypes]
4810 | find_get_pmu_context(struct pmu *pmu, struct perf_event_context *ctx,
| ^~~~~~~~~~~~~~~~~~~~
kernel/events/core.c: In function 'perf_event_set_output':
>> kernel/events/core.c:12047:9: error: implicit declaration of function 'mutex_lock_double'; did you mean 'mutex_lock_killable'? [-Werror=implicit-function-declaration]
12047 | mutex_lock_double(&event->mmap_mutex, &output_event->mmap_mutex);
| ^~~~~~~~~~~~~~~~~
| mutex_lock_killable
cc1: some warnings being treated as errors
vim +12047 kernel/events/core.c
974802eaa1afdc kernel/perf_counter.c Peter Zijlstra 2009-06-12 11992
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 11993 static int
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 11994 perf_event_set_output(struct perf_event *event, struct perf_event *output_event)
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 11995 {
56de4e8f914668 kernel/events/core.c Steven Rostedt (VMware 2019-12-13 11996) struct perf_buffer *rb = NULL;
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 11997 int ret = -EINVAL;
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 11998
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 11999 if (!output_event) {
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12000 mutex_lock(&event->mmap_mutex);
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12001 goto set;
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12002 }
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12003
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12004 /* don't allow circular references */
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12005 if (event == output_event)
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12006 goto out;
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12007
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12008 /*
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12009 * Don't allow cross-cpu buffers
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12010 */
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12011 if (output_event->cpu != event->cpu)
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12012 goto out;
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12013
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12014 /*
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 12015 * If its not a per-cpu rb, it must be the same task.
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12016 */
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12017 if (output_event->cpu == -1 && output_event->ctx != event->ctx)
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12018 goto out;
0f139300c9057c kernel/perf_event.c Peter Zijlstra 2010-05-20 12019
34f439278cef7b kernel/events/core.c Peter Zijlstra 2015-02-20 12020 /*
34f439278cef7b kernel/events/core.c Peter Zijlstra 2015-02-20 12021 * Mixing clocks in the same buffer is trouble you don't need.
34f439278cef7b kernel/events/core.c Peter Zijlstra 2015-02-20 12022 */
34f439278cef7b kernel/events/core.c Peter Zijlstra 2015-02-20 12023 if (output_event->clock != event->clock)
34f439278cef7b kernel/events/core.c Peter Zijlstra 2015-02-20 12024 goto out;
34f439278cef7b kernel/events/core.c Peter Zijlstra 2015-02-20 12025
9ecda41acb971e kernel/events/core.c Wang Nan 2016-04-05 12026 /*
9ecda41acb971e kernel/events/core.c Wang Nan 2016-04-05 12027 * Either writing ring buffer from beginning or from end.
9ecda41acb971e kernel/events/core.c Wang Nan 2016-04-05 12028 * Mixing is not allowed.
9ecda41acb971e kernel/events/core.c Wang Nan 2016-04-05 12029 */
9ecda41acb971e kernel/events/core.c Wang Nan 2016-04-05 12030 if (is_write_backward(output_event) != is_write_backward(event))
9ecda41acb971e kernel/events/core.c Wang Nan 2016-04-05 12031 goto out;
9ecda41acb971e kernel/events/core.c Wang Nan 2016-04-05 12032
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 12033 /*
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 12034 * If both events generate aux data, they must be on the same PMU
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 12035 */
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 12036 if (has_aux(event) && has_aux(output_event) &&
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 12037 event->pmu != output_event->pmu)
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 12038 goto out;
45bfb2e50471ab kernel/events/core.c Peter Zijlstra 2015-01-14 12039
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12040 /*
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12041 * Hold both mmap_mutex to serialize against perf_mmap_close(). Since
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12042 * output_event is already on rb->event_list, and the list iteration
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12043 * restarts after every removal, it is guaranteed this new event is
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12044 * observed *OR* if output_event is already removed, it's guaranteed we
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12045 * observe !rb->mmap_count.
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12046 */
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 @12047 mutex_lock_double(&event->mmap_mutex, &output_event->mmap_mutex);
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12048 set:
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12049 /* Can't redirect output if we've got an active mmap() */
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12050 if (atomic_read(&event->mmap_count))
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12051 goto unlock;
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12052
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12053 if (output_event) {
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 12054 /* get the rb we want to redirect to */
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 12055 rb = ring_buffer_get(output_event);
76369139ceb955 kernel/events/core.c Frederic Weisbecker 2011-05-19 12056 if (!rb)
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12057 goto unlock;
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12058
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12059 /* did we race against perf_mmap_close() */
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12060 if (!atomic_read(&rb->mmap_count)) {
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12061 ring_buffer_put(rb);
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12062 goto unlock;
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12063 }
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12064 }
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12065
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 12066 ring_buffer_attach(event, rb);
9bb5d40cd93c9d kernel/events/core.c Peter Zijlstra 2013-06-04 12067
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12068 ret = 0;
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12069 unlock:
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12070 mutex_unlock(&event->mmap_mutex);
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12071 if (output_event)
68e3c69803dada kernel/events/core.c Peter Zijlstra 2022-07-05 12072 mutex_unlock(&output_event->mmap_mutex);
ac9721f3f54b27 kernel/perf_event.c Peter Zijlstra 2010-05-27 12073
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12074 out:
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12075 return ret;
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12076 }
a4be7c2778d1fd kernel/perf_counter.c Peter Zijlstra 2009-08-19 12077
:::::: The code at line 12047 was first introduced by commit
:::::: 68e3c69803dada336893640110cb87221bb01dcf perf/core: Fix data race between perf_event_set_output() and perf_mmap_close()
:::::: TO: Peter Zijlstra <peterz@...radead.org>
:::::: CC: Peter Zijlstra <peterz@...radead.org>
--
0-DAY CI Kernel Test Service
https://01.org/lkp
Powered by blists - more mailing lists