Re: [PATCH v2 1/2] mm: disable LRU pagevec during the migration temporarily

[kernel test robot <lkp@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 6511 bytes --]

Hi Minchan,

I love your patch! Perhaps something to improve:

[auto build test WARNING on linux/master]
[also build test WARNING on linus/master v5.12-rc2 next-20210309]
[cannot apply to hnaz-linux-mm/master]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Minchan-Kim/mm-disable-LRU-pagevec-during-the-migration-temporarily/20210309-131826
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 144c79ef33536b4ecb4951e07dbc1f2b7fa99d32
config: openrisc-randconfig-r026-20210308 (attached as .config)
compiler: or1k-linux-gcc (GCC) 9.3.0
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://github.com/0day-ci/linux/commit/e746db1a2ab13441890fa2cad8604bbec190b401
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Minchan-Kim/mm-disable-LRU-pagevec-during-the-migration-temporarily/20210309-131826
        git checkout e746db1a2ab13441890fa2cad8604bbec190b401
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross ARCH=openrisc 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All warnings (new ones prefixed by >>):

   mm/swap.c: In function 'pagevec_add_and_need_flush':
   mm/swap.c:244:4: error: implicit declaration of function 'lru_cache_disabled'; did you mean 'lru_cache_disable'? [-Werror=implicit-function-declaration]
     244 |    lru_cache_disabled())
         |    ^~~~~~~~~~~~~~~~~~
         |    lru_cache_disable
   mm/swap.c: At top level:
>> mm/swap.c:743:6: warning: no previous prototype for '__lru_add_drain_all' [-Wmissing-prototypes]
     743 | void __lru_add_drain_all(bool force_all_cpus)
         |      ^~~~~~~~~~~~~~~~~~~
   mm/swap.c:858:6: warning: no previous prototype for 'lru_cache_disabled' [-Wmissing-prototypes]
     858 | bool lru_cache_disabled(void)
         |      ^~~~~~~~~~~~~~~~~~
   mm/swap.c:858:6: error: conflicting types for 'lru_cache_disabled'
   mm/swap.c:244:4: note: previous implicit declaration of 'lru_cache_disabled' was here
     244 |    lru_cache_disabled())
         |    ^~~~~~~~~~~~~~~~~~
   cc1: some warnings being treated as errors


vim +/__lru_add_drain_all +743 mm/swap.c

   742	
 > 743	void __lru_add_drain_all(bool force_all_cpus)
   744	{
   745		/*
   746		 * lru_drain_gen - Global pages generation number
   747		 *
   748		 * (A) Definition: global lru_drain_gen = x implies that all generations
   749		 *     0 < n <= x are already *scheduled* for draining.
   750		 *
   751		 * This is an optimization for the highly-contended use case where a
   752		 * user space workload keeps constantly generating a flow of pages for
   753		 * each CPU.
   754		 */
   755		static unsigned int lru_drain_gen;
   756		static struct cpumask has_work;
   757		static DEFINE_MUTEX(lock);
   758		unsigned cpu, this_gen;
   759	
   760		/*
   761		 * Make sure nobody triggers this path before mm_percpu_wq is fully
   762		 * initialized.
   763		 */
   764		if (WARN_ON(!mm_percpu_wq))
   765			return;
   766	
   767		/*
   768		 * Guarantee pagevec counter stores visible by this CPU are visible to
   769		 * other CPUs before loading the current drain generation.
   770		 */
   771		smp_mb();
   772	
   773		/*
   774		 * (B) Locally cache global LRU draining generation number
   775		 *
   776		 * The read barrier ensures that the counter is loaded before the mutex
   777		 * is taken. It pairs with smp_mb() inside the mutex critical section
   778		 * at (D).
   779		 */
   780		this_gen = smp_load_acquire(&lru_drain_gen);
   781	
   782		mutex_lock(&lock);
   783	
   784		/*
   785		 * (C) Exit the draining operation if a newer generation, from another
   786		 * lru_add_drain_all(), was already scheduled for draining. Check (A).
   787		 */
   788		if (unlikely(this_gen != lru_drain_gen && !force_all_cpus))
   789			goto done;
   790	
   791		/*
   792		 * (D) Increment global generation number
   793		 *
   794		 * Pairs with smp_load_acquire() at (B), outside of the critical
   795		 * section. Use a full memory barrier to guarantee that the new global
   796		 * drain generation number is stored before loading pagevec counters.
   797		 *
   798		 * This pairing must be done here, before the for_each_online_cpu loop
   799		 * below which drains the page vectors.
   800		 *
   801		 * Let x, y, and z represent some system CPU numbers, where x < y < z.
   802		 * Assume CPU #z is is in the middle of the for_each_online_cpu loop
   803		 * below and has already reached CPU #y's per-cpu data. CPU #x comes
   804		 * along, adds some pages to its per-cpu vectors, then calls
   805		 * lru_add_drain_all().
   806		 *
   807		 * If the paired barrier is done at any later step, e.g. after the
   808		 * loop, CPU #x will just exit@(C) and miss flushing out all of its
   809		 * added pages.
   810		 */
   811		WRITE_ONCE(lru_drain_gen, lru_drain_gen + 1);
   812		smp_mb();
   813	
   814		cpumask_clear(&has_work);
   815		for_each_online_cpu(cpu) {
   816			struct work_struct *work = &per_cpu(lru_add_drain_work, cpu);
   817	
   818			if (force_all_cpus ||
   819			    pagevec_count(&per_cpu(lru_pvecs.lru_add, cpu)) ||
   820			    data_race(pagevec_count(&per_cpu(lru_rotate.pvec, cpu))) ||
   821			    pagevec_count(&per_cpu(lru_pvecs.lru_deactivate_file, cpu)) ||
   822			    pagevec_count(&per_cpu(lru_pvecs.lru_deactivate, cpu)) ||
   823			    pagevec_count(&per_cpu(lru_pvecs.lru_lazyfree, cpu)) ||
   824			    need_activate_page_drain(cpu)) {
   825				INIT_WORK(work, lru_add_drain_per_cpu);
   826				queue_work_on(cpu, mm_percpu_wq, work);
   827				__cpumask_set_cpu(cpu, &has_work);
   828			}
   829		}
   830	
   831		for_each_cpu(cpu, &has_work)
   832			flush_work(&per_cpu(lru_add_drain_work, cpu));
   833	
   834	done:
   835		mutex_unlock(&lock);
   836	}
   837	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 24758 bytes --]