drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_bwc.c:708:17: warning: large atomic operation may incur significant performance penalty; the access size (4 bytes) exceeds the max lock-free size (0 bytes)

drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_bwc.c:708:17: warning: large atomic operation may incur significant performance penalty; the access size (4 bytes) exceeds the max lock-free size (0 bytes)

[kernel test robot]

Hi Yevgeny,

First bad commit (maybe != root cause):

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head:   3efc57369a0ce8f76bf0804f7e673982384e4ac9
commit: 510f9f61a1121a296a45962760d5e2824277fa37 net/mlx5: HWS, added API and enabled HWS support
date:   3 weeks ago
config: arm-randconfig-r052-20240929 (https://download.01.org/0day-ci/archive/20240929/202409291101.6NdtMFVC-lkp@intel.com/config)
compiler: clang version 20.0.0git (https://github.com/llvm/llvm-project 7773243d9916f98ba0ffce0c3a960e4aa9f03e81)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240929/202409291101.6NdtMFVC-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/202409291101.6NdtMFVC-lkp@intel.com/

All warnings (new ones prefixed by >>):

   In file included from drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_bwc.c:4:
   In file included from drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_internal.h:7:
   In file included from include/linux/mlx5/transobj.h:36:
   In file included from include/linux/mlx5/driver.h:38:
   In file included from include/linux/pci.h:1646:
   In file included from include/linux/dmapool.h:14:
   In file included from include/linux/scatterlist.h:8:
   In file included from include/linux/mm.h:2232:
   include/linux/vmstat.h:517:36: warning: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Wenum-enum-conversion]
     517 |         return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
         |                               ~~~~~~~~~~~ ^ ~~~
>> drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_bwc.c:708:17: warning: large atomic operation may incur significant performance penalty; the access size (4 bytes) exceeds the max lock-free size (0  bytes) [-Watomic-alignment]
     708 |         num_of_rules = __atomic_load_n(&bwc_matcher->num_of_rules, __ATOMIC_RELAXED);
         |                        ^
   2 warnings generated.


vim +708 drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_bwc.c

2111bb970c787b Yevgeny Kliteynik 2024-06-20  690  
2111bb970c787b Yevgeny Kliteynik 2024-06-20  691  static int
2111bb970c787b Yevgeny Kliteynik 2024-06-20  692  hws_bwc_matcher_rehash_size(struct mlx5hws_bwc_matcher *bwc_matcher)
2111bb970c787b Yevgeny Kliteynik 2024-06-20  693  {
2111bb970c787b Yevgeny Kliteynik 2024-06-20  694  	u32 num_of_rules;
2111bb970c787b Yevgeny Kliteynik 2024-06-20  695  	int ret;
2111bb970c787b Yevgeny Kliteynik 2024-06-20  696  
2111bb970c787b Yevgeny Kliteynik 2024-06-20  697  	/* If the current matcher size is already at its max size, we can't
2111bb970c787b Yevgeny Kliteynik 2024-06-20  698  	 * do the rehash. Skip it and try adding the rule again - perhaps
2111bb970c787b Yevgeny Kliteynik 2024-06-20  699  	 * there was some change.
2111bb970c787b Yevgeny Kliteynik 2024-06-20  700  	 */
2111bb970c787b Yevgeny Kliteynik 2024-06-20  701  	if (hws_bwc_matcher_size_maxed_out(bwc_matcher))
2111bb970c787b Yevgeny Kliteynik 2024-06-20  702  		return 0;
2111bb970c787b Yevgeny Kliteynik 2024-06-20  703  
2111bb970c787b Yevgeny Kliteynik 2024-06-20  704  	/* It is possible that other rule has already performed rehash.
2111bb970c787b Yevgeny Kliteynik 2024-06-20  705  	 * Need to check again if we really need rehash.
2111bb970c787b Yevgeny Kliteynik 2024-06-20  706  	 * If the reason for rehash was size, but not any more - skip rehash.
2111bb970c787b Yevgeny Kliteynik 2024-06-20  707  	 */
2111bb970c787b Yevgeny Kliteynik 2024-06-20 @708  	num_of_rules = __atomic_load_n(&bwc_matcher->num_of_rules, __ATOMIC_RELAXED);
2111bb970c787b Yevgeny Kliteynik 2024-06-20  709  	if (!hws_bwc_matcher_rehash_size_needed(bwc_matcher, num_of_rules))
2111bb970c787b Yevgeny Kliteynik 2024-06-20  710  		return 0;
2111bb970c787b Yevgeny Kliteynik 2024-06-20  711  
2111bb970c787b Yevgeny Kliteynik 2024-06-20  712  	/* Now we're done all the checking - do the rehash:
2111bb970c787b Yevgeny Kliteynik 2024-06-20  713  	 *  - extend match RTC size
2111bb970c787b Yevgeny Kliteynik 2024-06-20  714  	 *  - create new matcher
2111bb970c787b Yevgeny Kliteynik 2024-06-20  715  	 *  - move all the rules to the new matcher
2111bb970c787b Yevgeny Kliteynik 2024-06-20  716  	 *  - destroy the old matcher
2111bb970c787b Yevgeny Kliteynik 2024-06-20  717  	 */
2111bb970c787b Yevgeny Kliteynik 2024-06-20  718  
2111bb970c787b Yevgeny Kliteynik 2024-06-20  719  	ret = hws_bwc_matcher_extend_size(bwc_matcher);
2111bb970c787b Yevgeny Kliteynik 2024-06-20  720  	if (ret)
2111bb970c787b Yevgeny Kliteynik 2024-06-20  721  		return ret;
2111bb970c787b Yevgeny Kliteynik 2024-06-20  722  
2111bb970c787b Yevgeny Kliteynik 2024-06-20  723  	return hws_bwc_matcher_move(bwc_matcher);
2111bb970c787b Yevgeny Kliteynik 2024-06-20  724  }
2111bb970c787b Yevgeny Kliteynik 2024-06-20  725  

:::::: The code at line 708 was first introduced by commit
:::::: 2111bb970c787b16b002dc726c1d296ce87a00fb net/mlx5: HWS, added backward-compatible API handling

:::::: TO: Yevgeny Kliteynik <kliteyn@nvidia.com>
:::::: CC: Saeed Mahameed <saeedm@nvidia.com>

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki