Re: [PATCH v2 4/5] ext4: improve cr 0 / cr 1 group scanning

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

[-- Attachment #1: Type: text/plain, Size: 11533 bytes --]

CC: kbuild-all(a)lists.01.org
In-Reply-To: <20210209202857.4185846-5-harshadshirwadkar@gmail.com>
References: <20210209202857.4185846-5-harshadshirwadkar@gmail.com>
TO: Harshad Shirwadkar <harshadshirwadkar@gmail.com>
TO: linux-ext4(a)vger.kernel.org
CC: tytso(a)mit.edu
CC: bzzz(a)whamcloud.com
CC: artem.blagodarenko(a)gmail.com
CC: sihara(a)ddn.com
CC: adilger(a)dilger.ca
CC: Harshad Shirwadkar <harshadshirwadkar@gmail.com>

Hi Harshad,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on ext4/dev]
[also build test WARNING on v5.11-rc7 next-20210125]
[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/Harshad-Shirwadkar/ext4-drop-s_mb_bal_lock-and-convert-protected-fields-to-atomic/20210210-054647
base:   https://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git dev
:::::: branch date: 9 hours ago
:::::: commit date: 9 hours ago
config: h8300-randconfig-s031-20210209 (attached as .config)
compiler: h8300-linux-gcc (GCC) 9.3.0
reproduce:
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # apt-get install sparse
        # sparse version: v0.6.3-215-g0fb77bb6-dirty
        # https://github.com/0day-ci/linux/commit/ef4eebad9c018a972a470b7b41e68bc981b31d00
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Harshad-Shirwadkar/ext4-drop-s_mb_bal_lock-and-convert-protected-fields-to-atomic/20210210-054647
        git checkout ef4eebad9c018a972a470b7b41e68bc981b31d00
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' ARCH=h8300 

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


"sparse warnings: (new ones prefixed by >>)"
>> fs/ext4/mballoc.c:946:9: sparse: sparse: context imbalance in 'ext4_mb_choose_next_group_cr1' - wrong count at exit
   fs/ext4/mballoc.c:1188:9: sparse: sparse: context imbalance in 'ext4_mb_init_cache' - different lock contexts for basic block
   fs/ext4/mballoc.c:2091:5: sparse: sparse: context imbalance in 'ext4_mb_try_best_found' - different lock contexts for basic block
   fs/ext4/mballoc.c:2119:5: sparse: sparse: context imbalance in 'ext4_mb_find_by_goal' - different lock contexts for basic block
   fs/ext4/mballoc.c:2407:12: sparse: sparse: context imbalance in 'ext4_mb_good_group_nolock' - wrong count at exit
   fs/ext4/mballoc.c:2619:43: sparse: sparse: context imbalance in 'ext4_mb_regular_allocator' - different lock contexts for basic block
   fs/ext4/mballoc.c:3277:17: sparse: sparse: context imbalance in 'ext4_mb_release' - different lock contexts for basic block
   fs/ext4/mballoc.c:3396:26: sparse: sparse: context imbalance in 'ext4_free_data_in_buddy' - wrong count at exit
   fs/ext4/mballoc.c:3612:15: sparse: sparse: context imbalance in 'ext4_mb_mark_diskspace_used' - different lock contexts for basic block
   fs/ext4/mballoc.c:3620:6: sparse: sparse: context imbalance in 'ext4_mb_mark_bb' - different lock contexts for basic block
   fs/ext4/mballoc.c:3943:13: sparse: sparse: context imbalance in 'ext4_discard_allocated_blocks' - different lock contexts for basic block
   fs/ext4/mballoc.c:4245:13: sparse: sparse: context imbalance in 'ext4_mb_put_pa' - different lock contexts for basic block
   fs/ext4/mballoc.c:4582:9: sparse: sparse: context imbalance in 'ext4_mb_discard_group_preallocations' - different lock contexts for basic block
   fs/ext4/mballoc.c:4735:9: sparse: sparse: context imbalance in 'ext4_discard_preallocations' - different lock contexts for basic block
   fs/ext4/mballoc.c:5030:9: sparse: sparse: context imbalance in 'ext4_mb_discard_lg_preallocations' - different lock contexts for basic block
   fs/ext4/mballoc.c:5675:9: sparse: sparse: context imbalance in 'ext4_free_blocks' - different lock contexts for basic block
   fs/ext4/mballoc.c:5975:15: sparse: sparse: context imbalance in 'ext4_group_add_blocks' - different lock contexts for basic block
   fs/ext4/mballoc.c: note: in included file (through include/linux/atomic.h, include/asm-generic/bitops/lock.h, arch/h8300/include/asm/bitops.h, ...):
   arch/h8300/include/asm/atomic.h:92:31: sparse: sparse: context imbalance in 'ext4_trim_extent' - wrong count at exit
   fs/ext4/mballoc.c:6039:1: sparse: sparse: context imbalance in 'ext4_trim_all_free' - different lock contexts for basic block
   fs/ext4/mballoc.c:6205:1: sparse: sparse: context imbalance in 'ext4_mballoc_query_range' - different lock contexts for basic block

vim +/ext4_mb_choose_next_group_cr1 +946 fs/ext4/mballoc.c

ef4eebad9c018a Harshad Shirwadkar 2021-02-09  871  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  872  /*
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  873   * Choose next group by traversing average fragment size tree. Return 0 if next
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  874   * group was selected optimally. Return 1 if next group could not selected
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  875   * optimally (due to lock contention). Updates *new_cr if cr lvel needs an
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  876   * update.
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  877   */
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  878  static int ext4_mb_choose_next_group_cr1(struct ext4_allocation_context *ac,
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  879  		int *new_cr, ext4_group_t *group, ext4_group_t ngroups)
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  880  {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  881  	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  882  	int avg_fragment_size, best_so_far;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  883  	struct rb_node *node, *found;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  884  	struct ext4_group_info *grp;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  885  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  886  	/*
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  887  	 * If there is contention on the lock, instead of waiting for the lock
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  888  	 * to become available, just continue searching lineraly. We'll resume
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  889  	 * our rb tree search later starting at ac->ac_last_optimal_group.
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  890  	 */
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  891  	if (!read_trylock(&sbi->s_mb_rb_lock))
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  892  		return 1;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  893  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  894  	if (ac->ac_flags & EXT4_MB_CR1_OPTIMIZED) {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  895  		/* We have found something at CR 1 in the past */
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  896  		grp = ext4_get_group_info(ac->ac_sb, ac->ac_last_optimal_group);
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  897  		for (found = rb_next(&grp->bb_avg_fragment_size_rb); found != NULL;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  898  		     found = rb_next(found)) {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  899  			grp = rb_entry(found, struct ext4_group_info,
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  900  				       bb_avg_fragment_size_rb);
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  901  			/*
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  902  			 * Perform this check without locking, we'll lock later
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  903  			 * to confirm.
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  904  			 */
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  905  			if (likely(ext4_mb_good_group(ac, grp->bb_group, 1)))
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  906  				break;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  907  		}
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  908  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  909  		goto done;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  910  	}
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  911  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  912  	node = sbi->s_mb_avg_fragment_size_root.rb_node;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  913  	best_so_far = 0;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  914  	found = NULL;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  915  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  916  	while (node) {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  917  		grp = rb_entry(node, struct ext4_group_info,
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  918  			       bb_avg_fragment_size_rb);
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  919  		/*
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  920  		 * Perform this check without locking, we'll lock later to confirm.
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  921  		 */
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  922  		if (ext4_mb_good_group(ac, grp->bb_group, 1)) {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  923  			avg_fragment_size = grp->bb_fragments ?
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  924  				grp->bb_free / grp->bb_fragments : 0;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  925  			if (!best_so_far || avg_fragment_size < best_so_far) {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  926  				best_so_far = avg_fragment_size;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  927  				found = node;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  928  			}
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  929  		}
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  930  		if (avg_fragment_size > ac->ac_g_ex.fe_len)
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  931  			node = node->rb_right;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  932  		else
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  933  			node = node->rb_left;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  934  	}
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  935  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  936  done:
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  937  	if (found) {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  938  		grp = rb_entry(found, struct ext4_group_info,
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  939  			       bb_avg_fragment_size_rb);
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  940  		*group = grp->bb_group;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  941  		ac->ac_flags |= EXT4_MB_CR1_OPTIMIZED;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  942  	} else {
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  943  		*new_cr = 2;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  944  	}
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  945  
ef4eebad9c018a Harshad Shirwadkar 2021-02-09 @946  	read_unlock(&sbi->s_mb_rb_lock);
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  947  	ac->ac_last_optimal_group = *group;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  948  	return 0;
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  949  }
ef4eebad9c018a Harshad Shirwadkar 2021-02-09  950  

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

[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 23128 bytes --]