| 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: <202407260248.CBU1JMb1-lkp@intel.com>
Date: Fri, 26 Jul 2024 02:15:48 +0800
From: kernel test robot <lkp@...el.com>
To: Kinsey Ho <kinseyho@...gle.com>,
Andrew Morton <akpm@...ux-foundation.org>
Cc: oe-kbuild-all@...ts.linux.dev,
Linux Memory Management List <linux-mm@...ck.org>,
linux-kernel@...r.kernel.org, Yosry Ahmed <yosryahmed@...gle.com>,
Roman Gushchin <roman.gushchin@...ux.dev>,
Kinsey Ho <kinseyho@...gle.com>
Subject: Re: [PATCH mm-unstable v1 4/4] mm: clean up mem_cgroup_iter()
Hi Kinsey,
kernel test robot noticed the following build warnings:
[auto build test WARNING on akpm-mm/mm-everything]
[also build test WARNING on linus/master v6.10 next-20240725]
[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#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Kinsey-Ho/mm-don-t-hold-css-refcnt-during-traversal/20240725-030750
base: https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-everything
patch link: https://lore.kernel.org/r/20240724190214.1108049-5-kinseyho%40google.com
patch subject: [PATCH mm-unstable v1 4/4] mm: clean up mem_cgroup_iter()
config: x86_64-randconfig-121-20240725 (https://download.01.org/0day-ci/archive/20240726/202407260248.CBU1JMb1-lkp@intel.com/config)
compiler: gcc-7 (Ubuntu 7.5.0-6ubuntu2) 7.5.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240726/202407260248.CBU1JMb1-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@...el.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202407260248.CBU1JMb1-lkp@intel.com/
sparse warnings: (new ones prefixed by >>)
mm/memcontrol.c:1049:21: sparse: sparse: incorrect type in initializer (different address spaces) @@ expected struct mem_cgroup [noderef] __rcu *__old @@ got struct mem_cgroup *[assigned] pos @@
mm/memcontrol.c:1049:21: sparse: expected struct mem_cgroup [noderef] __rcu *__old
mm/memcontrol.c:1049:21: sparse: got struct mem_cgroup *[assigned] pos
>> mm/memcontrol.c:1049:21: sparse: sparse: incorrect type in initializer (different address spaces) @@ expected struct mem_cgroup [noderef] __rcu *__new @@ got struct mem_cgroup *[assigned] next @@
mm/memcontrol.c:1049:21: sparse: expected struct mem_cgroup [noderef] __rcu *__new
mm/memcontrol.c:1049:21: sparse: got struct mem_cgroup *[assigned] next
mm/memcontrol.c:1049:57: sparse: sparse: incompatible types in comparison expression (different address spaces):
mm/memcontrol.c:1049:57: sparse: struct mem_cgroup [noderef] __rcu *
mm/memcontrol.c:1049:57: sparse: struct mem_cgroup *
mm/memcontrol.c:1101:17: sparse: sparse: incorrect type in initializer (different address spaces) @@ expected struct mem_cgroup [noderef] __rcu *__old @@ got struct mem_cgroup *dead_memcg @@
mm/memcontrol.c:1101:17: sparse: expected struct mem_cgroup [noderef] __rcu *__old
mm/memcontrol.c:1101:17: sparse: got struct mem_cgroup *dead_memcg
mm/memcontrol.c: note: in included file (through include/linux/cgroup-defs.h):
include/linux/list.h:83:21: sparse: sparse: self-comparison always evaluates to true
mm/memcontrol.c: note: in included file:
include/linux/memcontrol.h:747:9: sparse: sparse: context imbalance in 'folio_lruvec_lock' - wrong count at exit
include/linux/memcontrol.h:747:9: sparse: sparse: context imbalance in 'folio_lruvec_lock_irq' - wrong count at exit
include/linux/memcontrol.h:747:9: sparse: sparse: context imbalance in 'folio_lruvec_lock_irqsave' - wrong count at exit
mm/memcontrol.c: note: in included file (through include/linux/cgroup-defs.h):
include/linux/list.h:83:21: sparse: sparse: self-comparison always evaluates to true
vim +1049 mm/memcontrol.c
974
975 /**
976 * mem_cgroup_iter - iterate over memory cgroup hierarchy
977 * @root: hierarchy root
978 * @prev: previously returned memcg, NULL on first invocation
979 * @reclaim: cookie for shared reclaim walks, NULL for full walks
980 *
981 * Returns references to children of the hierarchy below @root, or
982 * @root itself, or %NULL after a full round-trip.
983 *
984 * Caller must pass the return value in @prev on subsequent
985 * invocations for reference counting, or use mem_cgroup_iter_break()
986 * to cancel a hierarchy walk before the round-trip is complete.
987 *
988 * Reclaimers can specify a node in @reclaim to divide up the memcgs
989 * in the hierarchy among all concurrent reclaimers operating on the
990 * same node.
991 */
992 struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
993 struct mem_cgroup *prev,
994 struct mem_cgroup_reclaim_cookie *reclaim)
995 {
996 struct mem_cgroup_reclaim_iter *iter;
997 struct cgroup_subsys_state *css;
998 struct mem_cgroup *pos;
999 struct mem_cgroup *next = NULL;
1000
1001 if (mem_cgroup_disabled())
1002 return NULL;
1003
1004 if (!root)
1005 root = root_mem_cgroup;
1006
1007 rcu_read_lock();
1008 restart:
1009 if (reclaim) {
1010 int gen;
1011 int nid = reclaim->pgdat->node_id;
1012
1013 iter = &root->nodeinfo[nid]->iter;
1014 gen = atomic_read(&iter->generation);
1015
1016 /*
1017 * On start, join the current reclaim iteration cycle.
1018 * Exit when a concurrent walker completes it.
1019 */
1020 if (!prev)
1021 reclaim->generation = gen;
1022 else if (reclaim->generation != gen)
1023 goto out_unlock;
1024
1025 pos = rcu_dereference(iter->position);
1026 } else
1027 pos = prev;
1028
1029 css = pos ? &pos->css : NULL;
1030
1031 while ((css = css_next_descendant_pre(css, &root->css))) {
1032 /*
1033 * Verify the css and acquire a reference. The root
1034 * is provided by the caller, so we know it's alive
1035 * and kicking, and don't take an extra reference.
1036 */
1037 if (css == &root->css || css_tryget(css))
1038 break;
1039 }
1040
1041 next = mem_cgroup_from_css(css);
1042
1043 if (reclaim) {
1044 /*
1045 * The position could have already been updated by a competing
1046 * thread, so check that the value hasn't changed since we read
1047 * it to avoid reclaiming from the same cgroup twice.
1048 */
> 1049 if (cmpxchg(&iter->position, pos, next) != pos) {
1050 if (css && css != &root->css)
1051 css_put(css);
1052 goto restart;
1053 }
1054
1055 if (!next) {
1056 atomic_inc(&iter->generation);
1057
1058 /*
1059 * Reclaimers share the hierarchy walk, and a
1060 * new one might jump in right at the end of
1061 * the hierarchy - make sure they see at least
1062 * one group and restart from the beginning.
1063 */
1064 if (!prev)
1065 goto restart;
1066 }
1067 }
1068
1069 out_unlock:
1070 rcu_read_unlock();
1071 if (prev && prev != root)
1072 css_put(&prev->css);
1073
1074 return next;
1075 }
1076
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
Powered by blists - more mailing lists