aboutsummaryrefslogtreecommitdiffstats
path: root/kernel
diff options
context:
space:
mode:
authorDavid Rientjes <rientjes@google.com>2011-12-19 20:11:52 -0500
committerLinus Torvalds <torvalds@linux-foundation.org>2011-12-20 13:25:04 -0500
commitb246272ecc5ac68c743b15c9e41a2275f7ce70e2 (patch)
treed32b30f02501a6978c7c2d97f7c505604f9bac28 /kernel
parent511585a28e5b5fd1cac61e601e42efc4c5dd64b5 (diff)
cpusets: stall when updating mems_allowed for mempolicy or disjoint nodemask
Kernels where MAX_NUMNODES > BITS_PER_LONG may temporarily see an empty nodemask in a tsk's mempolicy if its previous nodemask is remapped onto a new set of allowed cpuset nodes where the two nodemasks, as a result of the remap, are now disjoint. c0ff7453bb5c ("cpuset,mm: fix no node to alloc memory when changing cpuset's mems") adds get_mems_allowed() to prevent the set of allowed nodes from changing for a thread. This causes any update to a set of allowed nodes to stall until put_mems_allowed() is called. This stall is unncessary, however, if at least one node remains unchanged in the update to the set of allowed nodes. This was addressed by 89e8a244b97e ("cpusets: avoid looping when storing to mems_allowed if one node remains set"), but it's still possible that an empty nodemask may be read from a mempolicy because the old nodemask may be remapped to the new nodemask during rebind. To prevent this, only avoid the stall if there is no mempolicy for the thread being changed. This is a temporary solution until all reads from mempolicy nodemasks can be guaranteed to not be empty without the get_mems_allowed() synchronization. Also moves the check for nodemask intersection inside task_lock() so that tsk->mems_allowed cannot change. This ensures that nothing can set this tsk's mems_allowed out from under us and also protects tsk->mempolicy. Reported-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: David Rientjes <rientjes@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Paul Menage <paul@paulmenage.org> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'kernel')
-rw-r--r--kernel/cpuset.c29
1 files changed, 24 insertions, 5 deletions
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 9fe58c46a426..0b1712dba587 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -123,6 +123,19 @@ static inline struct cpuset *task_cs(struct task_struct *task)
123 struct cpuset, css); 123 struct cpuset, css);
124} 124}
125 125
126#ifdef CONFIG_NUMA
127static inline bool task_has_mempolicy(struct task_struct *task)
128{
129 return task->mempolicy;
130}
131#else
132static inline bool task_has_mempolicy(struct task_struct *task)
133{
134 return false;
135}
136#endif
137
138
126/* bits in struct cpuset flags field */ 139/* bits in struct cpuset flags field */
127typedef enum { 140typedef enum {
128 CS_CPU_EXCLUSIVE, 141 CS_CPU_EXCLUSIVE,
@@ -949,7 +962,7 @@ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from,
949static void cpuset_change_task_nodemask(struct task_struct *tsk, 962static void cpuset_change_task_nodemask(struct task_struct *tsk,
950 nodemask_t *newmems) 963 nodemask_t *newmems)
951{ 964{
952 bool masks_disjoint = !nodes_intersects(*newmems, tsk->mems_allowed); 965 bool need_loop;
953 966
954repeat: 967repeat:
955 /* 968 /*
@@ -962,6 +975,14 @@ repeat:
962 return; 975 return;
963 976
964 task_lock(tsk); 977 task_lock(tsk);
978 /*
979 * Determine if a loop is necessary if another thread is doing
980 * get_mems_allowed(). If at least one node remains unchanged and
981 * tsk does not have a mempolicy, then an empty nodemask will not be
982 * possible when mems_allowed is larger than a word.
983 */
984 need_loop = task_has_mempolicy(tsk) ||
985 !nodes_intersects(*newmems, tsk->mems_allowed);
965 nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); 986 nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems);
966 mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1); 987 mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1);
967 988
@@ -981,11 +1002,9 @@ repeat:
981 1002
982 /* 1003 /*
983 * Allocation of memory is very fast, we needn't sleep when waiting 1004 * Allocation of memory is very fast, we needn't sleep when waiting
984 * for the read-side. No wait is necessary, however, if at least one 1005 * for the read-side.
985 * node remains unchanged.
986 */ 1006 */
987 while (masks_disjoint && 1007 while (need_loop && ACCESS_ONCE(tsk->mems_allowed_change_disable)) {
988 ACCESS_ONCE(tsk->mems_allowed_change_disable)) {
989 task_unlock(tsk); 1008 task_unlock(tsk);
990 if (!task_curr(tsk)) 1009 if (!task_curr(tsk))
991 yield(); 1010 yield();