[PATCH] sched/core: Don't use dying mm as active_mm for kernel threads

[PATCH] sched/core: Don't use dying mm as active_mm for kernel threads

[Waiman Long]

It was found that a dying mm_struct where the owning task has exited can
stay on as active_mm of kernel threads as long as no other user tasks
run on those CPUs that use it as active_mm. This prolongs the life time
of dying mm holding up memory and other resources that cannot be freed.

Fix that by forcing the kernel threads to use init_mm as the active_mm
if the previous active_mm is dying.

Signed-off-by: Waiman Long <longman@redhat.com>
---
 kernel/sched/core.c | 13 +++++++++++--
 mm/init-mm.c        |  2 ++
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 2b037f195473..ca348e1f5a1e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3233,13 +3233,22 @@ context_switch(struct rq *rq, struct task_struct *prev,
 	 * Both of these contain the full memory barrier required by
 	 * membarrier after storing to rq->curr, before returning to
 	 * user-space.
+	 *
+	 * If mm is NULL and oldmm is dying (!owner), we switch to
+	 * init_mm instead to make sure that oldmm can be freed ASAP.
 	 */
-	if (!mm) {
+	if (!mm && oldmm->owner) {
 		next->active_mm = oldmm;
 		mmgrab(oldmm);
 		enter_lazy_tlb(oldmm, next);
-	} else
+	} else {
+		if (!mm) {
+			mm = &init_mm;
+			next->active_mm = mm;
+			mmgrab(mm);
+		}
 		switch_mm_irqs_off(oldmm, mm, next);
+	}
 
 	if (!prev->mm) {
 		prev->active_mm = NULL;
diff --git a/mm/init-mm.c b/mm/init-mm.c
index a787a319211e..5bfc6bc333ca 100644
--- a/mm/init-mm.c
+++ b/mm/init-mm.c
@@ -5,6 +5,7 @@
 #include <linux/spinlock.h>
 #include <linux/list.h>
 #include <linux/cpumask.h>
+#include <linux/sched/task.h>
 
 #include <linux/atomic.h>
 #include <linux/user_namespace.h>
@@ -36,5 +37,6 @@ struct mm_struct init_mm = {
 	.mmlist		= LIST_HEAD_INIT(init_mm.mmlist),
 	.user_ns	= &init_user_ns,
 	.cpu_bitmap	= { [BITS_TO_LONGS(NR_CPUS)] = 0},
+	.owner		= &init_task,
 	INIT_MM_CONTEXT(init_mm)
 };
-- 
2.18.1

Re: [PATCH] sched/core: Don't use dying mm as active_mm for kernel threads

[Waiman Long]

On 7/26/19 7:45 PM, Waiman Long wrote:
> It was found that a dying mm_struct where the owning task has exited can
> stay on as active_mm of kernel threads as long as no other user tasks
> run on those CPUs that use it as active_mm. This prolongs the life time
> of dying mm holding up memory and other resources that cannot be freed.
>
> Fix that by forcing the kernel threads to use init_mm as the active_mm
> if the previous active_mm is dying.
>
> Signed-off-by: Waiman Long <longman@redhat.com>
> ---
>  kernel/sched/core.c | 13 +++++++++++--
>  mm/init-mm.c        |  2 ++
>  2 files changed, 13 insertions(+), 2 deletions(-)


Sorry, I didn't realize that mm->owner depends on CONFIG_MEMCG. I will
need to refresh the patch and send out v2 when I am done testing.

Cheers,
Longman

Re: [PATCH] sched/core: Don't use dying mm as active_mm for kernel threads

[kbuild test robot]

[-- Attachment #1: Type: text/plain, Size: 1158 bytes --]

Hi Waiman,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on linus/master]
[cannot apply to v5.3-rc1 next-20190726]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Waiman-Long/sched-core-Don-t-use-dying-mm-as-active_mm-for-kernel-threads/20190728-101948
config: x86_64-randconfig-s1-07280001 (attached as .config)
compiler: gcc-4.9 (Debian 4.9.2-10+deb8u1) 4.9.2
reproduce:
        # save the attached .config to linux build tree
        make ARCH=x86_64 

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

All errors (new ones prefixed by >>):

>> mm/init-mm.c:40:2: error: unknown field 'owner' specified in initializer
     .owner  = &init_task,
     ^
   mm/init-mm.c:40:2: warning: excess elements in struct initializer
   mm/init-mm.c:40:2: warning: (near initialization for 'init_mm')

vim +/owner +40 mm/init-mm.c

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 35017 bytes --]

Re: [PATCH] sched/core: Don't use dying mm as active_mm for kernel threads

[kbuild test robot]

[-- Attachment #1: Type: text/plain, Size: 1085 bytes --]

Hi Waiman,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on linus/master]
[cannot apply to v5.3-rc1 next-20190726]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Waiman-Long/sched-core-Don-t-use-dying-mm-as-active_mm-for-kernel-threads/20190728-101948
config: i386-allnoconfig (attached as .config)
compiler: gcc-7 (Debian 7.4.0-10) 7.4.0
reproduce:
        # save the attached .config to linux build tree
        make ARCH=i386 

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

All errors (new ones prefixed by >>):

   kernel/sched/core.c: In function 'context_switch':
>> kernel/sched/core.c:3240:18: error: 'struct mm_struct' has no member named 'owner'
     if (!mm && oldmm->owner) {
                     ^~

vim +3240 kernel/sched/core.c

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 7282 bytes --]