Commit 7cc78f8fa02c2485104b86434acbc1538a3bd807

Authored by Andy Lutomirski
Committed by Linus Torvalds
1 parent ebcd241a63

context_tracking: Restore previous state in schedule_user

It appears that some SCHEDULE_USER (asm for schedule_user) callers
in arch/x86/kernel/entry_64.S are called from RCU kernel context,
and schedule_user will return in RCU user context.  This causes RCU
warnings and possible failures.

This is intended to be a minimal fix suitable for 3.18.

Reported-and-tested-by: Dave Jones <davej@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Showing 1 changed file with 6 additions and 2 deletions Side-by-side Diff

... ... @@ -2874,10 +2874,14 @@
2874 2874 * or we have been woken up remotely but the IPI has not yet arrived,
2875 2875 * we haven't yet exited the RCU idle mode. Do it here manually until
2876 2876 * we find a better solution.
  2877 + *
  2878 + * NB: There are buggy callers of this function. Ideally we
  2879 + * should warn if prev_state != IN_USER, but that will trigger
  2880 + * too frequently to make sense yet.
2877 2881 */
2878   - user_exit();
  2882 + enum ctx_state prev_state = exception_enter();
2879 2883 schedule();
2880   - user_enter();
  2884 + exception_exit(prev_state);
2881 2885 }
2882 2886 #endif
2883 2887