Skip to content

Commit

Permalink
userfaultfd: avoid missing wakeups during refile in userfaultfd_read
Browse files Browse the repository at this point in the history
During the refile in userfaultfd_read both waitqueues could look empty to
the lockless wake_userfault().  Use a seqcount to prevent this false
negative that could leave an userfault blocked.

Signed-off-by: Andrea Arcangeli <[email protected]>
Cc: Pavel Emelyanov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Linus Torvalds <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
  • Loading branch information
aagit authored and torvalds committed Sep 4, 2015
1 parent 230c92a commit 2c5b7e1
Showing 1 changed file with 24 additions and 2 deletions.
26 changes: 24 additions & 2 deletions fs/userfaultfd.c
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,8 @@ struct userfaultfd_ctx {
wait_queue_head_t fault_wqh;
/* waitqueue head for the pseudo fd to wakeup poll/read */
wait_queue_head_t fd_wqh;
/* a refile sequence protected by fault_pending_wqh lock */
struct seqcount refile_seq;
/* pseudo fd refcounting */
atomic_t refcount;
/* userfaultfd syscall flags */
Expand Down Expand Up @@ -546,6 +548,15 @@ static ssize_t userfaultfd_ctx_read(struct userfaultfd_ctx *ctx, int no_wait,
spin_lock(&ctx->fault_pending_wqh.lock);
uwq = find_userfault(ctx);
if (uwq) {
/*
* Use a seqcount to repeat the lockless check
* in wake_userfault() to avoid missing
* wakeups because during the refile both
* waitqueue could become empty if this is the
* only userfault.
*/
write_seqcount_begin(&ctx->refile_seq);

/*
* The fault_pending_wqh.lock prevents the uwq
* to disappear from under us.
Expand All @@ -570,6 +581,8 @@ static ssize_t userfaultfd_ctx_read(struct userfaultfd_ctx *ctx, int no_wait,
list_del(&uwq->wq.task_list);
__add_wait_queue(&ctx->fault_wqh, &uwq->wq);

write_seqcount_end(&ctx->refile_seq);

/* careful to always initialize msg if ret == 0 */
*msg = uwq->msg;
spin_unlock(&ctx->fault_pending_wqh.lock);
Expand Down Expand Up @@ -647,6 +660,9 @@ static void __wake_userfault(struct userfaultfd_ctx *ctx,
static __always_inline void wake_userfault(struct userfaultfd_ctx *ctx,
struct userfaultfd_wake_range *range)
{
unsigned seq;
bool need_wakeup;

/*
* To be sure waitqueue_active() is not reordered by the CPU
* before the pagetable update, use an explicit SMP memory
Expand All @@ -662,8 +678,13 @@ static __always_inline void wake_userfault(struct userfaultfd_ctx *ctx,
* userfaults yet. So we take the spinlock only when we're
* sure we've userfaults to wake.
*/
if (waitqueue_active(&ctx->fault_pending_wqh) ||
waitqueue_active(&ctx->fault_wqh))
do {
seq = read_seqcount_begin(&ctx->refile_seq);
need_wakeup = waitqueue_active(&ctx->fault_pending_wqh) ||
waitqueue_active(&ctx->fault_wqh);
cond_resched();
} while (read_seqcount_retry(&ctx->refile_seq, seq));
if (need_wakeup)
__wake_userfault(ctx, range);
}

Expand Down Expand Up @@ -1219,6 +1240,7 @@ static void init_once_userfaultfd_ctx(void *mem)
init_waitqueue_head(&ctx->fault_pending_wqh);
init_waitqueue_head(&ctx->fault_wqh);
init_waitqueue_head(&ctx->fd_wqh);
seqcount_init(&ctx->refile_seq);
}

/**
Expand Down

0 comments on commit 2c5b7e1

Please sign in to comment.