timerfd: Protect the might cancel mechanism proper

The handling of the might_cancel queueing is not properly protected, so
parallel operations on the file descriptor can race with each other and
lead to list corruptions or use after free.

Protect the context for these operations with a seperate lock.

The wait queue lock cannot be reused for this because that would create a
lock inversion scenario vs. the cancel lock. Replacing might_cancel with an
atomic (atomic_t or atomic bit) does not help either because it still can
race vs. the actual list operation.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: "linux-fsdevel@vger.kernel.org"
Cc: syzkaller <syzkaller@googlegroups.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1701311521430.3457@nanos
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
(cherry picked from commit 1e38da300e1e395a15048b0af1e5305bd91402f6)

Orabug: 26673877
CVE: CVE-2017-10661

Signed-off-by: Tim Tianyang Chen <tianyang.chen@oracle.com>
Reviewed-by: Jack Vogel <jack.vogel@oracle.com>

diff --git a/fs/timerfd.c b/fs/timerfd.c
index b94fa6c3c6ebe8c7308aea0052f3afacad9745c4..4d2c8aa1fe909942d6696ddcadc916259c8e9af2 100644 (file)
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -41,6 +41,9 @@ struct timerfd_ctx {
        struct rcu_head rcu;
        struct list_head clist;
        bool might_cancel;
+#ifndef __GENKSYMS__
+       spinlock_t cancel_lock;
+#endif
 };
 
 static LIST_HEAD(cancel_list);
@@ -112,7 +115,7 @@ void timerfd_clock_was_set(void)
        rcu_read_unlock();
 }
 
-static void timerfd_remove_cancel(struct timerfd_ctx *ctx)
+static void __timerfd_remove_cancel(struct timerfd_ctx *ctx)
 {
        if (ctx->might_cancel) {
                ctx->might_cancel = false;
@@ -122,6 +125,13 @@ static void timerfd_remove_cancel(struct timerfd_ctx *ctx)
        }
 }
 
+static void timerfd_remove_cancel(struct timerfd_ctx *ctx)
+{
+       spin_lock(&ctx->cancel_lock);
+       __timerfd_remove_cancel(ctx);
+       spin_unlock(&ctx->cancel_lock);
+}
+
 static bool timerfd_canceled(struct timerfd_ctx *ctx)
 {
        if (!ctx->might_cancel || ctx->moffs.tv64 != KTIME_MAX)
@@ -132,6 +142,7 @@ static bool timerfd_canceled(struct timerfd_ctx *ctx)
 
 static void timerfd_setup_cancel(struct timerfd_ctx *ctx, int flags)
 {
+       spin_lock(&ctx->cancel_lock);
        if ((ctx->clockid == CLOCK_REALTIME ||
             ctx->clockid == CLOCK_REALTIME_ALARM) &&
            (flags & TFD_TIMER_ABSTIME) && (flags & TFD_TIMER_CANCEL_ON_SET)) {
@@ -141,9 +152,10 @@ static void timerfd_setup_cancel(struct timerfd_ctx *ctx, int flags)
                        list_add_rcu(&ctx->clist, &cancel_list);
                        spin_unlock(&cancel_lock);
                }
-       } else if (ctx->might_cancel) {
-               timerfd_remove_cancel(ctx);
+       } else {
+               __timerfd_remove_cancel(ctx);
        }
+       spin_unlock(&ctx->cancel_lock);
 }
 
 static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx)
@@ -395,6 +407,7 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags)
                return -ENOMEM;
 
        init_waitqueue_head(&ctx->wqh);
+       spin_lock_init(&ctx->cancel_lock);
        ctx->clockid = clockid;
 
        if (isalarm(ctx))