3.18-stable patches

added patches:
	epoll-fix-race-between-ep_poll_callback-pollfree-and-ep_free-ep_remove.patch

diff --git a/queue-3.18/epoll-fix-race-between-ep_poll_callback-pollfree-and-ep_free-ep_remove.patch b/queue-3.18/epoll-fix-race-between-ep_poll_callback-pollfree-and-ep_free-ep_remove.patch
new file mode 100644 (file)
index 0000000..0a2e8d0
--- /dev/null
+++ b/queue-3.18/epoll-fix-race-between-ep_poll_callback-pollfree-and-ep_free-ep_remove.patch
@@ -0,0 +1,110 @@
+From 138e4ad67afd5c6c318b056b4d17c17f2c0ca5c0 Mon Sep 17 00:00:00 2001
+From: Oleg Nesterov <oleg@redhat.com>
+Date: Fri, 1 Sep 2017 18:55:33 +0200
+Subject: epoll: fix race between ep_poll_callback(POLLFREE) and ep_free()/ep_remove()
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+From: Oleg Nesterov <oleg@redhat.com>
+
+commit 138e4ad67afd5c6c318b056b4d17c17f2c0ca5c0 upstream.
+
+The race was introduced by me in commit 971316f0503a ("epoll:
+ep_unregister_pollwait() can use the freed pwq->whead").  I did not
+realize that nothing can protect eventpoll after ep_poll_callback() sets
+->whead = NULL, only whead->lock can save us from the race with
+ep_free() or ep_remove().
+
+Move ->whead = NULL to the end of ep_poll_callback() and add the
+necessary barriers.
+
+TODO: cleanup the ewake/EPOLLEXCLUSIVE logic, it was confusing even
+before this patch.
+
+Hopefully this explains use-after-free reported by syzcaller:
+
+       BUG: KASAN: use-after-free in debug_spin_lock_before
+       ...
+        _raw_spin_lock_irqsave+0x4a/0x60 kernel/locking/spinlock.c:159
+        ep_poll_callback+0x29f/0xff0 fs/eventpoll.c:1148
+
+this is spin_lock(eventpoll->lock),
+
+       ...
+       Freed by task 17774:
+       ...
+        kfree+0xe8/0x2c0 mm/slub.c:3883
+        ep_free+0x22c/0x2a0 fs/eventpoll.c:865
+
+Fixes: 971316f0503a ("epoll: ep_unregister_pollwait() can use the freed pwq->whead")
+Reported-by: 范龙飞 <long7573@126.com>
+Signed-off-by: Oleg Nesterov <oleg@redhat.com>
+Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+
+---
+ fs/eventpoll.c |   37 ++++++++++++++++++++++++-------------
+ 1 file changed, 24 insertions(+), 13 deletions(-)
+
+--- a/fs/eventpoll.c
++++ b/fs/eventpoll.c
+@@ -518,8 +518,13 @@ static void ep_remove_wait_queue(struct
+       wait_queue_head_t *whead;
+ 
+       rcu_read_lock();
+-      /* If it is cleared by POLLFREE, it should be rcu-safe */
+-      whead = rcu_dereference(pwq->whead);
++      /*
++       * If it is cleared by POLLFREE, it should be rcu-safe.
++       * If we read NULL we need a barrier paired with
++       * smp_store_release() in ep_poll_callback(), otherwise
++       * we rely on whead->lock.
++       */
++      whead = smp_load_acquire(&pwq->whead);
+       if (whead)
+               remove_wait_queue(whead, &pwq->wait);
+       rcu_read_unlock();
+@@ -1006,17 +1011,6 @@ static int ep_poll_callback(wait_queue_t
+       struct epitem *epi = ep_item_from_wait(wait);
+       struct eventpoll *ep = epi->ep;
+ 
+-      if ((unsigned long)key & POLLFREE) {
+-              ep_pwq_from_wait(wait)->whead = NULL;
+-              /*
+-               * whead = NULL above can race with ep_remove_wait_queue()
+-               * which can do another remove_wait_queue() after us, so we
+-               * can't use __remove_wait_queue(). whead->lock is held by
+-               * the caller.
+-               */
+-              list_del_init(&wait->task_list);
+-      }
+-
+       spin_lock_irqsave(&ep->lock, flags);
+ 
+       /*
+@@ -1081,6 +1075,23 @@ out_unlock:
+       if (pwake)
+               ep_poll_safewake(&ep->poll_wait);
+ 
++
++      if ((unsigned long)key & POLLFREE) {
++              /*
++               * If we race with ep_remove_wait_queue() it can miss
++               * ->whead = NULL and do another remove_wait_queue() after
++               * us, so we can't use __remove_wait_queue().
++               */
++              list_del_init(&wait->task_list);
++              /*
++               * ->whead != NULL protects us from the race with ep_free()
++               * or ep_remove(), ep_remove_wait_queue() takes whead->lock
++               * held by the caller. Once we nullify it, nothing protects
++               * ep/epi or even wait.
++               */
++              smp_store_release(&ep_pwq_from_wait(wait)->whead, NULL);
++      }
++
+       return 1;
+ }
+ 

diff --git a/queue-3.18/series b/queue-3.18/series
index 11358e5ee69a2404ae22a8940bceecdfbe0fab6d..f048fd7c281e92faee2238e75ff119146d7c2574 100644 (file)
--- a/queue-3.18/series
+++ b/queue-3.18/series
@@ -6,3 +6,4 @@ cifs-remove-endian-related-sparse-warning.patch
 wl1251-add-a-missing-spin_lock_init.patch
 xfrm-policy-check-policy-direction-value.patch
 drm-ttm-fix-accounting-error-when-fail-to-get-pages-for-pool.patch
+epoll-fix-race-between-ep_poll_callback-pollfree-and-ep_free-ep_remove.patch