stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Matthew Wilcox <willy@linux.intel.com>,
	Hugh Dickins <hughd@google.com>, Ohad Ben-Cohen <ohad@wizery.com>,
	Konstantin Khlebnikov <khlebnikov@openvz.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: [PATCH 3.10 48/54] radix-tree: fix race in gang lookup
Date: Tue, 23 Feb 2016 19:34:05 -0800	[thread overview]
Message-ID: <20160224033354.266887421@linuxfoundation.org> (raw)
In-Reply-To: <20160224033352.613205610@linuxfoundation.org>

3.10-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Matthew Wilcox <willy@linux.intel.com>

commit 46437f9a554fbe3e110580ca08ab703b59f2f95a upstream.

If the indirect_ptr bit is set on a slot, that indicates we need to redo
the lookup.  Introduce a new function radix_tree_iter_retry() which
forces the loop to retry the lookup by setting 'slot' to NULL and
turning the iterator back to point at the problematic entry.

This is a pretty rare problem to hit at the moment; the lookup has to
race with a grow of the radix tree from a height of 0.  The consequences
of hitting this race are that gang lookup could return a pointer to a
radix_tree_node instead of a pointer to whatever the user had inserted
in the tree.

Fixes: cebbd29e1c2f ("radix-tree: rewrite gang lookup using iterator")
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ohad Ben-Cohen <ohad@wizery.com>
Cc: Konstantin Khlebnikov <khlebnikov@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 include/linux/radix-tree.h |   16 ++++++++++++++++
 lib/radix-tree.c           |   12 ++++++++++--
 2 files changed, 26 insertions(+), 2 deletions(-)

--- a/include/linux/radix-tree.h
+++ b/include/linux/radix-tree.h
@@ -322,6 +322,22 @@ void **radix_tree_next_chunk(struct radi
 			     struct radix_tree_iter *iter, unsigned flags);
 
 /**
+ * radix_tree_iter_retry - retry this chunk of the iteration
+ * @iter:	iterator state
+ *
+ * If we iterate over a tree protected only by the RCU lock, a race
+ * against deletion or creation may result in seeing a slot for which
+ * radix_tree_deref_retry() returns true.  If so, call this function
+ * and continue the iteration.
+ */
+static inline __must_check
+void **radix_tree_iter_retry(struct radix_tree_iter *iter)
+{
+	iter->next_index = iter->index;
+	return NULL;
+}
+
+/**
  * radix_tree_chunk_size - get current chunk size
  *
  * @iter:	pointer to radix tree iterator
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -1015,9 +1015,13 @@ radix_tree_gang_lookup(struct radix_tree
 		return 0;
 
 	radix_tree_for_each_slot(slot, root, &iter, first_index) {
-		results[ret] = indirect_to_ptr(rcu_dereference_raw(*slot));
+		results[ret] = rcu_dereference_raw(*slot);
 		if (!results[ret])
 			continue;
+		if (radix_tree_is_indirect_ptr(results[ret])) {
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
 		if (++ret == max_items)
 			break;
 	}
@@ -1094,9 +1098,13 @@ radix_tree_gang_lookup_tag(struct radix_
 		return 0;
 
 	radix_tree_for_each_tagged(slot, root, &iter, first_index, tag) {
-		results[ret] = indirect_to_ptr(rcu_dereference_raw(*slot));
+		results[ret] = rcu_dereference_raw(*slot);
 		if (!results[ret])
 			continue;
+		if (radix_tree_is_indirect_ptr(results[ret])) {
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
 		if (++ret == max_items)
 			break;
 	}



  parent reply	other threads:[~2016-02-24  3:54 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-24  3:33 [PATCH 3.10 00/54] 3.10.98-stable review Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 01/54] ALSA: seq: Fix double port list deletion Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 02/54] wan/x25: Fix use-after-free in x25_asy_open_tty() Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 03/54] staging/speakup: Use tty_ldisc_ref() for paste kworker Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 04/54] pty: fix possible use after free of tty->driver_data Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 05/54] pty: make sure super_block is still valid in final /dev/tty close Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 06/54] AIO: properly check iovec sizes Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 07/54] ext4: fix potential integer overflow Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 08/54] Btrfs: fix hang on extent buffer lock caused by the inode_paths ioctl Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 10/54] ptrace: use fsuid, fsgid, effective creds for fs access checks Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 11/54] tools lib traceevent: Fix output of %llu for 64 bit values read on 32 bit machines Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 12/54] tracing: Fix freak link error caused by branch tracer Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 13/54] klist: fix starting point removed bug in klist iterators Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 14/54] scsi: restart list search after unlock in scsi_remove_target Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 15/54] scsi_sysfs: Fix queue_ramp_up_period return code Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 16/54] iscsi-target: Fix rx_login_comp hang after login failure Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 17/54] Fix a memory leak in scsi_host_dev_release() Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 18/54] SCSI: Fix NULL pointer dereference in runtime PM Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 19/54] iscsi-target: Fix potential dead-lock during node acl delete Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 20/54] SCSI: fix crashes in sd and sr runtime PM Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 21/54] drivers/scsi/sg.c: mark VMA as VM_IO to prevent migration Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 22/54] scsi_dh_rdac: always retry MODE SELECT on command lock violation Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 23/54] scsi: fix soft lockup in scsi_remove_target() on module removal Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 24/54] iio:ad7793: Fix ad7785 product ID Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 25/54] iio: lpc32xx_adc: fix warnings caused by enabling unprepared clock Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 26/54] iio:ad5064: Make sure ad5064_i2c_write() returns 0 on success Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 27/54] iio: adis_buffer: Fix out-of-bounds memory access Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 28/54] iio: dac: mcp4725: set iio name property in sysfs Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 29/54] cifs: fix erroneous return value Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 30/54] nfs: Fix race in __update_open_stateid() Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 31/54] udf: limit the maximum number of indirect extents in a row Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 32/54] udf: Prevent buffer overrun with multi-byte characters Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 33/54] udf: Check output buffer length when converting name to CS0 Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 34/54] ARM: 8519/1: ICST: try other dividends than 1 Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 35/54] ARM: 8517/1: ICST: avoid arithmetic overflow in icst_hz() Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 36/54] fuse: break infinite loop in fuse_fill_write_pages() Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 37/54] mm: soft-offline: check return value in second __get_any_page() call Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 38/54] Input: elantech - add Fujitsu Lifebook U745 to force crc_enabled Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 39/54] Input: elantech - mark protocols v2 and v3 as semi-mt Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 41/54] iommu/vt-d: Fix 64-bit accesses to 32-bit DMAR_GSTS_REG Greg Kroah-Hartman
2016-02-24  3:33 ` [PATCH 3.10 42/54] mm/memory_hotplug.c: check for missing sections in test_pages_in_a_zone() Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 43/54] xhci: Fix list corruption in urb dequeue at host removal Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 44/54] m32r: fix m32104ut_defconfig build fail Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 45/54] dma-debug: switch check from _text to _stext Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 46/54] scripts/bloat-o-meter: fix python3 syntax error Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 47/54] memcg: only free spare array when readers are done Greg Kroah-Hartman
2016-02-24  3:34 ` Greg Kroah-Hartman [this message]
2016-02-24  3:34 ` [PATCH 3.10 49/54] radix-tree: fix oops after radix_tree_iter_retry Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 50/54] intel_scu_ipcutil: underflow in scu_reg_access() Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 51/54] x86/asm/irq: Stop relying on magic JMP behavior for early_idt_handlers Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 52/54] futex: Drop refcount if requeue_pi() acquired the rtmutex Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 53/54] ip6mr: call del_timer_sync() in ip6mr_free_table() Greg Kroah-Hartman
2016-02-24  3:34 ` [PATCH 3.10 54/54] module: wrapper for symbol name Greg Kroah-Hartman
2016-02-24 16:31 ` [PATCH 3.10 00/54] 3.10.98-stable review Willy Tarreau
2016-02-25 18:37   ` Greg Kroah-Hartman
2016-02-24 18:29 ` Shuah Khan
2016-02-25  4:57 ` Guenter Roeck
2016-02-25 16:33   ` Greg Kroah-Hartman
2016-02-25 17:01     ` Guenter Roeck

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20160224033354.266887421@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=akpm@linux-foundation.org \
    --cc=hughd@google.com \
    --cc=khlebnikov@openvz.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ohad@wizery.com \
    --cc=stable@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    --cc=willy@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).