Patch "Revert "ocfs2: should wait dio before inode lock in ocfs2_setattr()"" has been added to the 4.4-stable tree

Patch "Revert "ocfs2: should wait dio before inode lock in ocfs2_setattr()"" has been added to the 4.4-stable tree

[gregkh]

This is a note to let you know that I've just added the patch titled

    Revert "ocfs2: should wait dio before inode lock in ocfs2_setattr()"

to the 4.4-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     revert-ocfs2-should-wait-dio-before-inode-lock-in-ocfs2_setattr.patch
and it can be found in the queue-4.4 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From 02b359b1331a9fd96f43028cee5500dfac4454b9 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Date: Wed, 6 Dec 2017 17:33:18 +0100
Subject: Revert "ocfs2: should wait dio before inode lock in ocfs2_setattr()"

From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This reverts commit c4baa4a5870cb02f713def1620052bfca7a82bbb which is
commit 28f5a8a7c033cbf3e32277f4cc9c6afd74f05300 upstream.

It shouldn't be applied to the 4.4-stable tree.

Ben and Alex write:

> Now that ocfs2_setattr() calls this outside of the inode locked region,
> what prevents another task adding a new dio request immediately
> afterward?
>

In the kernel 4.6, firstly, we use the inode_lock() in do_truncate() to
prevent another bio to be issued from this node.
Furthermore, we use the ocfs2_rw_lock() and ocfs2_inode_lock() in ocfs2_setattr()
to guarantee no more bio will be issued from the other nodes in this cluster.

> Also, ocfs2_dio_end_io_write() was introduced in 4.6 and it looks like
> the dio completion path didn't previously take the inode lock.  So it
> doesn't look this fix is needed in 3.18 or 4.4.

Yes, ocfs2_dio_end_io_write() was introduced in 4.6 and the problem this patch
fixes is only exist in the kernel 4.6 and above 4.6.

Reported-by: Ben Hutchings <ben.hutchings@codethink.co.uk>
Cc: Alex Chen <alex.chen@huawei.com>
Cc: Jun Piao <piaojun@huawei.com>
Cc: Joseph Qi <jiangqi903@gmail.com>
Cc: Changwei Ge <ge.changwei@h3c.com>
Cc: Mark Fasheh <mfasheh@versity.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 fs/ocfs2/file.c |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1166,13 +1166,6 @@ int ocfs2_setattr(struct dentry *dentry,
 	}
 	size_change = S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE;
 	if (size_change) {
-		/*
-		 * Here we should wait dio to finish before inode lock
-		 * to avoid a deadlock between ocfs2_setattr() and
-		 * ocfs2_dio_end_io_write()
-		 */
-		inode_dio_wait(inode);
-
 		status = ocfs2_rw_lock(inode, 1);
 		if (status < 0) {
 			mlog_errno(status);
@@ -1193,6 +1186,8 @@ int ocfs2_setattr(struct dentry *dentry,
 		if (status)
 			goto bail_unlock;
 
+		inode_dio_wait(inode);
+
 		if (i_size_read(inode) >= attr->ia_size) {
 			if (ocfs2_should_order_data(inode)) {
 				status = ocfs2_begin_ordered_truncate(inode,


Patches currently in stable-queue which might be from gregkh@linuxfoundation.org are

queue-4.4/xen-netfront-improve-error-handling-during-initialization.patch
queue-4.4/net-systemport-pad-packet-before-inserting-tsb.patch
queue-4.4/kvm-arm-arm64-fix-occasional-warning-from-the-timer-work-function.patch
queue-4.4/s390-pci-do-not-require-ais-facility.patch
queue-4.4/ima-fix-hash-algorithm-initialization.patch
queue-4.4/usb-phy-tahvo-fix-error-handling-in-tahvo_usb_probe.patch
queue-4.4/spi-sh-msiof-fix-dma-transfer-size-check.patch
queue-4.4/usb-dwc2-error-out-of-dwc2_hsotg_ep_disable-if-we-re-in-host-mode.patch
queue-4.4/net-appletalk-fix-kernel-memory-disclosure.patch
queue-4.4/selftests-x86-ldt_get-add-a-few-additional-tests-for-limits.patch
queue-4.4/nfsv4-fix-client-recovery-when-server-reboots-multiple-times.patch
queue-4.4/tipc-fix-cleanup-at-module-unload.patch
queue-4.4/bcache-only-permit-to-recovery-read-error-when-cache-device-is-clean.patch
queue-4.4/ravb-remove-rx-overflow-log-messages.patch
queue-4.4/usb-quirks-add-no-lpm-quirk-for-ky-688-usb-3.1-type-c-hub.patch
queue-4.4/nfs-don-t-take-a-reference-on-fl-fl_file-for-lock-operation.patch
queue-4.4/serial-8250-preserve-dld-for-port_xr17v35x.patch
queue-4.4/arm-omap1-dma-correct-the-number-of-logical-channels.patch
queue-4.4/usb-dwc2-fix-udc-state-tracking.patch
queue-4.4/mm-avoid-returning-vm_fault_retry-from-page_mkwrite-handlers.patch
queue-4.4/dmaengine-pl330-fix-double-lock.patch
queue-4.4/revert-ocfs2-should-wait-dio-before-inode-lock-in-ocfs2_setattr.patch
queue-4.4/bcache-recover-data-from-backing-when-data-is-clean.patch
queue-4.4/perf-test-attr-fix-ignored-test-case-result.patch
queue-4.4/s390-runtime-instrumentation-simplify-task-exit-handling.patch
queue-4.4/serial-8250_fintek-fix-rs485-disablement-on-invalid-ioctl.patch
queue-4.4/net-systemport-utilize-skb_put_padto.patch
queue-4.4/usbip-tools-install-all-headers-needed-for-libusbip-development.patch
queue-4.4/serial-8250_pci-add-amazon-pci-serial-device-id.patch
queue-4.4/sysrq-fix-show-regs-call-trace-on-arm.patch
queue-4.4/tcp-correct-memory-barrier-usage-in-tcp_check_space.patch
queue-4.4/net-fec-fix-multicast-filtering-hardware-setup.patch
queue-4.4/x86-entry-use-syscall_define-macros-for-sys_modify_ldt.patch
queue-4.4/uas-always-apply-us_fl_no_ata_1x-quirk-to-seagate-devices.patch
queue-4.4/vti6-fix-device-register-to-report-ifla_info_kind.patch
queue-4.4/kprobes-x86-disable-preemption-in-ftrace-based-jprobes.patch
queue-4.4/usb-serial-option-add-quectel-bg96-id.patch
queue-4.4/drm-exynos-decon5433-set-standalone_update_f-on-output-enablement.patch
queue-4.4/edac-sb_edac-fix-missing-break-in-switch.patch
queue-4.4/net-sctp-fix-array-overrun-read-on-sctp_timer_tbl.patch