From patchwork Fri Sep 28 15:44:08 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: [08/11] ext4: endless truncate due to nonlocked dio readers V2 Date: Fri, 28 Sep 2012 05:44:08 -0000 From: Dmitri Monakho X-Patchwork-Id: 187819 Message-Id: <1348847051-6746-9-git-send-email-dmonakhov@openvz.org> To: linux-ext4@vger.kernel.org Cc: tytso@mit.edu, jack@suse.cz, lczerner@redhat.com, Dmitry Monakhov If we have enough aggressive DIO readers, truncate and other dio waiters will wait forever inside inode_dio_wait(). It is reasonable to disable nonlock DIO read optimization during truncate. Signed-off-by: Dmitry Monakhov Reviewed-by: Jan Kara --- fs/ext4/inode.c | 9 +++++++-- 1 files changed, 7 insertions(+), 2 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 583cb3f..c40a98d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4305,9 +4305,14 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid & ATTR_SIZE) { if (attr->ia_size != inode->i_size) { truncate_setsize(inode, attr->ia_size); - /* Inode size will be reduced, wait for dio in flight */ - if (orphan) + /* Inode size will be reduced, wait for dio in flight. + * Temporarily disable dioread_nolock to prevent + * livelock. */ + if (orphan) { + ext4_inode_block_unlocked_dio(inode); inode_dio_wait(inode); + ext4_inode_resume_unlocked_dio(inode); + } } ext4_truncate(inode); }