Re: [PATCH 6/7] ext4: endless truncate due to nonlocked dio readers V2
From: Zheng Liu <hidden>
Date: 2012-09-13 10:31:13
Hi Dmitry, Could you please provide more detailed workload to convince me? I am thinking about whether we really need to disable dioread_nolock feature in here. In our benchmarks, we don't see this problem. Regards, Zheng On Sun, Sep 09, 2012 at 09:27:13PM +0400, Dmitry Monakhov wrote:
quoted hunk ↗ jump to hunk
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 <redacted> --- fs/ext4/extents.c | 2 +- fs/ext4/fsync.c | 2 +- fs/ext4/inode.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-)diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 8252651..b5b801f 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c@@ -4853,7 +4853,7 @@ int ext4_ext_punch_hole(struct file *file, loff_t offset, loff_t length) } /* finish any pending end_io work */ - inode_dio_wait(inode); + ext4_inode_dio_wait(inode, 1); ext4_flush_completed_IO(inode); credits = ext4_writepage_trans_blocks(inode);diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 290c5cf..bdf6bfd 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c@@ -204,7 +204,7 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) if (inode->i_sb->s_flags & MS_RDONLY) goto out; - inode_dio_wait(inode); + ext4_inode_dio_wait(inode, 1); ret = ext4_flush_completed_IO(inode); if (ret < 0) goto out;diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 93e6b09..a850026 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c@@ -4335,7 +4335,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) truncate_setsize(inode, attr->ia_size); /* Inode size will be reduced, wait for dio in flight */ if (orphan) - inode_dio_wait(inode); + ext4_inode_dio_wait(inode, 1); } ext4_truncate(inode); }-- 1.7.7.6 -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html