linux_dsm_epyc7002/fs/nfs/io.c
Trond Myklebust e231c6879c NFS: Fix a race between mmap() and O_DIRECT
When locking the file in order to do O_DIRECT on it, we must unmap
any mmapped ranges on the pagecache so that we can flush out the
dirty data.

Fixes: a5864c999d ("NFS: Do not serialise O_DIRECT reads and writes")
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Cc: stable@vger.kernel.org # v4.8+
2018-01-28 22:00:15 -05:00

149 lines
4.1 KiB
C

// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (c) 2016 Trond Myklebust
*
* I/O and data path helper functionality.
*/
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/bitops.h>
#include <linux/rwsem.h>
#include <linux/fs.h>
#include <linux/nfs_fs.h>
#include "internal.h"
/* Call with exclusively locked inode->i_rwsem */
static void nfs_block_o_direct(struct nfs_inode *nfsi, struct inode *inode)
{
if (test_bit(NFS_INO_ODIRECT, &nfsi->flags)) {
clear_bit(NFS_INO_ODIRECT, &nfsi->flags);
inode_dio_wait(inode);
}
}
/**
* nfs_start_io_read - declare the file is being used for buffered reads
* @inode - file inode
*
* Declare that a buffered read operation is about to start, and ensure
* that we block all direct I/O.
* On exit, the function ensures that the NFS_INO_ODIRECT flag is unset,
* and holds a shared lock on inode->i_rwsem to ensure that the flag
* cannot be changed.
* In practice, this means that buffered read operations are allowed to
* execute in parallel, thanks to the shared lock, whereas direct I/O
* operations need to wait to grab an exclusive lock in order to set
* NFS_INO_ODIRECT.
* Note that buffered writes and truncates both take a write lock on
* inode->i_rwsem, meaning that those are serialised w.r.t. the reads.
*/
void
nfs_start_io_read(struct inode *inode)
{
struct nfs_inode *nfsi = NFS_I(inode);
/* Be an optimist! */
down_read(&inode->i_rwsem);
if (test_bit(NFS_INO_ODIRECT, &nfsi->flags) == 0)
return;
up_read(&inode->i_rwsem);
/* Slow path.... */
down_write(&inode->i_rwsem);
nfs_block_o_direct(nfsi, inode);
downgrade_write(&inode->i_rwsem);
}
/**
* nfs_end_io_read - declare that the buffered read operation is done
* @inode - file inode
*
* Declare that a buffered read operation is done, and release the shared
* lock on inode->i_rwsem.
*/
void
nfs_end_io_read(struct inode *inode)
{
up_read(&inode->i_rwsem);
}
/**
* nfs_start_io_write - declare the file is being used for buffered writes
* @inode - file inode
*
* Declare that a buffered read operation is about to start, and ensure
* that we block all direct I/O.
*/
void
nfs_start_io_write(struct inode *inode)
{
down_write(&inode->i_rwsem);
nfs_block_o_direct(NFS_I(inode), inode);
}
/**
* nfs_end_io_write - declare that the buffered write operation is done
* @inode - file inode
*
* Declare that a buffered write operation is done, and release the
* lock on inode->i_rwsem.
*/
void
nfs_end_io_write(struct inode *inode)
{
up_write(&inode->i_rwsem);
}
/* Call with exclusively locked inode->i_rwsem */
static void nfs_block_buffered(struct nfs_inode *nfsi, struct inode *inode)
{
if (!test_bit(NFS_INO_ODIRECT, &nfsi->flags)) {
set_bit(NFS_INO_ODIRECT, &nfsi->flags);
nfs_sync_mapping(inode->i_mapping);
}
}
/**
* nfs_end_io_direct - declare the file is being used for direct i/o
* @inode - file inode
*
* Declare that a direct I/O operation is about to start, and ensure
* that we block all buffered I/O.
* On exit, the function ensures that the NFS_INO_ODIRECT flag is set,
* and holds a shared lock on inode->i_rwsem to ensure that the flag
* cannot be changed.
* In practice, this means that direct I/O operations are allowed to
* execute in parallel, thanks to the shared lock, whereas buffered I/O
* operations need to wait to grab an exclusive lock in order to clear
* NFS_INO_ODIRECT.
* Note that buffered writes and truncates both take a write lock on
* inode->i_rwsem, meaning that those are serialised w.r.t. O_DIRECT.
*/
void
nfs_start_io_direct(struct inode *inode)
{
struct nfs_inode *nfsi = NFS_I(inode);
/* Be an optimist! */
down_read(&inode->i_rwsem);
if (test_bit(NFS_INO_ODIRECT, &nfsi->flags) != 0)
return;
up_read(&inode->i_rwsem);
/* Slow path.... */
down_write(&inode->i_rwsem);
nfs_block_buffered(nfsi, inode);
downgrade_write(&inode->i_rwsem);
}
/**
* nfs_end_io_direct - declare that the direct i/o operation is done
* @inode - file inode
*
* Declare that a direct I/O operation is done, and release the shared
* lock on inode->i_rwsem.
*/
void
nfs_end_io_direct(struct inode *inode)
{
up_read(&inode->i_rwsem);
}