| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Tue, 29 Mar 2016 19:59:50 -0600
From: Vishal Verma <vishal.l.verma@...el.com>
To: linux-nvdimm@...ts.01.org
Cc: Vishal Verma <vishal.l.verma@...el.com>,
linux-fsdevel@...r.kernel.org, linux-block@...r.kernel.org,
xfs@....sgi.com, linux-ext4@...r.kernel.org, linux-mm@...ck.org,
Matthew Wilcox <matthew.r.wilcox@...el.com>,
Ross Zwisler <ross.zwisler@...ux.intel.com>,
Dan Williams <dan.j.williams@...el.com>,
Dave Chinner <david@...morbit.com>, Jan Kara <jack@...e.cz>,
Jens Axboe <axboe@...com>, Al Viro <viro@...iv.linux.org.uk>,
Andrew Morton <akpm@...ux-foundation.org>,
linux-kernel@...r.kernel.org, Christoph Hellwig <hch@...radead.org>
Subject: [PATCH v2 5/5] dax: handle media errors in dax_do_io
dax_do_io (called for read() or write() for a dax file system) may fail
in the presence of bad blocks or media errors. Since we expect that a
write should clear media errors on nvdimms, make dax_do_io fall back to
the direct_IO path, which will send down a bio to the driver, which can
then attempt to clear the error.
Cc: Matthew Wilcox <matthew.r.wilcox@...el.com>
Cc: Dan Williams <dan.j.williams@...el.com>
Cc: Ross Zwisler <ross.zwisler@...ux.intel.com>
Cc: Dave Chinner <david@...morbit.com>
Cc: Jan Kara <jack@...e.cz>
Cc: Jens Axboe <axboe@...com>
Cc: Al Viro <viro@...iv.linux.org.uk>
Cc: Christoph Hellwig <hch@...radead.org>
Signed-off-by: Vishal Verma <vishal.l.verma@...el.com>
---
fs/block_dev.c | 17 ++++++++++++++---
fs/ext2/inode.c | 22 +++++++++++++++-------
fs/ext4/indirect.c | 18 +++++++++++++-----
fs/ext4/inode.c | 21 ++++++++++++++-------
fs/xfs/xfs_aops.c | 14 ++++++++++++--
5 files changed, 68 insertions(+), 24 deletions(-)
diff --git a/fs/block_dev.c b/fs/block_dev.c
index c5837fa..d6113b9 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -166,13 +166,24 @@ blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset)
{
struct file *file = iocb->ki_filp;
struct inode *inode = bdev_file_inode(file);
+ ssize_t ret, ret_saved = 0;
- if (IS_DAX(inode))
- return dax_do_io(iocb, inode, iter, offset, blkdev_get_block,
+ if (IS_DAX(inode)) {
+ ret = dax_do_io(iocb, inode, iter, offset, blkdev_get_block,
NULL, DIO_SKIP_DIO_COUNT);
- return __blockdev_direct_IO(iocb, inode, I_BDEV(inode), iter, offset,
+ if (ret == -EIO && (iov_iter_rw(iter) == WRITE))
+ ret_saved = ret;
+ else
+ return ret;
+ }
+
+ ret = __blockdev_direct_IO(iocb, inode, I_BDEV(inode), iter, offset,
blkdev_get_block, NULL, NULL,
DIO_SKIP_DIO_COUNT);
+ if (ret < 0 && ret_saved)
+ return ret_saved;
+
+ return ret;
}
int __sync_blockdev(struct block_device *bdev, int wait)
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 824f249..64792c6 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -859,14 +859,22 @@ ext2_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset)
struct address_space *mapping = file->f_mapping;
struct inode *inode = mapping->host;
size_t count = iov_iter_count(iter);
- ssize_t ret;
+ ssize_t ret, ret_saved = 0;
- if (IS_DAX(inode))
- ret = dax_do_io(iocb, inode, iter, offset, ext2_get_block, NULL,
- DIO_LOCKING);
- else
- ret = blockdev_direct_IO(iocb, inode, iter, offset,
- ext2_get_block);
+ if (IS_DAX(inode)) {
+ ret = dax_do_io(iocb, inode, iter, offset, ext2_get_block,
+ NULL, DIO_LOCKING | DIO_SKIP_HOLES);
+ if (ret == -EIO && iov_iter_rw(iter) == WRITE)
+ ret_saved = ret;
+ else
+ goto out;
+ }
+
+ ret = blockdev_direct_IO(iocb, inode, iter, offset, ext2_get_block);
+ if (ret < 0 && ret_saved)
+ ret = ret_saved;
+
+ out:
if (ret < 0 && iov_iter_rw(iter) == WRITE)
ext2_write_failed(mapping, offset + count);
return ret;
diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c
index 3027fa6..798f341 100644
--- a/fs/ext4/indirect.c
+++ b/fs/ext4/indirect.c
@@ -716,14 +716,22 @@ retry:
NULL, NULL, 0);
inode_dio_end(inode);
} else {
+ ssize_t ret_saved = 0;
+
locked:
- if (IS_DAX(inode))
+ if (IS_DAX(inode)) {
ret = dax_do_io(iocb, inode, iter, offset,
ext4_dio_get_block, NULL, DIO_LOCKING);
- else
- ret = blockdev_direct_IO(iocb, inode, iter, offset,
- ext4_dio_get_block);
-
+ if (ret == -EIO && iov_iter_rw(iter) == WRITE)
+ ret_saved = ret;
+ else
+ goto skip_dio;
+ }
+ ret = blockdev_direct_IO(iocb, inode, iter, offset,
+ ext4_get_block);
+ if (ret < 0 && ret_saved)
+ ret = ret_saved;
+skip_dio:
if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) {
loff_t isize = i_size_read(inode);
loff_t end = offset + count;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index dab84a2..27f07c2 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -3341,7 +3341,7 @@ static ssize_t ext4_ext_direct_IO(struct kiocb *iocb, struct iov_iter *iter,
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
- ssize_t ret;
+ ssize_t ret, ret_saved = 0;
size_t count = iov_iter_count(iter);
int overwrite = 0;
get_block_t *get_block_func = NULL;
@@ -3401,15 +3401,22 @@ static ssize_t ext4_ext_direct_IO(struct kiocb *iocb, struct iov_iter *iter,
#ifdef CONFIG_EXT4_FS_ENCRYPTION
BUG_ON(ext4_encrypted_inode(inode) && S_ISREG(inode->i_mode));
#endif
- if (IS_DAX(inode))
+ if (IS_DAX(inode)) {
ret = dax_do_io(iocb, inode, iter, offset, get_block_func,
ext4_end_io_dio, dio_flags);
- else
- ret = __blockdev_direct_IO(iocb, inode,
- inode->i_sb->s_bdev, iter, offset,
- get_block_func,
- ext4_end_io_dio, NULL, dio_flags);
+ if (ret == -EIO && iov_iter_rw(iter) == WRITE)
+ ret_saved = ret;
+ else
+ goto skip_dio;
+ }
+ ret = __blockdev_direct_IO(iocb, inode,
+ inode->i_sb->s_bdev, iter, offset,
+ get_block_func,
+ ext4_end_io_dio, NULL, dio_flags);
+ if (ret < 0 && ret_saved)
+ ret = ret_saved;
+ skip_dio:
if (ret > 0 && !overwrite && ext4_test_inode_state(inode,
EXT4_STATE_DIO_UNWRITTEN)) {
int err;
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index d445a64..7cfcf86 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -1413,6 +1413,7 @@ xfs_vm_direct_IO(
dio_iodone_t *endio = NULL;
int flags = 0;
struct block_device *bdev;
+ ssize_t ret, ret_saved = 0;
if (iov_iter_rw(iter) == WRITE) {
endio = xfs_end_io_direct_write;
@@ -1420,13 +1421,22 @@ xfs_vm_direct_IO(
}
if (IS_DAX(inode)) {
- return dax_do_io(iocb, inode, iter, offset,
+ ret = dax_do_io(iocb, inode, iter, offset,
xfs_get_blocks_direct, endio, 0);
+ if (ret == -EIO && iov_iter_rw(iter) == WRITE)
+ ret_saved = ret;
+ else
+ return ret;
}
bdev = xfs_find_bdev_for_inode(inode);
- return __blockdev_direct_IO(iocb, inode, bdev, iter, offset,
+ ret = __blockdev_direct_IO(iocb, inode, bdev, iter, offset,
xfs_get_blocks_direct, endio, NULL, flags);
+
+ if (ret < 0 && ret_saved)
+ ret = ret_saved;
+
+ return ret;
}
/*
--
2.5.5
Powered by blists - more mailing lists