The commit handling code is not safe against memory-pressure deadlocks
when writing to swap. In particular, nfs_commitdata_alloc() blocks
indefinitely waiting for memory, and this can consume all available
workqueue threads.
swap-out most likely uses STABLE writes anyway as COND_STABLE indicates
that a stable write should be used if the write fits in a single
request, and it normally does. However if we ever swap with a small
wsize, or gather unusually large numbers of pages for a single write,
this might change.
For safety, make it explicit in the code that direct writes use for swap
must always use FLUSH_COND_STABLE.
Signed-off-by: NeilBrown <neilb@xxxxxxx>
---
fs/nfs/direct.c | 12 +++++++-----
fs/nfs/file.c | 2 +-
include/linux/nfs_fs.h | 3 ++-
3 files changed, 10 insertions(+), 7 deletions(-)
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 1e80d243ba25..8d3b12402725 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -173,7 +173,7 @@ ssize_t nfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
if (iov_iter_rw(iter) == READ)
return nfs_file_direct_read(iocb, iter);
- return nfs_file_direct_write(iocb, iter);
+ return nfs_file_direct_write(iocb, iter, FLUSH_STABLE);
}
static void nfs_direct_release_pages(struct page **pages, unsigned int npages)
@@ -789,7 +789,7 @@ static const struct nfs_pgio_completion_ops nfs_direct_write_completion_ops = {
*/
static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
struct iov_iter *iter,
- loff_t pos)
+ loff_t pos, int ioflags)
{
struct nfs_pageio_descriptor desc;
struct inode *inode = dreq->inode;
@@ -797,7 +797,7 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
size_t requested_bytes = 0;
size_t wsize = max_t(size_t, NFS_SERVER(inode)->wsize, PAGE_SIZE);
- nfs_pageio_init_write(&desc, inode, FLUSH_COND_STABLE, false,
+ nfs_pageio_init_write(&desc, inode, ioflags, false,
&nfs_direct_write_completion_ops);
desc.pg_dreq = dreq;
get_dreq(dreq);
@@ -875,6 +875,7 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
* nfs_file_direct_write - file direct write operation for NFS files
* @iocb: target I/O control block
* @iter: vector of user buffers from which to write data
+ * @ioflags: flags for nfs_pageio_init_write()
*
* We use this function for direct writes instead of calling
* generic_file_aio_write() in order to avoid taking the inode
@@ -891,7 +892,8 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
* Note that O_APPEND is not supported for NFS direct writes, as there
* is no atomic O_APPEND write facility in the NFS protocol.
*/
-ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
+ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter,
+ int ioflags)
{
ssize_t result, requested;
size_t count;
@@ -935,7 +937,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
nfs_start_io_direct(inode);
- requested = nfs_direct_write_schedule_iovec(dreq, iter, pos);
+ requested = nfs_direct_write_schedule_iovec(dreq, iter, pos, ioflags);
if (mapping->nrpages) {
invalidate_inode_pages2_range(mapping,
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 59c271f42ea5..878a6a510a5e 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -630,7 +630,7 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from)
result = generic_write_checks(iocb, from);
if (result <= 0)
return result;
- return nfs_file_direct_write(iocb, from);
+ return nfs_file_direct_write(iocb, from, FLUSH_COND_STABLE);
}
dprintk("NFS: write(%pD2, %zu@%Ld)\n",
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 5a605e51f4b1..ca312aea6bec 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -509,7 +509,8 @@ extern ssize_t nfs_direct_IO(struct kiocb *, struct iov_iter *);
extern ssize_t nfs_file_direct_read(struct kiocb *iocb,
struct iov_iter *iter);
extern ssize_t nfs_file_direct_write(struct kiocb *iocb,
- struct iov_iter *iter);
+ struct iov_iter *iter,
+ int ioflags);
/*
* linux/fs/nfs/dir.c
