[PATCH] NFS: Fix O_DIRECT read problem when another write is going on

[PATCH] NFS: Fix O_DIRECT read problem when another write is going on

[Su Yanjun]

In xfstests generic/465 tests failed. Because O_DIRECT r/w use
async rpc calls, when r/w rpc calls are running concurrently we
may read partial data which is wrong.

For example as follows.
 user buffer
/--------\
|    |XXXX|
 rpc0 rpc1

When rpc0 runs it encounters eof so return 0, then another writes
something. When rpc1 runs it returns some data. The total data
buffer contains wrong data.

In this patch we check eof mark for each direct request. If encounters
eof then set eof mark in the request, when we meet it again report
-EAGAIN error. In nfs_direct_complete we convert -EAGAIN as if read
nothing. When the reader issue another read it will read ok.

Signed-off-by: Su Yanjun <suyj.fnst@cn.fujitsu.com>
---
 fs/nfs/direct.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 222d711..7f737a3 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -93,6 +93,7 @@ struct nfs_direct_req {
 				bytes_left,	/* bytes left to be sent */
 				error;		/* any reported error */
 	struct completion	completion;	/* wait for i/o completion */
+	int			eof;		/* eof mark in the req */
 
 	/* commit state */
 	struct nfs_mds_commit_info mds_cinfo;	/* Storage for cinfo */
@@ -380,6 +381,12 @@ static void nfs_direct_complete(struct nfs_direct_req *dreq)
 {
 	struct inode *inode = dreq->inode;
 
+	/* read partial data just as read nothing */
+	if (dreq->error == -EAGAIN) {
+		dreq->count = 0;
+		dreq->error = 0;
+	}
+
 	inode_dio_end(inode);
 
 	if (dreq->iocb) {
@@ -413,8 +420,13 @@ static void nfs_direct_read_completion(struct nfs_pgio_header *hdr)
 	if (hdr->good_bytes != 0)
 		nfs_direct_good_bytes(dreq, hdr);
 
-	if (test_bit(NFS_IOHDR_EOF, &hdr->flags))
+	if (dreq->eof)
+		dreq->error = -EAGAIN;
+
+	if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) {
 		dreq->error = 0;
+		dreq->eof = 1;
+	}
 
 	spin_unlock(&dreq->lock);
 
-- 
2.7.4

[PATCH 0/2] Fix O_DIRECT error handling

[Trond Myklebust]

These patches fix issues with how O_DIRECT handles read and write
errors, and also how it handles eof.

Trond Myklebust (2):
  NFS: Fix O_DIRECT accounting of number of bytes read/written
  NFS: Remove redundant mirror tracking in O_DIRECT

 fs/nfs/direct.c | 106 ++++++++++++++++--------------------------------
 1 file changed, 36 insertions(+), 70 deletions(-)

-- 
2.21.0

[PATCH 1/2] NFS: Fix O_DIRECT accounting of number of bytes read/written

[Trond Myklebust]

When a series of O_DIRECT reads or writes are truncated, either due to
eof or due to an error, then we should return the number of contiguous
bytes that were received/sent starting at the offset specified by the
application.

Currently, we are failing to correctly check contiguity, and so we're
failing the generic/465 in xfstests when the race between the read
and write RPCs causes the file to get extended while the 2 reads are
outstanding. If the first read RPC call wins the race and returns with
eof set, we should treat the second read RPC as being truncated.

Reported-by: Su Yanjun <suyj.fnst@cn.fujitsu.com>
Fixes: 1ccbad9f9f9bd ("nfs: fix DIO good bytes calculation")
Cc: stable@vger.kernel.org # 4.1+
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/direct.c | 78 +++++++++++++++++++++++++++----------------------
 1 file changed, 43 insertions(+), 35 deletions(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 222d7115db71..98a9a0bcdf38 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -123,32 +123,49 @@ static inline int put_dreq(struct nfs_direct_req *dreq)
 }
 
 static void
-nfs_direct_good_bytes(struct nfs_direct_req *dreq, struct nfs_pgio_header *hdr)
+nfs_direct_handle_truncated(struct nfs_direct_req *dreq,
+			    const struct nfs_pgio_header *hdr,
+			    ssize_t dreq_len)
 {
-	int i;
-	ssize_t count;
+	struct nfs_direct_mirror *mirror = &dreq->mirrors[hdr->pgio_mirror_idx];
+
+	if (!(test_bit(NFS_IOHDR_ERROR, &hdr->flags) ||
+	      test_bit(NFS_IOHDR_EOF, &hdr->flags)))
+		return;
+	if (dreq->max_count >= dreq_len) {
+		dreq->max_count = dreq_len;
+		if (dreq->count > dreq_len)
+			dreq->count = dreq_len;
+
+		if (test_bit(NFS_IOHDR_ERROR, &hdr->flags))
+			dreq->error = hdr->error;
+		else /* Clear outstanding error if this is EOF */
+			dreq->error = 0;
+	}
+	if (mirror->count > dreq_len)
+		mirror->count = dreq_len;
+}
 
-	WARN_ON_ONCE(dreq->count >= dreq->max_count);
+static void
+nfs_direct_count_bytes(struct nfs_direct_req *dreq,
+		       const struct nfs_pgio_header *hdr)
+{
+	struct nfs_direct_mirror *mirror = &dreq->mirrors[hdr->pgio_mirror_idx];
+	loff_t hdr_end = hdr->io_start + hdr->good_bytes;
+	ssize_t dreq_len = 0;
 
-	if (dreq->mirror_count == 1) {
-		dreq->mirrors[hdr->pgio_mirror_idx].count += hdr->good_bytes;
-		dreq->count += hdr->good_bytes;
-	} else {
-		/* mirrored writes */
-		count = dreq->mirrors[hdr->pgio_mirror_idx].count;
-		if (count + dreq->io_start < hdr->io_start + hdr->good_bytes) {
-			count = hdr->io_start + hdr->good_bytes - dreq->io_start;
-			dreq->mirrors[hdr->pgio_mirror_idx].count = count;
-		}
-		/* update the dreq->count by finding the minimum agreed count from all
-		 * mirrors */
-		count = dreq->mirrors[0].count;
+	if (hdr_end > dreq->io_start)
+		dreq_len = hdr_end - dreq->io_start;
 
-		for (i = 1; i < dreq->mirror_count; i++)
-			count = min(count, dreq->mirrors[i].count);
+	nfs_direct_handle_truncated(dreq, hdr, dreq_len);
 
-		dreq->count = count;
-	}
+	if (dreq_len > dreq->max_count)
+		dreq_len = dreq->max_count;
+
+	if (mirror->count < dreq_len)
+		mirror->count = dreq_len;
+	if (dreq->count < dreq_len)
+		dreq->count = dreq_len;
 }
 
 /*
@@ -402,20 +419,12 @@ static void nfs_direct_read_completion(struct nfs_pgio_header *hdr)
 	struct nfs_direct_req *dreq = hdr->dreq;
 
 	spin_lock(&dreq->lock);
-	if (test_bit(NFS_IOHDR_ERROR, &hdr->flags))
-		dreq->error = hdr->error;
-
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) {
 		spin_unlock(&dreq->lock);
 		goto out_put;
 	}
 
-	if (hdr->good_bytes != 0)
-		nfs_direct_good_bytes(dreq, hdr);
-
-	if (test_bit(NFS_IOHDR_EOF, &hdr->flags))
-		dreq->error = 0;
-
+	nfs_direct_count_bytes(dreq, hdr);
 	spin_unlock(&dreq->lock);
 
 	while (!list_empty(&hdr->pages)) {
@@ -652,6 +661,9 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo);
 
 	dreq->count = 0;
+	dreq->max_count = 0;
+	list_for_each_entry(req, &reqs, wb_list)
+		dreq->max_count += req->wb_bytes;
 	dreq->verf.committed = NFS_INVALID_STABLE_HOW;
 	nfs_clear_pnfs_ds_commit_verifiers(&dreq->ds_cinfo);
 	for (i = 0; i < dreq->mirror_count; i++)
@@ -791,17 +803,13 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 	nfs_init_cinfo_from_dreq(&cinfo, dreq);
 
 	spin_lock(&dreq->lock);
-
-	if (test_bit(NFS_IOHDR_ERROR, &hdr->flags))
-		dreq->error = hdr->error;
-
 	if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) {
 		spin_unlock(&dreq->lock);
 		goto out_put;
 	}
 
+	nfs_direct_count_bytes(dreq, hdr);
 	if (hdr->good_bytes != 0) {
-		nfs_direct_good_bytes(dreq, hdr);
 		if (nfs_write_need_commit(hdr)) {
 			if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES)
 				request_commit = true;
-- 
2.21.0

[PATCH 2/2] NFS: Remove redundant mirror tracking in O_DIRECT

[Trond Myklebust]

We no longer need the extra mirror length tracking in the O_DIRECT code,
as we are able to track the maximum contiguous length in dreq->max_count.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 fs/nfs/direct.c | 42 ------------------------------------------
 1 file changed, 42 deletions(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 98a9a0bcdf38..040a50fd9bf3 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -64,13 +64,6 @@
 
 static struct kmem_cache *nfs_direct_cachep;
 
-/*
- * This represents a set of asynchronous requests that we're waiting on
- */
-struct nfs_direct_mirror {
-	ssize_t count;
-};
-
 struct nfs_direct_req {
 	struct kref		kref;		/* release manager */
 
@@ -84,9 +77,6 @@ struct nfs_direct_req {
 	atomic_t		io_count;	/* i/os we're waiting for */
 	spinlock_t		lock;		/* protect completion state */
 
-	struct nfs_direct_mirror mirrors[NFS_PAGEIO_DESCRIPTOR_MIRROR_MAX];
-	int			mirror_count;
-
 	loff_t			io_start;	/* Start offset for I/O */
 	ssize_t			count,		/* bytes actually processed */
 				max_count,	/* max expected count */
@@ -127,8 +117,6 @@ nfs_direct_handle_truncated(struct nfs_direct_req *dreq,
 			    const struct nfs_pgio_header *hdr,
 			    ssize_t dreq_len)
 {
-	struct nfs_direct_mirror *mirror = &dreq->mirrors[hdr->pgio_mirror_idx];
-
 	if (!(test_bit(NFS_IOHDR_ERROR, &hdr->flags) ||
 	      test_bit(NFS_IOHDR_EOF, &hdr->flags)))
 		return;
@@ -142,15 +130,12 @@ nfs_direct_handle_truncated(struct nfs_direct_req *dreq,
 		else /* Clear outstanding error if this is EOF */
 			dreq->error = 0;
 	}
-	if (mirror->count > dreq_len)
-		mirror->count = dreq_len;
 }
 
 static void
 nfs_direct_count_bytes(struct nfs_direct_req *dreq,
 		       const struct nfs_pgio_header *hdr)
 {
-	struct nfs_direct_mirror *mirror = &dreq->mirrors[hdr->pgio_mirror_idx];
 	loff_t hdr_end = hdr->io_start + hdr->good_bytes;
 	ssize_t dreq_len = 0;
 
@@ -162,8 +147,6 @@ nfs_direct_count_bytes(struct nfs_direct_req *dreq,
 	if (dreq_len > dreq->max_count)
 		dreq_len = dreq->max_count;
 
-	if (mirror->count < dreq_len)
-		mirror->count = dreq_len;
 	if (dreq->count < dreq_len)
 		dreq->count = dreq_len;
 }
@@ -310,18 +293,6 @@ void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo,
 	cinfo->completion_ops = &nfs_direct_commit_completion_ops;
 }
 
-static inline void nfs_direct_setup_mirroring(struct nfs_direct_req *dreq,
-					     struct nfs_pageio_descriptor *pgio,
-					     struct nfs_page *req)
-{
-	int mirror_count = 1;
-
-	if (pgio->pg_ops->pg_get_mirror_count)
-		mirror_count = pgio->pg_ops->pg_get_mirror_count(pgio, req);
-
-	dreq->mirror_count = mirror_count;
-}
-
 static inline struct nfs_direct_req *nfs_direct_req_alloc(void)
 {
 	struct nfs_direct_req *dreq;
@@ -336,7 +307,6 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void)
 	INIT_LIST_HEAD(&dreq->mds_cinfo.list);
 	dreq->verf.committed = NFS_INVALID_STABLE_HOW;	/* not set yet */
 	INIT_WORK(&dreq->work, nfs_direct_write_schedule_work);
-	dreq->mirror_count = 1;
 	spin_lock_init(&dreq->lock);
 
 	return dreq;
@@ -655,7 +625,6 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	LIST_HEAD(reqs);
 	struct nfs_commit_info cinfo;
 	LIST_HEAD(failed);
-	int i;
 
 	nfs_init_cinfo_from_dreq(&cinfo, dreq);
 	nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo);
@@ -666,21 +635,12 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 		dreq->max_count += req->wb_bytes;
 	dreq->verf.committed = NFS_INVALID_STABLE_HOW;
 	nfs_clear_pnfs_ds_commit_verifiers(&dreq->ds_cinfo);
-	for (i = 0; i < dreq->mirror_count; i++)
-		dreq->mirrors[i].count = 0;
 	get_dreq(dreq);
 
 	nfs_pageio_init_write(&desc, dreq->inode, FLUSH_STABLE, false,
 			      &nfs_direct_write_completion_ops);
 	desc.pg_dreq = dreq;
 
-	req = nfs_list_entry(reqs.next);
-	nfs_direct_setup_mirroring(dreq, &desc, req);
-	if (desc.pg_error < 0) {
-		list_splice_init(&reqs, &failed);
-		goto out_failed;
-	}
-
 	list_for_each_entry_safe(req, tmp, &reqs, wb_list) {
 		/* Bump the transmission count */
 		req->wb_nio++;
@@ -698,7 +658,6 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
 	}
 	nfs_pageio_complete(&desc);
 
-out_failed:
 	while (!list_empty(&failed)) {
 		req = nfs_list_entry(failed.next);
 		nfs_list_remove_request(req);
@@ -931,7 +890,6 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
 				break;
 			}
 
-			nfs_direct_setup_mirroring(dreq, &desc, req);
 			if (desc.pg_error < 0) {
 				nfs_free_request(req);
 				result = desc.pg_error;
-- 
2.21.0

Re: [PATCH] NFS: Fix O_DIRECT read problem when another write is going on

[Trond Myklebust]

Hi Su,

On Mon, 2019-09-30 at 17:11 +0800, Su Yanjun wrote:
> In xfstests generic/465 tests failed. Because O_DIRECT r/w use
> async rpc calls, when r/w rpc calls are running concurrently we
> may read partial data which is wrong.
> 
> For example as follows.
>  user buffer
> /--------\
> >    |XXXX|
>  rpc0 rpc1
> 
> When rpc0 runs it encounters eof so return 0, then another writes
> something. When rpc1 runs it returns some data. The total data
> buffer contains wrong data.
> 
> In this patch we check eof mark for each direct request. If
> encounters
> eof then set eof mark in the request, when we meet it again report
> -EAGAIN error. In nfs_direct_complete we convert -EAGAIN as if read
> nothing. When the reader issue another read it will read ok.
> 
> Signed-off-by: Su Yanjun <suyj.fnst@cn.fujitsu.com>
> ---
>  fs/nfs/direct.c | 14 +++++++++++++-
>  1 file changed, 13 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
> index 222d711..7f737a3 100644
> --- a/fs/nfs/direct.c
> +++ b/fs/nfs/direct.c
> @@ -93,6 +93,7 @@ struct nfs_direct_req {
>  				bytes_left,	/* bytes left to be
> sent */
>  				error;		/* any reported error
> */
>  	struct completion	completion;	/* wait for i/o completion */
> +	int			eof;		/* eof mark in the
> req */
>  
>  	/* commit state */
>  	struct nfs_mds_commit_info mds_cinfo;	/* Storage for cinfo
> */
> @@ -380,6 +381,12 @@ static void nfs_direct_complete(struct
> nfs_direct_req *dreq)
>  {
>  	struct inode *inode = dreq->inode;
>  
> +	/* read partial data just as read nothing */
> +	if (dreq->error == -EAGAIN) {
> +		dreq->count = 0;
> +		dreq->error = 0;
> +	}
> +
>  	inode_dio_end(inode);
>  
>  	if (dreq->iocb) {
> @@ -413,8 +420,13 @@ static void nfs_direct_read_completion(struct
> nfs_pgio_header *hdr)
>  	if (hdr->good_bytes != 0)
>  		nfs_direct_good_bytes(dreq, hdr);
>  
> -	if (test_bit(NFS_IOHDR_EOF, &hdr->flags))
> +	if (dreq->eof)
> +		dreq->error = -EAGAIN;
> +
> +	if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) {
>  		dreq->error = 0;
> +		dreq->eof = 1;
> +	}
>  
>  	spin_unlock(&dreq->lock);
>  

Thanks for looking into this issue. I agree with your analysis of what
is going wrong in generic/465.

However, I think the problem is greater than just EOF. I think we also
need to look at the generic error handling, and ensure that it handles
a truncated RPC call in the middle of a series of calls correctly.

Please see the two patches I sent you just now and check if they fix
the problem for you.

-- 
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trond.myklebust@hammerspace.com

Re: [PATCH] NFS: Fix O_DIRECT read problem when another write is going on

[Su Yanjun]

在 2019/10/1 2:06, Trond Myklebust 写道:
> Hi Su,
>
> On Mon, 2019-09-30 at 17:11 +0800, Su Yanjun wrote:
>> In xfstests generic/465 tests failed. Because O_DIRECT r/w use
>> async rpc calls, when r/w rpc calls are running concurrently we
>> may read partial data which is wrong.
>>
>> For example as follows.
>>   user buffer
>> /--------\
>>>     |XXXX|
>>   rpc0 rpc1
>>
>> When rpc0 runs it encounters eof so return 0, then another writes
>> something. When rpc1 runs it returns some data. The total data
>> buffer contains wrong data.
>>
>> In this patch we check eof mark for each direct request. If
>> encounters
>> eof then set eof mark in the request, when we meet it again report
>> -EAGAIN error. In nfs_direct_complete we convert -EAGAIN as if read
>> nothing. When the reader issue another read it will read ok.
>>
>> Signed-off-by: Su Yanjun <suyj.fnst@cn.fujitsu.com>
>> ---
>>   fs/nfs/direct.c | 14 +++++++++++++-
>>   1 file changed, 13 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
>> index 222d711..7f737a3 100644
>> --- a/fs/nfs/direct.c
>> +++ b/fs/nfs/direct.c
>> @@ -93,6 +93,7 @@ struct nfs_direct_req {
>>   				bytes_left,	/* bytes left to be
>> sent */
>>   				error;		/* any reported error
>> */
>>   	struct completion	completion;	/* wait for i/o completion */
>> +	int			eof;		/* eof mark in the
>> req */
>>   
>>   	/* commit state */
>>   	struct nfs_mds_commit_info mds_cinfo;	/* Storage for cinfo
>> */
>> @@ -380,6 +381,12 @@ static void nfs_direct_complete(struct
>> nfs_direct_req *dreq)
>>   {
>>   	struct inode *inode = dreq->inode;
>>   
>> +	/* read partial data just as read nothing */
>> +	if (dreq->error == -EAGAIN) {
>> +		dreq->count = 0;
>> +		dreq->error = 0;
>> +	}
>> +
>>   	inode_dio_end(inode);
>>   
>>   	if (dreq->iocb) {
>> @@ -413,8 +420,13 @@ static void nfs_direct_read_completion(struct
>> nfs_pgio_header *hdr)
>>   	if (hdr->good_bytes != 0)
>>   		nfs_direct_good_bytes(dreq, hdr);
>>   
>> -	if (test_bit(NFS_IOHDR_EOF, &hdr->flags))
>> +	if (dreq->eof)
>> +		dreq->error = -EAGAIN;
>> +
>> +	if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) {
>>   		dreq->error = 0;
>> +		dreq->eof = 1;
>> +	}
>>   
>>   	spin_unlock(&dreq->lock);
>>   
> Thanks for looking into this issue. I agree with your analysis of what
> is going wrong in generic/465.
>
> However, I think the problem is greater than just EOF. I think we also
> need to look at the generic error handling, and ensure that it handles
> a truncated RPC call in the middle of a series of calls correctly.
>
> Please see the two patches I sent you just now and check if they fix
> the problem for you.

The patchset you sent works for generic/465.

Thanks a lot