[PATCH] io_uring: retry in case of short read on block device

[PATCH] io_uring: retry in case of short read on block device

[Ming Lei]

In case of buffered reading from block device, when short read happens,
we should retry to read more, otherwise the IO will be completed
partially, for example, the following fio expects to read 2MB, but it
can only read 1M or less bytes:

    fio --name=onessd --filename=/dev/nvme0n1 --filesize=2M \
	--rw=randread --bs=2M --direct=0 --overwrite=0 --numjobs=1 \
	--iodepth=1 --time_based=0 --runtime=2 --ioengine=io_uring \
	--registerfiles --fixedbufs --gtod_reduce=1 --group_reporting

Fix the issue by allowing short read retry for block device, which sets
FMODE_BUF_RASYNC really.

Fixes: 9a173346bd9e ("io_uring: fix short read retries for non-reg files")
Cc: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 fs/io_uring.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index bf548af0426c..bbcd1a9e75e5 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -3268,6 +3268,12 @@ static inline int io_iter_do_read(struct io_kiocb *req, struct iov_iter *iter)
 		return -EINVAL;
 }
 
+static bool need_read_all(struct io_kiocb *req)
+{
+	return req->flags & REQ_F_ISREG ||
+		S_ISBLK(file_inode(req->file)->i_mode);
+}
+
 static int io_read(struct io_kiocb *req, unsigned int issue_flags)
 {
 	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
@@ -3322,7 +3328,7 @@ static int io_read(struct io_kiocb *req, unsigned int issue_flags)
 	} else if (ret == -EIOCBQUEUED) {
 		goto out_free;
 	} else if (ret <= 0 || ret == io_size || !force_nonblock ||
-		   (req->flags & REQ_F_NOWAIT) || !(req->flags & REQ_F_ISREG)) {
+		   (req->flags & REQ_F_NOWAIT) || !need_read_all(req)) {
 		/* read all, failed, already did sync or don't want to retry */
 		goto done;
 	}
-- 
2.31.1

Re: [PATCH] io_uring: retry in case of short read on block device

[Ming Lei]

On Sat, Aug 21, 2021 at 11:07:51PM +0800, Ming Lei wrote:
> In case of buffered reading from block device, when short read happens,
> we should retry to read more, otherwise the IO will be completed
> partially, for example, the following fio expects to read 2MB, but it
> can only read 1M or less bytes:
> 
>     fio --name=onessd --filename=/dev/nvme0n1 --filesize=2M \
> 	--rw=randread --bs=2M --direct=0 --overwrite=0 --numjobs=1 \
> 	--iodepth=1 --time_based=0 --runtime=2 --ioengine=io_uring \
> 	--registerfiles --fixedbufs --gtod_reduce=1 --group_reporting
> 
> Fix the issue by allowing short read retry for block device, which sets
> FMODE_BUF_RASYNC really.
> 
> Fixes: 9a173346bd9e ("io_uring: fix short read retries for non-reg files")
> Cc: Pavel Begunkov <asml.silence@gmail.com>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>

Hello Jens and Pavel,

Any comments on this fix?


Thanks, 
Ming

Re: [PATCH] io_uring: retry in case of short read on block device

[Pavel Begunkov]

On 8/21/21 4:07 PM, Ming Lei wrote:
> In case of buffered reading from block device, when short read happens,
> we should retry to read more, otherwise the IO will be completed
> partially, for example, the following fio expects to read 2MB, but it
> can only read 1M or less bytes:
> 
>     fio --name=onessd --filename=/dev/nvme0n1 --filesize=2M \
> 	--rw=randread --bs=2M --direct=0 --overwrite=0 --numjobs=1 \
> 	--iodepth=1 --time_based=0 --runtime=2 --ioengine=io_uring \
> 	--registerfiles --fixedbufs --gtod_reduce=1 --group_reporting
> 
> Fix the issue by allowing short read retry for block device, which sets
> FMODE_BUF_RASYNC really.

Should note that overhead on touching inode shouldn't be of concern at
this point, so all looks good

Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>


> Fixes: 9a173346bd9e ("io_uring: fix short read retries for non-reg files")
> Cc: Pavel Begunkov <asml.silence@gmail.com>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> ---
>  fs/io_uring.c | 8 +++++++-
>  1 file changed, 7 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index bf548af0426c..bbcd1a9e75e5 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -3268,6 +3268,12 @@ static inline int io_iter_do_read(struct io_kiocb *req, struct iov_iter *iter)
>  		return -EINVAL;
>  }
>  
> +static bool need_read_all(struct io_kiocb *req)
> +{
> +	return req->flags & REQ_F_ISREG ||
> +		S_ISBLK(file_inode(req->file)->i_mode);
> +}
> +
>  static int io_read(struct io_kiocb *req, unsigned int issue_flags)
>  {
>  	struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
> @@ -3322,7 +3328,7 @@ static int io_read(struct io_kiocb *req, unsigned int issue_flags)
>  	} else if (ret == -EIOCBQUEUED) {
>  		goto out_free;
>  	} else if (ret <= 0 || ret == io_size || !force_nonblock ||
> -		   (req->flags & REQ_F_NOWAIT) || !(req->flags & REQ_F_ISREG)) {
> +		   (req->flags & REQ_F_NOWAIT) || !need_read_all(req)) {
>  		/* read all, failed, already did sync or don't want to retry */
>  		goto done;
>  	}
> 

-- 
Pavel Begunkov

Re: [PATCH] io_uring: retry in case of short read on block device

[Jens Axboe]

On 8/21/21 9:07 AM, Ming Lei wrote:
> In case of buffered reading from block device, when short read happens,
> we should retry to read more, otherwise the IO will be completed
> partially, for example, the following fio expects to read 2MB, but it
> can only read 1M or less bytes:
> 
>     fio --name=onessd --filename=/dev/nvme0n1 --filesize=2M \
> 	--rw=randread --bs=2M --direct=0 --overwrite=0 --numjobs=1 \
> 	--iodepth=1 --time_based=0 --runtime=2 --ioengine=io_uring \
> 	--registerfiles --fixedbufs --gtod_reduce=1 --group_reporting
> 
> Fix the issue by allowing short read retry for block device, which sets
> FMODE_BUF_RASYNC really.

Applied, thanks Ming.

-- 
Jens Axboe