1
The following changes since commit b0fbe46ad82982b289a44ee2495b59b0bad8a842:
1
The following changes since commit 3521ade3510eb5cefb2e27a101667f25dad89935:
2
2
3
Update version for v2.11.0-rc0 release (2017-11-07 16:05:28 +0000)
3
Merge remote-tracking branch 'remotes/thuth-gitlab/tags/pull-request-2021-07-29' into staging (2021-07-29 13:17:20 +0100)
4
4
5
are available in the git repository at:
5
are available in the Git repository at:
6
6
7
git://github.com/stefanha/qemu.git tags/block-pull-request
7
https://gitlab.com/stefanha/qemu.git tags/block-pull-request
8
8
9
for you to fetch changes up to 161d53bbed2be4c4336c098185de3afb13cc6bd5:
9
for you to fetch changes up to cc8eecd7f105a1dff5876adeb238a14696061a4a:
10
10
11
util/async: use atomic_mb_set in qemu_bh_cancel (2017-11-08 09:24:41 +0000)
11
MAINTAINERS: Added myself as a reviewer for the NVMe Block Driver (2021-07-29 17:17:34 +0100)
12
13
----------------------------------------------------------------
14
Pull request
15
16
The main fix here is for io_uring. Spurious -EAGAIN errors can happen and the
17
request needs to be resubmitted.
18
19
The MAINTAINERS changes carry no risk and we might as well include them in QEMU
20
6.1.
12
21
13
----------------------------------------------------------------
22
----------------------------------------------------------------
14
23
15
----------------------------------------------------------------
24
Fabian Ebner (1):
25
block/io_uring: resubmit when result is -EAGAIN
16
26
17
Sergio Lopez (1):
27
Philippe Mathieu-Daudé (1):
18
util/async: use atomic_mb_set in qemu_bh_cancel
28
MAINTAINERS: Added myself as a reviewer for the NVMe Block Driver
19
29
20
Stefan Hajnoczi (1):
30
Stefano Garzarella (1):
21
tests-aio-multithread: fix /aio/multi/schedule race condition
31
MAINTAINERS: add Stefano Garzarella as io_uring reviewer
22
32
23
tests/test-aio-multithread.c | 5 ++---
33
MAINTAINERS | 2 ++
24
util/async.c | 2 +-
34
block/io_uring.c | 16 +++++++++++++++-
25
2 files changed, 3 insertions(+), 4 deletions(-)
35
2 files changed, 17 insertions(+), 1 deletion(-)
26
36
27
--
37
--
28
2.13.6
38
2.31.1
29
39
30
diff view generated by jsdifflib
New patch
1
From: Stefano Garzarella <sgarzare@redhat.com>
1
2
3
I've been working with io_uring for a while so I'd like to help
4
with reviews.
5
6
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
7
Message-Id: <20210728131515.131045-1-sgarzare@redhat.com>
8
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9
---
10
MAINTAINERS | 1 +
11
1 file changed, 1 insertion(+)
12
13
diff --git a/MAINTAINERS b/MAINTAINERS
14
index XXXXXXX..XXXXXXX 100644
15
--- a/MAINTAINERS
16
+++ b/MAINTAINERS
17
@@ -XXX,XX +XXX,XX @@ Linux io_uring
18
M: Aarushi Mehta <mehta.aaru20@gmail.com>
19
M: Julia Suvorova <jusual@redhat.com>
20
M: Stefan Hajnoczi <stefanha@redhat.com>
21
+R: Stefano Garzarella <sgarzare@redhat.com>
22
L: qemu-block@nongnu.org
23
S: Maintained
24
F: block/io_uring.c
25
--
26
2.31.1
27
diff view generated by jsdifflib
New patch
1
From: Fabian Ebner <f.ebner@proxmox.com>
1
2
3
Linux SCSI can throw spurious -EAGAIN in some corner cases in its
4
completion path, which will end up being the result in the completed
5
io_uring request.
6
7
Resubmitting such requests should allow block jobs to complete, even
8
if such spurious errors are encountered.
9
10
Co-authored-by: Stefan Hajnoczi <stefanha@gmail.com>
11
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
12
Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
13
Message-id: 20210729091029.65369-1-f.ebner@proxmox.com
14
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
15
---
16
block/io_uring.c | 16 +++++++++++++++-
17
1 file changed, 15 insertions(+), 1 deletion(-)
18
19
diff --git a/block/io_uring.c b/block/io_uring.c
20
index XXXXXXX..XXXXXXX 100644
21
--- a/block/io_uring.c
22
+++ b/block/io_uring.c
23
@@ -XXX,XX +XXX,XX @@ static void luring_process_completions(LuringState *s)
24
total_bytes = ret + luringcb->total_read;
25
26
if (ret < 0) {
27
- if (ret == -EINTR) {
28
+ /*
29
+ * Only writev/readv/fsync requests on regular files or host block
30
+ * devices are submitted. Therefore -EAGAIN is not expected but it's
31
+ * known to happen sometimes with Linux SCSI. Submit again and hope
32
+ * the request completes successfully.
33
+ *
34
+ * For more information, see:
35
+ * https://lore.kernel.org/io-uring/20210727165811.284510-3-axboe@kernel.dk/T/#u
36
+ *
37
+ * If the code is changed to submit other types of requests in the
38
+ * future, then this workaround may need to be extended to deal with
39
+ * genuine -EAGAIN results that should not be resubmitted
40
+ * immediately.
41
+ */
42
+ if (ret == -EINTR || ret == -EAGAIN) {
43
luring_resubmit(s, luringcb);
44
continue;
45
}
46
--
47
2.31.1
48
diff view generated by jsdifflib
New patch
1
From: Philippe Mathieu-Daudé <philmd@redhat.com>
1
2
3
I'm interested in following the activity around the NVMe bdrv.
4
5
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
6
Message-id: 20210728183340.2018313-1-philmd@redhat.com
7
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
8
---
9
MAINTAINERS | 1 +
10
1 file changed, 1 insertion(+)
11
12
diff --git a/MAINTAINERS b/MAINTAINERS
13
index XXXXXXX..XXXXXXX 100644
14
--- a/MAINTAINERS
15
+++ b/MAINTAINERS
16
@@ -XXX,XX +XXX,XX @@ F: block/null.c
17
NVMe Block Driver
18
M: Stefan Hajnoczi <stefanha@redhat.com>
19
R: Fam Zheng <fam@euphon.net>
20
+R: Philippe Mathieu-Daudé <philmd@redhat.com>
21
L: qemu-block@nongnu.org
22
S: Supported
23
F: block/nvme*
24
--
25
2.31.1
26
diff view generated by jsdifflib