Hi all,
I create a vm with a macvtap network and a nat network.
```
netdev user,id=fl.1,hostfwd=tcp::-:22 \
-device e1000,netdev=fl.1 \
-net nic,model=virtio,macaddr=$(< /sys/class/net/macvtap0/address) \
-net tap,fd=3 3<>/dev/tap$(< /sys/class/net/macvtap0/ifindex)
```
I create two network
1 ↵
read 1073741824/1073741824 bytes at offset 0
1 GiB, 1 ops; 0:00:01.94 (526.770 MiB/sec and 0.5144 ops/sec)
Signed-off-by: Li Feng
---
block/nvme.c | 16 ++--
1 file changed, 6 insertions(+), 10 deletions(-)
diff --git a/block
Mark dirty as page, the step of each call is 1.
Signed-off-by: Li Feng
---
contrib/libvhost-user/libvhost-user.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/contrib/libvhost-user/libvhost-user.c
b/contrib/libvhost-user/libvhost-user.c
index e08d6c7b97..2689de6d1c 100644
,
Feng Li
Igor Mammedov 于2020年4月8日周三 下午6:40写道：
>
> On Wed, 8 Apr 2020 10:25:42 +0800
> Li Feng wrote:
>
> > Hi all,
> >
> > Hotplug of vhost-user-blk doesn't not work in qemu master branch and
> > all previous version.
> >
> > The action I insert a vhost
oned by user:f7
Container overflow: fc
Array cookie:ac
Intra object redzone:bb
ASan internal: fe
Left alloca redzone: ca
Right alloca redzone:cb
==7309==ABORTING
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 4
hw/vi
PEN)
-> vhost_user_blk_connect
-> vhost_dev_init
-> vhost_user_blk_event(CLOSE)
-> vhost_dev_cleanup
Li Feng (4):
vhost-user-blk: delay vhost_user_blk_disconnect
vhost-user-blk: fix invalid memory access
char-socket: avoid double call tcp_chr_
nected is true, otherwise the
hdev->host_ops will be nullptr.
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 47 +--
1 file changed, 25 insertions(+), 22 deletions(-)
diff --git a/hw/block/vhost-user-blk.c b/hw/block/vhost-user-blk.c
index 1
Since commit b0a335e351103bf92f3f9d0bd5759311be8156ac, a socket write
may trigger a disconnect events, calling vhost_user_blk_disconnect() and
clearing all the vhost_dev strutures. Then the next socket read will
encounter an invalid pointer to vhost_dev.
Signed-off-by: Li Feng
---
hw/block
double call tcp_chr_free_connection generates a crash.
Signed-off-by: Li Feng
---
chardev/char-socket.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/chardev/char-socket.c b/chardev/char-socket.c
index 185fe38dda..43aab8f24b 100644
--- a/chardev/char-socket.c
+++ b/chardev/char
Hi all,
Hotplug of vhost-user-blk doesn't not work in qemu master branch and
all previous version.
The action I insert a vhost-user-blk disk is:
(qemu) chardev-add socket,id=spdk_vhost_blk2,path=/vhost-blk.0,reconnect=1
(qemu) device_add
when writing tests
code(tests/test-char.c ).
Any tips?
Thanks,
Feng Li
Marc-André Lureau 于2020年4月15日周三 下午6:35写道：
>
> Hi
>
> On Wed, Apr 15, 2020 at 5:31 AM Li Feng wrote:
> >
> > double call tcp_chr_free_connection generates a crash.
> >
> > Signed-off-by: Li Feng
all the vhost_dev strutures holding data that vhost.c functions expect
to remain valid. Delay the cleanup to keep the vhost_dev structure
valid during the vhost.c functions.
Feng Li
Raphael Norwitz 于2020年4月17日周五 上午11:41写道：
>
> On Wed, Apr 15, 2020 at 11:28:23AM +0800, Li Feng
Add mail group list.
Thank you, Raphael .
Raphael Norwitz 于2020年4月17日周五 下午12:10写道：
>
> Mostly looks good - just a few superficial notes.
>
> On Wed, Apr 15, 2020 at 11:28:26AM +0800, Li Feng wrote:
> > 1. set s->connected to true after vhost_dev_init;
> > 2. call
OK, I will submit this patch "vhost-user-blk: fix invalid memory
access" firstly.
This is unrelated with other three and it has been acked.
Thanks,
Feng Li
Michael S. Tsirkin 于2020年4月17日周五 下午5:45写道：
>
> On Wed, Apr 15, 2020 at 11:28:22AM +0800, Li Feng wrote:
> > The
oned by user:f7
Container overflow: fc
Array cookie:ac
Intra object redzone:bb
ASan internal: fe
Left alloca redzone: ca
Right alloca redzone:cb
==7309==ABORTING
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 4
hw/vi
Nothing changed.
Just separate this patch from those patchset series.
Thanks,
Feng Li
Philippe Mathieu-Daudé 于2020年4月17日周五 下午6:55写道：
>
> Hi Li,
>
> On 4/17/20 12:17 PM, Li Feng wrote:
> > when s->inflight is freed, vhost_dev_free_inflight may try to access
This patch is trying to fix the same issue with me.
However, our fix is different.
I think that check the s->reconnect_timer is better.
Thanks,
Feng Li
Marc-André Lureau 于2020年4月24日周五 上午3:16写道：
>
> Hi
>
> On Thu, Apr 23, 2020 at 8:41 PM Dima Stepanov wrote:
> >
> > During vhost-user
0555a9556 in qemu_thread_start (args=) at
util/qemu-thread-posix.c:519
#10 0x75e61e25 in start_thread () from /lib64/libpthread.so.0
#11 0x75b8bbad in clone () from /lib64/libc.so.6
I think this is a new issue of qemu, not my test issue.
How do you think?
Thanks,
Feng Li
L
: qemu_chr_socket_restart_timer: Assertion
`!s->reconnect_timer' failed.
Signed-off-by: Li Feng
---
v2:
- Rewrite the solution.
- Add test to reproduce this issue.
chardev/char-socket.c | 2 +-
tests/test-char.c | 48 ++--
2 files changed, 39 insertions(+), 11 deleti
: qemu_chr_socket_restart_timer: Assertion
`!s->reconnect_timer' failed.
Signed-off-by: Li Feng
---
v2:
- Rewrite the solution.
- Add test to reproduce this issue.
chardev/char-socket.c | 2 +-
tests/test-char.c | 48 ++--
2 files changed, 39 insertions(+), 11 deleti
Hi, Dima.
If vhost_migration_log return < 0, then vhost_log_global_start will
trigger a crash.
Does your patch have process this abort?
If a disconnect happens in the migration stage, the correct operation
is to stop the migration, right?
841 static void vhost_log_global_start(MemoryListener
2:22PM +0800, Li Feng wrote:
> > Marc-André Lureau 于2020年5月8日周五 下午8:32写道：
> > >
> > > Hi
> > >
> > > On Fri, May 8, 2020 at 7:14 AM Li Feng wrote:
> > > >
> > > > Root cause:
> > > > From `man recvmsg`, the RETURN VAL
: qemu_chr_socket_restart_timer: Assertion
`!s->reconnect_timer' failed.
Signed-off-by: Li Feng
---
v4:
- remove the wrong patch
- fix the char_socket_ping_pong to support the reconnect exception test
v3:
- add a patch to fix a crash when recvmsg return 0
- make the tests reproduce the two crash
v2:
- add unit t
Marc-André Lureau 于2020年5月8日周五 下午8:32写道：
>
> Hi
>
> On Fri, May 8, 2020 at 7:14 AM Li Feng wrote:
> >
> > Root cause:
> > From `man recvmsg`, the RETURN VALUE says:
> > These calls return the number of bytes received, or -1 if an error
> > occurred.
Have you applied the first one?
io/channel: fix crash when qio_channel_readv_all return 0
Thanks,
Feng Li
Marc-André Lureau 于2020年5月8日周五 下午7:01写道：
>
> On Fri, May 8, 2020 at 7:14 AM Li Feng wrote:
> >
> > When the disconnect event is triggered in th
Thanks!
Feng Li
Marc-André Lureau 于2020年5月8日周五 下午8:29写道：
>
> Hi
>
> On Fri, May 8, 2020 at 2:16 PM Li Feng wrote:
> >
> > Have you applied the first one?
> > io/channel: fix crash when qio_channel_readv_all return 0
> >
>
> Sorry, I missed that one
Hi, Dima.
This abort is what I have mentioned in my previous email.
I have triggered this crash without any fix a week ago.
And I have written a test patch to let vhost_log_global_start return
int and propagate the error to up layer.
However, my change is a little large, because the origin
Dear all,
I have encountered a weird crash.
I remember before a few days it works well and I rebase my code from upstream.
This is the command:
/root/qemu-master/x86_64-softmmu/qemu-system-x86_64 -enable-kvm
-device virtio-balloon -cpu host -smp 4 -m 2G -drive
L1 kernel is :
(base) 20-05-14 23:31:32 root@31_216:~ uname -a
Linux 31_216 5.6.7-1.el7.centos.x86_64 #1 SMP Mon Apr 27 15:26:08 CST
2020 x86_64 x86_64 x86_64 GNU/Linux
The OS is CentOS 7.
Thanks,
Feng Li
Li Feng 于2020年5月14日周四 下午11:10写道：
>
> EXSi CPU is : Intel(R) Xeon(R) CPU E5-2
Dr. David Alan Gilbert 于2020年5月14日周四 下午11:16写道：
>
> * Li Feng (fen...@smartx.com) wrote:
> > EXSi CPU is : Intel(R) Xeon(R) CPU E5-2640 v3 @ 2.60GHz
> > This is my vm, I run qemu in it.
>
> Do you know what the real hardware is?
What information do you need? I could send
Cc'ing David/Paolo in case they have a clue...
> >
> > On 5/14/20 1:27 PM, Li Feng wrote:
> > > Dear all,
> > >
> > > I have encountered a weird crash.
> > > I remember before a few days it works well and I rebase my code from
> > > upstream.
&g
Dr. David Alan Gilbert 于2020年5月14日周四 下午11:31写道：
>
> * Li Feng (fen...@smartx.com) wrote:
> > Dr. David Alan Gilbert 于2020年5月14日周四 下午11:16写道：
> > >
> > > * Li Feng (fen...@smartx.com) wrote:
> > > > EXSi CPU is : Intel(R) Xeon(R) CPU E5-2640 v3 @
Thanks,
Feng Li
Dima Stepanov 于2020年4月30日周四 下午9:36写道：
>
> During testing of the vhost-user-blk reconnect functionality the qemu
> SIGSEGV was triggered:
> start qemu as:
> x86_64-softmmu/qemu-system-x86_64 -m 1024M -M q35 \
>-object
>
the G_IO_HUP is watched in tcp_chr_connect, and the callback
vhost_user_blk_watch is not needed, because tcp_chr_hup is registered as
callback. And it will close the tcp link.
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 19 ---
include/hw/virtio/vhost-user
: qemu_chr_socket_restart_timer: Assertion
`!s->reconnect_timer' failed.
Signed-off-by: Li Feng
---
chardev/char-socket.c | 2 +-
tests/test-char.c | 68 ---
2 files changed, 55 insertions(+), 15 deletions(-)
diff --git a/chardev/char-socket.c b/chardev/c
556ae12 in char_socket_client_server_thread
(data=data@entry=0x5582e350) at tests/test-char.c:891
#9 0x555a95b6 in qemu_thread_start (args=) at
util/qemu-thread-posix.c:519
#10 0x75e61e25 in start_thread () from /lib64/libpthread.so.0
#11 0x000075b8bbad in clone ()
Paolo, Million thanks.
I will try to dig into this code in my spare time.
Thanks,
Feng Li
Paolo Bonzini 于2020年5月20日周三 下午6:27写道：
>
> On 20/05/20 12:15, Li Feng wrote:
> > /root/qemu-master/x86_64-softmmu/qemu-system-x86_64 -enable-kvm
> > -device virtio-balloon -cpu
> &g
Hi, Any update about this issue?
Thanks,
Feng Li
Li Feng 于2020年5月14日周四 下午11:49写道：
>
> Dr. David Alan Gilbert 于2020年5月14日周四 下午11:31写道：
> >
> > * Li Feng (fen...@smartx.com) wrote:
> > > Dr. David Alan Gilbert 于2020年5月14日周四 下午11:16写道：
> > > >
>
: qemu_chr_socket_restart_timer: Assertion
`!s->reconnect_timer' failed.
Signed-off-by: Li Feng
---
v5:
- rebase to master
v4:
- remove the wrong patch
- fix the char_socket_ping_pong to support the reconnect exception test
v3:
- add a patch to fix a crash when recvmsg return 0
- make the tests reproduce the two cr
Qemu will send GET_INFLIGHT_FD and SET_INFLIGH_FD to backend, and
the backend setup the inflight memory to track the io.
Change-Id: I805d6189996f7a1b44c65f0b12ef7473b1789510
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 27 +++
include/hw/virtio
e based on the similar changes for the vhost-user-net
> > device:
> > commit e7c83a885f865128ae3cf1946f8cb538b63cbfba
> > "vhost-user: delay vhost_user_stop"
> >
> > Signed-off-by: Dima Stepanov
>
> Reviewed-by: Raphael Norwitz
>
> Li Feng - would you also like to sign
Paolo Bonzini 于2020年5月20日周三 下午3:28写道：
>
> On 20/05/20 05:50, Li Feng wrote:
> >>>>>>>>> qemu-system-x86_64: error: failed to set MSR 0x48f to
> >>>>>>>>> 0x7fefff00036dfb
>
> The high 32 bits are 0111 111
Paolo Bonzini 于2020年5月20日周三 下午4:32写道：
>
> On 20/05/20 10:22, Li Feng wrote:
> >> - bit 2, vmx-exit-nosave-debugctl
> >> - bit 9, host address space size, is handled automatically by QEMU
> >> - bit 15, vmx-exit-ack-intr
> >> - bit 17, vmx-exit-save-pat
&
Paolo Bonzini 于2020年5月20日周三 下午6:10写道：
>
> On 20/05/20 11:36, Li Feng wrote:
> >> Can you do "rdmsr 0x48f" (as root)? On my machine for example I get
> >> 1ff00036dfb.
> >>
> > $ rdmsr 0x48f
> > 3f00036dfb
> >
>
> Hmm,
Paolo Bonzini 于2020年5月20日周三 下午5:34写道：
>
> On 20/05/20 10:55, Li Feng wrote:
> >>> After computing xor, the result is 0b10.
> >>> The bit 9 is set.
> >> Are you running KVM with a 32-bit kernel.
> >>
> > No, I'm running 64 bit.
> &g
Hi Lureau,
When will this patch be merged?
Thanks, Feng Li.
-- Forwarded message -
发件人： Dima Stepanov
Date: 2020年6月15日周一 下午4:43
Subject: Re: [PATCH v5] char-socket: initialize reconnect timer only
when the timer doesn't start
To: Li Feng
Hi,
I don't see this change merged
ilesystem manually, and it's probably completely impossible in a
> test case without root permissions.
>
> So maybe just add a more detailed description of the bug to the commit
> message, and we'll have to apply it without a test.
>
> Kevin
>
> > Li Feng 于2020年12月4日周五 下午6
he syscall of 'fcnctl' will return non-zero.
Signed-off-by: Li Feng
---
block/file-posix.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index d5fd1dbcd2..806764f7e3 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -3104
Daniel P. Berrangé 于2020年12月8日周二 下午9:45写道：
>
> On Tue, Dec 08, 2020 at 08:59:37PM +0800, Li Feng wrote:
> > This patch addresses this issue:
> > When accessing a volume on an NFS filesystem without supporting the file
> > lock,
> > tools, like qemu-img, will com
Kevin Wolf 于2020年12月8日周二 下午10:38写道：
>
> Am 08.12.2020 um 13:59 hat Li Feng geschrieben:
> > This patch addresses this issue:
> > When accessing a volume on an NFS filesystem without supporting the file
> > lock,
> > tools, like qemu-img, will com
ly, the file.locking is per-drive property,
which depends on the underlay filesystem.
In this patch, make the 'qemu_has_ofd_lock' with a filename be more
generic and reasonable.
Signed-off-by: Li Feng
---
v2: remove the refactoring.
---
block/file-posix.c | 32 ++
ly, the file.locking is per-drive property,
which depends on the underlay filesystem.
In this patch, make the 'qemu_has_ofd_lock' with a filename be more generic
and reasonable.
Signed-off-by: Li Feng
---
block/file-posix.c | 32 +++-
include/qemu/osdep.h
Kevin Wolf 于2020年12月10日周四 上午1:43写道：
>
> Am 09.12.2020 um 10:33 hat Daniel P. Berrangé geschrieben:
> > On Tue, Dec 08, 2020 at 03:38:22PM +0100, Kevin Wolf wrote:
> > > Am 08.12.2020 um 13:59 hat Li Feng geschrieben:
> > > > This patch addresses this issue
When setting the file.locking = false, we shouldn't set the lock.
Signed-off-by: Li Feng
---
block/file-posix.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index d5fd1dbcd2..806764f7e3 100644
--- a/block/file-posix.c
+++ b/block
byte 100
I will look at the iotest 182 and try to add a test.
Thanks,
Feng Li
Kevin Wolf 于2020年12月4日周五 下午6:40写道：
>
> Am 04.12.2020 um 11:28 hat Li Feng geschrieben:
> > When setting the file.locking = false, we shouldn't set the lock.
> >
> > Signed-off-by: Li Feng
&g
ve is auto mode, use the 'qemu_has_file_lock' to set the toggle.
Signed-off-by: Li Feng
---
v3: don't call the qemu_has_ofd_lock, use a new function instead.
v2: remove the refactoring.
---
block/file-posix.c | 30 +-
include/qemu/osdep.h | 1 +
util/osdep.c
ly, the file.locking is per-drive property,
which depends on the underlay filesystem.
In this patch, add a new 'qemu_has_file_lock' to detect whether the
file supports the file lock. And disable the lock when the underlay file
system doesn't support locks.
Signed-off-by: Li Feng
Daniel P. Berrangé 于2020年12月15日周二 下午6:08写道：
>
> On Tue, Dec 15, 2020 at 03:09:28PM +0800, Li Feng wrote:
> > This patch addresses this issue:
> > When accessing a volume on an NFS filesystem without supporting the file
> > lock,
> > tools, like qemu-img, will com
ly, the file.locking is per-drive property,
which depends on the underlay filesystem.
In this patch, add a new 'qemu_has_file_lock' to detect whether the
file supports the file lock. And disable the lock when the underlay file
system doesn't support locks.
Signed-off-by: Li Feng
---
v5: simpli
ly, the file.locking is per-drive property,
which depends on the underlay filesystem.
In this patch, make the 'qemu_has_ofd_lock' with a filename be more
generic and reasonable.
Signed-off-by: Li Feng
---
v2: remove the refactoring.
---
block/file-posix.c | 32 ++
> On 1 Sep 2023, at 8:00 PM, Markus Armbruster wrote:
>
> Li Feng mailto:fen...@smartx.com>> writes:
>
>> Add a Error parameter to report the real error, like vhost-user-blk.
>>
>> Signed-off-by: Li Feng
>> ---
>> hw/scsi/vhost-scsi-common.c
the v1 patch to small separate patchset;
- New patch for fixing fd leak, which has sent to reviewers in another
mail;
- Implement the `vhost_user_scsi_handle_output`;
- Add the started_vu safe check;
- Fix error handler;
- Check the inflight before set/get inflight fd.
Li Feng (5):
vhost-user
Let's keep the same behavior as vhost-user-blk.
Some old guests kick virtqueue before setting VIRTIO_CONFIG_S_DRIVER_OK.
Signed-off-by: Li Feng
---
hw/scsi/vhost-user-scsi.c | 51 +++
1 file changed, 46 insertions(+), 5 deletions(-)
diff --git a/hw/scsi
.
This patch is a preparation for the following patches.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 37 ++---
1 file changed, 18 insertions(+), 19 deletions(-)
diff --git a/hw/scsi/vhost-scsi-common.c b/hw/scsi/vhost-scsi-common.c
index a06f01af26
a07d ("hw/virtio: generalise CHR_EVENT_CLOSED handling")
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 2 +-
hw/virtio/vhost-user-gpio.c| 2 +-
hw/virtio/vhost-user.c | 9 +++--
include/hw/virtio/vhost-user.h | 3 ++-
4 files changed, 11 insertions(+), 5 delet
Multiple devices need this macro, move it to a common header.
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/block/vhost-user-blk.c | 4 +---
hw/virtio/vhost-user-gpio.c | 3 +--
include/hw/virtio/vhost.h | 2 ++
3 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/hw
> On 1 Sep 2023, at 7:44 PM, Markus Armbruster wrote:
>
> Li Feng mailto:fen...@smartx.com>> writes:
>
>> Let's keep the same behavior as vhost-user-blk.
>>
>> Some old guests kick virtqueue before setting VIRTIO_CONFIG_S_DRIVER_OK.
>>
>> Sign
If the backend crashes and restarts, the device is broken.
This patch adds reconnect for vhost-user-scsi.
This patch also improves the error messages, and reports some silent errors.
Tested with spdk backend.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 16 +-
hw/scsi
> On 1 Sep 2023, at 8:00 PM, Markus Armbruster wrote:
>
> Li Feng mailto:fen...@smartx.com>> writes:
>
>> If the backend crashes and restarts, the device is broken.
>> This patch adds reconnect for vhost-user-scsi.
>>
>> Tested with spdk backend.
&
/
Changes for v3:
- Fix the code style.
Changes for v2:
- Add a event_cb in VhostAsyncCallback to be called when dev is NULL;
- Fix the error report message.
Li Feng (2):
vhost-user: Fix lost reconnect
vhost: Add Error parameter to vhost_scsi_common_start()
hw/block/vhost-user-blk.c
Add a Error parameter to report the real error, like vhost-user-blk.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 16 +---
hw/scsi/vhost-scsi.c | 5 +++--
hw/scsi/vhost-user-scsi.c | 14 --
include/hw/virtio/vhost-scsi
> On 30 Aug 2023, at 6:11 AM, Raphael Norwitz
> wrote:
>
>
>
>> On Aug 24, 2023, at 3:41 AM, Li Feng wrote:
>>
>> When the vhost-user is reconnecting to the backend, and if the vhost-user
>> fails
>> at the get_features in v
a07d ("hw/virtio: generalise CHR_EVENT_CLOSED handling")
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 2 +-
hw/scsi/vhost-user-scsi.c | 3 ++-
hw/virtio/vhost-user-gpio.c| 2 +-
hw/virtio/vhost-user.c | 9 +++--
include/hw/virtio/vhost-user.h | 3 ++-
.
This patch is a preparation for the following patches.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 37 ++---
1 file changed, 18 insertions(+), 19 deletions(-)
diff --git a/hw/scsi/vhost-scsi-common.c b/hw/scsi/vhost-scsi-common.c
index a06f01af26
If the backend crashes and restarts, the device is broken.
This patch adds reconnect for vhost-user-scsi.
This patch also improves the error messages, and reports some silent errors.
Tested with spdk backend.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 16 +-
hw/scsi
a07d ("hw/virtio: generalise CHR_EVENT_CLOSED handling")
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 2 +-
hw/scsi/vhost-user-scsi.c | 3 ++-
hw/virtio/vhost-user-gpio.c| 2 +-
hw/virtio/vhost-user.c | 10 --
include/hw/virtio/vhost-user.h | 3 +
Let's keep the same behavior as vhost-user-blk.
Some old guests kick virtqueue before setting VIRTIO_CONFIG_S_DRIVER_OK.
Signed-off-by: Li Feng
---
hw/scsi/vhost-user-scsi.c | 48 +++
1 file changed, 44 insertions(+), 4 deletions(-)
diff --git a/hw/scsi
Multiple devices need this macro, move it to a common header.
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/block/vhost-user-blk.c | 4 +---
hw/virtio/vhost-user-gpio.c | 3 +--
include/hw/virtio/vhost.h | 2 ++
3 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/hw
patch to small separate patchset;
- New patch for fixing fd leak, which has sent to reviewers in another
mail;
- Implement the `vhost_user_scsi_handle_output`;
- Add the started_vu safe check;
- Fix error handler;
- Check the inflight before set/get inflight fd.
Li Feng (5):
vhost-user-common
If the backend crashes and restarts, the device is broken.
This patch adds reconnect for vhost-user-scsi.
This patch also improves the error messages, and reports some silent errors.
Tested with spdk backend.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 16 +-
hw/scsi
.
This patch is a preparation for the following patches.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 37 ++---
1 file changed, 18 insertions(+), 19 deletions(-)
diff --git a/hw/scsi/vhost-scsi-common.c b/hw/scsi/vhost-scsi-common.c
index a06f01af26
Let's keep the same behavior as vhost-user-blk.
Some old guests kick virtqueue before setting VIRTIO_CONFIG_S_DRIVER_OK.
Signed-off-by: Li Feng
---
hw/scsi/vhost-user-scsi.c | 48 +++
1 file changed, 44 insertions(+), 4 deletions(-)
diff --git a/hw/scsi
a07d ("hw/virtio: generalise CHR_EVENT_CLOSED handling")
Signed-off-by: Li Feng
---
hw/block/vhost-user-blk.c | 2 +-
hw/scsi/vhost-user-scsi.c | 3 ++-
hw/virtio/vhost-user-gpio.c| 2 +-
hw/virtio/vhost-user.c | 9 +++--
include/hw/virtio/vhost-user.h | 3 ++-
;
- New patch for fixing fd leak, which has sent to reviewers in another
mail;
- Implement the `vhost_user_scsi_handle_output`;
- Add the started_vu safe check;
- Fix error handler;
- Check the inflight before set/get inflight fd.
Li Feng (5):
vhost-user-common: send get_inflight_fd once
vhost
Multiple devices need this macro, move it to a common header.
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/block/vhost-user-blk.c | 4 +---
hw/virtio/vhost-user-gpio.c | 3 +--
include/hw/virtio/vhost.h | 2 ++
3 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/hw
a07d ("hw/virtio: generalise CHR_EVENT_CLOSED handling")
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/block/vhost-user-blk.c | 2 +-
hw/scsi/vhost-user-scsi.c | 3 ++-
hw/virtio/vhost-user-gpio.c| 2 +-
hw/virtio/vhost-user.c | 10 --
incl
If the backend crashes and restarts, the device is broken.
This patch adds reconnect for vhost-user-scsi.
This patch also improves the error messages, and reports some silent errors.
Tested with spdk backend.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 16 +-
hw/scsi
.
This patch is a preparation for the following patches.
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/scsi/vhost-scsi-common.c | 37 ++---
1 file changed, 18 insertions(+), 19 deletions(-)
diff --git a/hw/scsi/vhost-scsi-common.c b/hw/scsi/vhost-scsi
Thanks for your comments, I will submit the v8.
> On 8 Oct 2023, at 6:46 PM, Manos Pitsidianakis
> wrote:
>
> Hello Li, I have some trivial style comments you could possibly address in a
> next version:
>
> On Sun, 08 Oct 2023 12:12, Li Feng wrote:
>> diff --git
Multiple devices need this macro, move it to a common header.
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/block/vhost-user-blk.c | 4 +---
hw/virtio/vhost-user-gpio.c | 3 +--
include/hw/virtio/vhost.h | 2 ++
3 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/hw
reviewers in another
mail;
- Implement the `vhost_user_scsi_handle_output`;
- Add the started_vu safe check;
- Fix error handler;
- Check the inflight before set/get inflight fd.
Li Feng (5):
vhost-user-common: send get_inflight_fd once
vhost: move and rename the conn retry times
vhost-user-scs
Let's keep the same behavior as vhost-user-blk.
Some old guests kick virtqueue before setting VIRTIO_CONFIG_S_DRIVER_OK.
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/scsi/vhost-user-scsi.c | 48 +++
1 file changed, 44 insertions(+), 4
On Fri, Sep 29, 2023 at 8:55 AM Raphael Norwitz
wrote:
>
>
>
> > On Sep 22, 2023, at 7:46 AM, Li Feng wrote:
> >
> > Currently the get_inflight_fd will be sent every time the device is
> > started, and
> > the backend will allocate shared memory to save
you? Otherwise do you think we should also
> fix up the vhost-user-blk realize function?
>
> > On Sep 22, 2023, at 7:46 AM, Li Feng wrote:
> >
> > If the backend crashes and restarts, the device is broken.
> > This patch adds reconnect for vhost-user-scsi.
> &
Hello Guys,
Ping…
These patches have been waiting for a long time. Can they be merged?
Best Regards,
li
> On 9 Oct 2023, at 12:46 PM, Li Feng wrote:
>
> Changes for v8:
> - [PATCH 3/5] vhost-user-scsi: support reconnect to backend
> - Fix code style suggested by Mano
On Sun, Oct 8, 2023 at 4:51 PM Michael S. Tsirkin wrote:
>
> On Sun, Oct 08, 2023 at 04:49:05PM +0800, Li Feng wrote:
> > On Fri, Sep 29, 2023 at 8:55 AM Raphael Norwitz
> > wrote:
> > >
> > >
> > >
> > > > On Sep 22, 20
On Sun, Oct 8, 2023 at 4:49 PM Michael S. Tsirkin wrote:
>
> On Fri, Sep 22, 2023 at 07:46:10PM +0800, Li Feng wrote:
> > Changes for v6:
> > - [PATCH] vhost-user: fix lost reconnect
> > - Fix missing assign event_cb.
>
>
> Pls don't make vN+1 a reply to vN
If the backend crashes and restarts, the device is broken.
This patch adds reconnect for vhost-user-scsi.
This patch also improves the error messages, and reports some silent errors.
Tested with spdk backend.
Signed-off-by: Li Feng
---
hw/scsi/vhost-scsi-common.c | 16 +-
hw/scsi
Multiple devices need this macro, move it to a common header.
Signed-off-by: Li Feng
Reviewed-by: Raphael Norwitz
---
hw/block/vhost-user-blk.c | 4 +---
hw/virtio/vhost-user-gpio.c | 3 +--
include/hw/virtio/vhost.h | 2 ++
3 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/hw
her
mail;
- Implement the `vhost_user_scsi_handle_output`;
- Add the started_vu safe check;
- Fix error handler;
- Check the inflight before set/get inflight fd.
Li Feng (5):
vhost-user-common: send get_inflight_fd once
vhost: move and rename the conn retry times
vhost-user-scsi: support
1 - 100 of 155 matches
Mail list logo
