On Mon, Jan 27, 2020 at 04:36:31PM +, Liam Merwick wrote:
> The second param in extract_from_deb() is 'path' not 'file'
>
> Signed-off-by: Liam Merwick
> ---
> tests/acceptance/boot_linux_console.py | 2 +-
> 1 file changed, 1 insertion(+), 1 deleti
est booting a
> compressed kernel using either machine class.
>
> Signed-off-by: Liam Merwick
> ---
> tests/acceptance/boot_linux_console.py | 15 +--
> 1 file changed, 13 insertions(+), 2 deletions(-)
Reviewed-by: Stefano Garzarella
>
> diff --git
rtion(+)
Reviewed-by: Stefano Garzarella
>
> diff --git a/.travis.yml b/.travis.yml
> index 1ae645e9fcec..3d8c2a38e679 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -278,6 +278,7 @@ matrix:
> - python3-pil
> - python3-pip
>
+process.run("rpm2cpio %s | cpio -id %s" % (rpm, path), shell=True)
> +os.chdir(cwd)
> +return self.workdir + '/' + path
^
Is the extra slash needed? (just because the extract_from_deb()
doesn't put
On Mon, Jan 27, 2020 at 04:36:34PM +, Liam Merwick wrote:
> Add tests to boot an uncompressed kernel using the x86/HVM direct boot ABI.
> The vmlinux binary is obtained from a small RPM for Kata containers and
> extracted using the new extract_from_rpm() method.
>
> Signed-off-by: Liam Merwick
On Thu, Jan 30, 2020 at 04:28:52PM +, Liam Merwick wrote:
> On 30/01/2020 15:34, Liam Merwick wrote:
> > On 30/01/2020 11:27, Stefano Garzarella wrote:
> > > Hi Liam,
> > >
> > > On Mon, Jan 27, 2020 at 04:36:30PM +, Liam Merwick wrote:
> > &g
On Thu, May 21, 2020 at 03:35:59PM +0100, Stefan Hajnoczi wrote:
> On Fri, May 15, 2020 at 05:21:07PM +0200, Stefano Garzarella wrote:
> > This series add vhost-user-vsock base device and its PCI version
> > (vhost-user-vsock-pci).
> > The vhost-user-vsock device can be
Rust using crates from Cloud Hypervisor.
The application implements the Firecracker hybrid vsock (vsock over
unix domain socket) [2]
The vhost-user-vsock-pci device can be tested following these steps:
# build vhost-user-vsock application
git clone https://github.com/stefano-garzar
Add the PCI version of vhost-user-vsock
Launch QEMU like this:
qemu -chardev socket,path=/tmp/vm.vsock,id=chr0 \
-device vhost-user-vsock-pci,chardev=chr0
Signed-off-by: Stefano Garzarella
---
hw/virtio/vhost-user-vsock-pci.c | 84
hw/virtio
This patch introduces a vhost-user device for vsock, using the
vhost-vsock-common parent class.
The vhost-user-vsock device can be used to implement the virtio-vsock
device emulation in user-space.
Signed-off-by: Stefano Garzarella
---
v1 -> v2:
- removed live migration code since vhost-u
This patch prepares the introduction of vhost-user-vsock, moving
the common code usable for both vhost-vsock and vhost-user-vsock
devices, in the new vhost-vsock-common parent class.
While moving the code, fixed checkpatch warnings about block comments.
Signed-off-by: Stefano Garzarella
Hi Yang,
On Tue, May 26, 2020 at 03:18:26PM +0800, Yang Zhong wrote:
> Hello all,
>
> The current PVH has been implemented in Qemu and Qboot/uefi, i am not
> familiar with the PVH implementation history, and found there is no
> PVH support in Seabios, my question is do we have plan to support
> P
declarations).
>
> Signed-off-by: Philippe Mathieu-Daudé
> ---
> scripts/git.orderfile | 3 +++
> 1 file changed, 3 insertions(+)
Reviewed-by: Stefano Garzarella
>
> diff --git a/scripts/git.orderfile b/scripts/git.orderfile
> index e89790941c..1f747b583a 100644
&
Hi Florian,
the patch LGTM but there are few style issues to fix:
On Tue, Jan 07, 2020 at 12:13:40PM +0100, Florian Florensa wrote:
> Starting from ceph Nautilus, RBD has support for namespaces, allowing
> for finer grain ACLs on images inside a pool, and tenant isolation.
>
> In the rbd cli tool
> block/rbd.c | 44 +++-
> qapi/block-core.json | 4
> 2 files changed, 35 insertions(+), 13 deletions(-)
Reviewed-by: Stefano Garzarella
Thanks,
Stefano
>
> diff --git a/block/rbd.c b/block/rbd.c
> index 027cbcc
M, but I don't have a lot of experience with io_uring until
now, so
Acked-by: Stefano Garzarella
I really interested on it and I'll try to contribute on this new AIO engine.
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 740401bcbb..fc7f53b229 100644
> --- a/MAINTAINERS
>
On Wed, Dec 18, 2019 at 04:32:21PM +, Stefan Hajnoczi wrote:
> From: Aarushi Mehta
>
> Signed-off-by: Aarushi Mehta
> Reviewed-by: Maxim Levitsky
> Signed-off-by: Stefan Hajnoczi
> ---
> block/file-posix.c | 95 --
> 1 file changed, 75 insertion
The series LGTM, just some comments on patches 4 and 8.
I succefully tried iotests on raw and qcow2 with io_uring, so
Acked-by: Stefano Garzarella
Thanks,
Stefano
On Wed, Dec 18, 2019 at 04:32:13PM +, Stefan Hajnoczi wrote:
> v12:
> * Reword BlockdevAioOptions QAPI schema
> hw/virtio/vhost-vsock.c | 9 +++--
> include/hw/virtio/vhost-vsock.h | 2 ++
> 2 files changed, 9 insertions(+), 2 deletions(-)
Reviewed-by: Stefano Garzarella
>
> diff --git a/hw/virtio/vhost-vsock.c b/hw/virtio/vhost-vsock.c
> index f5744363a8..896c0174c1 100
On Tue, Jan 14, 2020 at 5:45 PM Stefan Hajnoczi wrote:
>
> On Tue, Jan 14, 2020 at 03:52:29PM +0800, pannengy...@huawei.com wrote:
> > From: Pan Nengyuan
> >
> > Receive/transmit/event vqs forgot to cleanup in vhost_vsock_unrealize. This
> > patch save receive/transmit vq pointer in realize() and
Changes V2 to V1:
> - delete virtqueues after vhost cleanup to avoid use-after-free
> - aslo delete virtqueues in the error path of realize()
> ---
> hw/virtio/vhost-vsock.c | 12 ++--
> include/hw/virtio/vhost-vsock.h | 2 ++
> 2 files changed, 12 insertio
essage.
>
> Suggested-by: Philippe Mathieu-Daudé
> Signed-off-by: Alberto Garcia
> Reviewed-by: Alex Bennée
> ---
> block/qcow2.c | 23 +++
> 1 file changed, 11 insertions(+), 12 deletions(-)
Reviewed-by: Stefano Garzarella
>
> v2: Use g_autop
opening pool %s", opts->pool);
> goto failed_shutdown;
> }
> +/*
> + * Set the namespace after opening the io context on the pool,
> + * if nspace == NULL or if nspace == "", it is just as we did nothing
> + */
> +rados_ioctx_set
On Fri, Dec 20, 2019 at 09:56:51AM -0500, Jason Dillaman wrote:
> On Fri, Dec 20, 2019 at 9:11 AM Florian Florensa wrote:
> >
> > Hello Stefano and Jason,
> >
> > First of all thanks for the quick reply,
> > Response inline belowe
> > > Hi Florian,
> > >
> > > I think we need to add (Since: 5.0).
changed, 2 insertions(+), 1 deletion(-)
When I reviewed the Michael's patch for pci transport, I planned to do
the same but I postponed for several reason.
Thanks for doing that!
Reviewed-by: Stefano Garzarella
>
> diff --git a/hw/virtio/virtio-mmio.c b/hw/virtio/virtio-mmio.c
>
ile changed, 1 insertion(+)
Reviewed-by: Stefano Garzarella
>
> diff --git a/hw/rtc/Kconfig b/hw/rtc/Kconfig
> index 45daa8d655..3dc2dd6888 100644
> --- a/hw/rtc/Kconfig
> +++ b/hw/rtc/Kconfig
> @@ -17,6 +17,7 @@ config TWL92230
> depends on I2C
>
> confi
On Tue, Jul 13, 2021 at 03:51:15PM +0100, Stefan Hajnoczi wrote:
On Wed, Jul 07, 2021 at 05:00:18PM +0200, Stefano Garzarella wrote:
diff --git a/qapi/misc.json b/qapi/misc.json
index 156f98203e..f64bb69f74 100644
--- a/qapi/misc.json
+++ b/qapi/misc.json
@@ -86,6 +86,9 @@
# @poll-shrink: how
On Tue, Jul 13, 2021 at 03:58:04PM +0100, Stefan Hajnoczi wrote:
On Wed, Jul 07, 2021 at 05:00:19PM +0200, Stefano Garzarella wrote:
@@ -371,7 +375,7 @@ static int laio_do_submit(int fd, struct qemu_laiocb
*laiocb, off_t offset,
s->io_q.in_queue++;
if (!s->io_q.b
forward reference to the struct.
Signed-off-by: Paolo Bonzini
---
block/io_uring.c| 4 ++--
include/block/aio.h | 8
include/block/raw-aio.h | 1 -
3 files changed, 6 insertions(+), 7 deletions(-)
Reviewed-by: Stefano Garzarella
--filename=/dev/vdb \
--rw=${TEST} --bs=${BS} --iodepth=${IODEPTH} --numjobs=16
Next steps:
- benchmark io_uring and use `aio-max-batch` also there
- make MAX_EVENTS parametric adding a new `aio-max-events` parameter
Thanks,
Stefano
Stefano Garzarella (3):
iothread: generaliz
ks and it represents a good tradeoff between the latency
increase while a request is queued and the cost of the io_submit(2)
system call.
Signed-off-by: Stefano Garzarella
---
Notes:
v2:
- limit the batch with the number of available events [stefanha]
block/linux-aio.c | 9
Changes in preparation for next patches where we add a new
parameter not related to the poll mechanism.
Let's add two new generic functions (iothread_set_param and
iothread_get_param) that we use to set and get IOThread
parameters.
Signed-off-by: Stefano Garzarella
---
iothread.c
allows us to control the maximum batch size to reduce
the latency that requests might accumulate while queued in the AIO
engine queue.
If `aio-max-batch` is equal to 0 (default value), the AIO engine will
use its default maximum batch size value.
Signed-off-by: Stefano Garzarella
---
Notes:
v2
Kevin Wolf
Signed-off-by: Stefano Garzarella
---
iothread.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/iothread.c b/iothread.c
index ddbbde61f7..a73e560ba0 100644
--- a/iothread.c
+++ b/iothread.c
@@ -215,18 +215,18 @@ static void iothread_complete(UserCrea
aque` parameter with
`IOThreadParamInfo *info`.
Suggested-by: Kevin Wolf
Signed-off-by: Stefano Garzarella
---
iothread.c | 18 ++
1 file changed, 10 insertions(+), 8 deletions(-)
diff --git a/iothread.c b/iothread.c
index a73e560ba0..0f98af0f2a 100644
--- a/iothread.c
+++ b/iothread.
We recently added a new parameter (aio-max-batch) to IOThread.
This series cleans up the code a bit, no functional changes.
Stefano Garzarella (2):
iothread: rename PollParamInfo to IOThreadParamInfo
iothread: use IOThreadParamInfo in iothread_[set|get]_param()
iothread.c | 28
what all the
implications are, but retrying upon EAGAIN does not sound like
a bad thing to my inexperienced ears.
Yeah, that doesn't sound bad.
For kernels that don't have the patch applied, I don't think there's
much we can do about it, so this change seems okay to me:
Reviewed-by: Stefano Garzarella
I've been working with io_uring for a while so I'd like to help
with reviews.
Signed-off-by: Stefano Garzarella
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 42ac45c3e5..1776d0950b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
spurious errors are encountered.
Co-authored-by: Stefan Hajnoczi
Reviewed-by: Stefano Garzarella
Signed-off-by: Fabian Ebner
---
Changes from v1:
* Focus on what's relevant for the patch itself in the commit
message.
* Add Stefan's comment.
* Add Stefano's R-b tag (I h
| 4 +---
hw/scsi/vhost-user-scsi.c| 4 +---
hw/virtio/vhost-user-fs.c| 3 +--
hw/virtio/vhost-user-vsock.c | 3 +--
hw/virtio/vhost-vsock.c | 3 +--
hw/virtio/vhost.c| 16 ++--
11 files changed, 29 insertions(+), 25 deletions(-)
Reviewed-by: Stefano
irtio/vhost.c | 13 +-
5 files changed, 32 insertions(+), 29 deletions(-)
Reviewed-by: Stefano Garzarella
-user.c| 10 +++---
hw/virtio/vhost-vdpa.c| 4 +++-
hw/virtio/vhost.c | 8
4 files changed, 17 insertions(+), 9 deletions(-)
Reviewed-by: Stefano Garzarella
, 15 insertions(+), 16 deletions(-)
Reviewed-by: Stefano Garzarella
error_setg_errno(errp, -ret, "vhost_get_config failed");
+}
+return ret;
}
-return -1;
+error_setg(errp, "vhost_dev_get_config not implemented");
^
Maybe I'd replace s/vhost_dev_get_config/vhost_get_config
But it's not that important:
Reviewed-by: Stefano Garzarella
this while touching the code.
Signed-off-by: Kevin Wolf
---
hw/block/vhost-user-blk.c | 48 ++-
1 file changed, 32 insertions(+), 16 deletions(-)
Reviewed-by: Stefano Garzarella
On Wed, Jun 09, 2021 at 05:46:58PM +0200, Kevin Wolf wrote:
Commit dabefdd6 removed code that was supposed to try reconnecting
during .realize(), but actually just crashed and had several design
problems.
This adds the feature back without the crash in simple cases while also
fixing some design
On Thu, Jun 10, 2021 at 12:14:24AM +, Jiang Wang wrote:
Datagram sockets are connectionless and unreliable.
The sender does not know the capacity of the receiver
and may send more packets than the receiver can handle.
Add two more dedicate virtqueues for datagram sockets,
so that it will not
On Tue, Aug 03, 2021 at 11:58:27AM -0700, Jiang Wang . wrote:
On Wed, Jul 7, 2021 at 10:27 AM Stefano Garzarella wrote:
On Wed, Jul 07, 2021 at 09:52:46AM -0700, Jiang Wang . wrote:
>On Wed, Jul 7, 2021 at 1:33 AM Stefano Garzarella wrote:
>> On Tue, Jul 06, 2021 at 10:26:07PM +00
On Wed, Aug 4, 2021 at 8:41 AM Stefano Garzarella
wrote:
>
> On Tue, Aug 03, 2021 at 11:58:27AM -0700, Jiang Wang . wrote:
> >On Wed, Jul 7, 2021 at 10:27 AM Stefano Garzarella
> >wrote:
> >> On Wed, Jul 07, 2021 at 09:52:46AM -0700, Jiang Wang . wrote:
> >
On Tue, Aug 03, 2021 at 11:41:32PM +, Jiang Wang wrote:
Datagram sockets are connectionless and unreliable.
The sender does not know the capacity of the receiver
and may send more packets than the receiver can handle.
Add two more dedicate virtqueues for datagram sockets,
so that it will not
should allow block jobs to complete, even
if such spurious errors are encountered.
Co-authored-by: Stefan Hajnoczi
Reviewed-by: Stefano Garzarella
Signed-off-by: Fabian Ebner
---
Changes from v1:
* Focus on what's relevant for the patch itself in the commit
message.
* Add Ste
On Wed, Aug 04, 2021 at 06:52:15PM +0200, Kevin Wolf wrote:
Am 04.08.2021 um 16:50 hat Stefano Garzarella geschrieben:
On Mon, Aug 02, 2021 at 02:40:36PM +0200, Kevin Wolf wrote:
> Am 29.07.2021 um 11:10 hat Fabian Ebner geschrieben:
> > Linux SCSI can throw spurious -EAGAIN in so
On Thu, Aug 05, 2021 at 12:00:05PM -0700, Jiang Wang . wrote:
On Tue, Aug 3, 2021 at 11:49 PM Stefano Garzarella wrote:
On Wed, Aug 4, 2021 at 8:41 AM Stefano Garzarella
wrote:
>
> On Tue, Aug 03, 2021 at 11:58:27AM -0700, Jiang Wang . wrote:
> >On Wed, Jul 7, 2021 at 10:2
On Thu, Aug 05, 2021 at 12:07:02PM -0700, Jiang Wang . wrote:
On Wed, Aug 4, 2021 at 1:13 AM Stefano Garzarella wrote:
On Tue, Aug 03, 2021 at 11:41:32PM +, Jiang Wang wrote:
>Datagram sockets are connectionless and unreliable.
>The sender does not know the capacity of the receive
On Mon, Aug 09, 2021 at 03:41:36PM +0200, Peter Lieven wrote:
Please, can you add a description?
For example also describing what happens if RBD image does not support
RBD_FEATURE_FAST_DIFF.
Signed-off-by: Peter Lieven
---
block/rbd.c | 119 +++
g_free(name);
if (virtio_queue_set_host_notifier_mr(vdev, queue_idx, &n->mr, true)) {
+object_unparent(OBJECT(&n->mr));
munmap(addr, page_size);
return -1;
}
--
1.8.3.1
Reviewed-by: Stefano Garzarella
On Tue, Jul 06, 2021 at 04:10:22PM +0800, Jason Wang wrote:
在 2021/7/6 下午4:03, Jason Wang 写道:
在 2021/6/23 下午11:03, Stefano Garzarella 写道:
On Mon, Jun 21, 2021 at 12:16:39PM +0800, Jason Wang wrote:
Introduce new error label to avoid the unnecessary checking of net
pointer.
Fixes
On Tue, Jul 06, 2021 at 10:26:07PM +, Jiang Wang wrote:
Datagram sockets are connectionless and unreliable.
The sender does not know the capacity of the receiver
and may send more packets than the receiver can handle.
Add two more dedicate virtqueues for datagram sockets,
so that it will not
c adding a new `aio-max-events` parameter
Comments and suggestions are welcome :-)
Thanks,
Stefano
Stefano Garzarella (3):
iothread: generalize iothread_set_param/iothread_get_param
iothread: add aio-max-batch parameter
linux-aio: limit the batch size using `aio-max-batch` parameter
qapi
Changes in preparation for next patches where we add a new
parameter not related to the poll mechanism.
Let's add two new generic functions (iothread_set_param and
iothread_get_param) that we use to set and get IOThread
parameters.
Signed-off-by: Stefano Garzarella
---
iothread.c
allows us to control the maximum batch size to reduce
the latency that requests might accumulate while queued in the AIO
engine queue.
If `aio-max-batch` is equal to 0 (default value), the AIO engine will
use its default maximum batch size value.
Signed-off-by: Stefano Garzarella
---
qapi
ks and it represents a good tradeoff between the latency
increase while a request is queued and the cost of the io_submit(2)
system call.
Signed-off-by: Stefano Garzarella
---
block/linux-aio.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/block/linux-aio.c b/blo
On Wed, Jul 07, 2021 at 09:52:46AM -0700, Jiang Wang . wrote:
On Wed, Jul 7, 2021 at 1:33 AM Stefano Garzarella wrote:
On Tue, Jul 06, 2021 at 10:26:07PM +, Jiang Wang wrote:
>Datagram sockets are connectionless and unreliable.
>The sender does not know the capacity of the receive
On Mon, Jun 21, 2021 at 05:07:41PM +0300, Arseny Krasnov wrote:
This adds processing of VIRTIO_VSOCK_F_SEQPACKET features bit. Guest
negotiates it with vhost, thus both will know that SOCK_SEQPACKET
supported by peer.
Signed-off-by: Arseny Krasnov
---
hw/virtio/vhost-vsock.c
ude "standard-headers/linux/virtio_config.h"
+/* The feature bitmap for virtio vsock */
+#define VIRTIO_VSOCK_F_SEQPACKET 1 /* SOCK_SEQPACKET supported */
+
struct virtio_vsock_config {
uint64_t guest_cid;
} QEMU_PACKED;
--
2.25.1
LGTM:
Reviewed-by: Stefano Garzarella
headers/linux/virtio_ids.h"
#include "standard-headers/linux/virtio_config.h"
+/* The feature bitmap for virtio vsock */
+#define VIRTIO_VSOCK_F_SEQPACKET 1 /* SOCK_SEQPACKET supported */
+
struct virtio_vsock_config {
uint64_t guest_cid;
} QEMU_PACKED;
--
2.25.1
Reviewed-by: Stefano Garzarella
@@ static int vhost_vdpa_add(NetClientState *ncs, void *be)
options.net_backend = ncs;
options.opaque = be;
options.busyloop_timeout = 0;
+options.nvqs = 2;
net = vhost_net_init(&options);
if (!net) {
--
2.25.1
Reviewed-by: Stefano Garzarella
ailed to init vhost_net for queue");
goto err;
}
-if (s->vhost_net) {
-vhost_net_cleanup(s->vhost_net);
-g_free(s->vhost_net);
-}
Maybe we can add an assert() to discover future issues, but I don't have
a strong opinion.
It is fine:
Reviewed
e.
Signed-off-by: Jason Wang
---
net/vhost-vdpa.c | 11 ---
1 file changed, 11 deletions(-)
Reviewed-by: Stefano Garzarella
diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
index 728e63ff54..f5689a7c32 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -82,16 +82,6 @@
21f09c546f 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -111,6 +111,7 @@ static int vhost_vdpa_add(NetClientState *ncs, void *be)
err:
if (net) {
vhost_net_cleanup(net);
+g_free(net);
}
return -1;
}
--
2.25.1
Reviewed-by: Stefano Garzarella
On Mon, Jun 21, 2021 at 12:16:39PM +0800, Jason Wang wrote:
Introduce new error label to avoid the unnecessary checking of net
pointer.
Fixes: 1e0a84ea49b68 ("vhost-vdpa: introduce vhost-vdpa net client")
Signed-off-by: Jason Wang
---
net/vhost-vdpa.c | 13 ++---
1 file changed, 6 insert
_fd = vdpa_device_fd;
ret = vhost_vdpa_add(nc, (void *)&s->vhost_vdpa);
-assert(s->vhost_net);
+if (ret) {
+qemu_close(vdpa_device_fd);
+qemu_del_net_client(nc);
+}
Okay, I see now :-)
Reviewed-by: Stefano Garzarella
return ret;
}
--
2.25.1
deletions(-)
Reviewed-by: Stefano Garzarella
diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
index 572aed4ca2..e63a54a938 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -157,24 +157,19 @@ static NetClientInfo net_vhost_vdpa_info = {
};
static int net_vhost_vdpa_init(NetClientState *peer
deletion(-)
Reviewed-by: Stefano Garzarella
diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
index 87b181a74e..572aed4ca2 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -166,7 +166,6 @@ static int net_vhost_vdpa_init(NetClientState
*peer, const char *device,
assert(name);
nc
e, Sep 07, 2021 at 02:49:35PM +0200, Stefano Garzarella wrote:
> > > Commit 1e08fd0a46 ("vhost-vsock: SOCK_SEQPACKET feature bit support")
> > > enabled the SEQPACKET feature bit.
> > > This commit is released with QEMU 6.1, so if we try to migrate a VM where
> &
On Sun, Sep 12, 2021 at 06:46:03PM +, Jiang Wang wrote:
Datagram sockets are connectionless and unreliable.
The sender does not know the capacity of the receiver
and may send more packets than the receiver can handle.
Add two more dedicate virtqueues for datagram sockets,
so that it will not
On Mon, Sep 13, 2021 at 09:46:48AM -0400, Michael S. Tsirkin wrote:
On Mon, Sep 13, 2021 at 02:51:42PM +0200, Stefano Garzarella wrote:
On Fri, Sep 10, 2021 at 02:35:53AM -0400, Michael S. Tsirkin wrote:
> On Thu, Sep 09, 2021 at 10:02:12AM +0100, Daniel P. Berrangé wrote:
> > On Th
On Wed, Sep 15, 2021 at 08:59:17PM -0700, Jiang Wang . wrote:
On Tue, Sep 14, 2021 at 5:46 AM Stefan Hajnoczi wrote:
On Mon, Sep 13, 2021 at 10:18:43PM +, Jiang Wang wrote:
> Datagram sockets are connectionless and unreliable.
> The sender does not know the capacity of the receiver
> and m
On Thu, Sep 16, 2021 at 08:26:15AM +0200, Stefano Garzarella wrote:
On Wed, Sep 15, 2021 at 08:59:17PM -0700, Jiang Wang . wrote:
On Tue, Sep 14, 2021 at 5:46 AM Stefan Hajnoczi wrote:
On Mon, Sep 13, 2021 at 10:18:43PM +, Jiang Wang wrote:
Datagram sockets are connectionless and
simplify the
backport I left them separate.
v1:
https://lore.kernel.org/qemu-devel/20210907124935.147164-1-sgarz...@redhat.com/
v2:
- used OnOffAuto `seqpacket` property instead of boolean `features` [stefanha]
- moved `seqpacket` property to vhost-vsock-common class
Stefano Garzarella (2
, add
vhost_vsock_common_get_features() used by children, and disable
`seqpacket` for vhost-user-vsock device for machine types < 6.2.
The behavior of vhost-vsock device doesn't change; vhost-user-vsock
device now supports `seqpacket` property.
Signed-off-by: Stefano Garzarella
---
include/hw/virtio/vhost-v
ot;vhost-vsock: SOCK_SEQPACKET feature bit support")
Cc: qemu-sta...@nongnu.org
Reported-by: Jiang Wang
Signed-off-by: Stefano Garzarella
---
v2:
- used OnOffAuto `seqpacket` property instead of boolean `features` [stefanha]
---
include/hw/virtio/vhost-vsock.h | 3 +++
hw/core/machine.c
On Wed, Sep 22, 2021 at 12:00:24AM +, Jiang Wang wrote:
Datagram sockets are connectionless and unreliable.
The sender does not know the capacity of the receiver
and may send more packets than the receiver can handle.
Add two more dedicate virtqueues for datagram sockets,
so that it will not
On Wed, Sep 22, 2021 at 10:36:24AM -0700, Jiang Wang . wrote:
On Wed, Sep 22, 2021 at 2:23 AM Stefano Garzarella wrote:
On Wed, Sep 22, 2021 at 12:00:24AM +, Jiang Wang wrote:
>Datagram sockets are connectionless and unreliable.
>The sender does not know the capacity of the receive
size even
more to avoid increasing latency.
This series add the `aio-max-batch` option to the file backend,
and use it in laio_co_submit() and laio_io_unplug() to limit the
Linux AIO batch size more than the limit set by the AIO context.
Stefano Garzarella (3):
file-posix: add `aio-max-batch` opti
This new parameter can be used by block devices to limit the
Linux AIO batch size more than the limit set by the AIO context.
file-posix backend supports this, passing its `aio-max-batch` option
previously added.
Add an helper function to calculate the maximum batch size.
Signed-off-by: Stefano
size even
more to avoid increasing latency.
For this reason we add the `aio-max-batch` option to the file
backend, which will be used by the next commits to limit the size of
batches including requests generated by this device.
Suggested-by: Kevin Wolf
Signed-off-by: Stefano Garzarella
---
: Stefano Garzarella
---
include/block/raw-aio.h | 3 ++-
block/file-posix.c | 2 +-
block/linux-aio.c | 8 +---
3 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/include/block/raw-aio.h b/include/block/raw-aio.h
index ebd042fa27..21fc10c4c9 100644
--- a/include/block/raw
CCing Paolo and Thomas
On Wed, Aug 25, 2021 at 06:43:56PM +0200, Helge Konetzka wrote:
The appended patch fixes the link to Wiki Windows page on Website
Download Sources page.
Furthermore I would like to propose to change the instructions for
Native builds with MSYS2 on Wiki Windows page.
P
@@ static int virtio_queue_split_empty(VirtQueue *vq)
return empty;
}
+/* Called within rcu_read_lock(). */
static int virtio_queue_packed_empty_rcu(VirtQueue *vq)
{
struct VRingPackedDesc desc;
--
2.31.1
Reviewed-by: Stefano Garzarella
destroy(&caches->avail);
address_space_cache_destroy(&caches->used);
--
2.31.1
Reviewed-by: Stefano Garzarella
On Thu, Aug 26, 2021 at 07:26:58PM +0200, Philippe Mathieu-Daudé wrote:
Both virtqueue_packed_get_avail_bytes() and
virtqueue_split_get_avail_bytes() access the region cache, but
their caller also does. Simplify by having virtqueue_get_avail_bytes
calling both with RCU lock held, and passing the
On Thu, Sep 02, 2021 at 01:12:33PM +0100, Stefan Hajnoczi wrote:
On Wed, Sep 01, 2021 at 05:55:38PM +0200, Stefano Garzarella wrote:
On Thu, Aug 26, 2021 at 07:26:58PM +0200, Philippe Mathieu-Daudé wrote:
> Both virtqueue_packed_get_avail_bytes() and
> virtqueue_split_get_avail_bytes()
Ping :-)
Looks like it went into the crack during feature freeze,
should I resend it?
On Tue, Jul 27, 2021 at 04:59:34PM +0200, Stefano Garzarella wrote:
We recently added a new parameter (aio-max-batch) to IOThread.
This series cleans up the code a bit, no functional changes.
Stefano
On Sun, Sep 05, 2021 at 11:08:34AM -0700, Jiang Wang . wrote:
On Mon, Aug 9, 2021 at 3:58 AM Stefano Garzarella wrote:
On Thu, Aug 05, 2021 at 12:07:02PM -0700, Jiang Wang . wrote:
>On Wed, Aug 4, 2021 at 1:13 AM Stefano Garzarella wrote:
>> On Tue, Aug 03, 2021 at 11:41:32PM +00
On Tue, Sep 07, 2021 at 12:15:30PM +0200, Stefano Garzarella wrote:
On Sun, Sep 05, 2021 at 11:08:34AM -0700, Jiang Wang . wrote:
On Mon, Aug 9, 2021 at 3:58 AM Stefano Garzarella wrote:
On Thu, Aug 05, 2021 at 12:07:02PM -0700, Jiang Wang . wrote:
On Wed, Aug 4, 2021 at 1:13 AM Stefano
ure bit for machine types < 6.1, adding a
`features` field to VHostVSock to simplify the handling of upcoming
features we will support.
Fixes: 1e08fd0a46 ("vhost-vsock: SOCK_SEQPACKET feature bit support")
Cc: qemu-sta...@nongnu.org
Reported-by: Jiang Wang
Signed-off-by: Stefano Ga
On Tue, Sep 07, 2021 at 02:22:24PM +0100, Daniel P. Berrangé wrote:
On Tue, Sep 07, 2021 at 02:49:35PM +0200, Stefano Garzarella wrote:
Commit 1e08fd0a46 ("vhost-vsock: SOCK_SEQPACKET feature bit support")
enabled the SEQPACKET feature bit.
This commit is released with QEMU 6.1, so if
l()
hw/virtio: Have virtqueue_get_avail_bytes() pass caches arg to callees
include/hw/virtio/virtio.h | 7 +++
hw/virtio/virtio.c | 32 +++-
2 files changed, 22 insertions(+), 17 deletions(-)
--
2.31.1
Reviewed-by: Stefano Garzarella
Hi Denis,
I just found this discussion since we still have the following line in
hw/core/machine.c:
{ "vhost-blk-device", "seg_max_adjust", "off"}
IIUC it was a typo, and I think we should fix it since in the future we
can have `vhost-blk-device`.
So, I think we have 2 options:
1. remove
901 - 1000 of 1375 matches
Mail list logo
