yang Zhang
Signed-off-by: Haiyang Zhang
Co-developed-by: Shachar Raindel
Signed-off-by: Shachar Raindel
Signed-off-by: Dexuan Cui
---
drivers/net/hyperv/netvsc_drv.c | 14 --
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/
This version (v8)
posts them in 2 separate patches, as suggested by Stephen Hemminger.
Please refer to "[PATCH v8 net-next 2/2]" for the history of v1~v7.
Thanks,
Dexuan
Dexuan Cui (2):
hv_netvsc: Make netvsc/VF binding check both MAC and serial number
net: mana: Add a driver for
> From: Stephen Hemminger
> Sent: Friday, April 16, 2021 11:09 AM
> ...
> On Fri, 16 Apr 2021 17:58:45 +0000
> Dexuan Cui wrote:
>
> > > >
> > > > This probably should be a separate patch.
> > > > I think it is trying to address the
> From: Haiyang Zhang
> Sent: Friday, April 16, 2021 10:11 AM
> > From: Stephen Hemminger
> > > ...
> > > @@ -2319,8 +2320,17 @@ static struct net_device
> > *get_netvsc_byslot(const struct net_device *vf_netdev)
> > > if (!ndev_ctx->vf_alloc)
> > > continue;
> > >
> >
> From: Stephen Hemminger
> Sent: Thursday, April 15, 2021 2:15 PM
> > ...
> > + netif_carrier_off(ndev);
> > +
> > + get_random_bytes(apc->hashkey, MANA_HASH_KEY_SIZE);
>
> Current practice for network drivers is to use netdev_rss_key_fill() for this.
Will change to netdev_rss_key_fill(). T
> From: Jakub Kicinski
> Sent: Thursday, April 15, 2021 10:44 AM
> ...
> On Wed, 14 Apr 2021 22:45:19 -0700 Dexuan Cui wrote:
> > + buf = dma_alloc_coherent(gmi->dev, length, &dma_handle,
> > +GFP_KERNEL | __GFP_ZERO);
>
> No
> From: Jakub Kicinski
> Sent: Tuesday, April 13, 2021 12:03 PM
>
> On Mon, 12 Apr 2021 19:35:09 -0700 Dexuan Cui wrote:
> > + apc->port_st_save = apc->port_is_up;
> > + apc->port_is_up = false;
> > + apc->start_remove = true;
> > +
>
> From: Jakub Kicinski
> Sent: Monday, April 12, 2021 11:21 AM
> ...
> On Sun, 11 Apr 2021 19:34:55 -0700 Dexuan Cui wrote:
> > + for (i = 0; i < ANA_INDIRECT_TABLE_SIZE; i++)
> > + apc->indir_table[i] = i % apc->num_queues;
>
>
> From: Haiyang Zhang
> Sent: Monday, April 12, 2021 7:40 AM
> > From: Andrew Lunn
> > Sent: Monday, April 12, 2021 8:32 AM
> > > ...
> > > + /* At most num_online_cpus() + 1 interrupts are used. */
> > > + msix_index = queue->eq.msix_index;
> > > + if (WARN_ON(msix_index > num_online_cpus()))
>
> From: Leon Romanovsky
> Sent: Monday, April 12, 2021 12:46 AM
> To: Dexuan Cui
> > ...
> > +#define ANA_MAJOR_VERSION 0
> > +#define ANA_MINOR_VERSION 1
> > +#define ANA_MICRO_VERSION 1
>
> Please don't introduce drier versions.
This is not th
> From: Andrew Lunn
> Sent: Thursday, April 8, 2021 5:30 PM
> To: Stephen Hemminger
> ...
> > Linux kernel doesn't do namespaces in the code, so every new driver needs
> > to worry about global symbols clashing
>
> This driver is called mana, yet the code uses ana. It would be good to
> resolve
> From: David Miller
> Sent: Thursday, April 8, 2021 5:41 PM
> > ...
> > In the driver code, all the structs/unions marked by __packed are used to
> > talk with the hardware, so I think __packed is necessary here?
>
> It actually isan't in many cases, check with and without the __packed
> direct
> From: David Miller
> Sent: Thursday, April 8, 2021 4:46 PM
> ...
> > +struct gdma_msg_hdr {
> > + u32 hdr_type;
> > + u32 msg_type;
> > + u16 msg_version;
> > + u16 hwc_msg_id;
> > + u32 msg_size;
> > +} __packed;
> > +
> > +struct gdma_dev_id {
> > + union {
> > + struct {
> From: Stephen Hemminger
> Sent: Thursday, April 8, 2021 9:52 AM
Thanks all for your input! We'll make the below changes as suggested:
Microsoft Azure Network Device ==> Microsoft Network Devices
Drop the default m
validated ==> supported
We'll also fix some warnings reported by "kernel test r
> From: Leon Romanovsky
> Sent: Wednesday, April 7, 2021 5:45 AM
> > >
> > > BTW, you don't need to write { 0 }, the {} is enough.
> >
> > Thanks for the suggestion! I'll use {0} in v2.
>
> You missed the point, "{ 0 }" change to be "{}" without 0.
Got it. Will make the suggested change.
FWIW,
> From: Leon Romanovsky
> Sent: Wednesday, April 7, 2021 1:10 AM
>
> <...>
>
> > +int gdma_verify_vf_version(struct pci_dev *pdev)
> > +{
> > + struct gdma_context *gc = pci_get_drvdata(pdev);
> > + struct gdma_verify_ver_req req = { 0 };
> > + struct gdma_verify_ver_resp resp = { 0 };
> >
> From: Leon Romanovsky
> Sent: Wednesday, April 7, 2021 1:15 AM
> > ...
> > int gdma_test_eq(struct gdma_context *gc, struct gdma_queue *eq)
> > {
> > struct gdma_generate_test_event_req req = { 0 };
> > struct gdma_general_resp resp = { 0 };
>
> BTW, you don't need to write { 0
> From: kernel test robot
> Sent: Tuesday, April 6, 2021 6:31 PM
> ...
> Hi Dexuan,
> I love your patch! Perhaps something to improve:
>
> All warnings (new ones prefixed by >>):
>
>drivers/pci/controller/pci-hyperv.c: In function 'hv_irq_unmask':
>drivers/pci/controller/pci-hyperv.c:12
> From: Andrew Lunn
> Sent: Tuesday, April 6, 2021 6:08 PM
> To: Dexuan Cui
>
> > +static int gdma_query_max_resources(struct pci_dev *pdev)
> > +{
> > + struct gdma_context *gc = pci_get_drvdata(pdev);
> > + struct gdma_general_req req = { 0 };
> >
. Add a comment explaining why the code is written this
> way.
>
> Signed-off-by: Shachar Raindel
> Signed-off-by: Haiyang Zhang
> ---
Reviewed-by: Dexuan Cui
The previous change "hv_netvsc: Switch the data path at the right time
during hibernation" adds the call of netvsc_vf_changed() upon
NETDEV_CHANGE, so it's necessary to avoid the duplicate call and message
when the VF is brought UP or DOWN.
Signed-off-by: Dexuan Cui
---
dri
VF NIC has been resumed.
Fixes: 19162fd4063a ("hv_netvsc: Fix hibernation for mlx5 VF driver")
Signed-off-by: Dexuan Cui
---
drivers/net/hyperv/netvsc_drv.c | 11 +--
1 file changed, 1 insertion(+), 10 deletions(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drive
> From: Michael Kelley
> Sent: Tuesday, September 8, 2020 1:49 PM
> > @@ -2635,6 +2632,15 @@ static int netvsc_resume(struct hv_device *dev)
> > netvsc_devinfo_put(device_info);
> > net_device_ctx->saved_netvsc_dev_info = NULL;
> >
> > + /* A NIC driver (e.g. mlx5) may keep the VF networ
se in
netvsc_suspend()/resume() ndev_ctx->vf_netdev is NULL for mlx4.
Fixes: 0efeea5fb153 ("hv_netvsc: Add the support of hibernation")
Signed-off-by: Dexuan Cui
---
Changes in v2 (Thanks Jakub Kicinski ):
Added coments in the changelog and the code about the implicit
data pat
> From: Jakub Kicinski
> Sent: Saturday, September 5, 2020 4:27 PM
> [...]
> On Fri, 4 Sep 2020 19:52:18 -0700 Dexuan Cui wrote:
> > mlx5_suspend()/resume() keep the network interface, so during hibernation
> > netvsc_unregister_vf() and netvsc_register_vf() a
->vf_netdev is NULL for mlx4.
Fixes: 0efeea5fb153 ("hv_netvsc: Add the support of hibernation")
Signed-off-by: Dexuan Cui
---
drivers/net/hyperv/netvsc_drv.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv
> From: Dexuan Cui
> Sent: Monday, July 27, 2020 6:55 PM
> To: gre...@linuxfoundation.org; eduma...@google.com;
> sta...@vger.kernel.org
> Cc: w...@1wt.eu; Dexuan Cui ; Joseph Salisbury
> ; Michael Kelley ;
> v...@zeniv.linux.org.uk; netdev@vger.kernel.org; da...@davemloft.ne
corrupt packets earlier.
Signed-off-by: Eric Dumazet
Signed-off-by: Dexuan Cui
---
net/ipv4/udp.c | 3 +--
net/ipv6/udp.c | 6 ++
2 files changed, 3 insertions(+), 6 deletions(-)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index bb30699..49ab587 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
> From: Eric Dumazet
> Sent: Monday, July 27, 2020 11:40 AM
> To: Dexuan Cui
>
> On Mon, Jul 27, 2020 at 11:38 AM Dexuan Cui wrote:
> >
> > > From: Greg KH
> > > Sent: Friday, July 24, 2020 10:59 PM
> > > > [...]
> > > > Eri
> From: Greg KH
> Sent: Friday, July 24, 2020 10:59 PM
> > [...]
> > Eric Dumazet made an alternative that performs the csum validation earlier:
> >
> > --- a/net/ipv4/udp.c
> > +++ b/net/ipv4/udp.c
> > @@ -1589,8 +1589,7 @@ int udp_queue_rcv_skb(struct sock *sk, struct
> > sk_buff *skb)
> >
inter(sk->sk_filter) &&
- udp_lib_checksum_complete(skb))
+ if (udp_lib_checksum_complete(skb))
goto csum_error;
if (sk_rcvqueues_full(sk, sk->sk_rcvbuf)) {
I personally like Eric's fix and IMHO we'd better have it in v4.4 rather than
trying to backport 327868212381.
Looking forward to your comments!
Thanks,
Dexuan Cui
> From: Saeed Mahameed
> Sent: Tuesday, May 26, 2020 6:49 PM
> To: David S. Miller ; k...@kernel.org
> Cc: netdev@vger.kernel.org; Mark Bloch ; Dexuan Cui
> ; Moshe Shemesh ; Saeed
> Mahameed
> Subject: [net-next 12/16] net/mlx5: Add basic suspend/resume support
>
>
> From: linux-hyperv-ow...@vger.kernel.org
> On Behalf Of Dexuan Cui
> Sent: Monday, August 5, 2019 4:56 PM
> The warning is caused by the rcu_dereference_rtnl() :
>
> 1239 static void netvsc_get_stats64(struct net_device *net,
> 1240struct
Hi,
After the VM boots up, I always get the below call-trace when I run "nload" for
the first time:
[ 113.910911] WARNING: suspicious RCU usage
[ 113.913244] 5.2.0+ #19 Not tainted
[ 113.915216] -
[ 113.917521] drivers/net/hyperv/netvsc_drv.c:1243 suspicious
rcu_d
> From: linux-hyperv-ow...@vger.kernel.org
> On Behalf Of Andy Shevchenko
> Sent: Tuesday, July 23, 2019 9:40 AM
>
> The rest of Hyper-V code is using new types for UUID handling.
> Convert hv_sock as well.
>
> Signed-off-by: Andy Shevchenko
Reviewed-by: Dexuan Cui
cket option comes with the cost of that much memory being
> reserved/allocated by the kernel, for the lifetime of the connection.
Reviewed-by: Dexuan Cui
The patch looks good to me. Thanks, Sunil!
Thanks,
-- Dexuan
> From: Stefano Garzarella
> Sent: Tuesday, May 14, 2019 1:16 AM
> To: netdev@vger.kernel.org; Stefan Hajnoczi ; Dexuan
>
> Hi guys,
> I'm currently interested on implement a multi-transport support for VSOCK in
> order to handle nested VMs.
Hi Stefano,
Thanks for reviving the discussion! :-)
I
> From: Sunil Muthuswamy
> Sent: Thursday, May 16, 2019 11:11 AM
> > Hi Sunil,
> > To make it clear, your patch itself is good, and I was just talking about
> > the next change we're going to make. Once we make the next change,
> > IMO we need a further patch to schedule hvs_close_timeout() to the
> From: linux-hyperv-ow...@vger.kernel.org
> On Behalf Of Dexuan Cui
> Sent: Wednesday, May 15, 2019 9:34 PM
> ...
Hi Sunil,
To make it clear, your patch itself is good, and I was just talking about
the next change we're going to make. Once we make the next change,
IMO we need a
> From: Sunil Muthuswamy
> Sent: Tuesday, May 14, 2019 5:56 PM
> ...
> +static bool hvs_close_lock_held(struct vsock_sock *vsk)
> {
> ...
> + schedule_delayed_work(&vsk->close_work, HVS_CLOSE_TIMEOUT);
Reviewed-by: Dexuan Cui
The patch looks good to me. Th
> From: Sunil Muthuswamy
> Sent: Wednesday, May 8, 2019 4:11 PM
>
> Currently, when a hvsock socket is closed, the socket is shutdown and
> immediately a RST is sent. There is no wait for the FIN packet to arrive
> from the other end. This can lead to data loss since the connection is
> terminated
> From: linux-hyperv-ow...@vger.kernel.org
> On Behalf Of Michael Kelley
> Sent: Thursday, May 2, 2019 3:24 PM
> To: Juliana Rodrigueiro ;
> linux-hyp...@vger.kernel.org
> Cc: netdev@vger.kernel.org
> Subject: RE: Hyperv netvsc - regression for 32-PAE kernel
>
> From: Juliana Rodrigueiro Sent: T
> From: Stefan Hajnoczi
> Sent: Tuesday, December 11, 2018 8:27 AM
> To: Jorgen Hansen ; Dexuan Cui
>
> Cc: netdev@vger.kernel.org; Lepton Wu
> Subject: Re: [PATCH] VSOCK: bind to random port for VMADDR_PORT_ANY
>
> On Mon, Dec 10, 2018 at 11:02:35PM -0800, Lepton Wu
> From: David Miller
> Sent: Wednesday, August 29, 2018 17:49
>
> From: Dexuan Cui
> Date: Wed, 22 Aug 2018 21:20:03 +
>
> > ---
> > drivers/net/hyperv/netvsc_drv.c | 11 ++-
> > 1 file changed, 10 insertions(+), 1 deletion(-)
> >
> &
serspace, and next
when the userspace closes the connection quickly, hvs_release() may not see
the connection in the connected queue; finally hvs_open_connection()
inserts the connection into the queue, but we won't be able to purge the
connection for ever.
Signed-off-by: Dexuan Cui
Cc: K.
> From: Jorgen S. Hansen [mailto:jhan...@vmware.com]
> Sent: Wednesday, September 6, 2017 7:11 AM
>> ...
> > I'm currently working on NFS over AF_VSOCK and sock_diag support (for
> > ss(8) and netstat-like tools).
> >
> > Multi-transport support is lower priority for me at the moment. I'm
> > happ
> From: Stefan Hajnoczi [mailto:stefa...@redhat.com]
> Sent: Thursday, August 31, 2017 4:55 AM
> ...
> On Tue, Aug 29, 2017 at 03:37:07PM +, Jorgen S. Hansen wrote:
> > > On Aug 29, 2017, at 4:36 AM, Dexuan Cui wrote:
> > If we allow multiple host side transports, v
> From: Dexuan Cui
> Sent: Tuesday, August 22, 2017 21:21
> > ...
> > ...
> > The only problem here would be the potential for a guest and a host app
> to
> > have a conflict wrt port numbers, even though they would be able to
> > operate fine, if re
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Monday, August 28, 2017 15:39
> From: Dexuan Cui
> Date: Sat, 26 Aug 2017 04:52:43 +
>
> >
> > Hyper-V Sockets (hv_sock) supplies a byte-stream based communication
> > mechanism between the host and th
roducing a new
vsock transport for AF_VSOCK.
Signed-off-by: Dexuan Cui
Cc: K. Y. Srinivasan
Cc: Haiyang Zhang
Cc: Stephen Hemminger
Cc: Andy King
Cc: Dmitry Torokhov
Cc: George Zhang
Cc: Jorgen Hansen
Cc: Reilly Grant
Cc: Asias He
Cc: Stefan Hajnoczi
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Thursday, August 24, 2017 18:20
> > +#define VMBUS_PKT_TRAILER (sizeof(u64))
>
> This is not the packet trailer, it's the size of the packet trailer.
Thanks! I'll change it to VMBUS_PKT_TRAILER_SIZE.
> > + /* Have we sent the zero-lengt
roducing a new
vsock transport for AF_VSOCK.
Signed-off-by: Dexuan Cui
Cc: K. Y. Srinivasan
Cc: Haiyang Zhang
Cc: Stephen Hemminger
Cc: Andy King
Cc: Dmitry Torokhov
Cc: George Zhang
Cc: Jorgen Hansen
Cc: Reilly Grant
Cc: Stefan Hajnoczi
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
Cc: Rolf Neug
> From: Jorgen S. Hansen [mailto:jhan...@vmware.com]
> > On Aug 22, 2017, at 11:54 AM, Stefan Hajnoczi
> wrote:
> > ...
> > We *can* by looking at the destination CID. Please take a look at
> > drivers/misc/vmw_vmci/vmci_route.c:vmci_route() to see how VMCI
> handles
> > nested virt.
> >
> > It b
> From: Stefan Hajnoczi [mailto:stefa...@redhat.com]
> On Fri, Aug 18, 2017 at 10:23:54PM +, Dexuan Cui wrote:
> > > > +static bool hvs_stream_allow(u32 cid, u32 port)
> > > > +{
> > > > + static const u32 valid_cids[] = {
> > > > +
> From: Stefan Hajnoczi [mailto:stefa...@redhat.com]
> > CID is not really used by us, because we only support guest<->host
> communication,
> > and don't support guest<->guest communication. The Hyper-V host
> references
> > every VM by VmID (which is invisible to the VM), and a VM can only talk t
> From: Stefan Hajnoczi [mailto:stefa...@redhat.com]
> Sent: Thursday, August 17, 2017 07:56
> To: Dexuan Cui
> On Tue, Aug 15, 2017 at 10:18:41PM +0000, Dexuan Cui wrote:
> > +static u32 hvs_get_local_cid(void)
> > +{
> > + return VMADDR_CID_ANY;
> > +}
> From: Stefan Hajnoczi [mailto:stefa...@redhat.com]
> Sent: Thursday, August 17, 2017 07:06
>
> On Tue, Aug 15, 2017 at 10:15:39PM +0000, Dexuan Cui wrote:
> > With the current code, when vsock_dequeue_accept() is removing a sock
> > from the list, nothing prevents vso
> From: Jorgen S. Hansen [mailto:jhan...@vmware.com]
> Sent: Thursday, August 17, 2017 08:17
> >
> > Putting aside nested virtualization, I want to load the transport (vmci,
> > Hyper-V, vsock) for which there is paravirtualized hardware present
> > inside the guest.
>
> Good points. Completely ag
> > On Aug 16, 2017, at 12:15 AM, Dexuan Cui wrote:
> > With the current code, when vsock_dequeue_accept() is removing a sock
> > from the list, nothing prevents vsock_enqueue_accept() from adding a new
> > sock into the list concurrently. We should add a lock to protec
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Thursday, August 17, 2017 10:04
> I would avoid module parameters at all costs.
>
> It is the worst possible interface for users of your software.
>
> You really need to fundamentally solve the problems related to making
> sure the proper
> From: Dexuan Cui
> Sent: Wednesday, August 16, 2017 15:34
> > From: Jorgen S. Hansen [mailto:jhan...@vmware.com]
> > > Without the patch, vmw_vsock_vmci_transport.ko and vmw_vmci.ko can
> > > automatically load when an application creates an AF_VSOCK socket.
> &g
KVM's vsock_virtio_transport doesn't have the issue because it doesn't
define MODULE_ALIAS_NETPROTO(PF_VSOCK).
The patch also adds a module parameter "skip_hypervisor_check" for
vmw_vsock_vmci_transport.ko.
Signed-off-by: Dexuan Cui
Cc: Alok Kataria
Cc: Andy King
Cc: A
> From: Jorgen S. Hansen [mailto:jhan...@vmware.com]
> > Without the patch, vmw_vsock_vmci_transport.ko and vmw_vmci.ko can
> > automatically load when an application creates an AF_VSOCK socket.
> >
> > This is the expected good behavior on VMware hypervisor, but as we
> > are going to add hv_sock.
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Wednesday, August 16, 2017 11:07
>
> From: Dexuan Cui
> Date: Tue, 15 Aug 2017 22:13:29 +
>
> > + /*
> > +* Check if we are running on VMware's hypervisor and bail out
> > +* if we a
roducing a new
vsock transport for AF_VSOCK.
Signed-off-by: Dexuan Cui
Cc: K. Y. Srinivasan
Cc: Haiyang Zhang
Cc: Stephen Hemminger
Cc: Andy King
Cc: Dmitry Torokhov
Cc: George Zhang
Cc: Jorgen Hansen
Cc: Reilly Grant
Cc: Asias He
Cc: Stefan Hajnoczi
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
With the current code, when vsock_dequeue_accept() is removing a sock
from the list, nothing prevents vsock_enqueue_accept() from adding a new
sock into the list concurrently. We should add a lock to protect the list.
Signed-off-by: Dexuan Cui
Cc: Andy King
Cc: Dmitry Torokhov
Cc: George
-V in hv_acpi_init().
KVM's vsock_virtio_transport doesn't have the issue because it doesn't
define MODULE_ALIAS_NETPROTO(PF_VSOCK).
Signed-off-by: Dexuan Cui
Cc: Alok Kataria
Cc: Andy King
Cc: Adit Ranadive
Cc: George Zhang
Cc: Jorgen Hansen
Cc: K. Y. Srinivasan
Cc: Haiyan
d hence it manages to share the common vsock
infrastructure to greatly reduce duplicate code, and avoid adding a new
address family. The details are documented in PATCH 03.
Dexuan Cui (3):
VMCI: only load on VMware hypervisor
vsock: fix vsock_dequeue/enqueue_accept race
hv_sock: implemen
'nvdev' is freed in rndis_filter_device_remove -> netvsc_device_remove ->
free_netvsc_device, so we mustn't access it, before it's re-created in
rndis_filter_device_add -> netvsc_device_add.
Signed-off-by: Dexuan Cui
Cc: "K. Y. Srinivasan"
Cc: Haiyang Zha
Hi,
While trying SR-IOV with a Linux guest running on Hyper-V, I found this issue:
the VF driver can't work if the guest has 16 virtual CPUs (less vCPUs, e.g. 8,
can work fine):
[9.927820] mlx4_core: Mellanox ConnectX core driver v2.2-1 (Feb, 2014)
[9.927882] mlx4_core: Initializing b96
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Wednesday, July 27, 2016 1:45
> To: Dexuan Cui
>
> From: Dexuan Cui
> Date: Tue, 26 Jul 2016 07:09:41 +
>
> > I googled "S390 hypervisor socket" but didn't find anything related (I
>
> From: netdev-ow...@vger.kernel.org [mailto:netdev-
> ow...@vger.kernel.org] On Behalf Of Dexuan Cui
> Sent: Tuesday, July 26, 2016 21:22
> ...
> This is because, the design of AF_HYPERV in the Hyper-V host side is
> suboptimal IMHO (the current host side design requires the lea
> From: Michal Kubecek [mailto:mkube...@suse.cz]
> Sent: Tuesday, July 26, 2016 17:57
> ...
> On Tue, Jul 26, 2016 at 07:09:41AM +0000, Dexuan Cui wrote:
> > ... I don't think Michal
> > Kubecek was suggesting I build my code using the existing AF_VSOCK
> > cod
> From: devel [mailto:driverdev-devel-boun...@linuxdriverproject.org] On
> Behalf Of Dexuan Cui
> ...
> > From: David Miller [mailto:da...@davemloft.net]
> > ...
> > From: Dexuan Cui
> > Date: Tue, 26 Jul 2016 03:09:16 +
> >
> > > BTW, dur
> From: David Miller [mailto:da...@davemloft.net]
> ...
> From: Dexuan Cui
> Date: Tue, 26 Jul 2016 03:09:16 +
>
> > BTW, during the past month, at least 7 other people also reviewed
> > the patch and gave me quite a few good comments, which have
> > been ad
> From: David Miller [mailto:da...@davemloft.net]
>
> From: Dexuan Cui
> Date: Sat, 23 Jul 2016 01:35:51 +
>
> > +static struct sock *hvsock_create(struct net *net, struct socket *sock,
> > + gfp_t priority, unsigned short type)
> &
mt_service
The patch implements the necessary support in the guest side by introducing
a new socket address family AF_HYPERV.
Signed-off-by: Dexuan Cui
Cc: "K. Y. Srinivasan"
Cc: Haiyang Zhang
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
Cc: Olaf Hering
---
You can also get the patch by
re. :-(
3) remove the per-connection static send/recv buffers
Instead, we allocate and free the buffers dynamically only when we recv/send
data. This means: when a connection is idle, no memory is consumed as
recv/send buffers at all.
Dexuan Cui (1):
hv_sock: introduce Hyper-V Sockets
Changes si
> From: David Miller [mailto:da...@davemloft.net]
> >> From: kbuild test robot [mailto:l...@intel.com]
> >> [auto build test WARNING on net-next/master]
> >>
> >> url:https://github.com/0day-ci/linux/commits/Dexuan-Cui/introduce-
> >> Hyper-
> From: kbuild test robot [mailto:l...@intel.com]
> Sent: Wednesday, July 20, 2016 1:10
>
> Hi,
>
> [auto build test WARNING on net-next/master]
>
> url:https://github.com/0day-ci/linux/commits/Dexuan-Cui/introduce-
> Hyper-V-VM-Sockets-hv_sock/20160715-223433
>
mt_service
The patch implements the necessary support in the guest side by introducing
a new socket address family AF_HYPERV.
Signed-off-by: Dexuan Cui
Cc: "K. Y. Srinivasan"
Cc: Haiyang Zhang
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
Cc: Olaf Hering
---
You can also get the patch by
re. :-(
3) remove the per-connection static send/recv buffers
Instead, we allocate and free the buffers dynamically only when we recv/send
data. This means: when a connection is idle, no memory is consumed as
recv/send buffers at all.
Dexuan Cui (1):
hv_sock: introduce Hyper-V Sockets
Changes si
> From: Michal Kubecek [mailto:mkube...@suse.cz]
> > ..
> > However, though Hyper-V Sockets may seem conceptually similar to
> > AF_VOSCK, there are differences in the transportation layer, and IMO these
> > make the direct code reusing impractical:
> >
> > 1. In AF_VSOCK, the endpoint type is:
> From: Michal Kubecek [mailto:mkube...@suse.cz]
> > ..
> > +static struct sock *hvsock_find_connected_socket_by_channel(
> > + const struct vmbus_channel *channel)
> > +{
> > + struct hvsock_sock *hvsk;
> > +
> > + list_for_each_entry(hvsk, &hvsock_connected_list, connected_list) {
> > +
mt_service
The patch implements the necessary support in the guest side by introducing
a new socket address family AF_HYPERV.
Signed-off-by: Dexuan Cui
Cc: "K. Y. Srinivasan"
Cc: Haiyang Zhang
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
Cc: Olaf Hering
---
You can also get the patch by
boptimal considering memory consumption,
however unluckily we have to live with it, before the host comes up with
a new design in the future. :-(
3) remove the per-connection static send/recv buffers
Instead, we allocate and free the buffers dynamically only when we recv/send
data. This means: whe
> From: Vitaly Kuznetsov [mailto:vkuzn...@redhat.com]
> ...
> Some comments below. The vast majority of them are really minor, the
> only thing which bothers me a little bit is WARN() in hvsock_sendmsg()
> which I think shouldn't be there. But I may have missed something.
Thank you for the very
> From: Olaf Hering [mailto:o...@aepfle.de]
> Sent: Friday, July 8, 2016 0:02
> On Thu, Jun 30, Dexuan Cui wrote:
>
> > +/* The MTU is 16KB per the host side's design. */
> > +struct hvsock_recv_buf {
> > + unsigned int data_len;
> > + unsig
> From: Dexuan Cui
> Sent: Friday, July 8, 2016 15:47
>
> You can also get the patch here (2764221d):
> https://github.com/dcui/linux/commits/decui/hv_sock/net-next/20160708_v15
>
> In v14:
> fix some coding style issues pointed out by David.
>
> In v15:
> Just
mt_service
The patch implements the necessary support in the guest side by introducing
a new socket address family AF_HYPERV.
Signed-off-by: Dexuan Cui
Cc: "K. Y. Srinivasan"
Cc: Haiyang Zhang
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
---
You can also get the patch here (2764221d):
http
the per-connection static send/recv buffers
Instead, we allocate and free the buffers dynamically only when we recv/send
data. This means: when a connection is idle, no memory is consumed as
recv/send buffers at all.
Dexuan Cui (1):
hv_sock: introduce Hyper-V Sockets
Changes since v12:
return ENO
> From: Joe Perches [mailto:j...@perches.com]
> Sent: Tuesday, July 5, 2016 17:39
> To: Dexuan Cui ; da...@davemloft.net;
> gre...@linuxfoundation.org; netdev@vger.kernel.org; linux-
> ker...@vger.kernel.org; de...@linuxdriverproject.org; o...@aepfle.de;
> a...@canonical.com; j
> From: Joe Perches [mailto:j...@perches.com]
>
> > +#define sk_to_hvsock(__sk) ((struct hvsock_sock *)(__sk))
> > +#define hvsock_to_sk(__hvsk) ((struct sock *)(__hvsk))
>
> Might as well be static inlines
Hi Joe,
Thank you for the suggestions (again)! :-)
I'll change them to static inlines.
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Tuesday, July 5, 2016 14:27
> To: Dexuan Cui
> Subject: Re: [PATCH v14 net-next 1/1] hv_sock: introduce Hyper-V Sockets
>
> From: Dexuan Cui
> Date: Tue, 5 Jul 2016 01:58:31 +
>
> > Not sure if y
> From: linux-kernel-ow...@vger.kernel.org [mailto:linux-kernel-
> ow...@vger.kernel.org] On Behalf Of Dexuan Cui
> Sent: Thursday, June 30, 2016 23:59
> diff --git a/include/linux/socket.h b/include/linux/socket.h
> index b5cc5a6..0b68b58 100644
> --- a/include/linux/socket
> From: Olaf Hering [mailto:o...@aepfle.de]
> Sent: Friday, July 1, 2016 0:12
> To: Dexuan Cui
> Cc: da...@davemloft.net; gre...@linuxfoundation.org;
> netdev@vger.kernel.org; linux-ker...@vger.kernel.org;
> de...@linuxdriverproject.org; a...@canonical.com; jasow...@redhat.com;
the per-connection static send/recv buffers
Instead, we allocate and free the buffers dynamically only when we recv/send
data. This means: when a connection is idle, no memory is consumed as
recv/send buffers at all.
Dexuan Cui (1):
hv_sock: introduce Hyper-V Sockets
Changes since v12:
return ENO
mt_service
The patch implements the necessary support in the guest side by introducing
a new socket address family AF_HYPERV.
Signed-off-by: Dexuan Cui
Cc: "K. Y. Srinivasan"
Cc: Haiyang Zhang
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
---
You can also get the patch here (8ba95c8ec9):
http
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Thursday, June 30, 2016 20:45
> To: Dexuan Cui
> Cc: gre...@linuxfoundation.org; netdev@vger.kernel.org; linux-
> ker...@vger.kernel.org; de...@linuxdriverproject.org; o...@aepfle.de;
> a...@canonical.com; jasow...
mt_service
The patch implements the necessary support in the guest side by introducing
a new socket address family AF_HYPERV.
Signed-off-by: Dexuan Cui
Cc: "K. Y. Srinivasan"
Cc: Haiyang Zhang
Cc: Vitaly Kuznetsov
Cc: Cathy Avery
---
You can also get the patch here (ae3cbdabca):
http
1 - 100 of 230 matches
Mail list logo
