search for: hv_netvsc

...over the slave. This doesn't allow udev/systemd to do its device rename of the slave device. Netvsc uses a delayed work to workaround this. Secondly, the select queue needs to call queue selection in VF. The bonding/teaming logic doesn't work well for UDP flows. Commit b3bf5666a510 ("hv_netvsc: defer queue selection to VF") fixed this performance problem. Lastly, more indirection is bad in current climate. I am not completely adverse to this but it needs to be fast, simple and completely transparent.

...over the slave. This doesn't allow udev/systemd to do its device rename of the slave device. Netvsc uses a delayed work to workaround this. Secondly, the select queue needs to call queue selection in VF. The bonding/teaming logic doesn't work well for UDP flows. Commit b3bf5666a510 ("hv_netvsc: defer queue selection to VF") fixed this performance problem. Lastly, more indirection is bad in current climate. I am not completely adverse to this but it needs to be fast, simple and completely transparent.

...: Use bytes 4 and 5 from instance ID as the PCI domain numbers (Mohammed Gamal) [1766097 1634251] +- [pci] hv: Detect and fix Hyper-V PCI domain number collision (Mohammed Gamal) [1766097 1634251] +- [pci] hv: Serialize the present and eject work items (Mohammed Gamal) [1766097 1634251] +- [netdrv] hv_netvsc: fix network namespace issues with VF support (Mohammed Gamal) [1766093 1741334] +- [netdrv] hv_netvsc: move VF to same namespace as netvsc device (Mohammed Gamal) [1766093 1741334] +- [netdrv] hv_netvsc: set master device (Mohammed Gamal) [1766093 1741334] +- [pci] PCI: hv: Avoid use of hv_pci_dev...

...9;s built-in. How are you handling this? Is it recommended to install/update to the latest LIS from Microsoft provided as ISO or do I need to install the hyperv-daemons? Currently my system reports: [root at share ~]# lsmod | grep hv hv_utils 24418 2 hv_storvsc 18032 6 hv_netvsc 40261 0 hv_vmbus 397185 6 hyperv_keyboard,hv_netvsc,hid_hyperv,hv_utils,hyperv_fb,hv_storvsc and I think this time shifts on rebbots are related to missing LIS (?): [root at share ~]# last root pts/0 172.16.4.103 Mon Jan...

I am trying to take a CentOS 7 img and get hyperv to boot. First step. qemu-img convert "CentOS7.img" -O vhdx -o subformat=dynamic "CentOS7.vhdx" after adding new hyper-v I get dracut-timeout... So doing some searching it says add hv_vmbus hv_netvsc hv_storvsc to INITRD_MODULES. which doesnt really exist any more in CentOS 7. I did find /etc/dracut.conf and add_drivers... I uncommented the line for '#add_drivers += "" ' added the above modules and then ran: grub2-mkconfig -o /boot/grub2/grub.cfg Then redid the convert abov...

Hello group, I am running the Linux kernel 2.6.32-7 and using the GPL Hyper-V drivers (hv_vmbus, hv_storvsc, hv_netvsc, hv_blkvsc). The only thing that is not working is the mouse integration. This a Linux guest under Hyper-V on Windows Server 2008 R2. I had InputVSC (Citrix Project Satori) working under kernel 2.6.18-6 with Microsoft''s Linux integration components. This worked great under Windows Server...

...t customer from explicitly moving the net-failover slaves out of this netns, but we will not move them out of there automatically. >> >> >> The 2-device netvsc already handles case where master changes namespace. > > Is it by moving slave with it? See c0a41b887ce6 ("hv_netvsc: move VF to same namespace as netvsc device?). It seems that when NetVSC master netdev changes netns, the VF is moved to the same netns by the NetVSC driver. Kinda the opposite than what we are suggesting here to make sure that the net-failover master netdev is on a separate netns than it?s slaves....

On Thu, Mar 21, 2019 at 08:45:17AM -0700, Stephen Hemminger wrote: > On Thu, 21 Mar 2019 15:04:37 +0200 > Liran Alon <liran.alon at oracle.com> wrote: > > > > > > > OK. Now what happens if master is moved to another namespace? Do we need > > > to move the slaves too? > > > > No. Why would we move the slaves? The whole point is to make most

On Thu, Mar 21, 2019 at 08:45:17AM -0700, Stephen Hemminger wrote: > On Thu, 21 Mar 2019 15:04:37 +0200 > Liran Alon <liran.alon at oracle.com> wrote: > > > > > > > OK. Now what happens if master is moved to another namespace? Do we need > > > to move the slaves too? > > > > No. Why would we move the slaves? The whole point is to make most

On our CentOS 7 servers I see there is a new kernel available when doing yum update: kernel 3.10.0-1062.12.1.el7 kernel-devel 3.10.0-1062.12.1.el7 However I am not able to find any announcement at redhat.com for this kernel. The only announcement I find is this from CentOS: https://www.mail-archive.com/centos-announce at centos.org/msg11573.html However that CentOS announcement links to

mkinitrd relies on /sys/class/scsi_host/*/proc_name instead of /sys/block/sd*/device/../../../moalias to get the scsi driver module name. As a fallback the sysfs driver name could be used, which does not match the module name either ('storvsc' vs. 'hv_storvsc'). Signed-off-by: Olaf Hering <olaf at aepfle.de> --- drivers/staging/hv/storvsc_drv.c | 3 ++- 1 file changed,

mkinitrd relies on /sys/class/scsi_host/*/proc_name instead of /sys/block/sd*/device/../../../moalias to get the scsi driver module name. As a fallback the sysfs driver name could be used, which does not match the module name either ('storvsc' vs. 'hv_storvsc'). Signed-off-by: Olaf Hering <olaf at aepfle.de> --- drivers/staging/hv/storvsc_drv.c | 3 ++- 1 file changed,

I am attaching a clocksource driver for HyperV. Signed-off-by: K. Y. Srinivasan <ksrinivasan at novell.com> -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: hyperv_clocksource.patch Url: http://lists.linux-foundation.org/pipermail/virtualization/attachments/20100405/80968881/attachment-0001.txt

I am attaching a clocksource driver for HyperV. Signed-off-by: K. Y. Srinivasan <ksrinivasan at novell.com> -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: hyperv_clocksource.patch Url: http://lists.linux-foundation.org/pipermail/virtualization/attachments/20100405/80968881/attachment-0001.txt

I am enclosing a patch that implements the KVP (Key Value Pair) functionality for Linux guests on HyperV. This functionality allows Microsoft Management stack to query information from the guest. This functionality is implemented in two parts: (a) A kernel component that communicates with the host and (b) A user level daemon that implements data gathering. The attached patch (kvp.patch) implements

I am enclosing a patch that implements the KVP (Key Value Pair) functionality for Linux guests on HyperV. This functionality allows Microsoft Management stack to query information from the guest. This functionality is implemented in two parts: (a) A kernel component that communicates with the host and (b) A user level daemon that implements data gathering. The attached patch (kvp.patch) implements

...ce. Netvsc uses a delayed work to workaround this. Interesting. Does this mean udev must act within a specific time window then? > Secondly, the select queue needs to call queue selection in VF. > The bonding/teaming logic doesn't work well for UDP flows. > Commit b3bf5666a510 ("hv_netvsc: defer queue selection to VF") > fixed this performance problem. > > Lastly, more indirection is bad in current climate. > > I am not completely adverse to this but it needs to be fast, simple > and completely transparent.

...round this. Wait. Why the fact a device is enslaved has to affect the udev in any way? If it does, smells like a bug in udev. > >Secondly, the select queue needs to call queue selection in VF. >The bonding/teaming logic doesn't work well for UDP flows. >Commit b3bf5666a510 ("hv_netvsc: defer queue selection to VF") >fixed this performance problem. > >Lastly, more indirection is bad in current climate. > >I am not completely adverse to this but it needs to be fast, simple >and completely transparent.

...he net-failover slaves out of this netns, but we will not move them out of there automatically. > >> > >> > >> The 2-device netvsc already handles case where master changes namespace. > > > > Is it by moving slave with it? > > See c0a41b887ce6 ("hv_netvsc: move VF to same namespace as netvsc device?). > It seems that when NetVSC master netdev changes netns, the VF is moved to the same netns by the NetVSC driver. > Kinda the opposite than what we are suggesting here to make sure that the net-failover master netdev is on a separate > netns th...

The hv_netvsc gets RNDIS_STATUS_MEDIA_CONNECT event after the VM is live migrated. Adding call to netif_notify_peers() for this event to send GARP (Gratuitous ARP) to notify network peers. Otherwise, the VM's network connection may stop after a live migration. This patch should also be applied to stable ker...