[Kernel-packages] [Bug 1839673] Re: [linux-azure] Important InfiniBand patches for Ubuntu 18.04
I suggest we can just keep 142 as is. The extra module size on disk is not that bad at 140k. The original design was to deal with future ND version updates. -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux-azure in Ubuntu. https://bugs.launchpad.net/bugs/1839673 Title: [linux-azure] Important InfiniBand patches for Ubuntu 18.04 Status in linux-azure package in Ubuntu: New Status in linux-azure source package in Disco: In Progress Bug description: The following two patches are critical for unblocking kernel sign-off for the 5.0 kernel in Ubuntu 18.04: ("Don't wait in hvnd_query_gid after interface is already bound to ND") ("Expose extended attributes for user IB verbs QUERY_DEVICE, CREATE_CQ and CREATE_QP") The patches have not landed in linux-next yet, but are needed to unblock 18.04 IB. The patches have been attached to this bug report. Please consider including these patches as SAUCE until they land in mainline. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-azure/+bug/1839673/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
[Kernel-packages] [Bug 1839673] Re: [linux-azure] Important InfiniBand patches for Ubuntu 18.04
I don't think we have ND 142 cluster in use, they have all been upgraded to 144. So we only need the patch for ND 144. This is not needed for 4.15. Ubuntu 16.04 is running older user-mode drivers that don't cause problems. -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux-azure in Ubuntu. https://bugs.launchpad.net/bugs/1839673 Title: [linux-azure] Important InfiniBand patches for Ubuntu 18.04 Status in linux-azure package in Ubuntu: New Status in linux-azure source package in Disco: In Progress Bug description: The following two patches are critical for unblocking kernel sign-off for the 5.0 kernel in Ubuntu 18.04: ("Don't wait in hvnd_query_gid after interface is already bound to ND") ("Expose extended attributes for user IB verbs QUERY_DEVICE, CREATE_CQ and CREATE_QP") The patches have not landed in linux-next yet, but are needed to unblock 18.04 IB. The patches have been attached to this bug report. Please consider including these patches as SAUCE until they land in mainline. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-azure/+bug/1839673/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
[Kernel-packages] [Bug 1701744] Re: Add infiniband support for Azure HPC
** Patch added: "Infiniband driver for kernel 4.11" https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1701744/+attachment/4906980/+files/0001-Add-infiniband-driver-for-Azure-HPC.patch -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1701744 Title: [Hyper-V] Add infiniband support for Azure HPC Status in linux package in Ubuntu: New Bug description: This is the infiniband driver for Azure HPC. Windows Azure agent will provision an image for running infiniband RDMA via DAPL when "OS.EnableRDMA=y" is defined in waagent.conf. Note: Ubuntu image needs to load rdma_ucm on boot to expose the RDMA CM interface to user-mode library. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1701744/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
[Kernel-packages] [Bug 1701744] Re: Add infiniband support for Azure HPC
** Patch added: "Infiniband driver build configuration for kernel 4.11" https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1701744/+attachment/4906981/+files/0002-Add-driver-build-configuration-to-infiniband.patch -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1701744 Title: [Hyper-V] Add infiniband support for Azure HPC Status in linux package in Ubuntu: New Bug description: This is the infiniband driver for Azure HPC. Windows Azure agent will provision an image for running infiniband RDMA via DAPL when "OS.EnableRDMA=y" is defined in waagent.conf. Note: Ubuntu image needs to load rdma_ucm on boot to expose the RDMA CM interface to user-mode library. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1701744/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
[Kernel-packages] [Bug 1701744] [NEW] [Hyper-V] Add infiniband support for Azure HPC
Public bug reported: This is the infiniband driver for Azure HPC. Windows Azure agent will provision an image for running infiniband RDMA via DAPL when "OS.EnableRDMA=y" is defined in waagent.conf. Note: Ubuntu image needs to load rdma_ucm on boot to expose the RDMA CM interface to user-mode library. ** Affects: linux (Ubuntu) Importance: Undecided Status: New -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1701744 Title: [Hyper-V] Add infiniband support for Azure HPC Status in linux package in Ubuntu: New Bug description: This is the infiniband driver for Azure HPC. Windows Azure agent will provision an image for running infiniband RDMA via DAPL when "OS.EnableRDMA=y" is defined in waagent.conf. Note: Ubuntu image needs to load rdma_ucm on boot to expose the RDMA CM interface to user-mode library. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1701744/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
[Kernel-packages] [Bug 1689627] Re: add version to rdma driver to support new method to bind to ND IP interface
** Patch added: "Patch for RDMA driver" https://bugs.launchpad.net/ubuntu/+source/walinuxagent/+bug/1689627/+attachment/4874277/+files/0001-vmbus-rdma-add-version-number.patch ** Package changed: walinuxagent (Ubuntu) => linux (Ubuntu) -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1689627 Title: [Hyper-V] add version to rdma driver to support new method to bind to ND IP interface Status in linux package in Ubuntu: Confirmed Bug description: The Linux agent (WALA) checks for module version, then decides how to pass the InfiniBand IP address to the RDMA driver. The new RDMA driver supports discovery of IP by trying to bind to all possible IP address. The driver needs to export the module version to have WALA use the new behavior. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1689627/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
[Kernel-packages] [Bug 1581243] Re: [Hyper-V] PCI Passthrough kernel hang and explicit barriers
Tested. Everything worked as expected. -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1581243 Title: [Hyper-V] PCI Passthrough kernel hang and explicit barriers Status in linux package in Ubuntu: In Progress Status in linux source package in Xenial: In Progress Status in linux source package in Yakkety: In Progress Bug description: Two upstream commits (right now in Bjorn Helgaas's PCI tree, and heading to Linus's tree) address potential hangs in PCI passthrough. Please consider these upstream items for 16.10 and 16.04 (and HWE kernels based on lts-xenial). https://git.kernel.org/cgit/linux/kernel/git/helgaas/pci.git/commit/?h=pci /host-hv=deb22e5c84c884a129d801cf3bfde7411536998d PCI: hv: Report resources release after stopping the bus Kernel hang is observed when pci-hyperv module is release with device drivers still attached. E.g., when I do 'rmmod pci_hyperv' with BCM5720 device pass-through-ed (tg3 module) I see the following: NMI watchdog: BUG: soft lockup - CPU#1 stuck for 22s! [rmmod:2104] ... Call Trace: [] tg3_read_mem+0x87/0x100 [tg3] [] ? 0xa063f000 [] tg3_poll_fw+0x85/0x150 [tg3] [] tg3_chip_reset+0x357/0x8c0 [tg3] [] tg3_halt+0x3b/0x190 [tg3] [] tg3_stop+0x171/0x230 [tg3] ... [] tg3_remove_one+0x90/0x140 [tg3] [] pci_device_remove+0x39/0xc0 [] __device_release_driver+0xa1/0x160 [] device_release_driver+0x23/0x30 [] pci_stop_bus_device+0x8a/0xa0 [] pci_stop_root_bus+0x36/0x60 [] hv_pci_remove+0x238/0x260 [pci_hyperv] The problem seems to be that we report local resources release before stopping the bus and removing devices from it and device drivers may try to perform some operations with these resources on shutdown. Move resources release report after we do pci_stop_root_bus(). Signed-off-by: Vitaly KuznetsovSigned-off-by: Bjorn Helgaas Acked-by: Jake Oshins https://git.kernel.org/cgit/linux/kernel/git/helgaas/pci.git/commit/?h=pci /host-hv=bdd74440d9e887b1fa648eefa17421def5f5243c PCI: hv: Add explicit barriers to config space accesspci/host-hv I'm trying to pass-through Broadcom BCM5720 NIC (Dell device 1f5b) on a Dell R720 server. Everything works fine when the target VM has only one CPU, but SMP guests reboot when the NIC driver accesses PCI config space with hv_pcifront_read_config()/hv_pcifront_write_config(). The reboot appears to be induced by the hypervisor and no crash is observed. Windows event logs are not helpful at all ('Virtual machine ... has quit unexpectedly'). The particular access point is always different and putting debug between them (printk/mdelay/...) moves the issue further away. The server model affects the issue as well: on Dell R420 I'm able to pass-through BCM5720 NIC to SMP guests without issues. While I'm obviously failing to reveal the essence of the issue I was able to come up with a (possible) solution: if explicit barriers are added to hv_pcifront_read_config()/hv_pcifront_write_config() the issue goes away. The essential minimum is rmb() at the end on _hv_pcifront_read_config() and wmb() at the end of _hv_pcifront_write_config() but I'm not confident it will be sufficient for all hardware. I suggest the following barriers: 1) wmb()/mb() between choosing the function and writing to its space. 2) mb() before releasing the spinlock in both _hv_pcifront_read_config()/ _hv_pcifront_write_config() to ensure that consecutive reads/writes to the space won't get re-ordered as drivers may count on that. Config space access is not supposed to be performance-critical so these explicit barriers should not cause any slowdown. [bhelgaas: use Linux "barriers" terminology] Signed-off-by: Vitaly Kuznetsov Signed-off-by: Bjorn Helgaas Acked-by: Jake Oshins To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1581243/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
[Kernel-packages] [Bug 1570124] Re: [Hyper-V] In-flight PCI Passthrough Patches
We have tested the proposed kernel. It is looking good. -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1570124 Title: [Hyper-V] In-flight PCI Passthrough Patches Status in linux package in Ubuntu: Fix Committed Status in linux source package in Xenial: Fix Committed Bug description: Some additional patches for PCI Passthrough have been submitted but not accepted upstream that are necessary for accessing multiple GPUs on the host. These are expected to get accepted upstream, but since there is not much time left in 16.04 I'm submitting them before they are accepted upstream. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1570124/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp