Upstream Commits Needed for DPDK on Azure

Bug #1812123 reported by Joseph Salisbury on 2019-01-16
10
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux-azure (Ubuntu)
Undecided
Marcelo Cerri
Xenial
Undecided
Unassigned
Bionic
Undecided
Marcelo Cerri
Cosmic
Undecided
Marcelo Cerri

Bug Description

The following two patches are related to multi-queue. They are needed in the linux-azure kernels for DPDK to work properly on Azure.

b5679cebf780 ("vmbus: fix subchannel removal")
5e3c420dcca5 ("uio_hv_generic: set callbacks on open")

Both commits have also been cc'd to upstream stable. Commit 5e3c420dcca5 is in mainline as of v4.20. Commit b5679cebf780 is still in linux-next.

affects: linux (Ubuntu) → linux-azure (Ubuntu)
tags: added: kernel-hyper-v
Joshua R. Poulson (jrp) wrote :

Affects 4.15-azure in Xenial, 4.15-azure and 4.18-azure in Bionic.

Joseph Salisbury (jsalisbury) wrote :

Commit b5679cebf780 is now in mainline as of v5.0-rc4

Joshua R. Poulson (jrp) on 2019-01-31
Changed in linux-azure (Ubuntu):
status: New → Confirmed
Marcelo Cerri (mhcerri) on 2019-02-06
Changed in linux-azure (Ubuntu):
assignee: nobody → Marcelo Cerri (mhcerri)
Changed in linux-azure (Ubuntu Bionic):
assignee: nobody → Marcelo Cerri (mhcerri)
Changed in linux-azure (Ubuntu Cosmic):
assignee: nobody → Marcelo Cerri (mhcerri)
status: New → In Progress
Marcelo Cerri (mhcerri) wrote :

b5679cebf780 ("vmbus: fix subchannel removal") seems to be a fix to ae6935ed7d42 ("vmbus: split ring buffer allocation from open"), which we currently don't have in the azure kernel. Is it really necessary and we should pick ae6935ed7d42 as well, or is it possible to skip it?

If it's necessary I can provide a test kernel with just 5e3c420dcca5 ("uio_hv_generic: set callbacks on open") and its dependencies.

Marcelo Cerri (mhcerri) wrote :

Test kernel for cosmic: https://kernel.ubuntu.com/~mhcerri/azure/lp1812123-cosmic/

Please let's know if it's working as expected.

The 4.15 kernel will require more commits to be cherry picked, can you give us a better reasoning why those commits are necessary? That might be useful when backporting them to 4.15.

Joshua R. Poulson (jrp) wrote :

I believe these fixes from testing the scenario of rescinding/revoking the VF while DPDK is active, which is a scenario triggered by a host networking driver update.

Joshua R. Poulson (jrp) wrote :

Because these kernels are made for 18.10, but our DPDK PPAs are on 18.04 and 16.04, there has been difficulty testing this kernel. Can we get a one-off build for bionic?

Changed in linux-azure (Ubuntu Cosmic):
status: In Progress → Fix Committed
Marcelo Cerri (mhcerri) wrote :

Hi, Josh.

The changes are already applied and will be included in this current cycle. Let us know if you have found any issues with the test kernel.

Launchpad Janitor (janitor) wrote :
Download full text (15.2 KiB)

This bug was fixed in the package linux-azure - 4.18.0-1018.18

---------------
linux-azure (4.18.0-1018.18) cosmic; urgency=medium

  [ Ubuntu: 4.18.0-20.21 ]

  * CVE-2018-12126 // CVE-2018-12127 // CVE-2018-12130
    - Documentation/l1tf: Fix small spelling typo
    - x86/cpu: Sanitize FAM6_ATOM naming
    - kvm: x86: Report STIBP on GET_SUPPORTED_CPUID
    - locking/atomics, asm-generic: Move some macros from <linux/bitops.h> to a
      new <linux/bits.h> file
    - tools include: Adopt linux/bits.h
    - x86/msr-index: Cleanup bit defines
    - x86/speculation: Consolidate CPU whitelists
    - x86/speculation/mds: Add basic bug infrastructure for MDS
    - x86/speculation/mds: Add BUG_MSBDS_ONLY
    - x86/kvm: Expose X86_FEATURE_MD_CLEAR to guests
    - x86/speculation/mds: Add mds_clear_cpu_buffers()
    - x86/speculation/mds: Clear CPU buffers on exit to user
    - x86/kvm/vmx: Add MDS protection when L1D Flush is not active
    - x86/speculation/mds: Conditionally clear CPU buffers on idle entry
    - x86/speculation/mds: Add mitigation control for MDS
    - x86/speculation/mds: Add sysfs reporting for MDS
    - x86/speculation/mds: Add mitigation mode VMWERV
    - Documentation: Move L1TF to separate directory
    - Documentation: Add MDS vulnerability documentation
    - x86/speculation/mds: Add mds=full,nosmt cmdline option
    - x86/speculation: Move arch_smt_update() call to after mitigation decisions
    - x86/speculation/mds: Add SMT warning message
    - x86/speculation/mds: Fix comment
    - x86/speculation/mds: Print SMT vulnerable on MSBDS with mitigations off
    - x86/speculation/mds: Add 'mitigations=' support for MDS
  * CVE-2017-5715 // CVE-2017-5753
    - s390/speculation: Support 'mitigations=' cmdline option
  * CVE-2017-5715 // CVE-2017-5753 // CVE-2017-5754 // CVE-2018-3639
    - powerpc/speculation: Support 'mitigations=' cmdline option
  * CVE-2017-5715 // CVE-2017-5754 // CVE-2018-3620 // CVE-2018-3639 //
    CVE-2018-3646
    - cpu/speculation: Add 'mitigations=' cmdline option
    - x86/speculation: Support 'mitigations=' cmdline option
  * Packaging resync (LP: #1786013)
    - [Packaging] resync git-ubuntu-log

linux-azure (4.18.0-1017.17) cosmic; urgency=medium

  * linux-azure: 4.18.0-1017.17 -proposed tracker (LP: #1826166)

  * [linux-azure] Include mainline commits fc96df16a1ce and ba50bf1ce9a5 in
    Azure kernel (LP: #1821378)
    - Drivers: hv: vmbus: Return -EINVAL for the sys files for unopened channels
    - Drivers: hv: vmbus: Check for ring when getting debug info

  * [linux-azure] Commit To Improve NVMe Performance (LP: #1819689)
    - blk-mq: remove the request_list usage

  [ Ubuntu: 4.18.0-19.20 ]

  * linux: 4.18.0-19.20 -proposed tracker (LP: #1826171)
  * Packaging resync (LP: #1786013)
    - [Packaging] resync git-ubuntu-log
  * autopkgtests run too often, too much and don't skip enough (LP: #1823056)
    - [Debian] Set +x on rebuild testcase.
    - [Debian] Skip rebuild test, for regression-suite deps.
    - [Debian] Make ubuntu-regression-suite skippable on unbootable kernels.
    - [Debian] make rebuild use skippable error codes when skipping.
    - [Debian] Only run regression-suite...

Changed in linux-azure (Ubuntu Cosmic):
status: Fix Committed → Fix Released
Launchpad Janitor (janitor) wrote :
Download full text (15.7 KiB)

This bug was fixed in the package linux-azure - 4.18.0-1018.18~18.04.1

---------------
linux-azure (4.18.0-1018.18~18.04.1) bionic; urgency=medium

  [ Ubuntu: 4.18.0-1018.18 ]

  * CVE-2018-12126 // CVE-2018-12127 // CVE-2018-12130
    - Documentation/l1tf: Fix small spelling typo
    - x86/cpu: Sanitize FAM6_ATOM naming
    - kvm: x86: Report STIBP on GET_SUPPORTED_CPUID
    - locking/atomics, asm-generic: Move some macros from <linux/bitops.h> to a
      new <linux/bits.h> file
    - tools include: Adopt linux/bits.h
    - x86/msr-index: Cleanup bit defines
    - x86/speculation: Consolidate CPU whitelists
    - x86/speculation/mds: Add basic bug infrastructure for MDS
    - x86/speculation/mds: Add BUG_MSBDS_ONLY
    - x86/kvm: Expose X86_FEATURE_MD_CLEAR to guests
    - x86/speculation/mds: Add mds_clear_cpu_buffers()
    - x86/speculation/mds: Clear CPU buffers on exit to user
    - x86/kvm/vmx: Add MDS protection when L1D Flush is not active
    - x86/speculation/mds: Conditionally clear CPU buffers on idle entry
    - x86/speculation/mds: Add mitigation control for MDS
    - x86/speculation/mds: Add sysfs reporting for MDS
    - x86/speculation/mds: Add mitigation mode VMWERV
    - Documentation: Move L1TF to separate directory
    - Documentation: Add MDS vulnerability documentation
    - x86/speculation/mds: Add mds=full,nosmt cmdline option
    - x86/speculation: Move arch_smt_update() call to after mitigation decisions
    - x86/speculation/mds: Add SMT warning message
    - x86/speculation/mds: Fix comment
    - x86/speculation/mds: Print SMT vulnerable on MSBDS with mitigations off
    - x86/speculation/mds: Add 'mitigations=' support for MDS
  * CVE-2017-5715 // CVE-2017-5753
    - s390/speculation: Support 'mitigations=' cmdline option
  * CVE-2017-5715 // CVE-2017-5753 // CVE-2017-5754 // CVE-2018-3639
    - powerpc/speculation: Support 'mitigations=' cmdline option
  * CVE-2017-5715 // CVE-2017-5754 // CVE-2018-3620 // CVE-2018-3639 //
    CVE-2018-3646
    - cpu/speculation: Add 'mitigations=' cmdline option
    - x86/speculation: Support 'mitigations=' cmdline option
  * Packaging resync (LP: #1786013)
    - [Packaging] resync git-ubuntu-log

linux-azure (4.18.0-1017.17~18.04.1) bionic; urgency=medium

  * linux-azure: 4.18.0-1017.17~18.04.1 -proposed tracker (LP: #1826165)

  * Packaging resync (LP: #1786013)
    - [Packaging] update helper scripts

  [ Ubuntu: 4.18.0-1017.17 ]

  * linux-azure: 4.18.0-1017.17 -proposed tracker (LP: #1826166)
  * [linux-azure] Include mainline commits fc96df16a1ce and ba50bf1ce9a5 in
    Azure kernel (LP: #1821378)
    - Drivers: hv: vmbus: Return -EINVAL for the sys files for unopened channels
    - Drivers: hv: vmbus: Check for ring when getting debug info
  * [linux-azure] Commit To Improve NVMe Performance (LP: #1819689)
    - blk-mq: remove the request_list usage
  * linux: 4.18.0-19.20 -proposed tracker (LP: #1826171)
  * Packaging resync (LP: #1786013)
    - [Packaging] resync git-ubuntu-log
  * autopkgtests run too often, too much and don't skip enough (LP: #1823056)
    - [Debian] Set +x on rebuild testcase.
    - [Debian] Skip rebuild test, for regression-suite deps.
    - [Deb...

Changed in linux-azure (Ubuntu Bionic):
status: New → Fix Released

The verification of the Stable Release Update for linux-azure has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regressions.

Marcelo Cerri (mhcerri) wrote :

Hi, Joe and Josh.

I made a 4.15 test kernel available at https://kernel.ubuntu.com/~mhcerri/azure/lp1812123/

You can find the backported patches in the same directory.

Do you think you can test this kernel and verify if it's working properly?

Marcelo Cerri (mhcerri) on 2019-06-10
Changed in linux-azure (Ubuntu Xenial):
status: New → Fix Committed
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package linux-azure - 4.15.0-1049.54

---------------
linux-azure (4.15.0-1049.54) xenial; urgency=medium

  * linux-azure: 4.15.0-1049.54 -proposed tracker (LP: #1834091)

  * Upstream Commits Needed for DPDK on Azure (LP: #1812123)
    - uio: document uio_hv_generic regions
    - uio_hv_generic: create send and receive buffers
    - uio_hv_generic: add rescind support
    - uio_hv_generic: support sub-channels
    - uio_hv_generic: fix subchannel ring mmap
    - uio_hv_generic: use standard mmap for resources
    - vmbus: keep pointer to ring buffer page
    - uio: introduce UIO_MEM_IOVA
    - hv_uio_generic: map ringbuffer phys addr
    - uio_hv_generic: use ISR callback method
    - uio_hv_generic: use correct channel in isr
    - uio_hv_generic: make ring buffer attribute for primary channel
    - uio_hv_generic: defer opening vmbus until first use
    - uio_hv_generic: set callbacks on open
    - vmbus: pass channel to hv_process_channel_removal
    - vmbus: split ring buffer allocation from open
    - vmbus: fix subchannel removal

  * Enable eBPF JIT in the linux-azure kernels (LP: #1827916)
    - [Config] linux-azure: CONFIG_MLX{4,5}_INFINIBAND=m

  * [linux-azure] Please Include Mainline Commit ebaf39e6032f in the 16.04 and
    18.04 linux-azure kernels (LP: #1830266)
    - ipv4: ipv6: netfilter: Adjust the frag mem limit when truesize changes

  * [linux-azure] Commits Requested in 16.04 for the Azure Kernel (LP: #1830242)
    - blk-mq: remove the request_list usage
    - nvme-pci: remove cq check after submission
    - nvme-pci: split the nvme queue lock into submission and completion locks

  * [Packaging] Improve config annotations check on custom kernels
    (LP: #1820075)
    - [Config] linux-azure: Include custom annotations files

  [ Ubuntu: 4.15.0-54.58 ]

  * linux: 4.15.0-54.58 -proposed tracker (LP: #1833987)
  * Remote denial of service (resource exhaustion) caused by TCP SACK scoreboard
    manipulation (LP: #1831638) // CVE-2019-11478
    - tcp: refine memory limit test in tcp_fragment()
  * CVE-2019-11479
    - SAUCE: tcp: add tcp_min_snd_mss sysctl
    - SAUCE: tcp: enforce tcp_min_snd_mss in tcp_mtu_probing()

 -- Marcelo Henrique Cerri <email address hidden> Mon, 24 Jun 2019 17:44:20 -0300

Changed in linux-azure (Ubuntu Xenial):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers