Kernel fails to enable XSAVE when running in a “v5” AMD SEV-SNP VM

[John Cabaj]

SRU Justification

[Impact]

* Request from MSFT to include patch to disable CET Shadow Stack on SEV-SNP. Due to CPUID emulation bug not being fixed in VMs until October 2024, this feature is being disabled in the interim.

[Fix]

* Clean cherry-pick SAUCE patch from MSFT tree, https://github.com/dcui/linux
* cdf5cdd: "x86/hyperv: temporarily disable CET SS on SEV-SNP due to a paravisor bug"

[Test Plan]

* Compile tested
* Boot tested
* Microsoft tested

[Regression potential]

* Changes isolated, minimal regression risk

[Other info]

* SF #00387447

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This bug is awaiting verification that the linux-azure/6.8.0-1010.10 kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-noble-linux-azure' to 'verification-done-noble-linux-azure'. If the problem still exists, change the tag 'verification-needed-noble-linux-azure' to 'verification-failed-noble-linux-azure'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux-azure - 6.8.0-1010.10

---------------
linux-azure (6.8.0-1010.10) noble; urgency=medium

  * noble/linux-azure: 6.8.0-1010.10 -proposed tracker (LP: #2068298)

  * Kernel fails to enable XSAVE when running in a “v5” AMD SEV-SNP VM
    (LP: #2069204)
    - SAUCE: x86/hyperv: temporarily disable CET SS on SEV-SNP due to a paravisor
      bug

  [ Ubuntu: 6.8.0-38.38 ]

  * noble/linux: 6.8.0-38.38 -proposed tracker (LP: #2068318)
  * race_sched in ubuntu_stress_smoke_test will cause kernel panic on 6.8 with
    Azure Standard_A2_v2 instance (LP: #2068024)
    - sched/eevdf: Prevent vlag from going out of bounds in reweight_eevdf()
  * Noble: btrfs: re-introduce 'norecovery' mount option (LP: #2068591)
    - btrfs: re-introduce 'norecovery' mount option
  * Fix system hang while entering suspend with AMD Navi3x graphics
    (LP: #2063417)
    - drm/amdgpu/mes: fix use-after-free issue
  * Noble update: v6.8.8 upstream stable release (LP: #2068087)
    - io_uring: Fix io_cqring_wait() not restoring sigmask on get_timespec64()
      failure
    - drm/i915/cdclk: Fix voltage_level programming edge case
    - Revert "vmgenid: emit uevent when VMGENID updates"
    - SUNRPC: Fix rpcgss_context trace event acceptor field
    - selftests/ftrace: Limit length in subsystem-enable tests
    - random: handle creditable entropy from atomic process context
    - scsi: core: Fix handling of SCMD_FAIL_IF_RECOVERING
    - net: usb: ax88179_178a: avoid writing the mac address before first reading
    - btrfs: do not wait for short bulk allocation
    - btrfs: zoned: do not flag ZEROOUT on non-dirty extent buffer
    - r8169: fix LED-related deadlock on module removal
    - r8169: add missing conditional compiling for call to r8169_remove_leds
    - scsi: ufs: qcom: Add missing interconnect bandwidth values for Gear 5
    - netfilter: nf_tables: Fix potential data-race in __nft_expr_type_get()
    - netfilter: nf_tables: Fix potential data-race in __nft_obj_type_get()
    - netfilter: br_netfilter: skip conntrack input hook for promisc packets
    - netfilter: nft_set_pipapo: constify lookup fn args where possible
    - netfilter: nft_set_pipapo: walk over current view on netlink dump
    - netfilter: flowtable: validate pppoe header
    - netfilter: flowtable: incorrect pppoe tuple
    - af_unix: Call manage_oob() for every skb in unix_stream_read_generic().
    - af_unix: Don't peek OOB data without MSG_OOB.
    - net: sparx5: flower: fix fragment flags handling
    - net/mlx5: Lag, restore buckets number to default after hash LAG deactivation
    - net/mlx5: Restore mistakenly dropped parts in register devlink flow
    - net/mlx5e: Prevent deadlock while disabling aRFS
    - net: change maximum number of UDP segments to 128
    - octeontx2-pf: fix FLOW_DIS_IS_FRAGMENT implementation
    - selftests/tcp_ao: Make RST tests less flaky
    - selftests/tcp_ao: Zero-init tcp_ao_info_opt
    - selftests/tcp_ao: Fix fscanf() call for format-security
    - selftests/tcp_ao: Printing fixes to confirm with format-security
    - net: stmmac: Apply half-duplex-less constraint for DW QoS Eth only
    - net: stmmac: Fix max-speed being...

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This bug is awaiting verification that the linux-azure-6.8/6.8.0-1010.10~22.04.1 kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-jammy-linux-azure-6.8' to 'verification-done-jammy-linux-azure-6.8'. If the problem still exists, change the tag 'verification-needed-jammy-linux-azure-6.8' to 'verification-failed-jammy-linux-azure-6.8'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!