[SRU] RISC-V: Incorrect time-base frequency in KVM

[Heinrich Schuchardt]

[ Impact ]

When running on RISC-V under KVM the displayed time does not match the wall-clock.

On a specific system U-Boot's and GRUB's count-down before booting where 10 times slower than expected.

Please, apply upstream patch
385e575cd5ab ("target/riscv/kvm: fix timebase-frequency when using KVM acceleration")
to Ubuntu 22.04 Noble and 22.10 Oracular.

[ Test Plan ]

Boot a RISC-V virtual machine according to https://wiki.ubuntu.com/RISC-V/QEMU. Inside the virtual machine start a nested VM adding '-accel kvm' to the command line.

Do the same with '-accel tcg'

Check that the count-down in U-Boot and GRUB matches the wall time.

[ Where problems could occur ]

Only RISC-V code is changed.

VMs might not boot anymore.

The time-base frequency could still be wrong.

[ Other Info ]

n/a

[Sergio Durigan Junior (sergiodj)]

I've uploaded QEMU 9.0.2 to Oracular yesterday and it has this fix, so I'm marking its task as Fix Released.

[Sergio Durigan Junior (sergiodj)]

@Heinrich, I'm preparing a batch of bugfixes for QEMU on Noble. I see you assigned this bug to yourself, but I'm going to reassign it to me just to reflect the fact that I'll be doing the upload.

[Sergio Durigan Junior (sergiodj)]

Hi Heinrich,

Apologies for the delay; I'm preparing the SRU for Noble now. I've had to adjust the backported patch because of a build failure. Can you please double check it?

https://paste.debian.net/1330417/

The change affects only kvm_riscv_get_timebase_frequency. I wasn't entirely sure how to handle the case when env->kvm_timer_dirty is true, so I ignored it.

[Heinrich Schuchardt (xypron)]

Hello Sergio,

Thank you for taking care of this change.

In Linux time_init() sets riscv_timebase during initialization of the host. The value is either based on the device-tree's property /cpus/timebase-frequency or on the ACPI table RHTC.

With the patch this value is copied to /cpus/timebase-frequency in the client device-tree.

env->kvm_timer_dirty is not related to the value of riscv_timebase in the host. This is why it is ignored in the patch.

The only code difference I see between https://paste.debian.net/1330417/ and https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/2076927/+attachment/5805536/+files/0001-target-riscv-kvm-fix-timebase-frequency-when-using-K.patch seems to be the placement of this line:

uint64_t reg;

This line was move up 2 lines in the paste without functional change. Your paste looks correct.

[Sergio Durigan Junior (sergiodj)]

On Wednesday, September 25 2024, Heinrich Schuchardt wrote:

> Hello Sergio,
>
> Thank you for taking care of this change.
>
> In Linux time_init() sets riscv_timebase during initialization of the
> host. The value is either based on the device-tree's property
> /cpus/timebase-frequency or on the ACPI table RHTC.
>
> With the patch this value is copied to /cpus/timebase-frequency in the
> client device-tree.
>
> env->kvm_timer_dirty is not related to the value of riscv_timebase in
> the host. This is why it is ignored in the patch.
>
>
> The only code difference I see between https://paste.debian.net/1330417/ and https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/2076927/+attachment/5805536/+files/0001-target-riscv-kvm-fix-timebase-frequency-when-using-K.patch seems to be the placement of this line:
>
> uint64_t reg;
>
> This line was move up 2 lines in the paste without functional change.
> Your paste looks correct.

Thank you for confirming. I'm waiting for the build to finish in order
to run some tests, and will upload the package later today.

--
Sergio
GPG key ID: E92F D0B3 6B14 F1F4 D8E0 EB2F 106D A1C8 C3CB BF14

[Andreas Hasenack (ahasenack)]

Hello Heinrich, or anyone else affected,

Accepted qemu into noble-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/qemu/1:8.2.2+ds-0ubuntu1.3 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation on how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, what testing has been performed on the package and change the tag from verification-needed-noble to verification-done-noble. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed-noble. In either case, without details of your testing we will not be able to proceed.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance for helping!

N.B. The updated package will be released to -updates after the bug(s) fixed by this package have been verified and the package has been in -proposed for a minimum of 7 days.

[Ubuntu SRU Bot (ubuntu-sru-bot)]

All autopkgtests for the newly accepted qemu (1:8.2.2+ds-0ubuntu1.3) for noble have finished running.
The following regressions have been reported in tests triggered by the package:

cryptsetup/2:2.7.0-1ubuntu4.1 (ppc64el)

Please visit the excuses page listed below and investigate the failures, proceeding afterwards as per the StableReleaseUpdates policy regarding autopkgtest regressions [1].

https://people.canonical.com/~ubuntu-archive/proposed-migration/noble/update_excuses.html#qemu

[1] https://wiki.ubuntu.com/StableReleaseUpdates#Autopkgtest_Regressions

Thank you!

[Launchpad Janitor (janitor)]

This bug was fixed in the package qemu - 1:8.2.2+ds-0ubuntu1.4

---------------
qemu (1:8.2.2+ds-0ubuntu1.4) noble-security; urgency=medium

  * SECURITY UPDATE: denial of service
    - debian/patches/CVE-2024-4693-1.patch: virtio-pci: fix use of a
      released vector
    - debian/patches/CVE-2024-4693-2.patch: virtio-pci: Fix the use of
      an uninitialized irqfd
    - CVE-2024-4693
  * SECURITY UPDATE: heap buffer overflow
    - debian/patches/CVE-2024-7730.patch: add max size bounds check in
      input cb
    - CVE-2024-7730

 -- Bruce Cable <email address hidden> Tue, 22 Oct 2024 15:57:13 +1100