This commit appeared now in v5.15.46, which will be handled in the next
weeks by the kernel team.

But I change the title to 22.04, since 21.10 is EOL as of today, and this patch 
is not in jammy.
It got upstream accepted with v5.19-rc1 hence no need to patch kinetic, since 
kinetic's target kernel is 5.19.

** Changed in: ubuntu-z-systems
       Status: New => Triaged

** Changed in: linux (Ubuntu)
       Status: New => Triaged

** Also affects: linux (Ubuntu Jammy)
   Importance: Undecided
       Status: New

** Changed in: linux (Ubuntu Jammy)
       Status: New => Triaged

** Changed in: linux (Ubuntu)
       Status: Triaged => Invalid

** Changed in: linux (Ubuntu Jammy)
     Assignee: (unassigned) => Skipper Bug Screeners (skipper-screen-team)

** Changed in: linux (Ubuntu)
     Assignee: Skipper Bug Screeners (skipper-screen-team) => (unassigned)

-- 
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/1977962

Title:
  [UBUNTU 21.10] s390/perf: obtain sie_block from the right address

Status in Ubuntu on IBM z Systems:
  Triaged
Status in linux package in Ubuntu:
  Invalid
Status in linux source package in Jammy:
  Triaged

Bug description:
  Description:   s390/perf: obtain sie_block from the right address

  Symptom:       Read from unitialized stack memory, incorrect guest 
                 perf data reported, possibly system crash.

  Problem:       When perf is instructed to record a software-based event and
                 to trace the guest, i.e. using the following command:

                 perf kvm --guest --guestvmlinux=endless.elf top -e cpu-clock
                 
                 The reported guest instruction pointer is inaccurate or
                 the system may crash.

                 This is caused by the sie_block being taken from empty1[0]
                 in sie_block() in arch/s390/kernel/perf_event.c, but since 
                 1179f170b6f0 ("s390: fix fpu restore in entry.S")
                 it resides at empty1[1].

  Solution:      Obtain sie_block from the right address in sie_block().

  Reproduction:  1. Start a KVM guest executing an endless loop.
                 2. Determine the instruction address of the loop.
                 3. Start perf recording for any software event and trace 
                    the guest, for example using this command:

                    perf kvm --guest --guestvmlinux=endless.elf top -e
  cpu-clock

                 4. The reported guest instruction addresses don't match
                    the loop's instruction address or the system crashes.

  Upstream-ID:   c9bfb460c3e4da2462e16b0f0b200990b36b1dd2

  BBPF Probl.ID: 198450
  Preventive:    yes
  Reported:      -
  SupportTicket: -
  Reference:     -
  Date:          2022-06-02
  Author:        n...@linux.ibm.com
  Component:     kernel
  Link:          
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=c9bfb460c3e4da2462e16b0f0b200990b36b1dd2

  Problem occurs / affects impish and jammy

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu-z-systems/+bug/1977962/+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

Reply via email to