Prevent timer value 0 for MWAITX

[Kim Naru]

Newer hardware has uncovered a bug in the software implementation of using MWAITX for the delay function. A value of 0 for the timer is meant to indicate that a timeout will not be used to exit MWAITX. On newer hardware this can result in MWAITX never returning, resulting in NMI soft lockup messages being printed. On older hardware, some of the other conditions under which MWAITX can exit masked this issue.

The AMD APM does not currently document this and will be updated. Please refer to http://marc.info/?l=kvm&m=148950623231140 for information regarding NMI soft lockup messages on an AMD Ryzen 1800X. This has been root-caused as a 0 passed to MWAITX causing it to wait indefinitely. This change has the added benefit of avoiding the unnecessary setup of MONITORX/MWAITX when the delay value is zero.

https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=88d879d29f9cc0de2d930b584285638cdada6625

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

This bug is missing log files that will aid in diagnosing the problem. While running an Ubuntu kernel (not a mainline or third-party kernel) please enter the following command in a terminal window:

apport-collect 1729442

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Joseph Salisbury (jsalisbury)]

I built Xenial and Zesty test kernels, both with a pick of commit 88d879d29f9cc0d. The test kernels can be downloaded from:

http://kernel.ubuntu.com/~jsalisbury/lp1729442/xenial
http://kernel.ubuntu.com/~jsalisbury/lp1729442/zesty

Can you test this kernels and see if they resolve this bug?

[Kim Naru (kim-naru)]

Joe,
I will take a look.

-kim

-----Original Message-----
From: <email address hidden> [mailto:<email address hidden>] On Behalf Of Joseph Salisbury
Sent: Thursday, November 02, 2017 1:56 PM
To: Naru, Kim <email address hidden>
Subject: [Bug 1729442] Re: Prevent timer value 0 for MWAITX

I built Xenial and Zesty test kernels, both with a pick of commit 88d879d29f9cc0d. The test kernels can be downloaded from:

http://kernel.ubuntu.com/~jsalisbury/lp1729442/xenial
http://kernel.ubuntu.com/~jsalisbury/lp1729442/zesty

Can you test this kernels and see if they resolve this bug?

--
You received this bug notification because you are a member of AMD Team, which is subscribed to amd.
https://bugs.launchpad.net/bugs/1729442

Title:
  Prevent timer value 0 for MWAITX

Status in amd:
  New
Status in linux package in Ubuntu:
  In Progress
Status in linux source package in Xenial:
  In Progress
Status in linux source package in Zesty:
  In Progress

Bug description:
  Newer hardware has uncovered a bug in the software implementation of
  using MWAITX for the delay function. A value of 0 for the timer is
  meant to indicate that a timeout will not be used to exit MWAITX. On
  newer hardware this can result in MWAITX never returning, resulting in
  NMI soft lockup messages being printed. On older hardware, some of the
  other conditions under which MWAITX can exit masked this issue.

  The AMD APM does not currently document this and will be updated.
  Please refer to http://marc.info/?l=kvm&m=148950623231140 for
  information regarding NMI soft lockup messages on an AMD Ryzen 1800X.
  This has been root-caused as a 0 passed to MWAITX causing it to wait
  indefinitely. This change has the added benefit of avoiding the
  unnecessary setup of MONITORX/MWAITX when the delay value is zero.

  https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=88d879d29f9cc0de2d930b584285638cdada6625

To manage notifications about this bug go to:
https://bugs.launchpad.net/amd/+bug/1729442/+subscriptions