← Back to team overview

kernel-packages team mailing list archive

[Bug 1531768] Re: [arm64] multithreaded processes get locked up in futexes

 

Some good news: With bug 1534545 fixed I was now able to upgrade to the
Xenial 4.4 kernel. On the 4x CPU instance two parallel adt-run loops
have now run for about two hours without any dmesg spew. Stéphane has
run "lxc-test-concurrent -j 16 -i 10" twice on the 8x CPU instance
successfully too.

Bad news: I rebooted the 8x CPU instance (also xenial du jour with 4.4
kernel), and didn't do anything on it. After just sitting idle for an
hour or two ssh stopped responding and nova console-log shows
http://paste.ubuntu.com/14857144/ (only a hard reboot helped). So it wet
its pants without actually doing any action.

So it appears it's not fully fixed yet, but muuch better. I'll do some
more smoke testing, and if 4x CPU instances work, this is good enough to
put this into production. I'll keep the old Calxeda instances alive as a
fallback for a while, of course.

-- 
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/1531768

Title:
  [arm64] multithreaded processes get locked up in futexes

Status in linux package in Ubuntu:
  Confirmed

Bug description:
  I created an 8 CPU arm64 instance on Canonical's Scalingstack (which I
  want to use for armhf autopkgtesting in LXD). I started with wily as
  that has lxd available (it's not yet available in trusty nor the PPA
  for arm64).

  However, pretty much any LXD task that I do (I haven't tried much
  else) on this machine takes unbearably long. A simple "lxc profile set
  default raw.lxc lxc.seccomp=" or "lxc list" takes several minutes.

  I see tons of

  [ 1020.971955] rcu_sched kthread starved for 6000 jiffies! g1095 c1094 f0x0
  [ 1121.166926] INFO: task fsnotify_mark:69 blocked for more than 120 seconds.

  in dmesg (the attached apport info has the complete dmesg).

  ProblemType: Bug
  DistroRelease: Ubuntu 15.10
  Package: linux-image-4.2.0-22-generic 4.2.0-22.27
  ProcVersionSignature: User Name 4.2.0-22.27-generic 4.2.6
  Uname: Linux 4.2.0-22-generic aarch64
  AlsaDevices:
   total 0
   crw-rw---- 1 root audio 116,  1 Jan  7 09:18 seq
   crw-rw---- 1 root audio 116, 33 Jan  7 09:18 timer
  AplayDevices: Error: [Errno 2] No such file or directory: 'aplay'
  ApportVersion: 2.19.1-0ubuntu5
  Architecture: arm64
  ArecordDevices: Error: [Errno 2] No such file or directory: 'arecord'
  AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/seq', '/dev/snd/timer'] failed with exit code 1:
  CRDA: N/A
  Date: Thu Jan  7 09:24:01 2016
  IwConfig:
   eth0      no wireless extensions.

   lo        no wireless extensions.

   lxcbr0    no wireless extensions.
  Lspci:
   00:00.0 Host bridge [0600]: Red Hat, Inc. Device [1b36:0008]
    Subsystem: Red Hat, Inc Device [1af4:1100]
    Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
    Status: Cap- 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
  Lsusb: Error: command ['lsusb'] failed with exit code 1: unable to initialize libusb: -99
  PciMultimedia:

  ProcEnviron:
   TERM=screen
   PATH=(custom, no user)
   XDG_RUNTIME_DIR=<set>
   LANG=en_US.UTF-8
   SHELL=/bin/bash
  ProcFB:

  ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.2.0-22-generic root=LABEL=cloudimg-rootfs earlyprintk
  RelatedPackageVersions:
   linux-restricted-modules-4.2.0-22-generic N/A
   linux-backports-modules-4.2.0-22-generic  N/A
   linux-firmware                            1.149.3
  RfKill: Error: [Errno 2] No such file or directory: 'rfkill'
  SourcePackage: linux
  UdevLog: Error: [Errno 2] No such file or directory: '/var/log/udev'
  UpgradeStatus: No upgrade log present (probably fresh install)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1531768/+subscriptions


References