kernel-packages team mailing list archive

[Launchpad Bug Tracker <1545037@xxxxxxxxxxxxxxxxxx>]

This bug was fixed in the package linux - 4.4.0-6.21

---------------
linux (4.4.0-6.21) xenial; urgency=low

  [ Tim Gardner ]

  * Release Tracking Bug
    - LP: #1546283

  *  Naples/Zen, NTB Driver  (LP: #1542071)
    - [Config] CONFIG_NTB_AMD=m
    - NTB: Add support for AMD PCI-Express Non-Transparent Bridge

  * [Hyper-V] kernel panic occurs when installing Ubuntu Server x32 (LP: #1495983)
    - SAUCE: storvsc: use small sg_tablesize on x86

  * Enable arm64 emulation of removed ARMv7 instructions (LP: #1545542)
    - [Config] CONFIG_ARMV8_DEPRECATED=y

  * Surelock-GA2:kernel panic/ exception @ pcibios_set_pcie_reset_state+0x118/0x280 + cxl_reset+0x5c/0xc0 (LP: #1545037)
    - powerpc/eeh: Fix stale cached primary bus

  * Miscellaneous Ubuntu changes
    - SAUCE: fs: Add user namesapace member to struct super_block
    - SAUCE: fs: Limit file caps to the user namespace of the super block
    - SAUCE: Smack: Add support for unprivileged mounts from user namespaces
    - SAUCE: block_dev: Support checking inode permissions in lookup_bdev()
    - SAUCE: block_dev: Check permissions towards block device inode when mounting
    - SAUCE: fs: Treat foreign mounts as nosuid
    - SAUCE: selinux: Add support for unprivileged mounts from user namespaces
    - SAUCE: userns: Replace in_userns with current_in_userns
    - SAUCE: Smack: Handle labels consistently in untrusted mounts
    - SAUCE: fs: Check for invalid i_uid in may_follow_link()
    - SAUCE: cred: Reject inodes with invalid ids in set_create_file_as()
    - SAUCE: fs: Refuse uid/gid changes which don't map into s_user_ns
    - SAUCE: fs: Update posix_acl support to handle user namespace mounts
    - SAUCE: fs: Ensure the mounter of a filesystem is privileged towards its inodes
    - SAUCE: fs: Don't remove suid for CAP_FSETID in s_user_ns
    - SAUCE: fs: Allow superblock owner to access do_remount_sb()
    - SAUCE: capabilities: Allow privileged user in s_user_ns to set security.* xattrs
    - SAUCE: fuse: Add support for pid namespaces
    - SAUCE: fuse: Support fuse filesystems outside of init_user_ns
    - SAUCE: fuse: Restrict allow_other to the superblock's namespace or a descendant
    - SAUCE: fuse: Allow user namespace mounts
    - SAUCE: mtd: Check permissions towards mtd block device inode when mounting
    - SAUCE: fs: Update i_[ug]id_(read|write) to translate relative to s_user_ns
    - SAUCE: quota: Convert ids relative to s_user_ns
    - SAUCE: evm: Translate user/group ids relative to s_user_ns when computing HMAC
    - SAUCE: fs: Allow CAP_SYS_ADMIN in s_user_ns to freeze and thaw filesystems
    - SAUCE: quota: Treat superblock owner as privilged
    - SAUCE: ima/evm: Allow root in s_user_ns to set xattrs
    - SAUCE: block_dev: Forbid unprivileged mounting when device is opened for writing
    - SAUCE: ext4: Add support for unprivileged mounts from user namespaces
    - SAUCE: ext4: Add module parameter to enable user namespace mounts
    - SAUCE: fuse: Add module parameter to enable user namespace mounts

  * Miscellaneous upstream changes
    - megaraid: Fix possible NULL pointer deference in mraid_mm_ioctl
    - libahci: Implement the capability to override the generic ahci interrupt handler.
    - ata: Remove the AHCI_HFLAG_EDGE_IRQ support from libahci.
    - ahci_xgene: Implement the workaround to fix the missing of the edge interrupt for the HOST_IRQ_STAT.

 -- Tim Gardner <tim.gardner@xxxxxxxxxxxxx>  Fri, 12 Feb 2016 09:49:05
-0700

** Changed in: linux (Ubuntu Xenial)
       Status: Fix Committed => Fix Released

-- 
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/1545037

Title:
  Surelock-GA2:kernel panic/ exception @
  pcibios_set_pcie_reset_state+0x118/0x280 + cxl_reset+0x5c/0xc0

Status in linux package in Ubuntu:
  Fix Released
Status in linux source package in Xenial:
  Fix Released

Bug description:
  Requesting Canonical kernel team apply
  5b8255dd55f61ff1f65774a2084823415ec09a90 from upstream for 16.04

  Recent bug activity recap:

  == Comment: #17 - Guo Wen Shan  - 2016-02-08 22:40:35 ==
  Tentative patch sent to Andrew for verification. I'm putting a note here about the root cause: Each PE (struct eeh_pe) instance is tracking the primary bus by "pbus". When we have full hotplug as EEH recovery, the PE instance won't be refreshed. So the "pbus" is still dereferencing the stale PCI bus. Access to that at later pointer will cause kernel crash.

  The fix, as the patch sent to Andrew for verification, introduces
  another EEH PE flag (EEH_PE_PRI_BUS). That flag represents the
  validity of eeh_pe::pbus. Before unplugging in full hotplug case, the
  flag is cleared so that the eeh_pe::pbus can be updated at plugging
  time. I will send the patch to linux-ppc-dev for review after Andrew
  verified it helps.

  == Comment: #18 - Guo Wen Shan - 2016-02-08 23:55:00 ==
  The patch was sent to linux-ppc-dev for review/pickup: https://patchwork.ozlabs.org/patch/580626/ which reads "[v2,1/4] powerpc/eeh: Fix stale cached primary bus".

  == Comment: #19 - Andrew Donnellan - 2016-02-09 00:19:33 ==
  Patch has been submitted as per Gavin's comment. Please test!

  == Comment: #22 - Andrew Donnellan - 2016-02-11 01:41:06 ==
  Requesting Canonical kernel team apply 5b8255dd55f61ff1f65774a2084823415ec09a90 from upstream for 16.04

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1545037/+subscriptions