linux-libc-dev in proposed cause seccomp_bpf from seccomp in ubuntu_kernel_selftests to fail on F-oem-5.6

[Po-Hsu Lin]

Issue found on node glameow with 5.6.0-1028.28

# [ RUN ] global.user_notification_kill_in_middle
# [ FAIL ] global.user_notification_kill_in_middle

74 / 75 tests passed.

Since this might be the first time we're testing Focal OEM kernel with this dedicated node glameow, further investigation is needed.
---
ProblemType: Bug
AlsaDevices:
 total 0
 crw-rw---- 1 root audio 116, 1 Sep 18 06:34 seq
 crw-rw---- 1 root audio 116, 33 Sep 18 06:34 timer
AplayDevices: Error: [Errno 2] No such file or directory: 'aplay'
ApportVersion: 2.20.11-0ubuntu27.8
Architecture: amd64
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:
CasperMD5CheckResult: skip
DistroRelease: Ubuntu 20.04
IwConfig: Error: [Errno 2] No such file or directory: 'iwconfig'
MachineType: HP ProLiant DL360 Gen9
Package: linux (not installed)
PciMultimedia:

ProcFB: 0 mgag200drmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-5.6.0-1028-oem root=UUID=6422cfdd-2a69-4c0b-9784-6809a77ab980 ro console=ttyS0,96008n1
ProcVersionSignature: User Name 5.6.0-1028.28-oem 5.6.19
RelatedPackageVersions:
 linux-restricted-modules-5.6.0-1028-oem N/A
 linux-backports-modules-5.6.0-1028-oem N/A
 linux-firmware 1.187.3
RfKill: Error: [Errno 2] No such file or directory: 'rfkill'
Tags: focal uec-images
Uname: Linux 5.6.0-1028-oem x86_64
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm audio cdrom dialout dip floppy lxd netdev plugdev sudo video
_MarkForUpload: True
dmi.bios.date: 04/25/2017
dmi.bios.vendor: HP
dmi.bios.version: P89
dmi.board.name: ProLiant DL360 Gen9
dmi.board.vendor: HP
dmi.chassis.type: 23
dmi.chassis.vendor: HP
dmi.modalias: dmi:bvnHP:bvrP89:bd04/25/2017:svnHP:pnProLiantDL360Gen9:pvr:rvnHP:rnProLiantDL360Gen9:rvr:cvnHP:ct23:cvr:
dmi.product.family: ProLiant
dmi.product.name: ProLiant DL360 Gen9
dmi.product.sku: 780020-S01
dmi.sys.vendor: HP

[Po-Hsu Lin (cypressyew)]

Test passed with 5.6.0-1027-oem + source code from 5.6.0-1027-oem

[Po-Hsu Lin (cypressyew)]

Manually tested 5.6.0-1028 on node glameow, this test passed without any problem.

Reschedule this on jenkins

[Po-Hsu Lin (cypressyew)]

I have tested the following combination:

Kernel 5.6.0-1028 + Source code 5.6.0-1028
Kernel 5.6.0-1028 + Source code 5.6.0-1027
Kernel 5.6.0-1027 + Source code 5.6.0-1028
Kernel 5.6.0-1027 + Source code 5.6.0-1027

All these combination can pass without any issue, the key point is the dist-upgrade command
After investigation, it's the linux-libc-dev package in proposed.

linux-libc-dev 5.4.0-47.51 (updates) works well, but the one in proposed 5.4.0-48.52 will cause this issue.

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

[Po-Hsu Lin (cypressyew)]

apport information

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

This change was made by a bot.

[Po-Hsu Lin (cypressyew)]

Note that the dist-upgrade will be performed after the deployment with MAAS (before the test starts).

[Po-Hsu Lin (cypressyew)]

This test will pass with commit 47e33c05f9f07cac3de833e531bcac9ae052c7ca ("seccomp: Fix ioctl number for SECCOMP_IOCTL_NOTIF_ID_VALID")

Test kernel could be found here:
https://people.canonical.com/~phlin/kernel/lp-1896008-linux-libc-dev/

[Timo Aaltonen (tjaalton)]

should it be enough to add that commit next cycle?

[Thadeu Lima de Souza Cascardo (cascardo)]

So, the test is running with the newer definition as it is picked up from linux-libc-dev. But the kernel (linux-oem-5.6, in this case) doesn't know about it. So it fails.

However, any binaries out there would be using the old definition, which the linux-oem-5.6 should accept just fine. I guess we will be stuck with this double definition forever as "we don't break userspace ABI *ever*" *coff*.

IMHO, it's just fine leaving this for next cycle. But pick it up, cause we will start seeing binaries out there using the new definition.

Cascardo.

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

This bug is awaiting verification that the kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-focal' to 'verification-done-focal'. If the problem still exists, change the tag 'verification-needed-focal' to 'verification-failed-focal'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux-oem-5.6 - 5.6.0-1031.32

---------------
linux-oem-5.6 (5.6.0-1031.32) focal; urgency=medium

  * Packaging resync (LP: #1786013)
    - update dkms package versions

  * CVE-2020-16119
    - SAUCE: dccp: avoid double free of ccid on child socket

  * CVE-2020-16120
    - ovl: pass correct flags for opening real directory
    - ovl: switch to mounter creds in readdir
    - ovl: verify permissions in ovl_path_open()
    - ovl: call secutiry hook in ovl_real_ioctl()
    - ovl: check permission to open real file

linux-oem-5.6 (5.6.0-1030.30) focal; urgency=medium

  * focal/linux-oem-5.6: 5.6.0-1030.30 -proposed tracker (LP: #1898714)

  * Fix broken e1000e device after S3 (LP: #1897755)
    - SAUCE: e1000e: Increase polling timeout on MDIC ready bit

  * Fix headset jack detection on Dell UFF desktops (LP: #1897508)
    - Revert "ALSA: hda: call runtime_allow() for all hda controllers"

  * [SRU][OEM-5.6]Fix AMD system hang after re-plugin USBC-DP dongle
    (LP: #1896716)
    - drm/amd/display: Move USB-C workaround to after parameter variables are set

  * alsa/hda: mic doesn't work on a HP AIO machine (LP: #1897694)
    - SAUCE: ALSA: hda/realtek - set mic to auto detect on a HP AIO machine

  * CVE-2020-26088
    - net/nfc/rawsock.c: add CAP_NET_RAW check.

  * HP Zbook Studio G7 boots into corrupted screen with PSR featured panel
    (LP: #1897501)
    - SAUCE: drm/i915/psr: allow overriding PSR disable param by quirk
    - SAUCE: drm/dp: add DP_QUIRK_FORCE_PSR_CHIP_DEFAULT quirk to CMN prod-ID
      19-15

linux-oem-5.6 (5.6.0-1029.29) focal; urgency=medium

  * focal/linux-oem-5.6: 5.6.0-1029.29 -proposed tracker (LP: #1895982)

  * linux-libc-dev in proposed cause seccomp_bpf from seccomp in
    ubuntu_kernel_selftests to fail on F-oem-5.6 (LP: #1896008)
    - seccomp: Fix ioctl number for SECCOMP_IOCTL_NOTIF_ID_VALID

  * Enable LTR for endpoints behind VMD (LP: #1896598)
    - SAUCE: PCI/ASPM: Enable LTR for endpoints behind VMD

  * CVE-2020-14385
    - xfs: fix boundary test in xfs_attr_shortform_verify

  * Fix non-working NVMe after S3 (LP: #1895718)
    - SAUCE: PCI: Enable ACS quirk on CML root port

  * CVE-2020-12655
    - xfs: add agf freeblocks verify in xfs_agf_verify

  * Thunderbolt3 daisy chain sometimes doesn't work (LP: #1895606)
    - thunderbolt: Retry DROM read once if parsing fails

  * CVE-2020-12656
    - sunrpc: svcauth_gss_register_pseudoflavor must reject duplicate
      registrations.
    - sunrpc: clean up properly in gss_mech_unregister()

  * CVE-2020-15393
    - usb: usbtest: fix missing kfree(dev->buf) in usbtest_disconnect

  * CVE-2020-12771
    - bcache: fix potential deadlock problem in btree_gc_coalesce

  * CVE-2020-15780
    - ACPI: configfs: Disallow loading ACPI tables when locked down

  * CVE-2020-16166
    - random32: update the net random state on interrupt and activity

  * Get the extended DPCD according to the related field in the DPCD registers.
    (LP: #1893752)
    - drm: move DP_MAX_DOWNSTREAM_PORTS from i915 to drm core
    - drm/i915/dp: Extract drm_dp_read_dpcd_caps()
    - SAUCE: drm/dp_mst: Retrieve extended DPCD caps for topology manager
...

[Po-Hsu Lin (cypressyew)]

seccomp_bpf passed with 5.6.0-1035.37-oem