ubuntu-kernel-tests

net test from ubuntu_kernel_selftest will timeout on B-5.0

Bug #1888381 reported by Po-Hsu Lin on 2020-07-21

This bug affects 1 person

	Status	Importance	Assigned to
ubuntu-kernel-tests	Fix Released	Undecided	Unassigned
linux-oem-osp1 (Ubuntu)	Fix Released	Medium	Unassigned
Bionic	Fix Released	Undecided	Unassigned

Bug Description

[Impact]
Partial reads on TLS sockets will cause processes to wait indefinitively.

[Test case]
Run tls net selftests from the kernel source tree.

[Regression potential]
We only touch code on the TLS socket family itself, so regressions on subsystems other than the one we are fixing is highly unlikely.

--------------------

Issue found on:
linux-gke-5.0 - 5.0.0-1045.46 / 5.0.0-1044.45
linux-oem-osp1 - 5.0.0-1065.70 / 5.0.0-1064.69

The net test in ubuntu_kernel_selftests will timeout and gets killed, this issue does not exist in gke 5.0.0-1043.44 and oem-osp1 5.0.0-1063.68 (the net test passed with 5.0.0-1063.68 without any error)

The test will hang here:
Opening 127.0.0.1:9999
Opening INADDR_ANY:9999
bind: Address already in use
Opening in6addr_any:9999
Opening INADDR_ANY:9999
bind: Address already in use
Opening INADDR_ANY:9999 after closing ipv6 socket
bind: Address already in use
Timer expired (1800 sec.), nuking pid 27394
ERROR ubuntu_kernel_selftests.net ubuntu_kernel_selftests.net timestamp=1595327738 localtime=Jul 21 10:35:38 Test timeout expired, rc=15
END ERROR ubuntu_kernel_selftests.net ubuntu_kernel_selftests.net

Please find the attachment for the complete test log.

See original description

Tags:

Revision history for this message

Po-Hsu Lin (cypressyew) wrote on 2020-07-21:

gke-5.0.0-1045-kselftest.log Edit (890.2 KiB, text/plain)

Po-Hsu Lin (cypressyew) on 2020-07-21

tags:	added: 5.0 bionic kqa-blocker sru-20200629
tags:	added: gke

Po-Hsu Lin (cypressyew) on 2020-07-21

description:

updated

Revision history for this message

Sean Feole (sfeole) wrote on 2020-07-21:

gkeuks.tar Edit (18.4 MiB, application/x-tar)

Download full text (7.5 KiB)

I did see this failure in the logs, however was not able to reproduce it when ran manually on GKE.
I have attached the tarball of the logs.

Ftrace test output below:

07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:26 DEBUG| 07/21 13:32:28 DEBUG| 07/21 13:32:29 DEBUG| 07/21 13:32:29 DEBUG| 07/21 13:32:29 DEBUG| 07/21 13:32:29 DEBUG| 07/21 13:32:29 DEBUG| 07/21 13:32:29 DEBUG| 07/21 13:32:29 DEBUG| 07/21 13:32:32 DEBUG| 07/21 13:32:33 DEBUG| 07/21 13:32:34 DEBUG| 07/21 13:32:37 DEBUG| 07/21 13:32:41 DEBUG| 07/21 13:32:42 DEBUG| 07/21 13:32:43 DEBUG| 07/21 13:32:47 DEBUG| 07/21 13:32:48 DEBUG| utils:0116| Running 'sudo sh -c 'echo 1 > /proc/sys/net/ipv4/conf/all/accept_local''
utils:0116| Running 'sudo make -C linux/tools/testing/selftests TARGETS=ftrace run_tests'
utils:0153| [stdout] make: Entering directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests'
utils:0153| [stdout] make[1]: Entering directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests/ftrace'
utils:0153| [stdout] make[1]: Nothing to be done for 'all'.
utils:0153| [stdout] make[1]: Leaving directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests/ftrace'
utils:0153| [stdout] make[1]: Entering directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests/ftrace'
utils:0153| [stdout] TAP version 13
utils:0153| [stdout] selftests: ftracetest
utils:0153| [stdout] ========================================
utils:0153| [stdout] === Ftrace unit tests ===
utils:0153| [stdout] [1] Basic trace file check [PASS]
utils:0153| [stdout] [2] Basic test for tracers [PASS]
utils:0153| [stdout] [3] Basic trace clock test [PASS]
utils:0153| [stdout] [4] Basic event tracing check [PASS]
utils:0153| [stdout] [5] Snapshot and tracing setting [PASS]
utils:0153| [stdout] [6] event tracing - enable/disable with event level files [PASS]
utils:0153| [stdout] [7] event tracing - restricts events based on pid [PASS]
utils:0153| [stdout] [8] event tracing - enable/disable with subsystem level files [PASS]
utils:0153| [stdout] [9] event tracing - enable/disable with top level files [PASS]
utils:0153| [stdout] [10] ftrace - function graph filters with stack tracer [PASS]
utils:0153| [stdout] [11] ftrace - function graph filters [PASS]
utils:0153| [stdout] [12] ftrace - function pid filters [PASS]
utils:0153| [stdout] [13] ftrace - test for function event triggers [PASS]
utils:0153| [stdout] [14] ftrace - function profiler with function tracing [PASS]
utils:0153| [stdout] [15] ftrace - test reading of set_ftrace_filter [PASS]
utils:0153| [stdout] [16] ftrace - test for function traceon/off triggers [PASS]
utils:0153| [stdout] [17] Test creation and deletion of trace instances while setting an event [PASS]
utils:0153| [st...

I did see this failure in the logs, however was not able to reproduce it when ran manually on GKE. 
I have attached the tarball of the logs.

Ftrace test output below:

07/21 13:32:26 DEBUG|     utils:0116| Running 'sudo sh -c 'echo 1 > /proc/sys/net/ipv4/conf/all/accept_local''
07/21 13:32:26 DEBUG|     utils:0116| Running 'sudo make -C linux/tools/testing/selftests TARGETS=ftrace run_tests'
07/21 13:32:26 DEBUG|     utils:0153| [stdout] make: Entering directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests'
07/21 13:32:26 DEBUG|     utils:0153| [stdout] make[1]: Entering directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests/ftrace'
07/21 13:32:26 DEBUG|     utils:0153| [stdout] make[1]: Nothing to be done for 'all'.
07/21 13:32:26 DEBUG|     utils:0153| [stdout] make[1]: Leaving directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests/ftrace'
07/21 13:32:26 DEBUG|     utils:0153| [stdout] make[1]: Entering directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests/ftrace'
07/21 13:32:26 DEBUG|     utils:0153| [stdout] TAP version 13
07/21 13:32:26 DEBUG|     utils:0153| [stdout] selftests: ftracetest
07/21 13:32:26 DEBUG|     utils:0153| [stdout] ========================================
07/21 13:32:26 DEBUG|     utils:0153| [stdout] === Ftrace unit tests ===
07/21 13:32:26 DEBUG|     utils:0153| [stdout] [1] Basic trace file check	[PASS]
07/21 13:32:28 DEBUG|     utils:0153| [stdout] [2] Basic test for tracers	[PASS]
07/21 13:32:29 DEBUG|     utils:0153| [stdout] [3] Basic trace clock test	[PASS]
07/21 13:32:29 DEBUG|     utils:0153| [stdout] [4] Basic event tracing check	[PASS]
07/21 13:32:29 DEBUG|     utils:0153| [stdout] [5] Snapshot and tracing setting	[PASS]
07/21 13:32:29 DEBUG|     utils:0153| [stdout] [6] event tracing - enable/disable with event level files	[PASS]
07/21 13:32:29 DEBUG|     utils:0153| [stdout] [7] event tracing - restricts events based on pid	[PASS]
07/21 13:32:29 DEBUG|     utils:0153| [stdout] [8] event tracing - enable/disable with subsystem level files	[PASS]
07/21 13:32:29 DEBUG|     utils:0153| [stdout] [9] event tracing - enable/disable with top level files	[PASS]
07/21 13:32:32 DEBUG|     utils:0153| [stdout] [10] ftrace - function graph filters with stack tracer	[PASS]
07/21 13:32:33 DEBUG|     utils:0153| [stdout] [11] ftrace - function graph filters	[PASS]
07/21 13:32:34 DEBUG|     utils:0153| [stdout] [12] ftrace - function pid filters	[PASS]
07/21 13:32:37 DEBUG|     utils:0153| [stdout] [13] ftrace - test for function event triggers	[PASS]
07/21 13:32:41 DEBUG|     utils:0153| [stdout] [14] ftrace - function profiler with function tracing	[PASS]
07/21 13:32:42 DEBUG|     utils:0153| [stdout] [15] ftrace - test reading of set_ftrace_filter	[PASS]
07/21 13:32:43 DEBUG|     utils:0153| [stdout] [16] ftrace - test for function traceon/off triggers	[PASS]
07/21 13:32:47 DEBUG|     utils:0153| [stdout] [17] Test creation and deletion of trace instances while setting an event	[PASS]
07/21 13:32:48 DEBUG|     utils:0153| [stdout] [18] Test creation and deletion of trace instances	[PASS]
07/21 13:32:48 DEBUG|     utils:0153| [stdout] [19] Kprobe dynamic event - adding and removing	[PASS]
07/21 13:32:48 DEBUG|     utils:0153| [stdout] [20] Kprobe dynamic event - busy event check	[PASS]
07/21 13:32:48 DEBUG|     utils:0153| [stdout] [21] Kprobe dynamic event with arguments	[PASS]
07/21 13:32:48 DEBUG|     utils:0153| [stdout] [22] Kprobe event string type argument	[PASS]
07/21 13:32:49 DEBUG|     utils:0153| [stdout] [23] Kprobe event argument syntax	[PASS]
07/21 13:32:49 DEBUG|     utils:0153| [stdout] [24] Kprobes event arguments with types	[PASS]
07/21 13:32:49 DEBUG|     utils:0153| [stdout] [25] Kprobe event auto/manual naming	[PASS]
07/21 13:32:49 DEBUG|     utils:0153| [stdout] [26] Kprobe dynamic event with function tracer	[PASS]
07/21 13:32:49 DEBUG|     utils:0153| [stdout] [27] Kretprobe dynamic event with arguments	[PASS]
07/21 13:32:49 DEBUG|     utils:0153| [stdout] [28] Kretprobe dynamic event with maxactive	[PASS]
07/21 13:32:54 DEBUG|     utils:0153| [stdout] [29] Register/unregister many kprobe events	[PASS]
07/21 13:32:54 DEBUG|     utils:0153| [stdout] [30] event trigger - test multiple actions on hist trigger	[UNSUPPORTED]
07/21 13:32:54 DEBUG|     utils:0153| [stdout] [31] event trigger - test synthetic_events syntax parser	[UNSUPPORTED]
07/21 13:32:55 DEBUG|     utils:0153| [stdout] [32] event trigger - test event enable/disable trigger	[PASS]
07/21 13:32:55 DEBUG|     utils:0153| [stdout] [33] event trigger - test trigger filter	[PASS]
07/21 13:32:55 DEBUG|     utils:0153| [stdout] [34] event trigger - test histogram modifiers	[PASS]
07/21 13:32:56 DEBUG|     utils:0153| [stdout] [35] event trigger - test multiple histogram triggers	[PASS]
07/21 13:32:56 DEBUG|     utils:0153| [stdout] [36] event trigger - test snapshot-trigger	[PASS]
07/21 13:32:56 DEBUG|     utils:0153| [stdout] [37] event trigger - test stacktrace-trigger	[PASS]
07/21 13:32:56 DEBUG|     utils:0153| [stdout] [38] event trigger - test traceon/off trigger	[PASS]
07/21 13:32:58 DEBUG|     utils:0153| [stdout] [39] (instance)  Basic test for tracers	[PASS]
07/21 13:32:58 DEBUG|     utils:0153| [stdout] [40] (instance)  Basic trace clock test	[PASS]
07/21 13:32:58 DEBUG|     utils:0153| [stdout] [41] (instance)  Snapshot and tracing setting	[PASS]
07/21 13:32:59 DEBUG|     utils:0153| [stdout] [42] (instance)  event tracing - enable/disable with event level files	[PASS]
07/21 13:32:59 DEBUG|     utils:0153| [stdout] [43] (instance)  event tracing - restricts events based on pid	[PASS]
07/21 13:32:59 DEBUG|     utils:0153| [stdout] [44] (instance)  event tracing - enable/disable with subsystem level files	[PASS]
07/21 13:33:02 DEBUG|     utils:0153| [stdout] [45] (instance)  ftrace - test for function event triggers	[PASS]
07/21 13:33:03 DEBUG|     utils:0153| [stdout] [46] (instance)  ftrace - test for function traceon/off triggers	[PASS]
07/21 13:33:03 DEBUG|     utils:0153| [stdout] [47] (instance)  event trigger - test event enable/disable trigger	[PASS]
07/21 13:33:04 DEBUG|     utils:0153| [stdout] [48] (instance)  event trigger - test trigger filter	[PASS]
07/21 13:33:04 DEBUG|     utils:0153| [stdout] [49] (instance)  event trigger - test histogram modifiers	[PASS]
07/21 13:33:04 DEBUG|     utils:0153| [stdout] [50] (instance)  event trigger - test multiple histogram triggers	[PASS]
07/21 13:33:04 DEBUG|     utils:0153| [stdout] 
07/21 13:33:04 DEBUG|     utils:0153| [stdout] # of passed:  48
07/21 13:33:04 DEBUG|     utils:0153| [stdout] # of failed:  0
07/21 13:33:04 DEBUG|     utils:0153| [stdout] # of unresolved:  0
07/21 13:33:04 DEBUG|     utils:0153| [stdout] # of untested:  0
07/21 13:33:04 DEBUG|     utils:0153| [stdout] # of unsupported:  2
07/21 13:33:04 DEBUG|     utils:0153| [stdout] # of xfailed:  0
07/21 13:33:04 DEBUG|     utils:0153| [stdout] # of undefined(test bug):  0
07/21 13:33:04 DEBUG|     utils:0153| [stdout] ok 1..1 selftests: ftracetest [PASS]
07/21 13:33:04 DEBUG|     utils:0153| [stdout] make[1]: Leaving directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests/ftrace'
07/21 13:33:04 DEBUG|     utils:0153| [stdout] make: Leaving directory '/home/jenkins/autotest/client/tmp/ubuntu_kernel_selftests/src/linux/tools/testing/selftests'
07/21 13:33:04 INFO |ubuntu_ker:0223| ========== Summary ===========
07/21 13:33:04 INFO |ubuntu_ker:0234| No failed cases reported

Revision history for this message

Sean Feole (sfeole) wrote on 2020-07-21:

Download full text (95.2 KiB)

Revision history for this message

Po-Hsu Lin (cypressyew) wrote on 2020-07-21:

A manual test with B-oem-osp1 5.0.0-1065.70 with the source tree from 5.0.0-1063, the net test will hang with the tls test (no other significant output in syslog):

selftests: net: tls
========================================
[==========] Running 37 tests from 2 test cases.
[ RUN ] tls.sendfile
[ OK ] tls.sendfile
[ RUN ] tls.send_then_sendfile
[ OK ] tls.send_then_sendfile
[ RUN ] tls.recv_max
[ OK ] tls.recv_max
[ RUN ] tls.recv_small
[ OK ] tls.recv_small
[ RUN ] tls.msg_more
[ OK ] tls.msg_more
[ RUN ] tls.sendmsg_single
[ OK ] tls.sendmsg_single
[ RUN ] tls.sendmsg_fragmented
[ OK ] tls.sendmsg_fragmented
[ RUN ] tls.sendmsg_large
[ OK ] tls.sendmsg_large
[ RUN ] tls.sendmsg_multiple
[ OK ] tls.sendmsg_multiple
[ RUN ] tls.sendmsg_multiple_stress
[ OK ] tls.sendmsg_multiple_stress
[ RUN ] tls.splice_from_pipe
[ OK ] tls.splice_from_pipe
[ RUN ] tls.splice_from_pipe2
[ OK ] tls.splice_from_pipe2
[ RUN ] tls.send_and_splice
[ OK ] tls.send_and_splice
[ RUN ] tls.splice_to_pipe
[ OK ] tls.splice_to_pipe
[ RUN ] tls.recvmsg_single
[ OK ] tls.recvmsg_single
[ RUN ] tls.recvmsg_single_max
[ OK ] tls.recvmsg_single_max
[ RUN ] tls.recvmsg_multiple
[ OK ] tls.recvmsg_multiple
[ RUN ] tls.single_send_multiple_recv
[ OK ] tls.single_send_multiple_recv
[ RUN ] tls.multiple_send_single_recv
[ OK ] tls.multiple_send_single_recv
[ RUN ] tls.single_send_multiple_recv_non_align
(hang here)

Jul 21 13:56:42 gonzo systemd[1]: Starting Stop ureadahead data collection...
Jul 21 13:56:42 gonzo systemd[1]: Started Stop ureadahead data collection.
Jul 21 13:56:49 gonzo kernel: [ 213.788101] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [ 213.789053] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [ 213.790508] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [ 213.791065] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [ 213.791574] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [ 213.792045] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [ 213.792538] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:50 gonzo kernel: [ 214.892627] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:50 gonzo kernel: [ 214.894322] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:50 gonzo kernel: [ 214.894919] tls_set_device_offload_rx: netdev lo with no TLS offload

Next is to test with source -1065 tree

A manual test with B-oem-osp1 5.0.0-1065.70 with the source tree from 5.0.0-1063, the net test will hang with the tls test (no other significant output in syslog):

selftests: net: tls
========================================
[==========] Running 37 tests from 2 test cases.
[ RUN      ] tls.sendfile
[       OK ] tls.sendfile
[ RUN      ] tls.send_then_sendfile
[       OK ] tls.send_then_sendfile
[ RUN      ] tls.recv_max
[       OK ] tls.recv_max
[ RUN      ] tls.recv_small
[       OK ] tls.recv_small
[ RUN      ] tls.msg_more
[       OK ] tls.msg_more
[ RUN      ] tls.sendmsg_single
[       OK ] tls.sendmsg_single
[ RUN      ] tls.sendmsg_fragmented
[       OK ] tls.sendmsg_fragmented
[ RUN      ] tls.sendmsg_large
[       OK ] tls.sendmsg_large
[ RUN      ] tls.sendmsg_multiple
[       OK ] tls.sendmsg_multiple
[ RUN      ] tls.sendmsg_multiple_stress
[       OK ] tls.sendmsg_multiple_stress
[ RUN      ] tls.splice_from_pipe
[       OK ] tls.splice_from_pipe
[ RUN      ] tls.splice_from_pipe2
[       OK ] tls.splice_from_pipe2
[ RUN      ] tls.send_and_splice
[       OK ] tls.send_and_splice
[ RUN      ] tls.splice_to_pipe
[       OK ] tls.splice_to_pipe
[ RUN      ] tls.recvmsg_single
[       OK ] tls.recvmsg_single
[ RUN      ] tls.recvmsg_single_max
[       OK ] tls.recvmsg_single_max
[ RUN      ] tls.recvmsg_multiple
[       OK ] tls.recvmsg_multiple
[ RUN      ] tls.single_send_multiple_recv
[       OK ] tls.single_send_multiple_recv
[ RUN      ] tls.multiple_send_single_recv
[       OK ] tls.multiple_send_single_recv
[ RUN      ] tls.single_send_multiple_recv_non_align
(hang here)

Jul 21 13:56:42 gonzo systemd[1]: Starting Stop ureadahead data collection...
Jul 21 13:56:42 gonzo systemd[1]: Started Stop ureadahead data collection.
Jul 21 13:56:49 gonzo kernel: [  213.788101] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [  213.789053] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [  213.790508] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [  213.791065] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [  213.791574] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [  213.792045] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:49 gonzo kernel: [  213.792538] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:50 gonzo kernel: [  214.892627] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:50 gonzo kernel: [  214.894322] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 13:56:50 gonzo kernel: [  214.894919] tls_set_device_offload_rx: netdev lo with no TLS offload

Next is to test with source -1065 tree

Revision history for this message

Po-Hsu Lin (cypressyew) wrote on 2020-07-21:

With the 5.0.0-1065.70 kernel + 5.0.0-1065.70 tree on gonzo, it's still hanging on the same point:

[ RUN ] tls.single_send_multiple_recv_non_align

Revision history for this message

Po-Hsu Lin (cypressyew) wrote on 2020-07-21:

Sean, for your comment #3, there is no such tls net test.
selftests: net: tls

Which source tree you're using when testing this?

(I can see this test in 5.0.0-1043.44 too)

Revision history for this message

Po-Hsu Lin (cypressyew) wrote on 2020-07-21:

OK I have download the tar file provided in comment #2, it's for 4.15 GKE not for 5.0 GKE:
'apt-get source --download-only linux-modules-extra-4.15.0-1066-gke'

Can you give it a try on 5.0 GKE?
Thanks

Revision history for this message

Sean Feole (sfeole) wrote on 2020-07-21:

I was able to reproduce this with linux-gke-5.0 after another attempt and ensuring the correct packages were installed.

Confirmed the modules pkg and kernel is 5.0, Linux sfgke 5.0.0-58-generic #62~18.04.1-Ubuntu SMP Tue Jul 14 03:37:30 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux

[==========] Running 37 tests from 2 test cases.
[ RUN ] tls.sendfile
[ OK ] tls.sendfile
[ RUN ] tls.send_then_sendfile
[ OK ] tls.send_then_sendfile
[ RUN ] tls.recv_max
[ OK ] tls.recv_max
[ RUN ] tls.recv_small
[ OK ] tls.recv_small
[ RUN ] tls.msg_more
[ OK ] tls.msg_more
[ RUN ] tls.sendmsg_single
[ OK ] tls.sendmsg_single
[ RUN ] tls.sendmsg_fragmented
[ OK ] tls.sendmsg_fragmented
[ RUN ] tls.sendmsg_large
[ OK ] tls.sendmsg_large
[ RUN ] tls.sendmsg_multiple
[ OK ] tls.sendmsg_multiple
[ RUN ] tls.sendmsg_multiple_stress
[ OK ] tls.sendmsg_multiple_stress
[ RUN ] tls.splice_from_pipe
[ OK ] tls.splice_from_pipe
[ RUN ] tls.splice_from_pipe2
[ OK ] tls.splice_from_pipe2
[ RUN ] tls.send_and_splice
[ OK ] tls.send_and_splice
[ RUN ] tls.splice_to_pipe
[ OK ] tls.splice_to_pipe
[ RUN ] tls.recvmsg_single
[ OK ] tls.recvmsg_single
[ RUN ] tls.recvmsg_single_max
[ OK ] tls.recvmsg_single_max
[ RUN ] tls.recvmsg_multiple
[ OK ] tls.recvmsg_multiple
[ RUN ] tls.single_send_multiple_recv
[ OK ] tls.single_send_multiple_recv
[ RUN ] tls.multiple_send_single_recv
[ OK ] tls.multiple_send_single_recv
[ RUN ] tls.single_send_multiple_recv_non_align

Jul 21 19:33:25 sfgke kernel: [10842.325048] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:25 sfgke kernel: [10842.332940] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:25 sfgke kernel: [10842.340720] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:25 sfgke kernel: [10842.348469] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:25 sfgke kernel: [10842.356168] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:25 sfgke kernel: [10842.364229] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:25 sfgke kernel: [10842.372100] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:26 sfgke kernel: [10843.477618] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:27 sfgke kernel: [10843.529738] tls_set_device_offload_rx: netdev lo with no TLS offload
Jul 21 19:33:27 sfgke kernel: [10843.536771] tls_set_device_offload_rx: netdev lo with no TLS offload

I was able to reproduce this with linux-gke-5.0 after another attempt and ensuring the correct packages were installed.

Confirmed the modules pkg and kernel is 5.0, Linux sfgke 5.0.0-58-generic #62~18.04.1-Ubuntu SMP Tue Jul 14 03:37:30 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux

[==========] Running 37 tests from 2 test cases.
[ RUN      ] tls.sendfile
[       OK ] tls.sendfile
[ RUN      ] tls.send_then_sendfile
[       OK ] tls.send_then_sendfile
[ RUN      ] tls.recv_max
[       OK ] tls.recv_max
[ RUN      ] tls.recv_small
[       OK ] tls.recv_small
[ RUN      ] tls.msg_more
[       OK ] tls.msg_more
[ RUN      ] tls.sendmsg_single
[       OK ] tls.sendmsg_single
[ RUN      ] tls.sendmsg_fragmented
[       OK ] tls.sendmsg_fragmented
[ RUN      ] tls.sendmsg_large
[       OK ] tls.sendmsg_large
[ RUN      ] tls.sendmsg_multiple
[       OK ] tls.sendmsg_multiple
[ RUN      ] tls.sendmsg_multiple_stress
[       OK ] tls.sendmsg_multiple_stress
[ RUN      ] tls.splice_from_pipe
[       OK ] tls.splice_from_pipe
[ RUN      ] tls.splice_from_pipe2
[       OK ] tls.splice_from_pipe2
[ RUN      ] tls.send_and_splice
[       OK ] tls.send_and_splice
[ RUN      ] tls.splice_to_pipe
[       OK ] tls.splice_to_pipe
[ RUN      ] tls.recvmsg_single
[       OK ] tls.recvmsg_single
[ RUN      ] tls.recvmsg_single_max
[       OK ] tls.recvmsg_single_max
[ RUN      ] tls.recvmsg_multiple
[       OK ] tls.recvmsg_multiple
[ RUN      ] tls.single_send_multiple_recv
[       OK ] tls.single_send_multiple_recv
[ RUN      ] tls.multiple_send_single_recv
[       OK ] tls.multiple_send_single_recv
[ RUN      ] tls.single_send_multiple_recv_non_align

Revision history for this message

Kleber Sacilotto de Souza (kleber-souza) wrote on 2020-07-22:

#10

The messages on dmesg "tls_set_device_offload_rx: netdev lo with no TLS offload" doesn't seem to be related to the hung. These messages are logged while running another net/tls testcase and can be seen also on previous kernel versions in which the hang can't be reproduced.

Revision history for this message

Kleber Sacilotto de Souza (kleber-souza) wrote on 2020-07-22:

#11

The issue is reproducible by the following tls testcases:
- single_send_multiple_recv_non_align
- recv_lowat

Commenting these out from ./tools/testing/selftests/net/tls.c the tls test completes successfully.

Revision history for this message

Thadeu Lima de Souza Cascardo (cascardo) wrote on 2020-07-22:

#12

Kleber has identified the backport of commit 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records") as the responsible for this regression.

It turns out this commit has 4 fixups upstream, 3 of which can be easily cherry picked, while the fourth one seems related to TLS 1.3 support, which is not included in our 5.0 kernels yet.

Picking those 3 commits allow the tests to pass without issues.

description:

updated

Stefan Bader (smb) on 2020-07-23

Changed in linux-oem-osp1 (Ubuntu):
status:	New → Triaged
importance:	Undecided → Medium

Revision history for this message

Kleber Sacilotto de Souza (kleber-souza) wrote on 2020-07-23:

#13

SRU request:
https://lists.ubuntu.com/archives/kernel-team/2020-July/112279.html

Changed in linux-oem-osp1 (Ubuntu Bionic):
status:	New → In Progress

Kelsey Steele (kelsey-steele) on 2020-07-23

Changed in linux-oem-osp1 (Ubuntu Bionic):
status:	In Progress → Fix Committed
status:	Fix Committed → In Progress

Kelsey Steele (kelsey-steele) on 2020-07-23

Changed in linux-oem-osp1 (Ubuntu):
status:	Triaged → Fix Committed

Revision history for this message

Launchpad Janitor (janitor) wrote on 2020-08-17:

#14

Download full text (18.9 KiB)

This bug was fixed in the package linux-oem-osp1 - 5.0.0-1067.72

---------------
linux-oem-osp1 (5.0.0-1067.72) bionic; urgency=medium

* bionic/linux-oem-osp1: 5.0.0-1067.72 -proposed tracker (LP: #1889723)

* Packaging resync (LP: #1786013)
- [Packaging] resync dkms-build and family

[ Ubuntu: 5.0.0-60.64 ]

  * disco/linux: 5.0.0-60.64 -proposed tracker (LP: #1889602)
  * debian/scripts/file-downloader does not handle positive failures correctly
    (LP: #1878897)
    - [Packaging] file-downloader not handling positive failures correctly
  * dkms artifacts may expire from the pool (LP: #1850958)
    - [Packaging] autoreconstruct -- manage executable debian files
    - [packaging] handle downloads from the librarian better

[ Ubuntu: 5.0.0-59.63 ]

This bug was fixed in the package linux-oem-osp1 - 5.0.0-1067.72

---------------
linux-oem-osp1 (5.0.0-1067.72) bionic; urgency=medium

* bionic/linux-oem-osp1: 5.0.0-1067.72 -proposed tracker (LP: #1889723)

* Packaging resync (LP: #1786013)
    - [Packaging] resync dkms-build and family

[ Ubuntu: 5.0.0-60.64 ]

[ Ubuntu: 5.0.0-59.63 ]

* disco/linux: 5.0.0-59.63 -proposed tracker (LP: #1888293)
  * net test from ubuntu_kernel_selftest will timeout on B-5.0 (LP: #1888381)
    - net/tls: fix lowat calculation if some data came from previous record
    - net/tls: fix no wakeup on partial reads
    - net/tls: fix poll ignoring partially copied records
  * Regression in kernel 4.15.0-91 causes kernel panic with Bcache
    (LP: #1867916)
    - bcache: check and adjust logical block size for backing devices
  * use-after-free in af_alg_accept() due to bh_lock_sock() (LP: #1884766)
    - crypto: af_alg - fix use-after-free in af_alg_accept() due to bh_lock_sock()
  * Disco update: upstream stable patchset 2020-07-08 (LP: #1886876)
    - ARM: dts: renesas: Fix IOMMU device node names
    - scsi: core: free sgtables in case command setup fails
    - arm64: dts: meson: fixup SCP sram nodes
    - powerpc/kasan: Fix stack overflow by increasing THREAD_SHIFT
    - pinctrl: ocelot: Fix GPIO interrupt decoding on Jaguar2
    - clk: renesas: cpg-mssr: Fix STBCR suspend/resume handling
    - coresight: tmc: Fix TMC mode read in tmc_read_prepare_etb()
    - scsi: hisi_sas: Do not reset phy timer to wait for stray phy up
    - PCI: pci-bridge-emul: Fix PCIe bit conflicts
    - usb: gadget: core: sync interrupt before unbind the udc
    - clk: zynqmp: fix memory leak in zynqmp_register_clocks
    - scsi: vhost: Notify TCM about the maximum sg entries supported per command
    - IB/mlx5: Fix DEVX support for MLX5_CMD_OP_INIT2INIT_QP command
    - cifs: set up next DFS target before generic_ip_connect()
    - ASoC: qcom: q6asm-dai: kCFI fix
    - sparc32: mm: Don't try to free page-table pages if ctor() fails
    - net: dsa: lantiq_gswip: fix and improve the unsupported interface error
    - bpf, sockhash: Fix memory leak when unlinking sockets in sock_hash_free
    - RDMA/mlx5: Fix udata response upon SRQ creation
    - clk: meson: meson8b: Fix the polarity of the RESET_N lines
    - clk: meson: meson8b: Fix the vclk_div{1, 2, 4, 6, 12}_en gate bits
    - gpio: pca953x: fix handling of automatic address incrementing
    - ASoC: max98373: reorder max98373_reset() in resume
    - soundwire: slave: don't init debugfs on device registration error
    - arm64: dts: msm8996: Fix CSI IRQ types
    - scsi: target: loopback: Fix READ with data and sensebytes
    - SoC: rsnd: add interrupt support for SSI BUSIF buffer
    - ASoC: ux500: mop500: Fix some refcounted resources issues
    - ASoC: ti: omap-mcbsp: Fix an error handling path in 'asoc_mcbsp_probe()'
    - ALSA: usb-audio: Add duplex sound support for USB devices using implicit
      feedback
    - RDMA/hns: Fix cmdq parameter of querying pf timer resource
    - firmware: imx: scu: Fix possible memory leak in imx_scu_probe()
    - PCI: amlogic: meson: Don't use FAST_LINK_MODE to set up link
    - KVM: PPC: Book3S: Fix some RCU-list locks
    - input: i8042 - Remove special PowerPC handling
    - drm/nouveau/disp/gm200-: fix NV_PDISP_SOR_HDMI2_CTRL(n) selection
    - net: marvell: Fix OF_MDIO config check
    - bpf/sockmap: Fix kernel panic at __tcp_bpf_recvmsg
    - bpf, sockhash: Synchronize delete from bucket list on map free
    - nfs: set invalid blocks after NFSv4 writes
    - powerpc: Fix kernel crash in show_instructions() w/DEBUG_VIRTUAL
    - bnxt_en: Return from timer if interface is not in open state.
    - scsi: ufs-bsg: Fix runtime PM imbalance on error
    - jbd2: clean __jbd2_journal_abort_hard() and __journal_abort_soft()
    - drm/amd/display: Use swap() where appropriate
    - drm/msm: Check for powered down HW in the devfreq callbacks
    - drm/connector: notify userspace on hotplug after register complete
  * Disco update: upstream stable patchset 2020-07-02 (LP: #1886105)
    - power: supply: bq24257_charger: Replace depends on REGMAP_I2C with select
    - clk: sunxi: Fix incorrect usage of round_down()
    - ASoC: tegra: tegra_wm8903: Support nvidia, headset property
    - i2c: piix4: Detect secondary SMBus controller on AMD AM4 chipsets
    - iio: pressure: bmp280: Tolerate IRQ before registering
    - remoteproc: Fix IDR initialisation in rproc_alloc()
    - clk: qcom: msm8916: Fix the address location of pll->config_reg
    - backlight: lp855x: Ensure regulators are disabled on probe failure
    - ASoC: davinci-mcasp: Fix dma_chan refcnt leak when getting dma type
    - ARM: integrator: Add some Kconfig selections
    - scsi: qedi: Check for buffer overflow in qedi_set_path()
    - ALSA: hda/realtek - Introduce polarity for micmute LED GPIO
    - ALSA: isa/wavefront: prevent out of bounds write in ioctl
    - PCI: Allow pci_resize_resource() for devices on root bus
    - scsi: qla2xxx: Fix issue with adapter's stopping state
    - iio: bmp280: fix compensation of humidity
    - f2fs: report delalloc reserve as non-free in statfs for project quota
    - i2c: pxa: clear all master action bits in i2c_pxa_stop_message()
    - clk: samsung: Mark top ISP and CAM clocks on Exynos542x as critical
    - usblp: poison URBs upon disconnect
    - serial: 8250: Fix max baud limit in generic 8250 port
    - dm mpath: switch paths in dm_blk_ioctl() code path
    - PCI: aardvark: Don't blindly enable ASPM L0s and don't write to read-only
      register
    - ps3disk: use the default segment boundary
    - vfio/pci: fix memory leaks in alloc_perm_bits()
    - RDMA/mlx5: Add init2init as a modify command
    - m68k/PCI: Fix a memory leak in an error handling path
    - gpio: dwapb: Call acpi_gpiochip_free_interrupts() on GPIO chip de-
      registration
    - mfd: wm8994: Fix driver operation if loaded as modules
    - scsi: lpfc: Fix lpfc_nodelist leak when processing unsolicited event
    - clk: clk-flexgen: fix clock-critical handling
    - powerpc/perf/hv-24x7: Fix inconsistent output values incase multiple hv-24x7
      events run
    - nfsd: Fix svc_xprt refcnt leak when setup callback client failed
    - PCI: vmd: Filter resource type bits from shadow register
    - powerpc/crashkernel: Take "mem=" option into account
    - pwm: img: Call pm_runtime_put() in pm_runtime_get_sync() failed case
    - yam: fix possible memory leak in yam_init_driver
    - NTB: ntb_pingpong: Choose doorbells based on port number
    - NTB: Fix the default port and peer numbers for legacy drivers
    - mksysmap: Fix the mismatch of '.L' symbols in System.map
    - apparmor: fix introspection of of task mode for unconfined tasks
    - ASoC: meson: add missing free_irq() in error path
    - scsi: sr: Fix sr_probe() missing deallocate of device minor
    - scsi: ibmvscsi: Don't send host info in adapter info MAD after LPM
    - x86/purgatory: Disable various profiling and sanitizing options
    - staging: greybus: fix a missing-check bug in gb_lights_light_config()
    - arm64: dts: mt8173: fix unit name warnings
    - scsi: qedi: Do not flush offload work if ARP not resolved
    - ARM: dts: sun8i-h2-plus-bananapi-m2-zero: Fix led polarity
    - gpio: dwapb: Append MODULE_ALIAS for platform driver
    - scsi: qedf: Fix crash when MFW calls for protocol stats while function is
      still probing
    - pinctrl: rza1: Fix wrong array assignment of rza1l_swio_entries
    - firmware: qcom_scm: fix bogous abuse of dma-direct internals
    - staging: gasket: Fix mapping refcnt leak when put attribute fails
    - staging: gasket: Fix mapping refcnt leak when register/store fails
    - ALSA: usb-audio: Fix racy list management in output queue
    - s390/qdio: put thinint indicator after early error
    - slimbus: ngd: get drvdata from correct device
    - thermal/drivers/ti-soc-thermal: Avoid dereferencing ERR_PTR
    - usb: dwc3: gadget: Properly handle failed kick_transfer
    - staging: sm750fb: add missing case while setting FB_VISUAL
    - PCI: v3-semi: Fix a memory leak in v3_pci_probe() error handling paths
    - i2c: pxa: fix i2c_pxa_scream_blue_murder() debug output
    - serial: amba-pl011: Make sure we initialize the port.lock spinlock
    - drivers: base: Fix NULL pointer exception in __platform_driver_probe() if a
      driver developer is foolish
    - PCI: rcar: Fix incorrect programming of OB windows
    - PCI/ASPM: Allow ASPM on links to PCIe-to-PCI/PCI-X Bridges
    - scsi: qla2xxx: Fix warning after FC target reset
    - power: supply: lp8788: Fix an error handling path in
      'lp8788_charger_probe()'
    - power: supply: smb347-charger: IRQSTAT_D is volatile
    - scsi: mpt3sas: Fix double free warnings
    - pinctrl: rockchip: fix memleak in rockchip_dt_node_to_map
    - dlm: remove BUG() before panic()
    - clk: ti: composite: fix memory leak
    - PCI: Fix pci_register_host_bridge() device_register() error handling
    - powerpc/64: Don't initialise init_task->thread.regs
    - tty: n_gsm: Fix SOF skipping
    - tty: n_gsm: Fix waking up upper tty layer when room available
    - HID: Add quirks for Trust Panora Graphic Tablet
    - ipmi: use vzalloc instead of kmalloc for user creation
    - powerpc/pseries/ras: Fix FWNMI_VALID off by one
    - powerpc/ps3: Fix kexec shutdown hang
    - vfio-pci: Mask cap zero
    - usb/ohci-platform: Fix a warning when hibernating
    - drm/msm/mdp5: Fix mdp5_init error path for failed mdp5_kms allocation
    - ASoC: Intel: bytcr_rt5640: Add quirk for Toshiba Encore WT8-A tablet
    - USB: host: ehci-mxc: Add error handling in ehci_mxc_drv_probe()
    - tty: n_gsm: Fix bogus i++ in gsm_data_kick
    - fpga: dfl: afu: Corrected error handling levels
    - clk: samsung: exynos5433: Add IGNORE_UNUSED flag to sclk_i2s1
    - scsi: target: tcmu: Userspace must not complete queued commands
    - arm64: tegra: Fix ethernet phy-mode for Jetson Xavier
    - powerpc/64s/pgtable: fix an undefined behaviour
    - dm zoned: return NULL if dmz_get_zone_for_reclaim() fails to find a zone
    - PCI/PTM: Inherit Switch Downstream Port PTM settings from Upstream Port
    - PCI: dwc: Fix inner MSI IRQ domain registration
    - IB/cma: Fix ports memory leak in cma_configfs
    - watchdog: da9062: No need to ping manually before setting timeout
    - usb: dwc2: gadget: move gadget resume after the core is in L0 state
    - USB: gadget: udc: s3c2410_udc: Remove pointless NULL check in
      s3c2410_udc_nuke
    - usb: gadget: lpc32xx_udc: don't dereference ep pointer before null check
    - usb: gadget: fix potential double-free in m66592_probe.
    - usb: gadget: Fix issue with config_ep_by_speed function
    - RDMA/iw_cxgb4: cleanup device debugfs entries on ULD remove
    - x86/apic: Make TSC deadline timer detection message visible
    - ASoC: fix incomplete error-handling in img_i2s_in_probe.
    - scsi: target: tcmu: Fix a use after free in tcmu_check_expired_queue_cmd()
    - clk: bcm2835: Fix return type of bcm2835_register_gate
    - scsi: ufs-qcom: Fix scheduling while atomic issue
    - KVM: PPC: Book3S HV: Ignore kmemleak false positives
    - clk: sprd: return correct type of value for _sprd_pll_recalc_rate
    - net: sunrpc: Fix off-by-one issues in 'rpc_ntop6'
    - NFSv4.1 fix rpc_call_done assignment for BIND_CONN_TO_SESSION
    - of: Fix a refcounting bug in __of_attach_node_sysfs()
    - powerpc/4xx: Don't unmap NULL mbase
    - extcon: adc-jack: Fix an error handling path in 'adc_jack_probe()'
    - ASoC: fsl_asrc_dma: Fix dma_chan leak when config DMA channel failed
    - vfio/mdev: Fix reference count leak in add_mdev_supported_type
    - rxrpc: Adjust /proc/net/rxrpc/calls to display call->debug_id not user_ID
    - openrisc: Fix issue with argument clobbering for clone/fork
    - gfs2: Allow lock_nolock mount to specify jid=X
    - scsi: iscsi: Fix reference count leak in iscsi_boot_create_kobj
    - scsi: ufs: Don't update urgent bkops level when toggling auto bkops
    - pinctrl: imxl: Fix an error handling path in 'imx1_pinctrl_core_probe()'
    - pinctrl: freescale: imx: Fix an error handling path in 'imx_pinctrl_probe()'
    - crypto: omap-sham - add proper load balancing support for multicore
    - geneve: change from tx_error to tx_dropped on missing metadata
    - lib/zlib: remove outdated and incorrect pre-increment optimization
    - include/linux/bitops.h: avoid clang shift-count-overflow warnings
    - selftests/vm/pkeys: fix alloc_random_pkey() to make it really random
    - blktrace: use errno instead of bi_status
    - blktrace: fix endianness in get_pdu_int()
    - blktrace: fix endianness for blk_log_remap()
    - gfs2: fix use-after-free on transaction ail lists
    - ntb_perf: pass correct struct device to dma_alloc_coherent
    - ntb_tool: pass correct struct device to dma_alloc_coherent
    - NTB: ntb_tool: reading the link file should not end in a NULL byte
    - NTB: Revert the change to use the NTB device dev for DMA allocations
    - NTB: perf: Don't require one more memory window than number of peers
    - NTB: perf: Fix support for hardware that doesn't have port numbers
    - NTB: perf: Fix race condition when run with ntb_test
    - NTB: ntb_test: Fix bug when counting remote files
    - drivers/perf: hisi: Fix wrong value for all counters enable
    - selftests/net: in timestamping, strncpy needs to preserve null byte
    - afs: Fix memory leak in afs_put_sysnames()
    - ASoC: core: only convert non DPCM link to DPCM link
    - ASoC: Intel: bytcr_rt5640: Add quirk for Toshiba Encore WT10-A tablet
    - ASoC: rt5645: Add platform-data for Asus T101HA
    - drm/sun4i: hdmi ddc clk: Fix size of m divider
    - scsi: acornscsi: Fix an error handling path in acornscsi_probe()
    - x86/idt: Keep spurious entries unset in system_vectors
    - net/filter: Permit reading NET in load_bytes_relative when MAC not set
    - xdp: Fix xsk_generic_xmit errno
    - usb/xhci-plat: Set PM runtime as active on resume
    - usb: host: ehci-platform: add a quirk to avoid stuck
    - usb/ehci-platform: Set PM runtime as active on resume
    - perf report: Fix NULL pointer dereference in
      hists__fprintf_nr_sample_events()
    - ext4: stop overwrite the errcode in ext4_setup_super
    - bcache: fix potential deadlock problem in btree_gc_coalesce
    - afs: Fix non-setting of mtime when writing into mmap
    - afs: afs_write_end() should change i_size under the right lock
    - block: Fix use-after-free in blkdev_get()
    - arm64: hw_breakpoint: Don't invoke overflow handler on uaccess watchpoints
    - drm: encoder_slave: fix refcouting error for modules
    - drm/dp_mst: Reformat drm_dp_check_act_status() a bit
    - drm/qxl: Use correct notify port address when creating cursor ring
    - drm/amdgpu: Replace invalid device ID with a valid device ID
    - selinux: fix double free
    - ext4: fix partial cluster initialization when splitting extent
    - ext4: avoid race conditions when remounting with options that change dax
    - drm/dp_mst: Increase ACT retry timeout to 3s
    - x86/boot/compressed: Relax sed symbol type regex for LLVM ld.lld
    - block: nr_sects_write(): Disable preemption on seqcount write
    - mtd: rawnand: diskonchip: Fix the probe error path
    - mtd: rawnand: sharpsl: Fix the probe error path
    - mtd: rawnand: xway: Fix the probe error path
    - mtd: rawnand: orion: Fix the probe error path
    - mtd: rawnand: oxnas: Add of_node_put()
    - mtd: rawnand: oxnas: Fix the probe error path
    - mtd: rawnand: socrates: Fix the probe error path
    - mtd: rawnand: plat_nand: Fix the probe error path
    - mtd: rawnand: mtk: Fix the probe error path
    - mtd: rawnand: tmio: Fix the probe error path
    - s390: fix syscall_get_error for compat processes
    - drm/i915: Whitelist context-local timestamp in the gen9 cmdparser
    - drm/i915/icl+: Fix hotplug interrupt disabling after storm detection
    - crypto: algif_skcipher - Cap recv SG list at ctx->used
    - crypto: algboss - don't wait during notifier callback
    - kprobes: Fix to protect kick_kprobe_optimizer() by kprobe_mutex
    - e1000e: Do not wake up the system via WOL if device wakeup is disabled
    - net: octeon: mgmt: Repair filling of RX ring
    - kretprobe: Prevent triggering kretprobe from within kprobe_flush_task
    - sched/rt, net: Use CONFIG_PREEMPTION.patch
    - net: core: device_rename: Use rwsem instead of a seqcount
    - Revert "dpaa_eth: fix usage as DSA master, try 3"
    - kvm: x86: Move kvm_set_mmio_spte_mask() from x86.c to mmu.c
    - kvm: x86: Fix reserved bits related calculation errors caused by MKTME
    - KVM: x86/mmu: Set mmio_value to '0' if reserved #PF can't be generated
    - spi: spi-mem: Fix Dual/Quad modes on Octal-capable devices
    - arm64/kernel: Fix range on invalidating dcache for boot page tables
    - sched/fair: Refill bandwidth before scaling
    - ath10k: remove the max_sched_scan_reqs value
    - rtw88: fix an issue about leak system resources
    - block: reset mapping if failed to update hardware queue count
    - drm: rcar-du: Set primary plane zpos immutably at initializing
    - platform/x86: dell-laptop: don't register micmute LED if there is no token
    - mips: Fix cpu_has_mips64r1/2 activation for MIPS32 CPUs
    - media: dvbdev: Fix tuner->demod media controller link
    - ice: Fix memory leak
    - ice: Fix for memory leaks and modify ICE_FREE_CQ_BUFS
    - drm/mediatek: set dpi pin mode to gpio low to avoid leakage current
    - ath10k: add flush tx packets for SDIO chip
    - net: bcmgenet: Fix WoL with password after deep sleep
    - nvme-tcp: use bh_lock in data_ready
    - kgdboc: Use a platform device to handle tty drivers showing up late
    - sched: Defend cfs and rt bandwidth quota against overflow
    - MIPS: tools: Fix resource leak in elf-entry.c
    - ice: fix potential double free in probe unrolling
    - platform/x86: asus_wmi: Reserve more space for struct bias_args
    - bpf: Fix map permissions check
    - bpf: Refactor sockmap redirect code so its easy to reuse
    - ima: Remove __init annotation from ima_pcrread()
    - x86/amd_nb: Add AMD family 17h model 60h PCI IDs
    - ima: Remove redundant policy rule set in add_rules()
    - ima: Set again build_ima_appraise variable
    - media: cedrus: Program output format during each run
    - hwmon: (k10temp) Add AMD family 17h model 60h PCI match
    - soc/tegra: pmc: Select GENERIC_PINCONF
    - mtd: rawnand: Fix nand_gpio_waitrdy()
    - mtd: rawnand: onfi: Fix redundancy detection check
    - mtd: rawnand: sunxi: Fix the probe error path
  * Computer is frozen after suspend (LP: #1867983) // Disco update: upstream
    stable patchset 2020-07-02 (LP: #1886105)
    - libata: Use per port sync for detach

-- Chia-Lin Kao (AceLan) <acelan.kao@canonical.com>  Fri, 31 Jul 2020 14:05:16 +0800

Changed in linux-oem-osp1 (Ubuntu Bionic):
status:	In Progress → Fix Released

Po-Hsu Lin (cypressyew) on 2020-10-20

Changed in linux-oem-osp1 (Ubuntu):
status:	Fix Committed → Fix Released
Changed in ubuntu-kernel-tests:
status:	New → Fix Released

Report a bug

This report contains Public information

Everyone can see this information.

You are

Subscribing...

Edit bug mail

Other bug subscribers

Bug attachments

Add attachment

Remote bug watches

Bug watches keep track of this bug in other bug trackers.