[Thunderbolt][CML-U/S] After freeze sleep finished more than 30s, lspci trigger tbt in D0, "TX ring 0 is already enabled" Call Trace generated and then failed to send driver ready to ICM

Bug #1843790 reported by quanxian on 2019-09-12
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux (Ubuntu)

Bug Description

Platform: CML-U (CML-S no need TR AIC)

TR AIC: rev41

Linux kernel: 5.3.0-rc1

No TBT device connected.

Executed freeze sleep successfully, after freeze exit more than 30s,

cat /sys/bus/thunderbolt/devices/domain0/boot_acl and lspci to trigger TBT root port in D0.

Then "TX ring 0 is already enabled" Call Trace generated, then failed to send driver ready to ICM.

Thunderbolt driver was down.

[ 263.903342] LTP: starting TBT_XS_FUNC_RTD3_freeze_test (tbt_rtd3.sh -c 0 -s rtd3_host_freeze)
[ 264.010613] PM: suspend entry (s2idle)
[ 286.868539] PM: suspend exit
[ 313.202856] thunderbolt 0000:03:00.0: stopping RX ring 0
[ 313.204135] thunderbolt 0000:03:00.0: disabling interrupt at register 0x38200 bit 12 (0x1001 -> 0x1)
[ 313.205565] thunderbolt 0000:03:00.0: stopping TX ring 0
[ 313.206842] thunderbolt 0000:03:00.0: disabling interrupt at register 0x38200 bit 0 (0x1 -> 0x0)
[ 313.208018] thunderbolt 0000:03:00.0: control channel stopped
[ 321.432230] pcieport 0000:02:01.0: pciehp: Slot(1): Link Up
[ 321.433559] pcieport 0000:02:04.0: pciehp: Slot(4-1): Link Up
[ 322.550092] xhci_hcd 0000:39:00.0: Refused to change power state, currently in D3
[ 322.671084] thunderbolt 0000:03:00.0: control channel starting...
[ 322.672343] thunderbolt 0000:03:00.0: starting TX ring 0
[ 322.673601] thunderbolt 0000:03:00.0: enabling interrupt at register 0x38200 bit 0 (0xffffffff -> 0xffffffff)
[ 322.674912] -----------[ cut here ]-----------
[ 322.676173] thunderbolt 0000:03:00.0: interrupt for TX ring 0 is already enabled
[ 322.706970] Call Trace:
[ 322.708542] tb_ring_start+0x110/0x1a0 [thunderbolt]
[ 322.710166] tb_ctl_start+0x36/0xc0 [thunderbolt]
[ 322.711762] ? pci_restore_standard_config+0x50/0x50
[ 322.713333] ? pci_restore_standard_config+0x50/0x50
[ 322.748162] thunderbolt 0000:03:00.0: interrupt for RX ring 0 is already enabled
[ 322.788047] ? pci_restore_standard_config+0x50/0x50
[ 322.789901] ? pci_restore_standard_config+0x50/0x50
[ 404.405929] thunderbolt 0000:03:00.0: failed to send driver ready to ICM
[ 405.596681] LTP: starting TBT_XS_FUNC_RTD3_S3_test (tbt_rtd3.sh -c 0 -s rtd3_host_s3)

Target Release: 20.04
Target Kernel: TBD

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 1843790

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.

Changed in linux (Ubuntu):
status: New → Incomplete
quanxian (quanxian-wang) wrote :

5.3 mainline fixed. but no commit provided.

quanxian (quanxian-wang) wrote :

bios update fixes this issue. So if we come across this issue, please update BIOS.

Changed in intel:
status: New → Won't Fix
quanxian (quanxian-wang) wrote :

CML-U/CML-S all had this issue before.

Both cml-s and cml-u platforms found this issue as before.

And issue exist and issue not exist BIOS version was as below:

cml-s issue exist BIOS: CMLSFWR1.R00.1193
cml-s issue not exist BIOS: CMLSFWR1.R00.1344

cml-u issue exist BIOS: CMLSFWR1.R00.1263
cml-u issue not exist BIOS: CMLSFWR1.R00.1272

Anthony Wong (anthonywong) wrote :

Comment #2 says 5.3 fixes this issue.

If 5.3 fixes this issue, then we can close this bug because Ubuntu 19.10 (Eoan) will be using 5.3.

Launchpad Janitor (janitor) wrote :

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]

Changed in linux (Ubuntu):
status: Incomplete → Expired
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers