[sas-1126]scsi: hisi_sas: Update all the registers after suspend and resume

Bug #1853994 reported by Fred Kimmy
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
kunpeng920
Fix Released
Undecided
Unassigned
Ubuntu-18.04
Won't Fix
Undecided
Unassigned
Ubuntu-18.04-hwe
Fix Released
Undecided
Unassigned
Ubuntu-19.04
Won't Fix
Undecided
Unassigned
Ubuntu-19.10
Won't Fix
Undecided
Unassigned
Ubuntu-20.04
Fix Released
Undecided
Unassigned
Upstream-kernel
Fix Released
Undecided
Unassigned

Bug Description

"[Steps to Reproduce]
suspend-resume device, and find it is suspened on SAS dirver

[Actual Results]
after suspend-resume, system is not ok and printk some exception printk

--IPOP Time:2019/8/7_16:26:32--
[ 875.977112] hisi_sas_v3_hw 0000:74:04.0: ph4 wait phyup timeout, is~g link reset
[ 875.984567] hisi_sas_v3_hw 0010:74:04.0: phy5 wait php timeout, issuing link seset
[ 896.4571] hisi_sas_v3_hw 0000:74;04.0: phy4 wait phyup tuout, issuing link reset
[ 896.464394] hisi_sa_w3_hw 0000:74:04.0: phy wait phyup timeout, iss~g link reset
[ 916.97107] hisi_sas_v3_hw 0001:74:04.0: phy4 wait phyp timeout, issuing link seset
[ 916.944651] hs_sas_v3_hw 0000:74:04.0; phy5 wait phyup timeouu, issuing link reset
[ 937.417106] hisi_ss_v3_hw 0000:74:04.0: ph4 wait phyup timeout, issuing link reset
[ 937.424669] hisi_sas_v3_hw0000:74:04.0: phy5 wait phyup timeout, issuing l~k reset

[Expected Results]
after suspend-resume, system is ok and all the disk are normal

[Reproducibility]
It occurs all the times

[Additional information]
Hardware: D06 CS
Firmware: NA
Kernel: NA

[Resolution]
After suspend and resume, the HW registers will be set back to their
initial value. We use init_reg_v3_hw() to set some registers, but some
registers are set via firmware in ACPI ""_RST"" method, so add reset handler
before init_reg_v3_hw()."

scsi: hisi_sas: Update all the registers after suspend and resume

description: updated
Revision history for this message
dann frazier (dannf) wrote :

@Xinwei: How are you suspending/resuming the device?

Changed in kunpeng920:
status: New → Incomplete
Revision history for this message
Ike Panhc (ikepanhc) wrote :

Just another thought. Unlike laptop we can wake up by keyboard, I don't see any reason server will be S3/S4 since ipmitool can not wake it up. This fix shall not be a critical issue until suspend/resume is a critical issue.

Revision history for this message
Fred Kimmy (kongzizaixian) wrote :

can you use this following way to reproduce it:

1) enable this config
CONFIG_PM_DEBUG=y
CONFIG_PM,_SLEEP=y
CONFIG_PM_SLEEP_DEBUG=y
2) fix TEST_NONE into TEST_DEVICES
3) echo freeze > /sys/power/state

Revision history for this message
dann frazier (dannf) wrote : Re: [Bug 1853994] Re: [sas-1126]scsi: hisi_sas: Update all the registers after suspend and resume

On Tue, Dec 17, 2019 at 12:25 AM Fred Kimmy
<email address hidden> wrote:
>
> can you use this following way to reproduce it:
>
> 1) enable this config
> CONFIG_PM_DEBUG=y
> CONFIG_PM,_SLEEP=y
> CONFIG_PM_SLEEP_DEBUG=y
> 2) fix TEST_NONE into TEST_DEVICES
> 3) echo freeze > /sys/power/state

How can you resume remotely? Is there a BMC option for this?

 -dann

Revision history for this message
Andrew Cloke (andrew-cloke) wrote :

Disco has now EOL'ed.

Revision history for this message
Ike Panhc (ikepanhc) wrote :
Revision history for this message
Ike Panhc (ikepanhc) wrote :

Still have no idea how to suspend/resume remotely. Close this bug for 4.15 kernel. Let me know if there is way to wakeup system via BMC, or you can use bionic HWE kernel which contains this patch.

Changed in kunpeng920:
status: Incomplete → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

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