Bug 751320

Summary: hotplug failure with pciehp
Product: Red Hat Enterprise Linux 5 Reporter: Shahreen <shahreen_512>
Component: acpiAssignee: Prarit Bhargava <prarit>
Status: CLOSED WONTFIX QA Contact: Red Hat Kernel QE team <kernel-qe>
Severity: high Docs Contact:
Priority: unspecified    
Version: 5.3CC: jarod
Target Milestone: rc   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-11-04 16:51:13 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description Shahreen 2011-11-04 10:22:42 UTC
Description of problem:

Hotplugging fails with 'pciehp: hpc_check_lnk_status : Link
Training Error occurs'

Version-Release number of selected component (if applicable):

pciehp: PCI Express Hot Plug Controller
Driver version: 0.4

How reproducible:

I am trying to hotplug an AdvancedTCA Mezzanine card in Kontron AT8402 carrier. I tried in both RHEL5 up 0 and 3 as well as SLES 10 where hotplug fails is all three. The OS's were 32 bit.

Steps to Reproduce:

1. I enable hotplug in the ATCA carrier setting and push the hotplug module by modprobe pciehp pciehp_force=1

(pusing acpiphp givs, FATAL: Error inserting acpiphp (/lib/modules/2.6.9-55.ELsmp/kernel/drivers/pci/hotplug/acpiphp.ko): No such device
)

After pusing pciehp,

dmesg:

Jan  1 00:05:41 atca_node1 kernel: pciehp: HPC vendor_id 8086 device_id
2690 ss_vid 0 ss_did 0
Jan  1 00:05:41 atca_node1 kernel: pciehp: HPC vendor_id 10b5 device_id
8532 ss_vid 0 ss_did 0
Jan  1 00:05:44 atca_node1 last message repeated 4 times
Jan  1 00:05:44 atca_node1 kernel: pciehp: PCI Express Hot Plug Controller
Driver version: 0.4

lsmod:
pciehp                 41245  0

2. I hotplug the AMC card in a bay. The the card seems to get power as the status light goes green (also seen from dmesg below) but lspci does not list that board. The board is not usable.

dmesg:

Jan  1 00:07:14 atca_node1 kernel: pciehp: Latch close on Slot(0010_0000)
Jan  1 00:07:14 atca_node1 kernel: pciehp: Button pressed on
Slot(0010_0000)
Jan  1 00:07:14 atca_node1 kernel: pciehp: Card present on Slot(0010_0000)
Jan  1 00:07:14 atca_node1 kernel: pciehp: PCI slot #0010_0000 - powering
on due to button press.
Jan  1 00:07:24 atca_node1 kernel: pciehp: hpc_check_lnk_status : Link
Training Error occurs
Jan  1 00:07:24 atca_node1 kernel: pciehp: board_added: Failed to check
link status
  

Actual results:

Expected results:

It seems that the hotplug module is not reading any PCI interrupt and hence the card is disabled.

Please any idea/suggestion on what might be going wrong here and how to resole this?

Thanks very much,
Additional info:

Comment 1 Prarit Bhargava 2013-11-04 16:51:13 UTC
This Bugzilla has been reviewed by Red Hat and is not planned on being
addressed in Red Hat Enterprise Linux 5, and therefore is being closed.
If this bug is critical to production systems, please contact your Red
Hat support representative and provide a sufficient business justification
in order to re-open it.