lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <BAY111-W205B70ED6B226F2F9D8E2088C30@phx.gbl>
Date:	Wed, 4 Feb 2009 13:17:20 -0500
From:	Krishna Kothapalli <krishna_sunitha@...mail.com>
To:	<kaneshige.kenji@...fujitsu.com>
CC:	<ebiederm@...ssion.com>, <linux-pci@...r.kernel.org>,
	<jbarnes@...tuousgeek.org>, <achiang@...com>, <djwong@...ibm.com>,
	<xyzzy@...akeasy.org>, <linux-kernel@...r.kernel.org>
Subject: RE: Q: state of pci express hotplug




> Kenji Kaneshige wrote:
>
> Thank you for sending the information. I understood your
> hotplug slot can be handled by acpiphp or pciehp.
> More majordomo info at http://vger.kernel.org/majordomo-info.html

Tested this with 2.6.28.3 and details below:

>Could you give me the following additional information?
>o My understanding about the problem is:
>  - Bad IRQ happen when you hot-add the device,

   I don't see bad IRQ stack trace anymore, based on the steps you suggested. May be this is because I did not have hotplug driverloaded (and diff kernel build config) ?
  
>  - and the device is not found in the lspci output

   YES

>  - and the device didn't show up under sysfs
>  Am I correct?

   YES
 
> o Details about the steps to reproduce the problem.

  - boot the box with 2.6.28.3
  -   # /sbin/modprobe pciehp pciehp_debug
  - lspci 
  - Insert the PCI express card on the hot swap slot and wait for 2 mins
  - lspci <-- this does not show the PCI express device 
  (The same device shows up in lspci and sysfs during cold boot ) 

> o Does the problem happen with both acpiphp and pciehp?

   That is correct.

> o The pciehp's debug output. We can get this as follows.
  # /sbin/modprobe pciehp pciehp_debug
  # dmesg | grep pciehp 
pciehp 0000:00:09.0:pcie02: Hotplug Controller:
pciehp 0000:00:09.0:pcie02:   Seg/Bus/Dev/Func/IRQ : 0000:00:09.0 IRQ 2295
pciehp 0000:00:09.0:pcie02:   Vendor ID            : 0x8086
pciehp 0000:00:09.0:pcie02:   Device ID            : 0x3410
pciehp 0000:00:09.0:pcie02:   Subsystem ID         : 0x0000
pciehp 0000:00:09.0:pcie02:   Subsystem Vendor ID  : 0x0000
pciehp 0000:00:09.0:pcie02:   PCIe Cap offset      : 0x90
pciehp 0000:00:09.0:pcie02: Slot Capabilities      : 0x0302005f
pciehp 0000:00:09.0:pcie02:   Physical Slot Number : 96
pciehp 0000:00:09.0:pcie02:   Attention Button     : yes
pciehp 0000:00:09.0:pcie02:   Power Controller     : yes
pciehp 0000:00:09.0:pcie02:   MRL Sensor           : yes
pciehp 0000:00:09.0:pcie02:   Attention Indicator  : yes
pciehp 0000:00:09.0:pcie02:   Power Indicator      : yes
pciehp 0000:00:09.0:pcie02:   Hot-Plug Surprise    :  no
pciehp 0000:00:09.0:pcie02:   EMI Present          : yes
pciehp 0000:00:09.0:pcie02:   Command Completed    : yes
pciehp 0000:00:09.0:pcie02: Slot Status            : 0x0020
pciehp 0000:00:09.0:pcie02: Slot Control           : 0x078f
pciehp 0000:00:09.0:pcie02: Link Active Reporting supported
pciehp 0000:00:09.0:pcie02: Command not completed in 1000 msec
pciehp 0000:00:09.0:pcie02: HPC vendor_id 8086 device_id 3410 ss_vid 0 ss_did 0
pciehp 0000:00:09.0:pcie02: Command not completed in 1000 msec
pciehp 0000:00:09.0:pcie02: Registering domain:bus:dev=0000:09:00 hp_slot=0 sun=60 slot_device_offset=0
pciehp 0000:00:09.0:pcie02: get_power_status: physical_slot = 96
pciehp 0000:00:09.0:pcie02: hpc_get_power_status: SLOTCTRL a8 value read 7bf
pciehp 0000:00:09.0:pcie02: get_attention_status: physical_slot = 96
pciehp 0000:00:09.0:pcie02: hpc_get_attention_status: SLOTCTRL a8, value read 7bf
pciehp 0000:00:09.0:pcie02: get_latch_status: physical_slot = 96
pciehp 0000:00:09.0:pcie02: get_adapter_status: physical_slot = 96
pciehp 0000:00:09.0:pcie02: hpc_power_off_slot: slot->hp_slot 0
pciehp 0000:00:09.0:pcie02: Command not completed in 1000 msec
pciehp 0000:00:09.0:pcie02: hpc_power_off_slot: SLOTCTRL a8 write cmd 400
pciehp 0000:00:09.0:pcie02: service driver pciehp loaded
pciehp: pcie_port_service_register = 0
pciehp: PCI Express Hot Plug Controller Driver version: 0.4

-- Hot plug PCI express card and wait for 2 mins
-- no additional out put from dmesg

[root@...alhost test]# /sbin/rmmod pciehp 
[root@...alhost test]# dmesg | grep pciehp
-- Additional  output from dmesg  after pciehp is unloaded ( but the express card is still present in the slot):-
pciehp: unload_pciehpd()
pciehp 0000:00:09.0:pcie02: unloading service driver pciehp
pciehp 0000:00:09.0:pcie02: release_slot: physical_slot = 96
pciehp 0000:00:09.0:pcie02: Command not completed in 1000 msec
pciehp 0000:00:09.0:pcie02: pcie_isr: intr_loc 8
pciehp 0000:00:09.0:pcie02: Presence/Notify input change
pciehp 0000:00:09.0:pcie02: Card present on Slot(øu'?)

-- removed the PCI express card from the slot

[root@...alhost test]# ls -lt /sys/bus/pci/slots
total 0
[root@...alhost test]#  /sbin/modprobe acpiphp debug
[root@...alhost test]# dmesg | grep acpiphp
acpiphp: ACPI Hot Plug PCI Controller Driver version: 0.5
acpiphp_glue: found PCI-to-PCI bridge at PCI 0000:00:09.0
acpiphp_glue: found ACPI PCI Hotplug slot 2 at PCI 0000:09:00
acpiphp: Slot [2] registered
acpiphp_glue: Bus 0000:09 has 1 slot
acpiphp_glue: Total 1 slots
[root@...alhost test]# 
[root@...alhost test]# ls -lt /sys/bus/pci/slots
total 0
drwxr-xr-x 2 root root 0 2009-02-04 09:59 2
---  Insert PCI express card in the slot
[root@...alhost test]# dmesg | grep acpiphp
[root@...alhost test]# lspci | wc
    105    1284    9203
[root@...alhost test]# ls -lt /sys/bus/pci/slots/2/
total 0
-r--r--r-- 1 root root 4096 2009-02-04 10:01 adapter
-r--r--r-- 1 root root 4096 2009-02-04 10:01 address
-rw-r--r-- 1 root root 4096 2009-02-04 10:01 attention
-r--r--r-- 1 root root 4096 2009-02-04 10:01 latch
-rw-r--r-- 1 root root 4096 2009-02-04 10:01 power
[root@...alhost test]# 
[root@...alhost test]# /sbin/rmmod acpiphp
[root@...alhost test]# dmesg | grep acpiphp
acpiphp: Slot [2] unregistered
acpiphp: release_slot - physical_slot = 2
[root@...alhost test]# 



_________________________________________________________________
Windows Live™: Keep your life in sync. 
http://windowslive.com/howitworks?ocid=TXT_TAGLM_WL_t1_allup_howitworks_022009--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ