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]
Date:	Mon, 23 Jul 2012 17:53:34 -0300
From:	Kleber Sacilotto de Souza <klebers@...ux.vnet.ibm.com>
To:	Or Gerlitz <ogerlitz@...lanox.com>
CC:	David Miller <davem@...emloft.net>, netdev@...r.kernel.org,
	jackm@....mellanox.co.il, yevgenyp@...lanox.co.il,
	cascardo@...ux.vnet.ibm.com, brking@...ux.vnet.ibm.com,
	shlomop@...lanox.com
Subject: Re: [PATCH] mlx4: Add support for EEH error recovery

On 07/23/2012 03:12 PM, Kleber Sacilotto de Souza wrote:

> On 07/23/2012 10:45 AM, Or Gerlitz wrote:
> 
>> On 7/23/2012 4:18 PM, Kleber Sacilotto de Souza wrote:
>>> Exactly. The callbacks implemented are from standard PCI error recovery
>>> (Documentation/PCI/pci-error-recovery.txt) and the changes doesn't
>>> assume any platform in specific. The code was tested only on powerpc
>>> systems [...]
>>
>> So how did you test that? using the kernel provided error injection
>> support and user space tool (which?) or in another way? we've trying
>> quickly here to inject errors using /sbin/ear-inject from
>> ras-utils-6.1-1.el6.x86_64 on a kernel built with
>>
>> CONFIG_PCIEAER=y
>> CONFIG_PCIEAER_INJECT=m
> 
> 
> For powerpc we have an IBM internal user space tool that injects the
> error on the bus with the aid of the system firmware. The kernel used
> was built with the option:
> 
> CONFIG_EEH=y
> 
> and without the AER options. I will run some more tests with the AER
> options activated.


I tested the powerpc error injection with

CONFIG_EEH=y
CONFIG_PCIEAER=y
CONFIG_PCIEAER_INJECT=m

and with the aer_inject module loaded and it didn't affect the EEH
recovery, the adapter recovered as expected.

> 
>>
>> and it failed to inject errors, SB details.
>>
>> Or.
>>> since I don't have any mlx4 card on other platforms, however,
>>> these changes shouldn't make the error recover any worse than the
>>> current state.
>>
>>> # lspci | grep 08.00.1
>>> 08:00.1 Ethernet controller: Intel Corporation 82575EB Gigabit Network
>>> Connection (rev 02)
>>
>>> # cat /tmp/intel.aer
>>> AER
>>> BUS 8 DEV 0 FN 1
>>> COR_STATUS BAD_TLP
>>> HEADER_LOG 0 1 2 3
>>
>>> # /sbin/aer-inject < /tmp/intel.aer
>>> Error: Failed to write, Invalid argument
>>
>>
>>
>>> # strace -F -f /sbin/aer-inject < /tmp/intel.aer
>>> [...]
>>
>>> open("/dev/aer_inject", O_WRONLY)       = 3
>>> write(3, "\10\0\1\0\0\0\0\0@\0\0\0\0\0\0\0\1\0\0\0\2\0\0\0\3\0\0\0",
>>> 28) = -1 EINVAL (Invalid argument)
>>> write(2, "Error: ", 7Error: )                  = 7
>>> write(2, "Failed to write", 15Failed to write)         = 15
>>> write(2, ", Invalid argument\n", 19, Invalid argument
>>> )    = 19
>>> exit_group(-1)                          = ?
>>
>>
>>
>>
>> -- 
>> To unsubscribe from this list: send the line "unsubscribe netdev" in
>> the body of a message to majordomo@...r.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
> 
> 
> 



-- 
Kleber Sacilotto de Souza
IBM Linux Technology Center

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ