Re: marvell_cesa continues to fail on kirkwood with kernel 4.4
On Feb 10, 2016, at 3:49 AM, JM <fijam@archlinux.us> wrote:
> On Tue, Feb 9, 2016 at 8:27 PM, Rick Thomas <rbthomas@pobox.com> wrote:
>>
>> On Feb 9, 2016, at 4:12 AM, JM <fijam@archlinux.us> wrote:
>>
>>> On Fri, Feb 5, 2016 at 5:12 AM, Martin Michlmayr <tbm@cyrius.com> wrote:
>>>> * JM <fijam@archlinux.us> [2016-02-05 04:20]:
>>>>> To quickly bring you up to speed, the driver for Marvell's hardware
>>>>> crypto accelerator embedded in their Armada/Kirkwood SoCs has been
>>>>> rewritten as marvell_cesa and merged in kernel 4.2. The new driver
>>>>> received a number of patches since and has been enabled in Debian as
>>>>> a module in 4.4~rc4-1~exp1 (debian bug #807634), coexisting with the
>>>>> old driver (mv_cesa).
>>>>
>>>> I think it would be best if you directly reported this to
>>>> linux-arm-kernel@lists.infradead.org
>>>> and copied:
>>>> Boris BREZILLON <boris.brezillon@free-electrons.com>
>>>> Arnaud Ebalard <arno@natisbad.org>
>>>> Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
>>>> Russell King <rmk+kernel@arm.linux.org.uk>
>>>>
>>>> --
>>>> Martin Michlmayr
>>>> http://www.cyrius.com/
>>>
>>> Update:
>>>
>>> I reported the issue to upstream [1] but they cannot replicate it on
>>> their hardware. I've built a vanilla 4.4.1 kernel using Debian's
>>> .config as explained in [2] and the module still fails in the same
>>> fashion.
>>>
>>> For an independent confirmation, I'd appreciate if someone with a
>>> Kirkwood device (preferably a QNAP) could test the 4.4 kernel from
>>> experimental and try the following:
>>>
>>> rmmod mv_cesa
>>> modprobe marvell_cesa allhwsupport=1
>>> dmesg | tail -n 1
>>>
>>> I've been asked to build a vanilla kernel again using .config provided
>>> by Arnaud Ebalard for ReadyNAS Duo [3], but it ends up being too large
>>> to fit in QNAP's 2MB mtd kernel partition, and I am also not sure if
>>> it will work with the Debian userland, or which drivers I need to
>>> enable for the QNAP to work properly (flash, rtc?), any hints would be
>>> appreciated.
>>>
>>> Best regards,
>>> Jan
>>>
>>> 1 http://lists.infradead.org/pipermail/linux-arm-kernel/2016-February/thread.html#405279
>>> 2 http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s-kernel-org-package
>>> 3 http://lists.infradead.org/pipermail/linux-arm-kernel/2016-February/405530.html
>>
>> I have a sheevaplug reserved for testing that I could try installing the linux-image-4.4.0-trunk package from Debian experimental, if that would help...
>>
>> Rick
>>
>
> Please give it a shot!
>
> Best regards,
> Jan
Here you go...
> root@sheeva:~# rmmod mv_cesa
> root@sheeva:~# modprobe marvell_cesa allhwsupport=1
> root@sheeva:~# dmesg | tail -n 20
> [ 16.730426] systemd[1]: Mounted Debug File System.
> [ 16.756783] systemd[1]: Started Load Kernel Modules.
> [ 16.776948] systemd[1]: Started Create list of required static device nodes for the current kernel.
> [ 16.804146] systemd[1]: Started LSB: Set console font and keymap.
> [ 17.023950] systemd[1]: Started Journal Service.
> [ 18.514731] EXT4-fs (dm-0): re-mounted. Opts: errors=remount-ro
> [ 19.804133] orion_wdt: Initial timeout 21 sec
> [ 19.848469] MV-CESA:Could not register sha1 driver
> [ 19.848485] MV-CESA:Could not register hmac-sha1 driver
> [ 20.379009] systemd-journald[180]: Received request to flush runtime journal from PID 1
> [ 20.386082] sd 0:0:0:0: Attached scsi generic sg0 type 0
> [ 20.386886] ses 0:0:0:1: Attached scsi generic sg1 type 13
> [ 21.518406] EXT4-fs (mmcblk0p1): mounting ext2 file system using the ext4 subsystem
> [ 21.682097] EXT4-fs (mmcblk0p1): mounted filesystem without journal. Opts: (null)
> [ 22.402713] Adding 9764860k swap on /dev/mapper/sheeva-swap. Priority:-1 extents:1 across:9764860k FS
> [ 22.938703] EXT4-fs (dm-2): mounted filesystem with ordered data mode. Opts: (null)
> [ 23.631978] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
> [ 27.662122] mv643xx_eth_port mv643xx_eth_port.0 eth0: link up, 1000 Mb/s, full duplex, flow control disabled
> [ 27.662180] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
> [ 37.350057] warning: process `colord-sane' used the deprecated sysctl system call with 8.1.2.
> root@sheeva:~# uname -a
> Linux sheeva 4.4.0-trunk-marvell #1 Debian 4.4.1-1~exp1 (2016-02-10) armv5tel GNU/Linux
> root@sheeva:~#
For what it's worth, the system boots fine and appears to run OK with the 4.4 kernel.
Hope it helps!
Rick
Reply to: