WCN3680 support

[Eugene Krasnikov]

For some reason FW is crashing after a while.

On 16 декабря 2013 г., 22:04, Jason Mobarak <jam at cozybit.com> wrote:
> On Fri, Dec 13, 2013 at 12:40 AM, Pontus Fuchs <pontus.fuchs at gmail.com> wrote:
>> On 2013-12-12 13:01, Pontus Fuchs wrote:
>>> On 2013-12-11 00:06, Jason Mobarak wrote:
>>>>
>>>>
>>>> Some basic background on my setup: the phone is a C6903, my backports
>>>> is at tag backports-20131113 (af44a4a), and was generated from
>>>> next-20131113 (84ea88b) from the wireless testing tree.  The phone is
>>>> using build 14.1.G.1.534 [1] of the OS-- which is a 3.4.0 kernel
>>>> (which can be obtained from Sony's open source drop).  I modified the
>>>> kernel to support mac80211 [2] as a module so that backports would
>>>> work.
>>>
>>>
>>> I suspect the wcnss fw is a bit old in the OS you are using. Do you know
>>> if there is an updated OS available?
>>
>>
>> I think I found the solution to your problem now. Please try "wcn36xx: Wait
>> longer for SMD commands to complete"
>
> Thanks, this works to successfully bring up the chip on 14.1.G.1.534!!
>  However, I ran into some additional instability with 14.1.G.1.534!
> I've since updated to 14.1.G.2.257 and I'm able to bring up a mesh
> interface and run some iperf tests.  However I'm seeing some failures
> after a certain amount of data is pushed through the interface (log
> files also here
> https://gist.github.com/silverjam/fe677cf6bbc363ed6769):
>
> <7>[  288.771645] wcn36xx: tx skb e01af240 len 1544 fc 0388 sn 1143 low ucast
> <3>[  289.098795] Fatal error on the wcnss.
> <3>[  289.098873] wcnss subsystem failure reason: dog.c:1290:Dog bark timeout.
> <6>[  289.098949] subsys-restart: subsystem_restart_dev(): Restart
> sequence requested for wcnss, restart_level = RELATED.
> <6>[  289.101082] subsys-restart: subsystem_shutdown(): [e4854740]:
> Shutting down wcnss
> <3>[  289.137119] smem_find(22, 40): wrong size 88
> <3>[  289.137771] smem_find(22, 40): wrong size 88
> <6>[  289.138208] Bluetooth: Closing HCI-SMD channel :APPS_RIVA_BT_ACL
> <6>[  289.138278] Bluetooth: Closing HCI-SMD channel :APPS_RIVA_BT_CMD
> <6>[  289.138319] Bluetooth: SSR state is : 1
> <6>[  289.138609] Bluetooth: SSR state is : 2
> <6>[  289.138644] Bluetooth: HCI device un-registration going on
> <6>[  289.149217] Bluetooth: SSR state is : 4
> <3>[  289.163309] Ramdump(ramdump_smem): No consumers. Aborting..
> <3>[  289.163359] restart_notifier_cb: unable to dump smem -32
> <3>[  289.165434] wcnss_notifier_cb: sysmon_send_event error -38
> <6>[  289.167685] subsys-restart: subsystem_powerup(): [e4854740]:
> Powering up wcnss
> <6>[  290.391430] pil_pronto fb21b000.qcom,pronto: wcnss: Brought out of reset
> <4>[  290.391522] ------------[ cut here ]------------
> <4>[  290.391587] WARNING: at
> /opt/jam/dev/xperiaz/kernel/xz1/kernel/irq/manage.c:428
> enable_irq+0x50/0x6c()
> <4>[  290.391640] Unbalanced enable for IRQ 181
> <4>[  290.391669] Modules linked in: wcn36xx_msm(O) wcn36xx(O)
> mac80211(O) cfg80211(O) compat(O) ccm ctr texfat(PO) seqiv adsprpc
> adsp_loader hid_sony(O)
> <4>[  290.391905] [<c010c124>] (unwind_backtrace+0x0/0x11c) from
> [<c018d294>] (warn_slowpath_common+0x4c/0x64)
> <4>[  290.391987] [<c018d294>] (warn_slowpath_common+0x4c/0x64) from
> [<c018d32c>] (warn_slowpath_fmt+0x2c/0x3c)
> <4>[  290.392066] [<c018d32c>] (warn_slowpath_fmt+0x2c/0x3c) from
> [<c01de874>] (enable_irq+0x50/0x6c)
> <4>[  290.392151] [<c01de874>] (enable_irq+0x50/0x6c) from
> [<c0132ca0>] (wcnss_powerup+0x64/0x80)
> <4>[  290.392226] [<c0132ca0>] (wcnss_powerup+0x64/0x80) from
> [<c0161c70>] (subsystem_restart_wq_func+0x3d0/0x508)
> <4>[  290.392314] [<c0161c70>] (subsystem_restart_wq_func+0x3d0/0x508)
> from [<c01a5a20>] (process_one_work+0x260/0x440)
> <4>[  290.392397] [<c01a5a20>] (process_one_work+0x260/0x440) from
> [<c01a5dc8>] (worker_thread+0x1c8/0x314)
> <4>[  290.392476] [<c01a5dc8>] (worker_thread+0x1c8/0x314) from
> [<c01a9c50>] (kthread+0x84/0x90)
> <4>[  290.392553] [<c01a9c50>] (kthread+0x84/0x90) from [<c0106908>]
> (kernel_thread_exit+0x0/0x8)
> <4>[  290.392607] ---[ end trace 9e7044c5ea84957b ]---
> <6>[  290.749339] subsys-restart: subsys_err_ready_intr_handler():
> Error ready interrupt occured for wcnss
> <6>[  290.750764] subsys-restart: subsystem_restart_wq_func():
> [e4854740]: Restart sequence for wcnss completed.
> <6>[  290.869719] wcnss: version 01050102
> <6>[  290.869767] wcnss: schedule dnld work for pronto
> <6>[  290.869883] wcnss: NV download
> <6>[  290.875260] wcnss: NV bin size: 29812, total_fragments: 10
> <3>[  290.875567] wcnss: no space available for smd frame
>
> Thanks,
> -Jason M
>
> _______________________________________________
> wcn36xx mailing list
> wcn36xx at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/wcn36xx



-- 
Best regards,
Eugene