2017-05-18 19:03:45

by Larry Finger

[permalink] [raw]
Subject: Re: crash in r8723bs

On 05/17/2017 10:05 AM, Stefan Assmann wrote:
> Hi Larry,
>
> the rtl8723bs made it into staging, awesome.
> I built a kernel to test it but it goes down the drain pretty quickly
> on modprobe. Going forward who should I discuss bugs with?
>
> Here's what's going on atm.
>
> Stefan
>
> [ 125.244195] r8723bs: module is from the staging directory, the quality is unknown, you have been warned.
> [ 125.262538] RTL8723BS: module init start
> [ 125.269377] RTL8723BS: rtl8723bs v4.3.5.5_12290.20140916_BTCOEX20140507-4E40
> [ 125.275955] RTL8723BS: rtl8723bs BT-Coex version = BTCOEX20140507-4E40
> [ 125.350236] pnetdev = e421c800
> [ 125.444211] RTL8723BS: rtw_ndev_init(wlan0)
> [ 125.451896] ------------[ cut here ]------------
> [ 125.459543] kernel BUG at net/wireless/core.h:111!
> [ 125.466098] invalid opcode: 0000 [#1] SMP
> [ 125.472524] Modules linked in: r8723bs(C+) cfg80211 rfkill hid_logitech_hidpp mmc_block crc32c_intel hid_logitech_dj i2c_hid sdhci_acpi sdhci mmc_core
> [ 125.485622] CPU: 1 PID: 456 Comm: modprobe Tainted: G C 4.12.0-rc1+ #1
> [ 125.492260] Hardware name: AXDIA International GmbH Wintab Gen 8/Type2 - Board Product Name, BIOS AD890.1.02.022 09/25/2014
> [ 125.505696] task: f5b46480 task.stack: e4724000
> [ 125.512515] EIP: cfg80211_netdev_notifier_call+0x4f0/0x6b0 [cfg80211]
> [ 125.519329] EFLAGS: 00010246 CPU: 1
> [ 125.526049] EAX: 00000000 EBX: f5a88400 ECX: e4725ce8 EDX: 00000010
> [ 125.532748] ESI: 00000000 EDI: ffffffed EBP: e4725c94 ESP: e4725bcc
> [ 125.539356] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
> [ 125.545890] CR0: 80050033 CR2: b7648a70 CR3: 24111000 CR4: 001006d0
> [ 125.552462] Call Trace:
> [ 125.558947] ? smp_apic_timer_interrupt+0x38/0x50
> [ 125.565407] ? ida_get_new_above+0xdd/0x2c0
> [ 125.571792] ? inetdev_event+0x33/0x4e0
> [ 125.578163] notifier_call_chain+0x4e/0x70
> [ 125.584546] raw_notifier_call_chain+0x11/0x20
> [ 125.590908] call_netdevice_notifiers_info+0x2a/0x60
> [ 125.597268] register_netdevice+0x213/0x4d0
> [ 125.603564] register_netdev+0x17/0x30
> [ 125.609807] rtw_drv_register_netdev+0x5a/0x90 [r8723bs]
> [ 125.616013] rtw_drv_init+0x1e8/0x220 [r8723bs]
> [ 125.622122] sdio_bus_probe+0xe0/0x100 [mmc_core]
> [ 125.628166] driver_probe_device+0x249/0x430
> [ 125.634174] __driver_attach+0xd9/0xe0
> [ 125.640086] ? unbind_store+0x100/0x100
> [ 125.645894] ? driver_probe_device+0x430/0x430
> [ 125.651662] bus_for_each_dev+0x57/0xa0
> [ 125.657354] driver_attach+0x19/0x20
> [ 125.662972] ? driver_probe_device+0x430/0x430
> [ 125.668553] bus_add_driver+0x15f/0x250
> [ 125.674041] ? 0xf7f7f000
> [ 125.679435] driver_register+0x56/0xd0
> [ 125.684795] ? 0xf7f7f000
> [ 125.690097] sdio_register_driver+0x1b/0x20 [mmc_core]
> [ 125.695440] rtw_drv_entry+0x52/0x1000 [r8723bs]
> [ 125.700729] do_one_initcall+0x46/0x170
> [ 125.705981] ? kvfree+0x45/0x50
> [ 125.711172] ? kmem_cache_alloc_trace+0x155/0x1c0
> [ 125.716317] ? do_init_module+0x21/0x1c7
> [ 125.721414] ? do_init_module+0x21/0x1c7
> [ 125.726431] do_init_module+0x50/0x1c7
> [ 125.731411] load_module+0x2170/0x2780
> [ 125.736349] ? kernel_read_file+0x28f/0x2c0
> [ 125.741262] SyS_finit_module+0xa7/0xe0
> [ 125.746217] do_fast_syscall_32+0x6c/0x130
> [ 125.751535] entry_SYSENTER_32+0x4e/0x7c
> [ 125.756495] EIP: 0xb77d8cc9
> [ 125.761445] EFLAGS: 00000246 CPU: 1
> [ 125.766417] EAX: ffffffda EBX: 00000006 ECX: 800459da EDX: 00000000
> [ 125.771486] ESI: 81c08f60 EDI: 81c094e8 EBP: 81c081d8 ESP: bfaf1b4c
> [ 125.776599] DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 007b
> [ 125.781752] Code: ff b9 01 00 00 00 ba 03 00 00 00 83 a8 a4 fe ff ff 01 6a 00 2d 58 01 00 00 e8 4d 1e 5d d5 58 e9 97 fb ff ff 8d b4 26 00 00 00 00 <0f> 0b 8d b6 00 00 00 00 0f ff e9 5f fb ff ff 90 80 7b 38 00 0f
> [ 125.797796] EIP: cfg80211_netdev_notifier_call+0x4f0/0x6b0 [cfg80211] SS:ESP: 0068:e4725bcc
> [ 125.803343] ---[ end trace 7ffc5b2894f962d1 ]---

The best place to discuss this is in the linux-wireless ML, which is Cc'd here.

The BUG is triggered by a NULL value for the struct wiphy object. I have no idea
why that is happening on your machine. Have you tried the driver at
http://github.com/hadess/rtl8723bs.git? It may give the same result, but if not,
then that would help a lot.

Larry


2017-05-18 22:50:27

by Ian W MORRISON

[permalink] [raw]
Subject: Re: crash in r8723bs

On 19 May 2017 at 05:03, Larry Finger <[email protected]> wrote:
> On 05/17/2017 10:05 AM, Stefan Assmann wrote:
>>
>> Hi Larry,
>>
>> the rtl8723bs made it into staging, awesome.
>> I built a kernel to test it but it goes down the drain pretty quickly
>> on modprobe. Going forward who should I discuss bugs with?
>>
>> Here's what's going on atm.
>>
>> Stefan
>>

>> [ 125.485622] CPU: 1 PID: 456 Comm: modprobe Tainted: G C
>> 4.12.0-rc1+ #1

If the kernel is built from v4.12-rc1 then you need to add
https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git/commit/?h=staging-linus&id=ca6e8cdbe1865caf7b05483e1a242e72d9bc919f

Regards,
Ian

2017-05-19 05:57:56

by Stefan Assmann

[permalink] [raw]
Subject: Re: crash in r8723bs

On 19.05.2017 00:50, Ian W MORRISON wrote:
> On 19 May 2017 at 05:03, Larry Finger <[email protected]> wrote:
>> On 05/17/2017 10:05 AM, Stefan Assmann wrote:
>>>
>>> Hi Larry,
>>>
>>> the rtl8723bs made it into staging, awesome.
>>> I built a kernel to test it but it goes down the drain pretty quickly
>>> on modprobe. Going forward who should I discuss bugs with?
>>>
>>> Here's what's going on atm.
>>>
>>> Stefan
>>>
>
>>> [ 125.485622] CPU: 1 PID: 456 Comm: modprobe Tainted: G C
>>> 4.12.0-rc1+ #1
>
> If the kernel is built from v4.12-rc1 then you need to add
> https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git/commit/?h=staging-linus&id=ca6e8cdbe1865caf7b05483e1a242e72d9bc919f

Thanks Ian, that works.

Stefan