Re: [PATCH 2/2] ipmi_ssif: Fix crash seen while ipmi_unregister_smi

[George Cherian <gcherian@caviumnetworks.com>]

Hi Corey,


On 08/24/2018 06:38 PM, Corey Minyard wrote:
> 
> On 08/24/2018 06:10 AM, George Cherian wrote:
>> Dont set ssif_info->intf to NULL before ipmi_unresgiter_smi.
>> shutdown_ssif will anyways free ssif_info.
> 
> This is correct, but it goes a little deeper.  I just sent out a
> patch yesterday that included this.

Yes I saw the patch now, 
https://sourceforge.net/p/openipmi/mailman/message/36397896/
I will test and update in that thread.

> 
> Thanks,
> 
> -corey
> 
>> Following crash is obsearved if ssif_info->intf is set to NULL
>> before ipmi_unregister_smi.
>>
>>   CPU: 119 PID: 7317 Comm: kssif000e Not tainted 4.18.0+ #80
>>   Hardware name: Cavium Inc. Saber/Saber, BIOS Cavium reference 
>> firmware version 7.0 08/04/2018
>>   pstate: 20400009 (nzCv daif +PAN -UAO)
>>   pc : ipmi_smi_msg_received+0x44/0x3bc [ipmi_msghandler]
>>   lr : deliver_recv_msg+0x30/0x5c [ipmi_ssif]
>>   sp : ffff000037a0fd20
>>   x29: ffff000037a0fd20 x28: 0000000000000000
>>   x27: ffff0000047e08f0 x26: ffff800ed9375800
>>   x25: ffff000037a0fe00 x24: ffff000009073000
>>   x23: 0000000000000013 x22: 0000000000000000
>>   x21: 0000000000007000 x20: ffff800adce18400
>>   x19: 0000000000000000 x18: ffff00003742fd38
>>   x17: ffff0000089960f0 x16: 000000000000000e
>>   x15: 0000000000000007 x14: 0000000000000000
>>   x13: 0000000000000000 x12: 0000000000000033
>>   x11: 0000000000000381 x10: 0000000000000ba0
>>   x9 : 0000000000000000 x8 : ffff800ac001fc00
>>   x7 : ffff7fe003b4d800 x6 : ffff800adce1854b
>>   x5 : 0000000000000014 x4 : 0000000000000004
>>   x3 : 0000000000000000 x2 : 0000000000000002
>>   x1 : 567cb12f8b916b00 x0 : 0000000000000002
>>   Process kssif000e (pid: 7317, stack limit = 0x0000000041077d8a)
>>   Call trace:
>>    ipmi_smi_msg_received+0x44/0x3bc [ipmi_msghandler]
>>    deliver_recv_msg+0x30/0x5c [ipmi_ssif]
>>    msg_done_handler+0x2f0/0x66c [ipmi_ssif]
>>    ipmi_ssif_thread+0x108/0x124 [ipmi_ssif]
>>    kthread+0x108/0x134
>>    ret_from_fork+0x10/0x18
>>   Code: b9402280 91401e75 f90037a1 7100041f (b945bab6)
>>   ---[ end trace fb7d748bc7b17490 ]---
>>   Kernel panic - not syncing: Fatal exception
>>   SMP: stopping secondary CPUs
>>   Kernel Offset: disabled
>>   CPU features: 0x23800c38
>>   Memory Limit: none
>>   ---[ end Kernel panic - not syncing: Fatal exception ]---
>>
>> Signed-off-by: George Cherian <george.cherian@cavium.com>
>> ---
>>   drivers/char/ipmi/ipmi_ssif.c | 5 +----
>>   1 file changed, 1 insertion(+), 4 deletions(-)
>>
>> diff --git a/drivers/char/ipmi/ipmi_ssif.c 
>> b/drivers/char/ipmi/ipmi_ssif.c
>> index ccdf6b1..1490636 100644
>> --- a/drivers/char/ipmi/ipmi_ssif.c
>> +++ b/drivers/char/ipmi/ipmi_ssif.c
>> @@ -1226,7 +1226,6 @@ static void shutdown_ssif(void *send_info)
>>   static int ssif_remove(struct i2c_client *client)
>>   {
>>       struct ssif_info *ssif_info = i2c_get_clientdata(client);
>> -     struct ipmi_smi *intf;
>>       struct ssif_addr_info *addr_info;
>>
>>       if (!ssif_info)
>> @@ -1236,9 +1235,7 @@ static int ssif_remove(struct i2c_client *client)
>>        * After this point, we won't deliver anything asychronously
>>        * to the message handler.  We can unregister ourself.
>>        */
>> -     intf = ssif_info->intf;
>> -     ssif_info->intf = NULL;
>> -     ipmi_unregister_smi(intf);
>> +     ipmi_unregister_smi(ssif_info->intf);
>>
>>       list_for_each_entry(addr_info, &ssif_infos, link) {
>>               if (addr_info->client == client) {
> 
>