svn commit: r291911 - head/sys/dev/iscsi

[Steven Hartland]

On 07/12/2015 06:41, Andriy Gapon wrote:
> On 07/12/2015 04:56, Steven Hartland wrote:
>> Author: smh
>> Date: Mon Dec  7 02:56:08 2015
>> New Revision: 291911
>> URL: https://svnweb.freebsd.org/changeset/base/291911
>>
>> Log:
>>    Fix panic on shutdown due to iscsi event priority
>>    
>>    iscsi's shutdown_pre_sync prio was SHUTDOWN_PRI_FIRST which caused it to
>>    run before other high priority handlers such as filesystems e.g. ZFS.
>>    
>>    This meant the iscsi sessions where removed before the ZFS geom consumer
>>    was closed, resulting in a panic from g_access calls on debug kernels
>>    due to negative acr.
>>    
>>    Instead use the same as the old iscsi_initiator SHUTDOWN_PRI_DEFAULT-1
>>    which allows it to run before dashutdown etc but after filesystems.
>>    
>>    MFC after:	2 weeks
>>    Sponsored by:	Multiplay
>>
>> Modified:
>>    head/sys/dev/iscsi/iscsi.c
>>
>> Modified: head/sys/dev/iscsi/iscsi.c
>> ==============================================================================
>> --- head/sys/dev/iscsi/iscsi.c	Sun Dec  6 21:31:09 2015	(r291910)
>> +++ head/sys/dev/iscsi/iscsi.c	Mon Dec  7 02:56:08 2015	(r291911)
>> @@ -2365,7 +2365,7 @@ iscsi_load(void)
>>   	sc->sc_cdev->si_drv1 = sc;
>>   
>>   	sc->sc_shutdown_eh = EVENTHANDLER_REGISTER(shutdown_pre_sync,
>> -	    iscsi_shutdown, sc, SHUTDOWN_PRI_FIRST);
>> +	    iscsi_shutdown, sc, SHUTDOWN_PRI_DEFAULT-1);
>>   
>>   	return (0);
>>   }
> It would be nice to also add a comment here explaining the priority.
> And spaces around '-' operator.
Ah, there was actually already a comment there but seems that's been 
removed in HEAD by r286226 which also changed the post_sync to a 
pre_sync so its currently really broken in HEAD and stable/10 compared 
to 10.2. Surprisingly this was already mentioned by mav in the review of 
said commit (https://reviews.freebsd.org/D3052).

The style was copied from iscsi_initiator, I'll fix them both.

I'm retesting this to validate down session recovery now and will update 
once I've confirmed that.