Re: EH method APIs

[Hannes Reinecke <hare@suse.de>]

On 04/04/2014 09:24 AM, Christoph Hellwig wrote:
> On Fri, Apr 04, 2014 at 09:17:21AM +0200, Hannes Reinecke wrote:
>> Plus we don't meddle with block request allocation intrinsics
>> anymore; the SCSI EH command is allocate within the SCSI midlayer,
>> and requests originating from the block layer won't be messed with.
> 
> Any chance we could also switch to a new scsi_tmf structure instead
> of reusing struct scsi_cmnd for that?  It should be a lot smaller
> without all the baggage, and make it clear we're not dealing with a
> command.  That would also solve the scsi_reset_provide issue.
> 
Have to look into this. Problem is that some eh_XXX callback
implementations actually are using the command to send the TMF;
will need to look into them how and if they could be changed.

Actually, alongside with that change I would update the arguments
to the callback functions to align with the strategy level;
eh_device_reset() would be getting struct scsi_device as argument,
eh_target_reset() would get getting struct scsi_target as argument etc.

So the actual SCSI EH would not be tied to struct scsi_cmnd, and we
would just be needing something to send TUR etc in the course of
SCSI EH. Whether that needs to be a struct scsi_cmnd or something
else can discussed.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		      zSeries & Storage
hare@suse.de			      +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: J. Hawn, J. Guild, F. Imendörffer, HRB 16746 (AG Nürnberg)
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html