On 07/04/14 16:12, Hannes Reinecke wrote: > On 07/04/2014 03:48 PM, Christoph Hellwig wrote: >> I think storing the struct scsi_lun in the scsi_device is the right way >> to go ahead. Any "accessors" for 8 or 32-bit LUNs should be simply >> enough by just ignoring bits in the array, so there's very little >> performance overhead. >> >> If we can get rid of the old scalar LUN field that would be great, >> and given that we know the printk format fallout already it doesn't look >> like too much work. Do you want to look into this? > > Already working on it. That sounds great. Regarding the SRP initiator driver: if the "__be64 lun" declarations in <scsi/srp.h> are changed into "struct scsi_lun lun" then that allows to encode / decode LUNs inside the SRP initiator without having to cast the address of these "lun" members into struct scsi_lun *. In case you would prefer me to have a look into this, please let me know. Bart. -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
