On Thu, Mar 05, 2015 at 03:58:47PM +0100, Bart Van Assche wrote:
> Thanks for the review. Whether scsi_lun.h is separate or not is not
> important to me. The only reason I had proposed to create a separate
> header file for struct scsi_lun is because some other header files only
> need the definiti
On 03/05/15 15:26, Christoph Hellwig wrote:
> In general this looks fine, but:
>
> - why do you need a separate scsi_lun.h?
> - I'd really prefer to only have the protocol defintion here,
>not prototypes for helpers like scsi_device_type, int_to_scsilun
>and scsilun_to_int. The target c
In general this looks fine, but:
- why do you need a separate scsi_lun.h?
- I'd really prefer to only have the protocol defintion here,
not prototypes for helpers like scsi_device_type, int_to_scsilun
and scsilun_to_int. The target code should not depend on the
initiator for helpers.
On Wed, 2015-02-25 at 09:40 +0100, Bart Van Assche wrote:
> Move the constants and functions that are used by both initiator and
> target drivers into new header files. This change will allow to modify
> the SCSI target code such that the initiator SCSI header files are no
> longer included. Note:
Move the constants and functions that are used by both initiator and
target drivers into new header files. This change will allow to modify
the SCSI target code such that the initiator SCSI header files are no
longer included. Note: the SCSI target driver patch is available for
review at
https://g
5 matches
Mail list logo