public inbox for linux-scsi@vger.kernel.org
 help / color / mirror / Atom feed
* [RFC] Set functions for scsi_cmnd result
@ 2008-06-12  9:29 Christof Schmitt
  2008-06-12 15:55 ` James Bottomley
  0 siblings, 1 reply; 3+ messages in thread
From: Christof Schmitt @ 2008-06-12  9:29 UTC (permalink / raw)
  To: linux-scsi

zfcp currently has helper functions to set the host and driver bytes
in the scsi_cmnd result. These functions do not belong in the
low-level driver. The attached patch moves the functions to the global
SCSI header file, so that every driver can use them.

If nobody wants ths, i will remove them from zfcp and directly
manipulate the status like other drivers do.

---
 drivers/s390/scsi/zfcp_scsi.c |   20 --------------------
 include/scsi/scsi.h           |   10 ++++++++++
 2 files changed, 10 insertions(+), 20 deletions(-)

--- a/drivers/s390/scsi/zfcp_scsi.c	2008-06-12 11:00:16.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_scsi.c	2008-06-12 11:06:37.000000000 +0200
@@ -107,27 +107,7 @@ zfcp_set_fcp_dl(struct fcp_cmnd_iu *fcp_
 	*zfcp_get_fcp_dl_ptr(fcp_cmd) = fcp_dl;
 }
 
-/*
- * note: it's a bit-or operation not an assignment
- * regarding the specified byte
- */
-static inline void
-set_byte(int *result, char status, char pos)
-{
-	*result |= status << (pos * 8);
-}
 
-void
-set_host_byte(int *result, char status)
-{
-	set_byte(result, status, 2);
-}
-
-void
-set_driver_byte(int *result, char status)
-{
-	set_byte(result, status, 3);
-}
 
 static int
 zfcp_scsi_slave_alloc(struct scsi_device *sdp)
--- a/include/scsi/scsi.h	2008-06-10 17:30:49.000000000 +0200
+++ b/include/scsi/scsi.h	2008-06-12 11:06:34.000000000 +0200
@@ -425,6 +425,16 @@ struct scsi_lun {
 #define driver_byte(result) (((result) >> 24) & 0xff)
 #define suggestion(result)  (driver_byte(result) & SUGGEST_MASK)
 
+static inline void set_host_byte(int *result, char status)
+{
+	*result |= status << 16;
+}
+
+static inline void set_driver_byte(int *result, char status)
+{
+	*result |= status << 24;
+}
+
 #define sense_class(sense)  (((sense) >> 4) & 0x7)
 #define sense_error(sense)  ((sense) & 0xf)
 #define sense_valid(sense)  ((sense) & 0x80);

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [RFC] Set functions for scsi_cmnd result
  2008-06-12  9:29 [RFC] Set functions for scsi_cmnd result Christof Schmitt
@ 2008-06-12 15:55 ` James Bottomley
  2008-06-16  8:58   ` Christof Schmitt
  0 siblings, 1 reply; 3+ messages in thread
From: James Bottomley @ 2008-06-12 15:55 UTC (permalink / raw)
  To: Christof Schmitt; +Cc: linux-scsi

On Thu, 2008-06-12 at 11:29 +0200, Christof Schmitt wrote:
> zfcp currently has helper functions to set the host and driver bytes
> in the scsi_cmnd result. These functions do not belong in the
> low-level driver. The attached patch moves the functions to the global
> SCSI header file, so that every driver can use them.
> 
> If nobody wants ths, i will remove them from zfcp and directly
> manipulate the status like other drivers do.
> 
> ---
>  drivers/s390/scsi/zfcp_scsi.c |   20 --------------------
>  include/scsi/scsi.h           |   10 ++++++++++
>  2 files changed, 10 insertions(+), 20 deletions(-)
> 
> --- a/drivers/s390/scsi/zfcp_scsi.c	2008-06-12 11:00:16.000000000 +0200
> +++ b/drivers/s390/scsi/zfcp_scsi.c	2008-06-12 11:06:37.000000000 +0200
> @@ -107,27 +107,7 @@ zfcp_set_fcp_dl(struct fcp_cmnd_iu *fcp_
>  	*zfcp_get_fcp_dl_ptr(fcp_cmd) = fcp_dl;
>  }
>  
> -/*
> - * note: it's a bit-or operation not an assignment
> - * regarding the specified byte
> - */
> -static inline void
> -set_byte(int *result, char status, char pos)
> -{
> -	*result |= status << (pos * 8);
> -}
>  
> -void
> -set_host_byte(int *result, char status)
> -{
> -	set_byte(result, status, 2);
> -}
> -
> -void
> -set_driver_byte(int *result, char status)
> -{
> -	set_byte(result, status, 3);
> -}
>  
>  static int
>  zfcp_scsi_slave_alloc(struct scsi_device *sdp)
> --- a/include/scsi/scsi.h	2008-06-10 17:30:49.000000000 +0200
> +++ b/include/scsi/scsi.h	2008-06-12 11:06:34.000000000 +0200
> @@ -425,6 +425,16 @@ struct scsi_lun {
>  #define driver_byte(result) (((result) >> 24) & 0xff)
>  #define suggestion(result)  (driver_byte(result) & SUGGEST_MASK)
>  
> +static inline void set_host_byte(int *result, char status)
> +{
> +	*result |= status << 16;
> +}
> +
> +static inline void set_driver_byte(int *result, char status)
> +{
> +	*result |= status << 24;
> +}
> +

If we're making accessors, it might be better if they took a struct
scsi_cmnd rather than a pointer to one of its elements.  Other than
that, this looks fine.

James



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [RFC] Set functions for scsi_cmnd result
  2008-06-12 15:55 ` James Bottomley
@ 2008-06-16  8:58   ` Christof Schmitt
  0 siblings, 0 replies; 3+ messages in thread
From: Christof Schmitt @ 2008-06-16  8:58 UTC (permalink / raw)
  To: James Bottomley; +Cc: linux-scsi

On Thu, Jun 12, 2008 at 10:55:41AM -0500, James Bottomley wrote:
> > --- a/include/scsi/scsi.h	2008-06-10 17:30:49.000000000 +0200
> > +++ b/include/scsi/scsi.h	2008-06-12 11:06:34.000000000 +0200
> > @@ -425,6 +425,16 @@ struct scsi_lun {
> >  #define driver_byte(result) (((result) >> 24) & 0xff)
> >  #define suggestion(result)  (driver_byte(result) & SUGGEST_MASK)
> >  
> > +static inline void set_host_byte(int *result, char status)
> > +{
> > +	*result |= status << 16;
> > +}
> > +
> > +static inline void set_driver_byte(int *result, char status)
> > +{
> > +	*result |= status << 24;
> > +}
> > +
> 
> If we're making accessors, it might be better if they took a struct
> scsi_cmnd rather than a pointer to one of its elements.  Other than
> that, this looks fine.

Ok. I will update the patch and send it with the next batch of zfcp
patches.

Christof

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2008-06-16  8:58 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-06-12  9:29 [RFC] Set functions for scsi_cmnd result Christof Schmitt
2008-06-12 15:55 ` James Bottomley
2008-06-16  8:58   ` Christof Schmitt

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox