[PATCH] scsi/NCR5380: dprintk macro

[PATCH] scsi/NCR5380: dprintk macro

[Finn Thain]

This is the delta between the two submissions:
[PATCH 00/12] scsi/NCR5380: fix debugging macros and #include structure
and
[PATCH v2 00/12] scsi/NCR5380: fix debugging macros and #include structure

The macro definition changes were discussed on the mailing list during 
review. The idea is to get the compiler to check the parameters of 
disabled printk() calls so that the debugging code doesn't rot again.

Signed-off-by: Finn Thain <fthain@telegraphics.com.au>

---

I know the whitespace is wrong throughout these drivers but when I put 
this patch series together it seemed best to defer the whitespace cleanup. 
(I still intend to do that cleanup.)

diff -ru v1/drivers/scsi/NCR5380.h v2/drivers/scsi/NCR5380.h
--- v1/drivers/scsi/NCR5380.h	2014-05-29 00:07:28.525627086 +1000
+++ v2/drivers/scsi/NCR5380.h	2014-05-29 00:06:41.335557873 +1000
@@ -295,9 +295,10 @@
 #define NDEBUG (0)
 #endif
 
+#define dprintk(flg, fmt, ...) \
+	do { if ((NDEBUG) & (flg)) pr_debug(fmt, ## __VA_ARGS__); } while (0)
+
 #if NDEBUG
-#define dprintk(flg, fmt, args...) \
-	do { if ((NDEBUG) & (flg)) pr_debug(fmt, ## args); } while (0)
 #define NCR5380_dprint(flg, arg) \
 	do { if ((NDEBUG) & (flg)) NCR5380_print(arg); } while (0)
 #define NCR5380_dprint_phase(flg, arg) \
@@ -305,7 +306,6 @@
 static void NCR5380_print_phase(struct Scsi_Host *instance);
 static void NCR5380_print(struct Scsi_Host *instance);
 #else
-#define dprintk(flg, fmt, args...)     do {} while (0)
 #define NCR5380_dprint(flg, arg)       do {} while (0)
 #define NCR5380_dprint_phase(flg, arg) do {} while (0)
 #endif
diff -ru v1/drivers/scsi/sun3_NCR5380.c v2/drivers/scsi/sun3_NCR5380.c
--- v1/drivers/scsi/sun3_NCR5380.c	2014-05-29 00:07:28.525627086 +1000
+++ v2/drivers/scsi/sun3_NCR5380.c	2014-05-29 00:06:41.355557902 +1000
@@ -1006,12 +1006,8 @@
 	    for (tmp = (struct scsi_cmnd *) hostdata->issue_queue,
 		 prev = NULL; tmp; prev = tmp, tmp = NEXT(tmp) ) {
 
-#if (NDEBUG & NDEBUG_LISTS)
 		if (prev != tmp)
-		    printk("MAIN tmp=%p   target=%d   busy=%d lun=%d\n",
-			   tmp, tmp->target, hostdata->busy[tmp->target],
-			   tmp->lun);
-#endif
+			dprintk(NDEBUG_LISTS, "MAIN tmp=%p   target=%d   busy=%d lun=%d\n", tmp, tmp->device->id, hostdata->busy[tmp->device->id], tmp->device->lun);
 		/*  When we find one, remove it from the issue queue. */
 		/* ++guenther: possible race with Falcon locking */
 		if (

[PATCH] scsi/NCR5380: dprintk macro

[Christoph Hellwig]

On Thu, May 29, 2014 at 12:43:43AM +1000, Finn Thain wrote:
> 
> This is the delta between the two submissions:
> [PATCH 00/12] scsi/NCR5380: fix debugging macros and #include structure
> and
> [PATCH v2 00/12] scsi/NCR5380: fix debugging macros and #include structure
> 
> The macro definition changes were discussed on the mailing list during 
> review. The idea is to get the compiler to check the parameters of 
> disabled printk() calls so that the debugging code doesn't rot again.
> 
> Signed-off-by: Finn Thain <fthain@telegraphics.com.au>

Thanks a lot Finn!

> I know the whitespace is wrong throughout these drivers but when I put 
> this patch series together it seemed best to defer the whitespace cleanup. 
>(I still intend to do that cleanup.)

Yes, I ignored the whitespace warnings in the debug macro cleanups.  I
fixed them up for the "newly" added code in the sun3 driver merge.