PATCH: fix "ide_scan_direction defined but not used" in ide.c

PATCH: fix "ide_scan_direction defined but not used" in ide.c

[Muli Ben-Yehuda]

ide_scan_drection is only used if CONFIG_BLK_DEV_IDEPCI is defined,
giving a compilation warning otherwise. Against 2.5.54-bk. 

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.974   -> 1.975  
#	   drivers/ide/ide.c	1.48    -> 1.49   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/01/07	mulix@alhambra.mulix.org	1.975
# fix "defined but not used" warning
# --------------------------------------------
#
diff -Nru a/drivers/ide/ide.c b/drivers/ide/ide.c
--- a/drivers/ide/ide.c	Tue Jan  7 14:17:42 2003
+++ b/drivers/ide/ide.c	Tue Jan  7 14:17:42 2003
@@ -179,7 +179,9 @@
 
 spinlock_t ide_lock __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
 
+#ifdef CONFIG_BLK_DEV_IDEPCI
 static int ide_scan_direction; /* THIS was formerly 2.2.x pci=reverse */
+#endif /* CONFIG_BLK_DEV_IDEPCI */ 
 
 #ifdef CONFIG_IDEDMA_AUTO
 int noautodma = 0;

-- 
Muli Ben-Yehuda

my opinions may seem crazy. But they all make sense. Insane sense, but
sense nontheless. -- Shlomi Fish on #offtopic.

Re: PATCH: fix "ide_scan_direction defined but not used" in ide.c

[Muli Ben-Yehuda]

On Tue, Jan 07, 2003 at 04:36:17PM +0000, Alan Cox wrote:
> On Tue, 2003-01-07 at 13:10, Muli Ben-Yehuda wrote:
> > ide_scan_drection is only used if CONFIG_BLK_DEV_IDEPCI is defined,
> > giving a compilation warning otherwise. Against 2.5.54-bk. 
> 
> Please reject. This is uglier than the warning and not the right
> approach

Alan, I bow to your superior knowledge and judgement. However,
'ide_scan_direction' is only used in two places in ide.c, and both of
those are only compiled in if CONFIG_BLK_DEV_IDEPCI:  

in ide_setup(): 

#ifdef CONFIG_BLK_DEV_IDEPCI
        if (!strcmp(s, "ide=reverse")) {
                ide_scan_direction = 1;
                printk(" : Enabled support for IDE inverse scan order.\n");
                return 1;
        }
#endif /* CONFIG_BLK_DEV_IDEPCI */

in probe_for_hwifs(): 

#ifdef CONFIG_BLK_DEV_IDEPCI
        if (pci_present())
        {
                ide_scan_pcibus(ide_scan_direction);
        }
#endif /* CONFIG_BLK_DEV_IDEPCI */

So protecting the variable with #ifdef .. #endif seems the simplest
and least intrusive way to fix it. How would you prefer I do it? 
-- 
Muli Ben-Yehuda

my opinions may seem crazy. But they all make sense. Insane sense, but
sense nontheless. -- Shlomi Fish on #offtopic.

Re: PATCH: fix "ide_scan_direction defined but not used" in ide.c

[Alan Cox]

On Tue, 2003-01-07 at 13:10, Muli Ben-Yehuda wrote:
> ide_scan_drection is only used if CONFIG_BLK_DEV_IDEPCI is defined,
> giving a compilation warning otherwise. Against 2.5.54-bk. 

Please reject. This is uglier than the warning and not the right approach

Re: PATCH: fix "ide_scan_direction defined but not used" in ide.c

[Alan Cox]

On Tue, 2003-01-07 at 16:01, Muli Ben-Yehuda wrote:
> Alan, I bow to your superior knowledge and judgement. However,
> 'ide_scan_direction' is only used in two places in ide.c, and both of
> those are only compiled in if CONFIG_BLK_DEV_IDEPCI:  

Move the ide_scan_direction variable into the setup-pci.c code I suspect
is the best way.