[PATCH] 2.6.14 include/asm-arm/arch-ixp4xx/io.h: eliminate warning for pointer passed to integral function argument

All of lore.kernel.org
 help / color / mirror / Atom feed

* [PATCH] 2.6.14 include/asm-arm/arch-ixp4xx/io.h: eliminate warning for pointer passed to integral function argument
@ 2005-10-28  6:19 John Bowler
  2005-10-28  7:57 ` Russell King
  0 siblings, 1 reply; 3+ messages in thread
From: John Bowler @ 2005-10-28  6:19 UTC (permalink / raw)
  To: 'Deepak Saxena'; +Cc: linux-kernel, linux-arm-kernel

Fix for a compiler warning, this wasn't apparent in 2.6.12, I
believe the compiler options have been changed (somewhere) so
that passing a (void*) to a (u32) argument is now warned.

This accounts for the majority of the warnings in my builds of
the 2.6.14 kernel for NSLU2.

This is the trivial fix, but may be controversial in that it
simply casts away the warning - the alternative is to change
all the static inlines and potentially some of the callers
thereof.

Signed-off-by: John Bowler <jbowler@acm.org>

--- linux-2.6.14-rc5/include/asm-arm/arch-ixp4xx/io.h	2005-10-26 08:37:22.844417060 -0700
+++ patched/include/asm-arm/arch-ixp4xx/io.h	2005-10-26 12:14:20.771583571 -0700
@@ -80,21 +80,21 @@ __ixp4xx_iounmap(void __iomem *addr)
 #define __arch_ioremap(a, s, f, x)	__ixp4xx_ioremap(a, s, f, x)
 #define	__arch_iounmap(a)		__ixp4xx_iounmap(a)
 
-#define	writeb(p, v)			__ixp4xx_writeb(p, v)
-#define	writew(p, v)			__ixp4xx_writew(p, v)
-#define	writel(p, v)			__ixp4xx_writel(p, v)
-
-#define	writesb(p, v, l)		__ixp4xx_writesb(p, v, l)
-#define	writesw(p, v, l)		__ixp4xx_writesw(p, v, l)
-#define	writesl(p, v, l)		__ixp4xx_writesl(p, v, l)
+#define	writeb(v, p)			__ixp4xx_writeb(v, (u32)p)
+#define	writew(v, p)			__ixp4xx_writew(v, (u32)p)
+#define	writel(v, p)			__ixp4xx_writel(v, (u32)p)
+
+#define	writesb(p, v, l)		__ixp4xx_writesb((u32)p, v, l)
+#define	writesw(p, v, l)		__ixp4xx_writesw((u32)p, v, l)
+#define	writesl(p, v, l)		__ixp4xx_writesl((u32)p, v, l)
 	
-#define	readb(p)			__ixp4xx_readb(p)
-#define	readw(p)			__ixp4xx_readw(p)
-#define	readl(p)			__ixp4xx_readl(p)
+#define	readb(p)			__ixp4xx_readb((u32)p)
+#define	readw(p)			__ixp4xx_readw((u32)p)
+#define	readl(p)			__ixp4xx_readl((u32)p)
 	
-#define	readsb(p, v, l)			__ixp4xx_readsb(p, v, l)
-#define	readsw(p, v, l)			__ixp4xx_readsw(p, v, l)
-#define	readsl(p, v, l)			__ixp4xx_readsl(p, v, l)
+#define	readsb(p, v, l)			__ixp4xx_readsb((u32)p, v, l)
+#define	readsw(p, v, l)			__ixp4xx_readsw((u32)p, v, l)
+#define	readsl(p, v, l)			__ixp4xx_readsl((u32)p, v, l)
 
 static inline void 
 __ixp4xx_writeb(u8 value, u32 addr)


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

* Re: [PATCH] 2.6.14 include/asm-arm/arch-ixp4xx/io.h: eliminate warning for pointer passed to integral function argument
  2005-10-28  6:19 [PATCH] 2.6.14 include/asm-arm/arch-ixp4xx/io.h: eliminate warning for pointer passed to integral function argument John Bowler
@ 2005-10-28  7:57 ` Russell King
  2005-10-31  1:49   ` John Bowler
  0 siblings, 1 reply; 3+ messages in thread
From: Russell King @ 2005-10-28  7:57 UTC (permalink / raw)
  To: John Bowler; +Cc: 'Deepak Saxena', linux-kernel

On Thu, Oct 27, 2005 at 11:19:14PM -0700, John Bowler wrote:
> Fix for a compiler warning, this wasn't apparent in 2.6.12, I
> believe the compiler options have been changed (somewhere) so
> that passing a (void*) to a (u32) argument is now warned.

readb/writeb and friends take a "void __iomem *".  You should
arrange the __ixp4xx versions to take that.

> This accounts for the majority of the warnings in my builds of
> the 2.6.14 kernel for NSLU2.

If you want to see why, grab sparse and build your kernel with C=1.

PS, please do not copy both linux-kernel and linux-arm-kernel.  It's
bad taste to send a message to an open mailing list and a members only
list.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core

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

* RE: [PATCH] 2.6.14 include/asm-arm/arch-ixp4xx/io.h: eliminate warning for pointer passed to integral function argument
  2005-10-28  7:57 ` Russell King
@ 2005-10-31  1:49   ` John Bowler
  0 siblings, 0 replies; 3+ messages in thread
From: John Bowler @ 2005-10-31  1:49 UTC (permalink / raw)
  To: 'Russell King'; +Cc: 'Deepak Saxena', linux-kernel

From: Russell King
>On Thu, Oct 27, 2005 at 11:19:14PM -0700, John Bowler wrote:
>> Fix for a compiler warning, this wasn't apparent in 2.6.12, I
>> believe the compiler options have been changed (somewhere) so
>> that passing a (void*) to a (u32) argument is now warned.
>
>readb/writeb and friends take a "void __iomem *".  You should
>arrange the __ixp4xx versions to take that.

After some delay I've done that.  One use of these APIs actually
passes a (volatile void*) and this generates a warning with the
change.  In general I changed the 'read' p (pointer) parameter
to be:

const volatile void __iomem *

and the write parameter to be:

volatile void __iomem *

I've appended the full patch, however I will submit this via the
ARM patch system after an appropriate time for commments...

I did have to change 'check_signature' in the header file and I
changed it to (const unsigned char __iomem *) (it increments the
address).  Other instances of this function have a variety of
definitions, but the callers all seem to use a base type of void.

I note that the callers are happy adding offsets to a (void*) (this
never worked when I was young) - advice?  (I.e. can I rely on adding
numbers to a (void*), in which case I should probably change this?)

John Bowler <jbowler@acm.org>

--- linux-2.6.14/include/asm-arm/arch-ixp4xx/io.h	2005-10-29 23:33:21.757679882 -0700
+++ linux-2.6.14/include/asm-arm/arch-ixp4xx/io.h	2005-10-29 23:47:02.581331058 -0700
@@ -80,9 +80,9 @@ __ixp4xx_iounmap(void __iomem *addr)
 #define __arch_ioremap(a, s, f, x)	__ixp4xx_ioremap(a, s, f, x)
 #define	__arch_iounmap(a)		__ixp4xx_iounmap(a)
 
-#define	writeb(p, v)			__ixp4xx_writeb(p, v)
-#define	writew(p, v)			__ixp4xx_writew(p, v)
-#define	writel(p, v)			__ixp4xx_writel(p, v)
+#define	writeb(v, p)			__ixp4xx_writeb(v, p)
+#define	writew(v, p)			__ixp4xx_writew(v, p)
+#define	writel(v, p)			__ixp4xx_writel(v, p)
 
 #define	writesb(p, v, l)		__ixp4xx_writesb(p, v, l)
 #define	writesw(p, v, l)		__ixp4xx_writesw(p, v, l)
@@ -97,8 +97,9 @@ __ixp4xx_iounmap(void __iomem *addr)
 #define	readsl(p, v, l)			__ixp4xx_readsl(p, v, l)
 
 static inline void 
-__ixp4xx_writeb(u8 value, u32 addr)
+__ixp4xx_writeb(u8 value, volatile void __iomem *p)
 {
+	u32 addr = (u32)p;
 	u32 n, byte_enables, data;
 
 	if (addr >= VMALLOC_START) {
@@ -113,15 +114,16 @@ __ixp4xx_writeb(u8 value, u32 addr)
 }
 
 static inline void
-__ixp4xx_writesb(u32 bus_addr, const u8 *vaddr, int count)
+__ixp4xx_writesb(volatile void __iomem *bus_addr, const u8 *vaddr, int count)
 {
 	while (count--)
 		writeb(*vaddr++, bus_addr);
 }
 
 static inline void 
-__ixp4xx_writew(u16 value, u32 addr)
+__ixp4xx_writew(u16 value, volatile void __iomem *p)
 {
+	u32 addr = (u32)p;
 	u32 n, byte_enables, data;
 
 	if (addr >= VMALLOC_START) {
@@ -136,15 +138,16 @@ __ixp4xx_writew(u16 value, u32 addr)
 }
 
 static inline void
-__ixp4xx_writesw(u32 bus_addr, const u16 *vaddr, int count)
+__ixp4xx_writesw(volatile void __iomem *bus_addr, const u16 *vaddr, int count)
 {
 	while (count--)
 		writew(*vaddr++, bus_addr);
 }
 
 static inline void 
-__ixp4xx_writel(u32 value, u32 addr)
+__ixp4xx_writel(u32 value, volatile void __iomem *p)
 {
+	u32 addr = (u32)p;
 	if (addr >= VMALLOC_START) {
 		__raw_writel(value, addr);
 		return;
@@ -154,15 +157,16 @@ __ixp4xx_writel(u32 value, u32 addr)
 }
 
 static inline void
-__ixp4xx_writesl(u32 bus_addr, const u32 *vaddr, int count)
+__ixp4xx_writesl(volatile void __iomem *bus_addr, const u32 *vaddr, int count)
 {
 	while (count--)
 		writel(*vaddr++, bus_addr);
 }
 
 static inline unsigned char 
-__ixp4xx_readb(u32 addr)
+__ixp4xx_readb(const volatile void __iomem *p)
 {
+	u32 addr = (u32)p;
 	u32 n, byte_enables, data;
 
 	if (addr >= VMALLOC_START)
@@ -177,15 +181,16 @@ __ixp4xx_readb(u32 addr)
 }
 
 static inline void
-__ixp4xx_readsb(u32 bus_addr, u8 *vaddr, u32 count)
+__ixp4xx_readsb(const volatile void __iomem *bus_addr, u8 *vaddr, u32 count)
 {
 	while (count--)
 		*vaddr++ = readb(bus_addr);
 }
 
 static inline unsigned short 
-__ixp4xx_readw(u32 addr)
+__ixp4xx_readw(const volatile void __iomem *p)
 {
+	u32 addr = (u32)p;
 	u32 n, byte_enables, data;
 
 	if (addr >= VMALLOC_START)
@@ -200,15 +205,16 @@ __ixp4xx_readw(u32 addr)
 }
 
 static inline void 
-__ixp4xx_readsw(u32 bus_addr, u16 *vaddr, u32 count)
+__ixp4xx_readsw(const volatile void __iomem *bus_addr, u16 *vaddr, u32 count)
 {
 	while (count--)
 		*vaddr++ = readw(bus_addr);
 }
 
 static inline unsigned long 
-__ixp4xx_readl(u32 addr)
+__ixp4xx_readl(const volatile void __iomem *p)
 {
+	u32 addr = (u32)p;
 	u32 data;
 
 	if (addr >= VMALLOC_START)
@@ -221,7 +227,7 @@ __ixp4xx_readl(u32 addr)
 }
 
 static inline void 
-__ixp4xx_readsl(u32 bus_addr, u32 *vaddr, u32 count)
+__ixp4xx_readsl(const volatile void __iomem *bus_addr, u32 *vaddr, u32 count)
 {
 	while (count--)
 		*vaddr++ = readl(bus_addr);
@@ -239,7 +245,7 @@ __ixp4xx_readsl(u32 bus_addr, u32 *vaddr
 				eth_copy_and_sum((s),__mem_pci(c),(l),(b))
 
 static inline int
-check_signature(unsigned long bus_addr, const unsigned char *signature,
+check_signature(const unsigned char __iomem *bus_addr, const unsigned char *signature,
 		int length)
 {
 	int retval = 0;
@@ -389,7 +395,7 @@ __ixp4xx_insl(u32 io_addr, u32 *vaddr, u
 #define	__is_io_address(p)	(((unsigned long)p >= PIO_OFFSET) && \
 					((unsigned long)p <= (PIO_MASK + PIO_OFFSET)))
 static inline unsigned int
-__ixp4xx_ioread8(void __iomem *addr)
+__ixp4xx_ioread8(const void __iomem *addr)
 {
 	unsigned long port = (unsigned long __force)addr;
 	if (__is_io_address(port))
@@ -398,12 +404,12 @@ __ixp4xx_ioread8(void __iomem *addr)
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		return (unsigned int)__raw_readb(port);
 #else
-		return (unsigned int)__ixp4xx_readb(port);
+		return (unsigned int)__ixp4xx_readb(addr);
 #endif
 }
 
 static inline void
-__ixp4xx_ioread8_rep(void __iomem *addr, void *vaddr, u32 count)
+__ixp4xx_ioread8_rep(const void __iomem *addr, void *vaddr, u32 count)
 {
 	unsigned long port = (unsigned long __force)addr;
 	if (__is_io_address(port))
@@ -412,12 +418,12 @@ __ixp4xx_ioread8_rep(void __iomem *addr,
 #ifndef	CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_readsb(addr, vaddr, count);
 #else
-		__ixp4xx_readsb(port, vaddr, count);
+		__ixp4xx_readsb(addr, vaddr, count);
 #endif
 }
 
 static inline unsigned int
-__ixp4xx_ioread16(void __iomem *addr)
+__ixp4xx_ioread16(const void __iomem *addr)
 {
 	unsigned long port = (unsigned long __force)addr;
 	if (__is_io_address(port))
@@ -426,12 +432,12 @@ __ixp4xx_ioread16(void __iomem *addr)
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		return le16_to_cpu(__raw_readw((u32)port));
 #else
-		return (unsigned int)__ixp4xx_readw((u32)port);
+		return (unsigned int)__ixp4xx_readw(addr);
 #endif
 }
 
 static inline void
-__ixp4xx_ioread16_rep(void __iomem *addr, void *vaddr, u32 count)
+__ixp4xx_ioread16_rep(const void __iomem *addr, void *vaddr, u32 count)
 {
 	unsigned long port = (unsigned long __force)addr;
 	if (__is_io_address(port))
@@ -440,12 +446,12 @@ __ixp4xx_ioread16_rep(void __iomem *addr
 #ifndef	CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_readsw(addr, vaddr, count);
 #else
-		__ixp4xx_readsw(port, vaddr, count);
+		__ixp4xx_readsw(addr, vaddr, count);
 #endif
 }
 
 static inline unsigned int
-__ixp4xx_ioread32(void __iomem *addr)
+__ixp4xx_ioread32(const void __iomem *addr)
 {
 	unsigned long port = (unsigned long __force)addr;
 	if (__is_io_address(port))
@@ -454,13 +460,13 @@ __ixp4xx_ioread32(void __iomem *addr)
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		return le32_to_cpu(__raw_readl((u32)port));
 #else
-		return (unsigned int)__ixp4xx_readl((u32)port);
+		return (unsigned int)__ixp4xx_readl(addr);
 #endif
 	}
 }
 
 static inline void
-__ixp4xx_ioread32_rep(void __iomem *addr, void *vaddr, u32 count)
+__ixp4xx_ioread32_rep(const void __iomem *addr, void *vaddr, u32 count)
 {
 	unsigned long port = (unsigned long __force)addr;
 	if (__is_io_address(port))
@@ -469,7 +475,7 @@ __ixp4xx_ioread32_rep(void __iomem *addr
 #ifndef	CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_readsl(addr, vaddr, count);
 #else
-		__ixp4xx_readsl(port, vaddr, count);
+		__ixp4xx_readsl(addr, vaddr, count);
 #endif
 }
 
@@ -483,7 +489,7 @@ __ixp4xx_iowrite8(u8 value, void __iomem
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_writeb(value, port);
 #else
-		__ixp4xx_writeb(value, port);
+		__ixp4xx_writeb(value, addr);
 #endif
 }
 
@@ -497,7 +503,7 @@ __ixp4xx_iowrite8_rep(void __iomem *addr
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_writesb(addr, vaddr, count);
 #else
-		__ixp4xx_writesb(port, vaddr, count);
+		__ixp4xx_writesb(addr, vaddr, count);
 #endif
 }
 
@@ -511,7 +517,7 @@ __ixp4xx_iowrite16(u16 value, void __iom
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_writew(cpu_to_le16(value), addr);
 #else
-		__ixp4xx_writew(value, port);
+		__ixp4xx_writew(value, addr);
 #endif
 }
 
@@ -525,7 +531,7 @@ __ixp4xx_iowrite16_rep(void __iomem *add
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_writesw(addr, vaddr, count);
 #else
-		__ixp4xx_writesw(port, vaddr, count);
+		__ixp4xx_writesw(addr, vaddr, count);
 #endif
 }
 
@@ -539,7 +545,7 @@ __ixp4xx_iowrite32(u32 value, void __iom
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_writel(cpu_to_le32(value), port);
 #else
-		__ixp4xx_writel(value, port);
+		__ixp4xx_writel(value, addr);
 #endif
 }
 
@@ -553,7 +559,7 @@ __ixp4xx_iowrite32_rep(void __iomem *add
 #ifndef CONFIG_IXP4XX_INDIRECT_PCI
 		__raw_writesl(addr, vaddr, count);
 #else
-		__ixp4xx_writesl(port, vaddr, count);
+		__ixp4xx_writesl(addr, vaddr, count);
 #endif
 }
 


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

end of thread, other threads:[~2005-10-31  1:49 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-10-28  6:19 [PATCH] 2.6.14 include/asm-arm/arch-ixp4xx/io.h: eliminate warning for pointer passed to integral function argument John Bowler
2005-10-28  7:57 ` Russell King
2005-10-31  1:49   ` John Bowler

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.