--- zzzz-none-000/linux-2.6.28.10/arch/mips/include/asm/delay.h	2009-05-02 18:54:43.000000000 +0000
+++ puma5-6360-529/linux-2.6.28.10/arch/mips/include/asm/delay.h	2011-12-06 10:11:23.000000000 +0000
@@ -17,83 +17,26 @@
 #include <asm/compiler.h>
 #include <asm/war.h>
 
-static inline void __delay(unsigned long loops)
+
+static inline void __delay(unsigned int loops)
 {
-	if (sizeof(long) == 4)
-		__asm__ __volatile__ (
-		"	.set	noreorder				\n"
-		"	.align	3					\n"
-		"1:	bnez	%0, 1b					\n"
-		"	subu	%0, 1					\n"
-		"	.set	reorder					\n"
-		: "=r" (loops)
-		: "0" (loops));
-	else if (sizeof(long) == 8 && !DADDI_WAR)
-		__asm__ __volatile__ (
-		"	.set	noreorder				\n"
-		"	.align	3					\n"
-		"1:	bnez	%0, 1b					\n"
-		"	dsubu	%0, 1					\n"
-		"	.set	reorder					\n"
-		: "=r" (loops)
-		: "0" (loops));
-	else if (sizeof(long) == 8 && DADDI_WAR)
-		__asm__ __volatile__ (
-		"	.set	noreorder				\n"
-		"	.align	3					\n"
-		"1:	bnez	%0, 1b					\n"
-		"	dsubu	%0, %2					\n"
-		"	.set	reorder					\n"
-		: "=r" (loops)
-		: "0" (loops), "r" (1));
+	__asm__ __volatile__ (
+	"	.set	noreorder				\n"
+	"	.align	3					\n"
+	"1:	bnez	%0, 1b					\n"
+	"	subu	%0, 1					\n"
+	"	.set	reorder					\n"
+	: "=r" (loops)
+	: "0" (loops));
 }
 
 
-/*
- * Division by multiplication: you don't have to worry about
- * loss of precision.
- *
- * Use only for very small delays ( < 1 msec).  Should probably use a
- * lookup table, really, as the multiplications take much too long with
- * short delays.  This is a "reasonable" implementation, though (and the
- * first constant multiplications gets optimized away if the delay is
- * a constant)
- */
-
-static inline void __udelay(unsigned long usecs, unsigned long lpj)
-{
-	unsigned long hi, lo;
-
-	/*
-	 * The rates of 128 is rounded wrongly by the catchall case
-	 * for 64-bit.  Excessive precission?  Probably ...
-	 */
-#if defined(CONFIG_64BIT) && (HZ == 128)
-	usecs *= 0x0008637bd05af6c7UL;		/* 2**64 / (1000000 / HZ) */
-#elif defined(CONFIG_64BIT)
-	usecs *= (0x8000000000000000UL / (500000 / HZ));
-#else /* 32-bit junk follows here */
-	usecs *= (unsigned long) (((0x8000000000000000ULL / (500000 / HZ)) +
-	                           0x80000000ULL) >> 32);
-#endif
-
-	if (sizeof(long) == 4)
-		__asm__("multu\t%2, %3"
-		: "=h" (usecs), "=l" (lo)
-		: "r" (usecs), "r" (lpj)
-		: GCC_REG_ACCUM);
-	else if (sizeof(long) == 8 && !R4000_WAR)
-		__asm__("dmultu\t%2, %3"
-		: "=h" (usecs), "=l" (lo)
-		: "r" (usecs), "r" (lpj)
-		: GCC_REG_ACCUM);
-	else if (sizeof(long) == 8 && R4000_WAR)
-		__asm__("dmultu\t%3, %4\n\tmfhi\t%0"
-		: "=r" (usecs), "=h" (hi), "=l" (lo)
-		: "r" (usecs), "r" (lpj)
-		: GCC_REG_ACCUM);
+static inline void __udelay(unsigned long us, unsigned long lpj) {
+	__delay((us * 0x000010c7ull * HZ * lpj) >> 32);
+}
 
-	__delay(usecs);
+static inline void __ndelay(unsigned long ns, unsigned long lpj) {
+	__delay((ns * 0x00000005ull * HZ * lpj) >> 32);
 }
 
 #define __udelay_val cpu_data[raw_smp_processor_id()].udelay_val