delay: Rework udelay and ndelay

udelay() as well as ndelay() are defines and no functions and are using
constants to be able to transform a sleep time into loops and to prevent
too long udelays/ndelays. There was a compiler error with non-const 8 bit
arguments which was fixed by commit a87e553 ("asm-generic: delay.h fix
udelay and ndelay for 8 bit args"). When using a function, the non-const 8
bit argument is type casted and the problem would be gone.

Transform udelay() and ndelay() into proper functions, remove the no longer
and confusing division, add defines for the magic values and add some
explanations as well.

Suggested-by: Thomas Gleixner <[email protected]>
Signed-off-by: Anna-Maria Behnsen <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Reviewed-by: Frederic Weisbecker <[email protected]>
Link: https://lore.kernel.org/all/20241014-devel-anna-maria-b4-timers-flseep-v3-6-dc8b907cb62f@linutronix.de

Author: anna-marialx

include/asm-generic/delay.h

@@ -2,6 +2,9 @@
 #ifndef __ASM_GENERIC_DELAY_H
 #define __ASM_GENERIC_DELAY_H
 
+#include <linux/math.h>
+#include <vdso/time64.h>
+
 /* Undefined functions to get compile-time errors */
 extern void __bad_udelay(void);
 extern void __bad_ndelay(void);
@@ -12,13 +15,18 @@ extern void __const_udelay(unsigned long xloops);
 extern void __delay(unsigned long loops);
 
 /*
- * Implementation details:
- *
- * * The weird n/20000 thing suppresses a "comparison is always false due to
- *   limited range of data type" warning with non-const 8-bit arguments.
- * * 0x10c7 is 2**32 / 1000000 (rounded up) -> udelay
- * * 0x5 is 2**32 / 1000000000 (rounded up) -> ndelay
+ * The microseconds/nanosecond delay multiplicators are used to convert a
+ * constant microseconds/nanoseconds value to a value which can be used by the
+ * architectures specific implementation to transform it into loops.
+ */
+#define UDELAY_CONST_MULT	((unsigned long)DIV_ROUND_UP(1ULL << 32, USEC_PER_SEC))
+#define NDELAY_CONST_MULT	((unsigned long)DIV_ROUND_UP(1ULL << 32, NSEC_PER_SEC))
+
+/*
+ * The maximum constant udelay/ndelay value picked out of thin air to prevent
+ * too long constant udelays/ndelays.
  */
+#define DELAY_CONST_MAX   20000
 
 /**
  * udelay - Inserting a delay based on microseconds with busy waiting
@@ -45,34 +53,35 @@ extern void __delay(unsigned long loops);
  * #. cache behaviour affecting the time it takes to execute the loop function.
  * #. CPU clock rate changes.
  */
-#define udelay(n)							\
-	({								\
-		if (__builtin_constant_p(n)) {				\
-			if ((n) / 20000 >= 1)				\
-				 __bad_udelay();			\
-			else						\
-				__const_udelay((n) * 0x10c7ul);		\
-		} else {						\
-			__udelay(n);					\
-		}							\
-	})
+static __always_inline void udelay(unsigned long usec)
+{
+	if (__builtin_constant_p(usec)) {
+		if (usec >= DELAY_CONST_MAX)
+			__bad_udelay();
+		else
+			__const_udelay(usec * UDELAY_CONST_MULT);
+	} else {
+		__udelay(usec);
+	}
+}
 
 /**
  * ndelay - Inserting a delay based on nanoseconds with busy waiting
  * @nsec:	requested delay in nanoseconds
  *
  * See udelay() for basic information about ndelay() and it's variants.
  */
-#define ndelay(n)							\
-	({								\
-		if (__builtin_constant_p(n)) {				\
-			if ((n) / 20000 >= 1)				\
-				__bad_ndelay();				\
-			else						\
-				__const_udelay((n) * 5ul);		\
-		} else {						\
-			__ndelay(n);					\
-		}							\
-	})
+static __always_inline void ndelay(unsigned long nsec)
+{
+	if (__builtin_constant_p(nsec)) {
+		if (nsec >= DELAY_CONST_MAX)
+			__bad_udelay();
+		else
+			__const_udelay(nsec * NDELAY_CONST_MULT);
+	} else {
+		__udelay(nsec);
+	}
+}
+#define ndelay(x) ndelay(x)
 
 #endif /* __ASM_GENERIC_DELAY_H */