From 102fa15c3f14565f2edb9f08f08ea3f2bf123dc9 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 16 Feb 2007 17:18:50 +0000
Subject: [PATCH] [MIPS] Use MIPS R2 instructions for bitops.

Add R2 optimized variants of clear_bit, set_bit and test_and_clear_bit.
With gcc 4.1.1 this saves 1592 bytes on a defconfig (minus IPv6) kernel.
Turns out that R2 bitop instructions are no gain for the other bitop
functions.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/bitops.h | 52 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 51 insertions(+), 1 deletion(-)

diff --git a/include/asm-mips/bitops.h b/include/asm-mips/bitops.h
index 06c08228a52..89436b96ad6 100644
--- a/include/asm-mips/bitops.h
+++ b/include/asm-mips/bitops.h
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (c) 1994 - 1997, 1999, 2000, 06  Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (c) 1994 - 1997, 99, 2000, 06, 07  Ralf Baechle (ralf@linux-mips.org)
  * Copyright (c) 1999, 2000  Silicon Graphics, Inc.
  */
 #ifndef _ASM_BITOPS_H
@@ -24,11 +24,15 @@
 #define SZLONG_MASK 31UL
 #define __LL		"ll	"
 #define __SC		"sc	"
+#define __INS		"ins    "
+#define __EXT		"ext    "
 #elif (_MIPS_SZLONG == 64)
 #define SZLONG_LOG 6
 #define SZLONG_MASK 63UL
 #define __LL		"lld	"
 #define __SC		"scd	"
+#define __INS		"dins    "
+#define __EXT		"dext    "
 #endif
 
 /*
@@ -62,6 +66,19 @@ static inline void set_bit(unsigned long nr, volatile unsigned long *addr)
 		"	.set	mips0					\n"
 		: "=&r" (temp), "=m" (*m)
 		: "ir" (1UL << (nr & SZLONG_MASK)), "m" (*m));
+#ifdef CONFIG_CPU_MIPSR2
+	} else if (__builtin_constant_p(nr)) {
+		__asm__ __volatile__(
+		"1:	" __LL "%0, %1			# set_bit	\n"
+		"	" __INS "%0, %4, %2, 1				\n"
+		"	" __SC "%0, %1					\n"
+		"	beqz	%0, 2f					\n"
+		"	.subsection 2					\n"
+		"2:	b	1b					\n"
+		"	.previous					\n"
+		: "=&r" (temp), "=m" (*m)
+		: "ir" (nr & SZLONG_MASK), "m" (*m), "r" (~0));
+#endif /* CONFIG_CPU_MIPSR2 */
 	} else if (cpu_has_llsc) {
 		__asm__ __volatile__(
 		"	.set	mips3					\n"
@@ -113,6 +130,19 @@ static inline void clear_bit(unsigned long nr, volatile unsigned long *addr)
 		"	.set	mips0					\n"
 		: "=&r" (temp), "=m" (*m)
 		: "ir" (~(1UL << (nr & SZLONG_MASK))), "m" (*m));
+#ifdef CONFIG_CPU_MIPSR2
+	} else if (__builtin_constant_p(nr)) {
+		__asm__ __volatile__(
+		"1:	" __LL "%0, %1			# clear_bit	\n"
+		"	" __INS "%0, $0, %2, 1				\n"
+		"	" __SC "%0, %1					\n"
+		"	beqz	%0, 2f					\n"
+		"	.subsection 2					\n"
+		"2:	b	1b					\n"
+		"	.previous					\n"
+		: "=&r" (temp), "=m" (*m)
+		: "ir" (nr & SZLONG_MASK), "m" (*m));
+#endif /* CONFIG_CPU_MIPSR2 */
 	} else if (cpu_has_llsc) {
 		__asm__ __volatile__(
 		"	.set	mips3					\n"
@@ -291,6 +321,26 @@ static inline int test_and_clear_bit(unsigned long nr,
 		: "memory");
 
 		return res != 0;
+#ifdef CONFIG_CPU_MIPSR2
+	} else if (__builtin_constant_p(nr)) {
+		unsigned long *m = ((unsigned long *) addr) + (nr >> SZLONG_LOG);
+		unsigned long temp, res;
+
+		__asm__ __volatile__(
+		"1:	" __LL	"%0, %1		# test_and_clear_bit	\n"
+		"	" __EXT "%2, %0, %3, 1				\n"
+		"	" __INS	"%0, $0, %3, 1				\n"
+		"	" __SC 	"%0, %1					\n"
+		"	beqz	%0, 2f					\n"
+		"	.subsection 2					\n"
+		"2:	b	1b					\n"
+		"	.previous					\n"
+		: "=&r" (temp), "=m" (*m), "=&r" (res)
+		: "ri" (nr & SZLONG_MASK), "m" (*m)
+		: "memory");
+
+		return res;
+#endif
 	} else if (cpu_has_llsc) {
 		unsigned long *m = ((unsigned long *) addr) + (nr >> SZLONG_LOG);
 		unsigned long temp, res;