[PATCH 1/2] bitops: use find_first_zero_bit() instead of find_next_zero_bit(addr, size, 0)

[Akinobu Mita]

The little-endian bitops patch series introduced inefficiency to some
little-endian architectures.

include/asm-generic/bitops/le.h defines find_first_zero_le_bit() like this:

	#define find_first_zero_le_bit(addr, size) \
		find_next_zero_le_bit((addr), (size), 0)

All little-endian architectures include le.h and little-endian bitops
are just aliases for their native endian bitops. So above macro definision
is converted to:

	#define find_first_zero_le_bit(addr, size) \
		find_next_zero_bit(addr, size, 0)

This is inefficient for some architectures which have faster
find_first_zero_bit() than find_next_zero_bit(addr, size, 0).

Fix it by aliasing find_first_zero_bit() to find_first_zero_le_bit()
on little-endian architectures.

Signed-off-by: Akinobu Mita <akinobu.mita@xxxxxxxxx>
Cc: Arnd Bergmann <arnd@xxxxxxxx>
---
 include/asm-generic/bitops/le.h |    8 +++++---
 1 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/include/asm-generic/bitops/le.h b/include/asm-generic/bitops/le.h
index 6ad46ce..17a7119 100644
--- a/include/asm-generic/bitops/le.h
+++ b/include/asm-generic/bitops/le.h
@@ -12,6 +12,8 @@
 	find_next_zero_bit(addr, size, offset)
 #define find_next_le_bit(addr, size, offset) \
 	find_next_bit(addr, size, offset)
+#define find_first_zero_le_bit(addr, size) \
+	find_first_zero_bit(addr, size)
 
 #elif defined(__BIG_ENDIAN)
 
@@ -22,6 +24,9 @@ extern unsigned long find_next_zero_le_bit(const unsigned long *addr,
 extern unsigned long find_next_le_bit(const unsigned long *addr,
 		unsigned long size, unsigned long offset);
 
+#define find_first_zero_le_bit(addr, size) \
+	find_next_zero_le_bit((addr), (size), 0)
+
 #else
 #error "Please fix <asm/byteorder.h>"
 #endif
@@ -43,7 +48,4 @@ extern unsigned long find_next_le_bit(const unsigned long *addr,
 #define __test_and_clear_le_bit(nr, addr) \
 	__test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
 
-#define find_first_zero_le_bit(addr, size) \
-	find_next_zero_le_bit((addr), (size), 0)
-
 #endif /* _ASM_GENERIC_BITOPS_LE_H_ */
-- 
1.7.3.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/