From 3af722cb735d212554027ec81e2aa2e6bf1ee34d Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Tue, 8 Mar 2022 17:12:10 +0100 Subject: powerpc/net: Implement powerpc specific csum_shift() to remove branch Today's implementation of csum_shift() leads to branching based on parity of 'offset' 000002f8 : 2f8: 70 a5 00 01 andi. r5,r5,1 2fc: 41 a2 00 08 beq 304 300: 54 84 c0 3e rotlwi r4,r4,24 304: 7c 63 20 14 addc r3,r3,r4 308: 7c 63 01 94 addze r3,r3 30c: 4e 80 00 20 blr Use first bit of 'offset' directly as input of the rotation instead of branching. 000002f8 : 2f8: 54 a5 1f 38 rlwinm r5,r5,3,28,28 2fc: 20 a5 00 20 subfic r5,r5,32 300: 5c 84 28 3e rotlw r4,r4,r5 304: 7c 63 20 14 addc r3,r3,r4 308: 7c 63 01 94 addze r3,r3 30c: 4e 80 00 20 blr And change to left shift instead of right shift to skip one more instruction. This has no impact on the final sum. 000002f8 : 2f8: 54 a5 1f 38 rlwinm r5,r5,3,28,28 2fc: 5c 84 28 3e rotlw r4,r4,r5 300: 7c 63 20 14 addc r3,r3,r4 304: 7c 63 01 94 addze r3,r3 308: 4e 80 00 20 blr Seems like only powerpc benefits from a branchless implementation. Other main architectures like ARM or X86 get better code with the generic implementation and its branch. Signed-off-by: Christophe Leroy Signed-off-by: David S. Miller --- include/net/checksum.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/checksum.h b/include/net/checksum.h index 79c67f14c448..6bc783b7a06c 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -80,6 +80,7 @@ static __always_inline __sum16 csum16_sub(__sum16 csum, __be16 addend) return csum16_add(csum, ~addend); } +#ifndef HAVE_ARCH_CSUM_SHIFT static __always_inline __wsum csum_shift(__wsum sum, int offset) { /* rotate sum to align it with a 16b boundary */ @@ -87,6 +88,7 @@ static __always_inline __wsum csum_shift(__wsum sum, int offset) return (__force __wsum)ror32((__force u32)sum, 8); return sum; } +#endif static __always_inline __wsum csum_block_add(__wsum csum, __wsum csum2, int offset) -- cgit v1.2.3