Files
linux/arch/powerpc/include/asm
Christophe Leroy 3af722cb73 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 <csum_block_add>:
	     2f8:	70 a5 00 01 	andi.   r5,r5,1
	     2fc:	41 a2 00 08 	beq     304 <csum_block_add+0xc>
	     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 <csum_block_add>:
	     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 <csum_block_add>:
	     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 <christophe.leroy@csgroup.eu>
Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-11 10:57:22 +00:00
..
2021-09-08 15:32:35 -07:00
2021-08-10 23:14:55 +10:00
2020-12-05 21:49:52 +11:00
2021-03-24 14:09:29 +11:00
2021-12-09 22:41:19 +11:00
2021-03-24 14:09:30 +11:00
2021-07-01 11:06:03 -07:00
2021-02-09 01:09:44 +11:00
2018-05-25 12:04:43 +10:00
2021-12-09 22:41:21 +11:00
2021-07-08 11:48:23 -07:00
2018-07-30 22:48:21 +10:00
2019-05-01 16:45:05 +10:00
2021-08-23 19:28:20 +10:00
2020-12-04 01:01:19 +11:00