Attachment 44386 Details for Bug 86504 – bad vectorizatoin result for boundary size 16

bad vectorizatoin result for boundary size 16

test_loop_inner_16.s (text/plain), 1.85 KB, created by Jiangning Liu on 2018-07-12 07:35:15 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Jiangning Liu

Created: 2018-07-12 07:35:15 UTC

Size: 1.85 KB

patch

obsolete

>	.arch armv8-a
>	.file	"test_loop.cpp"
>	.text
>	.align	2
>	.p2align 3,,7
>	.global	_Z9test_loopv
>	.type	_Z9test_loopv, %function
>_Z9test_loopv:
>.LFB0:
>	.cfi_startproc
>	adrp	x1, .LANCHOR0
>	add	x3, x1, :lo12:.LANCHOR0
>	add	x0, x3, 2048
>	ldrb	w1, [x1, #:lo12:.LANCHOR0]
>	add	x2, x0, 2048
>	ldrb	w7, [x3, 1024]
>	ldrb	w6, [x3, 1025]
>	ldrb	w5, [x3, 1]
>	add	w1, w1, w7
>	fmov	s0, w1
>	ldrb	w7, [x3, 1026]
>	add	w5, w5, w6
>	ldrb	w1, [x3, 2]
>	ldrb	w6, [x3, 1027]
>	ins	v0.b[1], w5
>	add	w1, w1, w7
>	ldrb	w7, [x3, 3]
>	ldrb	w5, [x3, 1028]
>	add	w7, w7, w6
>	ldrb	w6, [x3, 4]
>	ins	v0.b[2], w1
>	ldrb	w8, [x3, 1029]
>	add	w6, w6, w5
>	ldrb	w5, [x3, 5]
>	ldrb	w9, [x3, 1030]
>	add	w5, w5, w8
>	ldrb	w1, [x3, 6]
>	ins	v0.b[3], w7
>	ldrb	w8, [x3, 1031]
>	add	w1, w1, w9
>	ldrb	w11, [x3, 7]
>	ldrb	w7, [x3, 1032]
>	add	w11, w11, w8
>	ldrb	w10, [x3, 8]
>	ins	v0.b[4], w6
>	ldrb	w8, [x3, 1033]
>	add	w10, w10, w7
>	ldrb	w9, [x3, 9]
>	ldrb	w7, [x3, 1034]
>	add	w9, w9, w8
>	ldrb	w8, [x3, 10]
>	ins	v0.b[5], w5
>	ldrb	w6, [x3, 1035]
>	add	w8, w8, w7
>	ldrb	w7, [x3, 11]
>	ldrb	w5, [x3, 1036]
>	add	w7, w7, w6
>	ldrb	w6, [x3, 12]
>	ins	v0.b[6], w1
>	ldrb	w12, [x3, 1037]
>	add	w6, w6, w5
>	ldrb	w5, [x3, 13]
>	ldrb	w1, [x3, 14]
>	add	w5, w5, w12
>	ldrb	w13, [x3, 1038]
>	ins	v0.b[7], w11
>	ldrb	w12, [x3, 1039]
>	add	w4, w1, w13
>	ldrb	w1, [x3, 15]
>	add	w1, w1, w12
>	ins	v0.b[8], w10
>	ins	v0.b[9], w9
>	ins	v0.b[10], w8
>	ins	v0.b[11], w7
>	ins	v0.b[12], w6
>	ins	v0.b[13], w5
>	ins	v0.b[14], w4
>	ins	v0.b[15], w1
>	.p2align 3,,7
>.L2:
>	str	q0, [x0], 16
>	cmp	x2, x0
>	bne	.L2
>	ret
>	.cfi_endproc
>.LFE0:
>	.size	_Z9test_loopv, .-_Z9test_loopv
>	.global	g_s2
>	.global	g_s1
>	.global	g_d
>	.bss
>	.align	4
>	.set	.LANCHOR0,. + 0
>	.type	g_s1, %object
>	.size	g_s1, 1024
>g_s1:
>	.zero	1024
>	.type	g_s2, %object
>	.size	g_s2, 1024
>g_s2:
>	.zero	1024
>	.type	g_d, %object
>	.size	g_d, 1024
>g_d:
>	.zero	1024
>	.ident	"GCC: (GNU) 9.0.0 20180710 (experimental)"
>	.section	.note.GNU-stack,"",@progbits

Actions: View

Attachments on bug 86504: 44386 | 44387