Commit bf4e4e36 authored by James Zern's avatar James Zern Committed by Gerrit Code Review
Browse files

Merge "Remove the vp8_vpxyv12_copy_y_neon."

parents edb0c26d 1f755894
; Copyright (c) 2010 The WebM project authors. All Rights Reserved.
; Use of this source code is governed by a BSD-style license
; that can be found in the LICENSE file in the root of the source
; tree. An additional intellectual property rights grant can be found
; in the file PATENTS. All contributing project authors may
; be found in the AUTHORS file in the root of the source tree.
EXPORT |vpx_yv12_copy_y_neon|
INCLUDE vpx_scale_asm_offsets.asm
;void vpx_yv12_copy_y_neon(const YV12_BUFFER_CONFIG *src_ybc,
; YV12_BUFFER_CONFIG *dst_ybc)
|vpx_yv12_copy_y_neon| PROC
push {r4 - r11, lr}
vpush {d8-d15}
ldr r4, [r0, #yv12_buffer_config_y_height]
ldr r5, [r0, #yv12_buffer_config_y_width]
ldr r6, [r0, #yv12_buffer_config_y_stride]
ldr r7, [r1, #yv12_buffer_config_y_stride]
ldr r2, [r0, #yv12_buffer_config_y_buffer] ;srcptr1
ldr r3, [r1, #yv12_buffer_config_y_buffer] ;dstptr1
; copy two rows at one time
mov lr, r4, lsr #1
mov r8, r2
mov r9, r3
add r10, r2, r6
add r11, r3, r7
movs r12, r5, lsr #7
ble extra_copy_needed ; y_width < 128
vld1.8 {q0, q1}, [r8]!
vld1.8 {q8, q9}, [r10]!
vld1.8 {q2, q3}, [r8]!
vld1.8 {q10, q11}, [r10]!
vld1.8 {q4, q5}, [r8]!
vld1.8 {q12, q13}, [r10]!
vld1.8 {q6, q7}, [r8]!
vld1.8 {q14, q15}, [r10]!
subs r12, r12, #1
vst1.8 {q0, q1}, [r9]!
vst1.8 {q8, q9}, [r11]!
vst1.8 {q2, q3}, [r9]!
vst1.8 {q10, q11}, [r11]!
vst1.8 {q4, q5}, [r9]!
vst1.8 {q12, q13}, [r11]!
vst1.8 {q6, q7}, [r9]!
vst1.8 {q14, q15}, [r11]!
bne cp_src_to_dst_width_loop1
subs lr, lr, #1
add r2, r2, r6, lsl #1
add r3, r3, r7, lsl #1
bne cp_src_to_dst_height_loop1
ands r10, r5, #0x7f ;check to see if extra copy is needed
sub r11, r5, r10
ldr r2, [r0, #yv12_buffer_config_y_buffer] ;srcptr1
ldr r3, [r1, #yv12_buffer_config_y_buffer] ;dstptr1
bne extra_cp_src_to_dst_width1
vpop {d8 - d15}
pop {r4-r11, pc}
add r2, r2, r11
add r3, r3, r11
add r0, r8, r6
add r11, r9, r7
mov lr, r4, lsr #1
mov r8, r2
mov r9, r3
add r0, r8, r6
add r11, r9, r7
mov r12, r10
vld1.8 {q0}, [r8]!
vld1.8 {q1}, [r0]!
subs r12, r12, #16
vst1.8 {q0}, [r9]!
vst1.8 {q1}, [r11]!
bne extra_cp_src_to_dst_width_loop1
subs lr, lr, #1
add r2, r2, r6, lsl #1
add r3, r3, r7, lsl #1
bne extra_cp_src_to_dst_height_loop1
b end_of_cp_src_to_dst1
......@@ -11,7 +11,6 @@ SCALE_SRCS-yes +=
SCALE_SRCS-$(HAVE_NEON) += arm/neon/vp8_vpxyv12_copyframe_func_neon$(ASM)
SCALE_SRCS-$(HAVE_NEON) += arm/neon/vp8_vpxyv12_copy_y_neon$(ASM)
SCALE_SRCS-$(HAVE_NEON) += arm/neon/vp8_vpxyv12_copysrcframe_func_neon$(ASM)
SCALE_SRCS-$(HAVE_NEON) += arm/neon/vp8_vpxyv12_extendframeborders_neon$(ASM)
SCALE_SRCS-$(HAVE_NEON) += arm/neon/yv12extend_arm.c
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment