@/*****************************************************************************
@*
@* Copyright (C) 2012 Ittiam Systems Pvt Ltd, Bangalore
@*
@* Licensed under the Apache License, Version 2.0 (the "License");
@* you may not use this file except in compliance with the License.
@* You may obtain a copy of the License at:
@*
@* http://www.apache.org/licenses/LICENSE-2.0
@*
@* Unless required by applicable law or agreed to in writing, software
@* distributed under the License is distributed on an "AS IS" BASIS,
@* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
@* See the License for the specific language governing permissions and
@* limitations under the License.
@*
@*****************************************************************************/
@/*******************************************************************************
@* @file
@* ihevc_deblk_luma_vert.s
@*
@* @brief
@* contains function definitions for inter prediction interpolation.
@* functions are coded using neon intrinsics and can be compiled using
@* rvct
@*
@* @author
@* anand s
@*
@* @par list of functions:
@*
@*
@* @remarks
@* none
@*
@*******************************************************************************/
.equ qp_q_offset, 108
.equ beta_offset_div2_offset, 112
.equ tc_offset_div2_offset, 116
.equ filter_p_offset, 120
.equ filter_q_offset, 124
.text
.align 4
.extern gai4_ihevc_tc_table
.extern gai4_ihevc_beta_table
.globl ihevc_deblk_luma_horz_a9q
gai4_ihevc_tc_table_addr:
.long gai4_ihevc_tc_table - ulbl1 - 8
gai4_ihevc_beta_table_addr:
.long gai4_ihevc_beta_table - ulbl2 - 8
.type ihevc_deblk_luma_horz_a9q, %function
ihevc_deblk_luma_horz_a9q:
stmfd sp!, {r3-r12,lr}
vpush {d8 - d15}
ldr r4,[sp,#qp_q_offset]
ldr r5,[sp,#beta_offset_div2_offset]
add r3,r3,r4
add r3,r3,#1
ldr r6, [sp,#tc_offset_div2_offset]
asr r3,r3,#1
add r7,r3,r5,lsl #1
add r3,r3,r6,lsl #1
cmp r7,#0x33
movgt r7,#0x33
bgt l1.1532
cmp r7,#0x0
movlt r7,#0x0 @ r7 has the beta_index value
l1.1532:
@ bic r2,r2,#1
asr r2,r2,#1
add r3,r3,r2,lsl #1
cmp r3,#0x35
movgt r3,#0x35
bgt l1.1564
cmp r3,#0x0
movlt r3,#0x0 @ r3 has the tc_index value
@ qp_luma = (quant_param_p + quant_param_q + 1) >> 1@
@ beta_indx = clip3(qp_luma + (beta_offset_div2 << 1), 0, 51)@
@ tc_indx = clip3(qp_luma + (2 * (bs >> 1)) + (tc_offset_div2 << 1), 0, 53)@
l1.1564:
ldr r2,gai4_ihevc_beta_table_addr
ulbl2:
add r2,r2,pc
ldr r4,gai4_ihevc_tc_table_addr
ulbl1:
add r4,r4,pc
ldr r5,[r2,r7,lsl #2] @ beta
ldr r6,[r4,r3,lsl #2] @ tc
cmp r6,#0
beq l1.2404
vmov.i16 d0,#0x2
lsl r7,r6,#1
add r14,r1,r1,lsl #1
ldr r8,[r0,-r14] @ -3 value
vdup.8 d1,r7
ldr r10,[r0,-r1,lsl #1] @-2 value
vdup.32 d23,r8 @ -3 value
ldr r11,[r0,-r1] @-1 value
vdup.32 d24,r10 @ -2 value
and r8,#0xff
ldr r12,[r0,#0] @ 0 value
vdup.32 d25, r11 @-1 value
and r10,#0xff
ldr r9,[r0,r1] @ 1 value
vdup.32 d26,r12 @ 0 value
and r11,#0xff
ldr r2,[r0,r1,lsl #1] @ 2 value
vdup.32 d27,r9 @ 1value
and r12,#0xff
vdup.32 d28,r2 @ 2 value
and r9,#0xff
and r2,#0xff
add r12,r12,r2
subs r9,r12,r9,lsl #1 @ dq0 value is stored in r9
rsbmi r9,r9,#0
@dq0 = abs( pu1_src[2] - 2 * pu1_src[1] + pu1_src[0] )@
add r8,r8,r11
subs r8,r8,r10,lsl #1
rsbmi r8,r8,#0 @ dp0 value is stored in r8
@ dp0 = abs( pu1_src[-3] - 2 * pu1_src[-2] + pu1_src[-1] )@
add r3,r1,r1,lsl #1
add r14,r0,#3
ldrb r2,[r14,-r3] @ -2 value
ldrb r10,[r14,-r1,lsl #1] @ -2 value
ldrb r11,[r14,-r1] @ -1 value
ldrb r12,[r14,#0] @ 0 value
ldrb r3,[r14,r1] @ 1 value
ldrb r4,[r14,r1,lsl #1] @ 2 value
add r12,r12,r4
subs r12,r12,r3,lsl #1 @ dq3value is stored in r12
rsbmi r12,r12,#0
@ dq3 = abs( pu1_src[3 * src_strd + 2] - 2 * pu1_src[3 * src_strd + 1]+ pu1_src[3 * src_strd + 0] )@
add r2,r2,r11
subs r11,r2,r10,lsl #1
rsbmi r11,r11,#0 @ dp3 value is stored in r8
@ dp3 = abs( pu1_src[3 * src_strd - 3] - 2 * pu1_src[3 * src_strd - 2] + pu1_src[3 * src_strd - 1] )@
add r3,r8,r9 @ r3 has the d0 value
add r4,r11,r12 @ r4 has the d3 value
@ d0 = dp0 + dq0@
@ d3 = dp3 + dq3@
add r14,r8,r11 @ r13 has the value dp
add r12,r12,r9 @ r12 has the value dq
@ dp = dp0 + dp3@
@ dq = dq0 + dq3@
add r11, r3, r4 @ r3 has the value d
@ d = d0 + d3@
cmp r11,r5
bge l1.2404
@ if(d < beta)
@ registers which cannont be altered : r3,r4 r5,r6,r12,r13,r0,r1,r11
@ registers for use: r2,r7,r8,r9,r10,
asr r10,r5,#2
vqadd.u8 d30,d26,d1
cmp r10,r3,lsl #1
vqsub.u8 d31,d26,d1
ble l1.1840
add r10,r1,r1,lsl #1
vaddl.u8 q3,d25,d26
ldr r2,[r0,-r1,lsl #2] @ has the -4 value
ldrb r7,[r0,-r1] @ has the -1 value
vdup.32 d22,r2 @ -4 value
vaddw.u8 q4,q3,d27
ldrb r3,[r0,#0] @ r4 has the 0 value
vqadd.u8 d16,d27,d1
and r2,#0xff
vmul.i16 q6,q4,d0[0]
ldr r8,[r0,r10] @ has the 3 value
vaddl.u8 q5,d24,d28
subs r2,r2,r7
vqsub.u8 d17,d27,d1
vdup.32 d29,r8 @ 3 value
and r8,#0xff
vadd.i16 q6,q6,q5
rsbmi r2,r2,#0
vrshrn.i16 d20,q6,#3
subs r8,r8,r3
rsbmi r8,r8,#0
vmin.u8 d18,d20,d30
add r8,r8,r2
cmp r8,r5,asr #3
bge l1.1840
vaddw.u8 q7,q4,d28
subs r7,r3,r7
vmax.u8 d4,d18,d31
rsbmi r7,r7,#0
vqadd.u8 d30,d28,d1
mov r10,#5
vrshrn.i16 d21,q7,#2
mul r10,r10,r6
vqsub.u8 d31,d28,d1
add r10,#1
cmp r7,r10,asr #1
vmin.u8 d18,d21,d16
bge l1.1840
@ if( (2 * d3 < (beta >> 2)&& ( abs(pu1_src[3] - pu1_src[0]) + abs(pu1_src[-1] - pu1_src[-4]) < (beta >> 3) )
@ && abs(pu1_src[0] - pu1_src[-1]) < ( (5 * tc + 1) >> 1 ) )
vmax.u8 d5,d18,d17
asr r10,r5,#2
vaddl.u8 q8,d29,d28
cmp r10,r4,lsl #1
ble l1.1840
add r10,r1,r1,lsl #1
vmul.i16 q8,q8,d0[0]
add r4,r0,#3
ldrb r2,[r4,-r1,lsl #2]
vadd.i16 q8,q8,q7
ldrb r7,[r4,-r1]
vrshrn.i16 d19,q8,#3
ldrb r3,[r4,#0]
ldrb r8,[r4,r10]
@ ubfx r7,r2,#24,#8 @ has the -1 value
@ and r2,#0xff @ has the -4 value
@ ubfx r8,r3,#24,#8 @ has the 3 value
@ and r3,#0xff @ r4 has the 0 value
subs r8,r8,r3
vmin.u8 d18,d19,d30
rsbmi r8,r8,#0
vaddl.u8 q3,d25,d24
subs r2,r2,r7
vmax.u8 d3,d18,d31
rsbmi r2,r2,#0
vaddw.u8 q4,q3,d26
add r8,r8,r2
vqadd.u8 d30,d25,d1
cmp r8,r5,asr #3
vqsub.u8 d31,d25,d1
bge l1.1840
vmul.i16 q6,q4,d0[0]
subs r7,r3,r7
vqadd.u8 d16,d24,d1
rsbmi r7,r7,#0
vaddl.u8 q5,d23,d27
mov r10,#5
vqsub.u8 d17,d24,d1
mul r10,r10,r6
vadd.i16 q6,q6,q5
add r10,#1
vrshrn.i16 d20,q6,#3
cmp r7,r10,asr #1
vaddw.u8 q7,q4,d23
bge l1.1840
vmin.u8 d18,d20,d30
mov r2,#2
vqadd.u8 d30,d23,d1
ldr r4,[sp,#filter_p_offset] @ loading the filter_flag_p
vmax.u8 d2,d18,d31
ldr r5,[sp,#filter_q_offset] @ loading the filter_flag_q
vrshrn.i16 d21,q7,#2
b end_dep_deq_decision_horz
@ r2 has the value of de
@ r6 has teh value of tc
@ r5 has the value of beta
@ r14 has the value of dp
@ r12 has the value of dq
@ r0 has the value of source address
@ r1 has the src stride
l1.1840:
mov r2,#1
mov r11,r5
ldr r4,[sp,#filter_p_offset] @ loading the filter_flag_p
ldr r5,[sp,#filter_q_offset] @ loading the filter_flag_q
cmp r6,#1
moveq r9,#0
moveq r10,#0
beq end_dep_deq_decision_horz
and r7,r4,r5
cmp r7,#1
beq both_flags_set_horz
cmp r4,#0
beq set_flag_dep_zero_horz
add r8,r11,r11,asr #1
mov r10,#0
asr r8,#3
cmp r8,r14
movgt r9,#1
movle r9,#0
b end_dep_deq_decision_horz
set_flag_dep_zero_horz:
add r8,r11,r11,asr #1
mov r9,#0
asr r8,#3
cmp r8,r12
movgt r10,#1
movle r10,#0
b end_dep_deq_decision_horz
both_flags_set_horz:
add r8,r11,r11,asr #1
asr r8,#3
cmp r8,r14
movgt r9,#1
movle r9,#0
cmp r8,r12
movgt r10,#1
movle r10,#0
end_dep_deq_decision_horz:
@r0=source address
@r1=stride
@ r2 =de
@ r4=flag p
@r5= flag q
@r6 =tc
@ r9 =dep
@ r10=deq
@ add r14,r1,r1,lsl #1
@ lsl r7,r6,#1
@ vdup.8 d1,r7
@ vmov.i16 d0,#0x2
vmin.u8 d18,d21,d16
cmp r2,#1
vqsub.u8 d31,d23,d1
beq l1.2408
vaddl.u8 q4,d23,d22
cmp r5,#1
bne strong_filtering_p
strong_filtering_q:
mov r12,r0
vst1.32 d4[0],[r12],r1
vst1.32 d5[0],[r12],r1
vst1.32 d3[0],[r12]
cmp r4,#1
bne l1.2404
strong_filtering_p:
vmax.u8 d5,d18,d17
mov r12,r0
vmul.i16 q4,q4,d0[0]
rsb r11,r1,#0
vadd.i16 q8,q4,q7
add r12,r12,r11
vrshrn.i16 d19,q8,#3
vst1.32 d2[0],[r12],r11
vmin.u8 d18,d19,d30
vst1.32 d5[0],[r12],r11
vmax.u8 d3,d18,d31
vst1.32 d3[0],[r12]
l1.2404:
vpop {d8 - d15}
ldmfd sp!, {r3-r12,pc}
@ r4=flag p
@r5= flag q
@r6 =tc
@ r9 =dep
@ r10=deq
@ d22 -4 value
@d23 @ -3 value
@ vdup.32 d24,r11 @ -2 value
@ vdup.32 d25, r11 @-1 value
@ vdup.32 d26,r11 @ 0 value
@ vdup.32 d27,r11 @ 1value
@ vdup.32 d28,r11 @ 2 value
@ vdup.32 d29,r11 @ 3 value
l1.2408:
vmov.i16 d0,#0x9
vsubl.u8 q5,d26,d25
vmul.i16 q5,q5,d0[0]
vmov.i16 d0,#0x3
vsubl.u8 q6,d27,d24
vmul.i16 q6,q6,d0[0]
vdup.8 d30,r6 @ duplicating the +tc value
rsb r12,r6,#0
vdup.8 d31,r12 @ duplicating the -tc value
vsub.i16 q5,q5,q6
vrshr.s16 q5,q5,#4
@ delta = ( 9 * (pu1_src[0] - pu1_src[-1]) - 3 * (pu1_src[1] - pu1_src[-2]) + 8 ) >> 4@
vabs.s16 q4,q5
vmovn.i16 d9,q4
@ storing the absolute values of delta in d9
vqmovn.s16 d10,q5
@ storing the clipped values of delta in d16
vmin.s8 d11,d10,d30
vmax.s8 d8,d31,d11 @ d8 has the value delta = clip3(delta, -tc, tc)@
vmovl.u8 q3,d25
vaddw.s8 q2,q3,d8
vqmovun.s16 d12,q2
vmovl.u8 q3,d26
vsubw.s8 q2,q3,d8
vqmovun.s16 d13,q2
mov r11,#0xa
mul r12,r11,r6
vdup.8 d2,r12 @ d2 has the 10*tc value
vmov d18,d24
vdup.8 d0,r6
vshr.s8 d0,#1
vneg.s8 d1,d0
cmp r4,#1
bne l1.2724
cmp r9,#1
bne l1.2700
@ d12 and d13 have the value temp_p0 and temp_q0
vaddl.u8 q7,d23,d25
vrshrn.u16 d14,q7,#1
vsubl.u8 q7,d14,d24
vaddw.s8 q7,q7,d8
vqshrn.s16 d14,q7,#1
vmin.s8 d15,d14,d0
vmax.s8 d14,d1,d15
@ d14 has the delta p value
vmovl.u8 q8,d24
vaddw.s8 q8,q8,d14
vqmovun.s16 d14,q8
@ d14 =tmp_p1 = clip_u8(pu1_src[-2 * src_strd] + delta_p)@
vcge.u8 d18,d9,d2
vbsl d18,d24,d14
l1.2700:
mov r12,r0
rsb r11,r1,#0
add r12,r11
vcge.u8 d19,d9,d2
vbsl d19,d25,d12
vst1.32 {d19[0]},[r12],r11
vst1.32 {d18[0]},[r12]
l1.2724:
cmp r5,#1
bne l1.2404
cmp r10,#1
vmov d18, d27
bne l1.2852
vaddl.u8 q7,d26,d28
vrshrn.u16 d14,q7,#1
vsubl.u8 q7,d14,d27
vsubw.s8 q7,q7,d8
vqshrn.s16 d14,q7,#1
vmin.s8 d15,d14,d0
vmax.s8 d14,d1,d15
@ d14 has the delta p value
vmovl.u8 q8,d27
vaddw.s8 q8,q8,d14
vqmovun.s16 d14,q8
vcge.u8 d18,d9,d2
vbsl d18,d27,d14
l1.2852:
mov r12,r0
vcge.u8 d19,d9,d2
vbsl d19,d26,d13
vst1.32 {d19[0]},[r12],r1
vst1.32 {d18[0]},[r12]
vpop {d8 - d15}
ldmfd sp!, {r3-r12,r15}