;*******************************************************************************
;* SIMD-optimized IDCT functions for HEVC decoding
;* Copyright (c) 2014 Pierre-Edouard LEPERE
;* Copyright (c) 2014 James Almer
;*
;* This file is part of Libav.
;*
;* Libav is free software; you can redistribute it and/or
;* modify it under the terms of the GNU Lesser General Public
;* License as published by the Free Software Foundation; either
;* version 2.1 of the License, or (at your option) any later version.
;*
;* Libav is distributed in the hope that it will be useful,
;* but WITHOUT ANY WARRANTY; without even the implied warranty of
;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
;* Lesser General Public License for more details.
;*
;* You should have received a copy of the GNU Lesser General Public
;* License along with Libav; if not, write to the Free Software
;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
;******************************************************************************

%include "libavutil/x86/x86util.asm"

section .text

; void ff_hevc_idctHxW_dc_{8,10}_<opt>(int16_t *coeffs)
; %1 = HxW
; %2 = number of loops
; %3 = bitdepth
%macro IDCT_DC 3
cglobal hevc_idct_%1x%1_dc_%3, 1, 2, 1, coeff, tmp
    movsx             tmpd, word [coeffq]
    add               tmpd, (1 << (14 - %3)) + 1
    sar               tmpd, (15 - %3)
    movd               xm0, tmpd
    SPLATW              m0, xm0
    DEFINE_ARGS coeff, cnt
    mov               cntd, %2
.loop:
    mova [coeffq+mmsize*0], m0
    mova [coeffq+mmsize*1], m0
    mova [coeffq+mmsize*2], m0
    mova [coeffq+mmsize*3], m0
    add  coeffq, mmsize*8
    mova [coeffq+mmsize*-4], m0
    mova [coeffq+mmsize*-3], m0
    mova [coeffq+mmsize*-2], m0
    mova [coeffq+mmsize*-1], m0
    dec  cntd
    jg  .loop
    RET
%endmacro

; %1 = HxW
; %2 = bitdepth
%macro IDCT_DC_NL 2 ; No loop
cglobal hevc_idct_%1x%1_dc_%2, 1, 2, 1, coeff, tmp
    movsx             tmpd, word [coeffq]
    add               tmpd, (1 << (14 - %2)) + 1
    sar               tmpd, (15 - %2)
    movd                m0, tmpd
    SPLATW              m0, xm0
    mova [coeffq+mmsize*0], m0
    mova [coeffq+mmsize*1], m0
    mova [coeffq+mmsize*2], m0
    mova [coeffq+mmsize*3], m0
%if mmsize == 16
    mova [coeffq+mmsize*4], m0
    mova [coeffq+mmsize*5], m0
    mova [coeffq+mmsize*6], m0
    mova [coeffq+mmsize*7], m0
%endif
    RET
%endmacro

; 8-bit
INIT_MMX mmxext
IDCT_DC_NL  4,      8
IDCT_DC     8,  2,  8

INIT_XMM sse2
IDCT_DC_NL  8,      8
IDCT_DC    16,  4,  8
IDCT_DC    32, 16,  8

%if HAVE_AVX2_EXTERNAL
INIT_YMM avx2
IDCT_DC    16,  2,  8
IDCT_DC    32,  8,  8
%endif ;HAVE_AVX2_EXTERNAL

; 10-bit
INIT_MMX mmxext
IDCT_DC_NL  4,     10
IDCT_DC     8,  2, 10

INIT_XMM sse2
IDCT_DC_NL  8,     10
IDCT_DC    16,  4, 10
IDCT_DC    32, 16, 10

%if HAVE_AVX2_EXTERNAL
INIT_YMM avx2
IDCT_DC    16,  2, 10
IDCT_DC    32,  8, 10
%endif ;HAVE_AVX2_EXTERNAL