@@ -28,6 +28,7 @@ version <next>:
- support for decoding through D3D11VA in ffmpeg
- limiter video filter
- libvmaf video filter
+- aac 960/120 frame length support
version 3.3:
- CrystalHD decoder moved to new decode API
@@ -306,6 +306,7 @@ struct AACContext {
ChannelElement *tag_che_map[4][MAX_ELEM_ID];
int tags_mapped;
int warned_remapping_once;
+ int warned_960_sbr;
/** @} */
/**
@@ -327,7 +328,9 @@ struct AACContext {
#if USE_FIXED
AVFixedDSPContext *fdsp;
#else
+ MDCT15Context *mdct120;
MDCT15Context *mdct480;
+ MDCT15Context *mdct960;
AVFloatDSPContext *fdsp;
#endif /* USE_FIXED */
int random_state;
@@ -811,11 +811,21 @@ static int decode_ga_specific_config(AACContext *ac, AVCodecContext *avctx,
uint8_t layout_map[MAX_ELEM_ID*4][3];
int tags = 0;
- if (get_bits1(gb)) { // frameLengthFlag
- avpriv_request_sample(avctx, "960/120 MDCT window");
+#if USE_FIXED
+ if (get_bits1(gb)) {
+ avpriv_report_missing_feature(avctx, "960/120 MDCT window");
return AVERROR_PATCHWELCOME;
}
- m4ac->frame_length_short = 0;
+#else
+ m4ac->frame_length_short = get_bits1(gb);
+#endif
+ if (m4ac->frame_length_short) {
+ if (!ac->warned_960_sbr)
+ avpriv_report_missing_feature(avctx, "SBR with a 960 frame length");
+ ac->warned_960_sbr = 1;
+ m4ac->sbr = 0;
+ m4ac->ps = 0;
+ }
if (get_bits1(gb)) // dependsOnCoreCoder
skip_bits(gb, 14); // coreCoderDelay
@@ -1126,6 +1136,10 @@ static av_cold void aac_static_table_init(void)
// window initialization
AAC_RENAME(ff_kbd_window_init)(AAC_RENAME(ff_aac_kbd_long_1024), 4.0, 1024);
AAC_RENAME(ff_kbd_window_init)(AAC_RENAME(ff_aac_kbd_short_128), 6.0, 128);
+ AAC_RENAME(ff_kbd_window_init)(AAC_RENAME(ff_aac_kbd_long_960), 4.0, 960);
+ AAC_RENAME(ff_kbd_window_init)(AAC_RENAME(ff_aac_kbd_short_120), 6.0, 120);
+ AAC_RENAME(ff_sine_window_init)(AAC_RENAME(ff_sine_960), 960);
+ AAC_RENAME(ff_sine_window_init)(AAC_RENAME(ff_sine_120), 120);
AAC_RENAME(ff_init_ff_sine_windows)(10);
AAC_RENAME(ff_init_ff_sine_windows)( 9);
AAC_RENAME(ff_init_ff_sine_windows)( 7);
@@ -1214,6 +1228,12 @@ static av_cold int aac_decode_init(AVCodecContext *avctx)
ret = ff_mdct15_init(&ac->mdct480, 1, 5, 1.0f/(16*1024*960));
if (ret < 0)
return ret;
+ ret = ff_mdct15_init(&ac->mdct120, 1, 3, 1.0 / RANGE15(120.0));
+ if (ret < 0)
+ return ret;
+ ret = ff_mdct15_init(&ac->mdct960, 1, 6, 1.0 / RANGE15(960.0));
+ if (ret < 0)
+ return ret;
#endif
return 0;
@@ -1314,8 +1334,13 @@ static int decode_ics_info(AACContext *ac, IndividualChannelStream *ics,
}
}
ics->num_windows = 8;
- ics->swb_offset = ff_swb_offset_128[sampling_index];
- ics->num_swb = ff_aac_num_swb_128[sampling_index];
+ if (m4ac->frame_length_short) {
+ ics->swb_offset = ff_swb_offset_120[sampling_index];
+ ics->num_swb = ff_aac_num_swb_120[sampling_index];
+ } else {
+ ics->swb_offset = ff_swb_offset_128[sampling_index];
+ ics->num_swb = ff_aac_num_swb_128[sampling_index];
+ }
ics->tns_max_bands = ff_tns_max_bands_128[sampling_index];
ics->predictor_present = 0;
} else {
@@ -1334,8 +1359,13 @@ static int decode_ics_info(AACContext *ac, IndividualChannelStream *ics,
if (!ics->num_swb || !ics->swb_offset)
return AVERROR_BUG;
} else {
- ics->swb_offset = ff_swb_offset_1024[sampling_index];
- ics->num_swb = ff_aac_num_swb_1024[sampling_index];
+ if (m4ac->frame_length_short) {
+ ics->num_swb = ff_aac_num_swb_960[sampling_index];
+ ics->swb_offset = ff_swb_offset_960[sampling_index];
+ } else {
+ ics->num_swb = ff_aac_num_swb_1024[sampling_index];
+ ics->swb_offset = ff_swb_offset_1024[sampling_index];
+ }
ics->tns_max_bands = ff_tns_max_bands_1024[sampling_index];
}
if (aot != AOT_ER_AAC_ELD) {
@@ -2348,6 +2378,12 @@ static int decode_extension_payload(AACContext *ac, GetBitContext *gb, int cnt,
if (!che) {
av_log(ac->avctx, AV_LOG_ERROR, "SBR was found before the first channel element.\n");
return res;
+ } else if (ac->oc[1].m4ac.frame_length_short) {
+ if (!ac->warned_960_sbr)
+ avpriv_report_missing_feature(ac->avctx, "SBR with a 960 frame length");
+ ac->warned_960_sbr = 1;
+ skip_bits_long(gb, 8 * cnt - 4);
+ return res;
} else if (!ac->oc[1].m4ac.sbr) {
av_log(ac->avctx, AV_LOG_ERROR, "SBR signaled to be not-present but was found in the bitstream.\n");
skip_bits_long(gb, 8 * cnt - 4);
@@ -2607,6 +2643,73 @@ static void imdct_and_windowing(AACContext *ac, SingleChannelElement *sce)
}
}
+/**
+ * Conduct IMDCT and windowing.
+ */
+static void imdct_and_windowing_960(AACContext *ac, SingleChannelElement *sce)
+{
+#if !USE_FIXED
+ IndividualChannelStream *ics = &sce->ics;
+ INTFLOAT *in = sce->coeffs;
+ INTFLOAT *out = sce->ret;
+ INTFLOAT *saved = sce->saved;
+ const INTFLOAT *swindow = ics->use_kb_window[0] ? AAC_RENAME(ff_aac_kbd_short_120) : AAC_RENAME(ff_sine_120);
+ const INTFLOAT *lwindow_prev = ics->use_kb_window[1] ? AAC_RENAME(ff_aac_kbd_long_960) : AAC_RENAME(ff_sine_960);
+ const INTFLOAT *swindow_prev = ics->use_kb_window[1] ? AAC_RENAME(ff_aac_kbd_short_120) : AAC_RENAME(ff_sine_120);
+ INTFLOAT *buf = ac->buf_mdct;
+ INTFLOAT *temp = ac->temp;
+ int i;
+
+ // imdct
+ if (ics->window_sequence[0] == EIGHT_SHORT_SEQUENCE) {
+ for (i = 0; i < 8; i++)
+ ac->mdct120->imdct_half(ac->mdct120, buf + i * 120, in + i * 128, 1);
+ } else {
+ ac->mdct960->imdct_half(ac->mdct960, buf, in, 1);
+ }
+
+ /* window overlapping
+ * NOTE: To simplify the overlapping code, all 'meaningless' short to long
+ * and long to short transitions are considered to be short to short
+ * transitions. This leaves just two cases (long to long and short to short)
+ * with a little special sauce for EIGHT_SHORT_SEQUENCE.
+ */
+
+ if ((ics->window_sequence[1] == ONLY_LONG_SEQUENCE || ics->window_sequence[1] == LONG_STOP_SEQUENCE) &&
+ (ics->window_sequence[0] == ONLY_LONG_SEQUENCE || ics->window_sequence[0] == LONG_START_SEQUENCE)) {
+ ac->fdsp->vector_fmul_window( out, saved, buf, lwindow_prev, 480);
+ } else {
+ memcpy( out, saved, 420 * sizeof(*out));
+
+ if (ics->window_sequence[0] == EIGHT_SHORT_SEQUENCE) {
+ ac->fdsp->vector_fmul_window(out + 420 + 0*120, saved + 420, buf + 0*120, swindow_prev, 60);
+ ac->fdsp->vector_fmul_window(out + 420 + 1*120, buf + 0*120 + 60, buf + 1*120, swindow, 60);
+ ac->fdsp->vector_fmul_window(out + 420 + 2*120, buf + 1*120 + 60, buf + 2*120, swindow, 60);
+ ac->fdsp->vector_fmul_window(out + 420 + 3*120, buf + 2*120 + 60, buf + 3*120, swindow, 60);
+ ac->fdsp->vector_fmul_window(temp, buf + 3*120 + 60, buf + 4*120, swindow, 60);
+ memcpy( out + 420 + 4*120, temp, 60 * sizeof(*out));
+ } else {
+ ac->fdsp->vector_fmul_window(out + 420, saved + 420, buf, swindow_prev, 60);
+ memcpy( out + 540, buf + 60, 420 * sizeof(*out));
+ }
+ }
+
+ // buffer update
+ if (ics->window_sequence[0] == EIGHT_SHORT_SEQUENCE) {
+ memcpy( saved, temp + 60, 60 * sizeof(*saved));
+ ac->fdsp->vector_fmul_window(saved + 60, buf + 4*120 + 60, buf + 5*120, swindow, 60);
+ ac->fdsp->vector_fmul_window(saved + 180, buf + 5*120 + 60, buf + 6*120, swindow, 60);
+ ac->fdsp->vector_fmul_window(saved + 300, buf + 6*120 + 60, buf + 7*120, swindow, 60);
+ memcpy( saved + 420, buf + 7*120 + 60, 60 * sizeof(*saved));
+ } else if (ics->window_sequence[0] == LONG_START_SEQUENCE) {
+ memcpy( saved, buf + 480, 420 * sizeof(*saved));
+ memcpy( saved + 420, buf + 7*120 + 60, 60 * sizeof(*saved));
+ } else { // LONG_STOP or ONLY_LONG
+ memcpy( saved, buf + 480, 480 * sizeof(*saved));
+ }
+#endif
+}
+
static void imdct_and_windowing_ld(AACContext *ac, SingleChannelElement *sce)
{
IndividualChannelStream *ics = &sce->ics;
@@ -2758,7 +2861,10 @@ static void spectral_to_sample(AACContext *ac, int samples)
imdct_and_window = imdct_and_windowing_eld;
break;
default:
- imdct_and_window = ac->imdct_and_windowing;
+ if (ac->oc[1].m4ac.frame_length_short)
+ imdct_and_window = imdct_and_windowing_960;
+ else
+ imdct_and_window = ac->imdct_and_windowing;
}
for (type = 3; type >= 0; type--) {
for (i = 0; i < MAX_ELEM_ID; i++) {
@@ -3002,7 +3108,7 @@ static int aac_decode_frame_int(AVCodecContext *avctx, void *data,
err = AVERROR_INVALIDDATA;
goto fail;
}
- samples = 1024;
+ samples = ac->oc[1].m4ac.frame_length_short ? 960 : 1024;
che->present = 1;
}
@@ -3229,7 +3335,9 @@ static av_cold int aac_decode_close(AVCodecContext *avctx)
ff_mdct_end(&ac->mdct_ld);
ff_mdct_end(&ac->mdct_ltp);
#if !USE_FIXED
+ ff_mdct15_uninit(&ac->mdct120);
ff_mdct15_uninit(&ac->mdct480);
+ ff_mdct15_uninit(&ac->mdct960);
#endif
av_freep(&ac->fdsp);
return 0;
@@ -620,8 +620,7 @@ static int read_sbr_grid(AACContext *ac, SpectralBandReplication *sbr,
{
int i;
int bs_pointer = 0;
- // frameLengthFlag ? 15 : 16; 960 sample length frames unsupported; this value is numTimeSlots
- int abs_bord_trail = 16;
+ int abs_bord_trail = ac->oc[1].m4ac.frame_length_short ? 15 : 16;
int num_rel_lead, num_rel_trail;
unsigned bs_num_env_old = ch_data->bs_num_env;
int bs_frame_class, bs_num_env;
@@ -37,6 +37,10 @@ float ff_aac_pow34sf_tab[428];
DECLARE_ALIGNED(32, float, ff_aac_kbd_long_1024)[1024];
DECLARE_ALIGNED(32, float, ff_aac_kbd_short_128)[128];
+DECLARE_ALIGNED(32, float, ff_aac_kbd_long_960)[960];
+DECLARE_ALIGNED(32, float, ff_aac_kbd_short_120)[120];
+DECLARE_ALIGNED(32, int, ff_aac_kbd_long_960_fixed)[960];
+DECLARE_ALIGNED(32, int, ff_aac_kbd_short_120_fixed)[120];
DECLARE_ALIGNED(32, int, ff_aac_kbd_long_1024_fixed)[1024];
DECLARE_ALIGNED(32, int, ff_aac_kbd_short_128_fixed)[128];
@@ -44,6 +48,10 @@ const uint8_t ff_aac_num_swb_1024[] = {
41, 41, 47, 49, 49, 51, 47, 47, 43, 43, 43, 40, 40
};
+const uint8_t ff_aac_num_swb_960[] = {
+ 40, 40, 46, 49, 49, 49, 46, 46, 42, 42, 42, 40, 40
+};
+
const uint8_t ff_aac_num_swb_512[] = {
0, 0, 0, 36, 36, 37, 31, 31, 0, 0, 0, 0, 0
};
@@ -56,6 +64,10 @@ const uint8_t ff_aac_num_swb_128[] = {
12, 12, 12, 14, 14, 14, 15, 15, 15, 15, 15, 15, 15
};
+const uint8_t ff_aac_num_swb_120[] = {
+ 12, 12, 12, 14, 14, 14, 15, 15, 15, 15, 15, 15, 15
+};
+
const uint8_t ff_aac_pred_sfb_max[] = {
33, 33, 38, 40, 40, 40, 41, 41, 37, 37, 37, 34, 34
};
@@ -1229,6 +1241,100 @@ static const uint16_t swb_offset_128_8[] = {
36, 44, 52, 60, 72, 88, 108, 128
};
+static const uint16_t swb_offset_960_96[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 32, 36,
+ 40, 44, 48, 52, 56, 64, 72, 80, 88, 96,
+ 108, 120, 132, 144, 156, 172, 188, 212, 240, 276,
+ 320, 384, 448, 512, 576, 640, 704, 768, 832, 896,
+ 960
+};
+
+static const uint16_t swb_offset_960_64[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 32, 36,
+ 40, 44, 48, 52, 56, 64, 72, 80, 88, 100,
+ 112, 124, 140, 156, 172, 192, 216, 240, 268, 304,
+ 344, 384, 424, 464, 504, 544, 584, 624, 664, 704,
+ 744, 784, 824, 864, 904, 944, 960
+};
+
+static const uint16_t swb_offset_960_48[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 32, 36,
+ 40, 48, 56, 64, 72, 80, 88, 96, 108, 120,
+ 132, 144, 160, 176, 196, 216, 240, 264, 292, 320,
+ 352, 384, 416, 448, 480, 512, 544, 576, 608, 640,
+ 672, 704, 736, 768, 800, 832, 864, 896, 928, 960
+};
+
+static const uint16_t swb_offset_960_32[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 32, 36,
+ 40, 48, 56, 64, 72, 80, 88, 96, 108, 120,
+ 132, 144, 160, 176, 196, 216, 240, 264, 292, 320,
+ 352, 384, 416, 448, 480, 512, 544, 576, 608, 640,
+ 672, 704, 736, 768, 800, 832, 864, 896, 928, 960
+};
+
+static const uint16_t swb_offset_960_24[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 32, 36,
+ 40, 44, 52, 60, 68, 76, 84, 92, 100, 108,
+ 116, 124, 136, 148, 160, 172, 188, 204, 220, 240,
+ 260, 284, 308, 336, 364, 396, 432, 468, 508, 552,
+ 600, 652, 704, 768, 832, 896, 960
+};
+
+static const uint16_t swb_offset_960_16[] =
+{
+ 0, 8, 16, 24, 32, 40, 48, 56, 64, 72,
+ 80, 88, 100, 112, 124, 136, 148, 160, 172, 184,
+ 196, 212, 228, 244, 260, 280, 300, 320, 344, 368,
+ 396, 424, 456, 492, 532, 572, 616, 664, 716, 772,
+ 832, 896, 960
+};
+
+static const uint16_t swb_offset_960_8[] =
+{
+ 0, 12, 24, 36, 48, 60, 72, 84, 96, 108,
+ 120, 132, 144, 156, 172, 188, 204, 220, 236, 252,
+ 268, 288, 308, 328, 348, 372, 396, 420, 448, 476,
+ 508, 544, 580, 620, 664, 712, 764, 820, 880, 944,
+ 960
+};
+
+
+static const uint16_t swb_offset_120_96[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 32, 40, 48, 64, 92, 120
+};
+
+static const uint16_t swb_offset_120_64[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 32, 40, 48, 64, 92, 120
+};
+
+static const uint16_t swb_offset_120_48[] =
+{
+ 0, 4, 8, 12, 16, 20, 28, 36, 44, 56, 68, 80, 96, 112, 120
+};
+
+static const uint16_t swb_offset_120_24[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 36, 44, 52, 64, 76, 92, 108, 120
+};
+
+static const uint16_t swb_offset_120_16[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 32, 40, 48, 60, 72, 88, 108, 120
+};
+
+static const uint16_t swb_offset_120_8[] =
+{
+ 0, 4, 8, 12, 16, 20, 24, 28, 36, 44, 52, 60, 72, 88, 108, 120
+};
+
const uint16_t * const ff_swb_offset_1024[] = {
swb_offset_1024_96, swb_offset_1024_96, swb_offset_1024_64,
swb_offset_1024_48, swb_offset_1024_48, swb_offset_1024_32,
@@ -1237,6 +1343,14 @@ const uint16_t * const ff_swb_offset_1024[] = {
swb_offset_1024_8
};
+const uint16_t * const ff_swb_offset_960[] = {
+ swb_offset_960_96, swb_offset_960_96, swb_offset_960_64,
+ swb_offset_960_48, swb_offset_960_48, swb_offset_960_32,
+ swb_offset_960_24, swb_offset_960_24, swb_offset_960_16,
+ swb_offset_960_16, swb_offset_960_16, swb_offset_960_8,
+ swb_offset_960_8
+};
+
const uint16_t * const ff_swb_offset_512[] = {
NULL, NULL, NULL,
swb_offset_512_48, swb_offset_512_48, swb_offset_512_32,
@@ -1263,6 +1377,14 @@ const uint16_t * const ff_swb_offset_128[] = {
swb_offset_128_8
};
+const uint16_t * const ff_swb_offset_120[] = {
+ swb_offset_120_96, swb_offset_120_96, swb_offset_120_96,
+ swb_offset_120_48, swb_offset_120_48, swb_offset_120_48,
+ swb_offset_120_24, swb_offset_120_24, swb_offset_120_16,
+ swb_offset_120_16, swb_offset_120_16, swb_offset_120_8,
+ swb_offset_120_8
+};
+
// @}
/* @name ff_tns_max_bands
@@ -136,6 +136,10 @@ static const INTFLOAT * const tns_tmp2_map[4] = {
*/
DECLARE_ALIGNED(32, extern float, ff_aac_kbd_long_1024)[1024];
DECLARE_ALIGNED(32, extern float, ff_aac_kbd_short_128)[128];
+DECLARE_ALIGNED(32, extern float, ff_aac_kbd_long_960)[960];
+DECLARE_ALIGNED(32, extern float, ff_aac_kbd_short_120)[120];
+DECLARE_ALIGNED(32, extern int, ff_aac_kbd_long_960_fixed)[960];
+DECLARE_ALIGNED(32, extern int, ff_aac_kbd_short_120_fixed)[120];
DECLARE_ALIGNED(32, extern int, ff_aac_kbd_long_1024_fixed)[1024];
DECLARE_ALIGNED(32, extern int, ff_aac_kbd_long_512_fixed)[512];
DECLARE_ALIGNED(32, extern int, ff_aac_kbd_short_128_fixed)[128];
@@ -149,9 +153,11 @@ DECLARE_ALIGNED(32, extern const int, ff_aac_eld_window_480_fixed)[1800];
* @{
*/
extern const uint8_t ff_aac_num_swb_1024[];
+extern const uint8_t ff_aac_num_swb_960 [];
extern const uint8_t ff_aac_num_swb_512 [];
extern const uint8_t ff_aac_num_swb_480 [];
extern const uint8_t ff_aac_num_swb_128 [];
+extern const uint8_t ff_aac_num_swb_120 [];
// @}
extern const uint8_t ff_aac_pred_sfb_max [];
@@ -168,9 +174,11 @@ extern const float *ff_aac_codebook_vector_vals[];
extern const uint16_t *ff_aac_codebook_vector_idx[];
extern const uint16_t * const ff_swb_offset_1024[13];
+extern const uint16_t * const ff_swb_offset_960 [13];
extern const uint16_t * const ff_swb_offset_512 [13];
extern const uint16_t * const ff_swb_offset_480 [13];
extern const uint16_t * const ff_swb_offset_128 [13];
+extern const uint16_t * const ff_swb_offset_120 [13];
extern const uint8_t ff_tns_max_bands_1024[13];
extern const uint8_t ff_tns_max_bands_512 [13];
@@ -52,14 +52,16 @@ void AAC_RENAME(ff_init_ff_sine_windows)(int index);
extern SINETABLE( 32);
extern SINETABLE( 64);
+extern SINETABLE( 120);
extern SINETABLE( 128);
extern SINETABLE( 256);
extern SINETABLE( 512);
+extern SINETABLE( 960);
extern SINETABLE(1024);
extern SINETABLE(2048);
extern SINETABLE(4096);
extern SINETABLE(8192);
-extern SINETABLE_CONST INTFLOAT * const AAC_RENAME(ff_sine_windows)[14];
+extern SINETABLE_CONST INTFLOAT * const AAC_RENAME(ff_sine_windows)[16];
#endif /* AVCODEC_SINEWIN_H */
@@ -34,9 +34,11 @@
#if !CONFIG_HARDCODED_TABLES
SINETABLE( 32);
SINETABLE( 64);
+SINETABLE( 120);
SINETABLE( 128);
SINETABLE( 256);
SINETABLE( 512);
+SINETABLE( 960);
SINETABLE(1024);
SINETABLE(2048);
SINETABLE(4096);
@@ -59,7 +61,8 @@ SINETABLE_CONST INTFLOAT * const AAC_RENAME(ff_sine_windows)[] = {
NULL, NULL, NULL, NULL, NULL, // unused
AAC_RENAME(ff_sine_32) , AAC_RENAME(ff_sine_64), AAC_RENAME(ff_sine_128),
AAC_RENAME(ff_sine_256), AAC_RENAME(ff_sine_512), AAC_RENAME(ff_sine_1024),
- AAC_RENAME(ff_sine_2048), AAC_RENAME(ff_sine_4096), AAC_RENAME(ff_sine_8192)
+ AAC_RENAME(ff_sine_2048), AAC_RENAME(ff_sine_4096), AAC_RENAME(ff_sine_8192),
+ AAC_RENAME(ff_sine_120), AAC_RENAME(ff_sine_960),
};
// Generate a sine window.