@ -1479,38 +1479,38 @@ itxfm_wrap(16, 6)
static av_always_inline void idct32_1d ( const dctcoef * in , ptrdiff_t stride ,
static av_always_inline void idct32_1d ( const dctcoef * in , ptrdiff_t stride ,
dctcoef * out , int pass )
dctcoef * out , int pass )
{
{
dctint t0a = ( ( IN ( 0 ) + IN ( 16 ) ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
dctint t0a = ( dctint ) ( ( IN ( 0 ) + IN ( 16 ) ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
dctint t1a = ( ( IN ( 0 ) - IN ( 16 ) ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
dctint t1a = ( dctint ) ( ( IN ( 0 ) - IN ( 16 ) ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
dctint t2a = ( IN ( 8 ) * 6270 - IN ( 24 ) * 15137 + ( 1 < < 13 ) ) > > 14 ;
dctint t2a = ( dctint ) ( IN ( 8 ) * 6270U - IN ( 24 ) * 15137U + ( 1 < < 13 ) ) > > 14 ;
dctint t3a = ( IN ( 8 ) * 15137 + IN ( 24 ) * 6270 + ( 1 < < 13 ) ) > > 14 ;
dctint t3a = ( dctint ) ( IN ( 8 ) * 15137U + IN ( 24 ) * 6270U + ( 1 < < 13 ) ) > > 14 ;
dctint t4a = ( IN ( 4 ) * 3196 - IN ( 28 ) * 16069 + ( 1 < < 13 ) ) > > 14 ;
dctint t4a = ( dctint ) ( IN ( 4 ) * 3196U - IN ( 28 ) * 16069U + ( 1 < < 13 ) ) > > 14 ;
dctint t7a = ( IN ( 4 ) * 16069 + IN ( 28 ) * 3196 + ( 1 < < 13 ) ) > > 14 ;
dctint t7a = ( dctint ) ( IN ( 4 ) * 16069U + IN ( 28 ) * 3196U + ( 1 < < 13 ) ) > > 14 ;
dctint t5a = ( IN ( 20 ) * 13623 - IN ( 12 ) * 9102 + ( 1 < < 13 ) ) > > 14 ;
dctint t5a = ( dctint ) ( IN ( 20 ) * 13623U - IN ( 12 ) * 9102U + ( 1 < < 13 ) ) > > 14 ;
dctint t6a = ( IN ( 20 ) * 9102 + IN ( 12 ) * 13623 + ( 1 < < 13 ) ) > > 14 ;
dctint t6a = ( dctint ) ( IN ( 20 ) * 9102U + IN ( 12 ) * 13623U + ( 1 < < 13 ) ) > > 14 ;
dctint t8a = ( IN ( 2 ) * 1606 - IN ( 30 ) * 16305 + ( 1 < < 13 ) ) > > 14 ;
dctint t8a = ( dctint ) ( IN ( 2 ) * 1606U - IN ( 30 ) * 16305U + ( 1 < < 13 ) ) > > 14 ;
dctint t15a = ( IN ( 2 ) * 16305 + IN ( 30 ) * 1606 + ( 1 < < 13 ) ) > > 14 ;
dctint t15a = ( dctint ) ( IN ( 2 ) * 16305U + IN ( 30 ) * 1606U + ( 1 < < 13 ) ) > > 14 ;
dctint t9a = ( IN ( 18 ) * 12665 - IN ( 14 ) * 10394 + ( 1 < < 13 ) ) > > 14 ;
dctint t9a = ( dctint ) ( IN ( 18 ) * 12665U - IN ( 14 ) * 10394U + ( 1 < < 13 ) ) > > 14 ;
dctint t14a = ( IN ( 18 ) * 10394 + IN ( 14 ) * 12665 + ( 1 < < 13 ) ) > > 14 ;
dctint t14a = ( dctint ) ( IN ( 18 ) * 10394U + IN ( 14 ) * 12665U + ( 1 < < 13 ) ) > > 14 ;
dctint t10a = ( IN ( 10 ) * 7723 - IN ( 22 ) * 14449 + ( 1 < < 13 ) ) > > 14 ;
dctint t10a = ( dctint ) ( IN ( 10 ) * 7723U - IN ( 22 ) * 14449U + ( 1 < < 13 ) ) > > 14 ;
dctint t13a = ( IN ( 10 ) * 14449 + IN ( 22 ) * 7723 + ( 1 < < 13 ) ) > > 14 ;
dctint t13a = ( dctint ) ( IN ( 10 ) * 14449U + IN ( 22 ) * 7723U + ( 1 < < 13 ) ) > > 14 ;
dctint t11a = ( IN ( 26 ) * 15679 - IN ( 6 ) * 4756 + ( 1 < < 13 ) ) > > 14 ;
dctint t11a = ( dctint ) ( IN ( 26 ) * 15679U - IN ( 6 ) * 4756U + ( 1 < < 13 ) ) > > 14 ;
dctint t12a = ( IN ( 26 ) * 4756 + IN ( 6 ) * 15679 + ( 1 < < 13 ) ) > > 14 ;
dctint t12a = ( dctint ) ( IN ( 26 ) * 4756U + IN ( 6 ) * 15679U + ( 1 < < 13 ) ) > > 14 ;
dctint t16a = ( IN ( 1 ) * 804 - IN ( 31 ) * 16364 + ( 1 < < 13 ) ) > > 14 ;
dctint t16a = ( dctint ) ( IN ( 1 ) * 804U - IN ( 31 ) * 16364U + ( 1 < < 13 ) ) > > 14 ;
dctint t31a = ( IN ( 1 ) * 16364 + IN ( 31 ) * 804 + ( 1 < < 13 ) ) > > 14 ;
dctint t31a = ( dctint ) ( IN ( 1 ) * 16364U + IN ( 31 ) * 804U + ( 1 < < 13 ) ) > > 14 ;
dctint t17a = ( IN ( 17 ) * 12140 - IN ( 15 ) * 11003 + ( 1 < < 13 ) ) > > 14 ;
dctint t17a = ( dctint ) ( IN ( 17 ) * 12140U - IN ( 15 ) * 11003U + ( 1 < < 13 ) ) > > 14 ;
dctint t30a = ( IN ( 17 ) * 11003 + IN ( 15 ) * 12140 + ( 1 < < 13 ) ) > > 14 ;
dctint t30a = ( dctint ) ( IN ( 17 ) * 11003U + IN ( 15 ) * 12140U + ( 1 < < 13 ) ) > > 14 ;
dctint t18a = ( IN ( 9 ) * 7005 - IN ( 23 ) * 14811 + ( 1 < < 13 ) ) > > 14 ;
dctint t18a = ( dctint ) ( IN ( 9 ) * 7005U - IN ( 23 ) * 14811U + ( 1 < < 13 ) ) > > 14 ;
dctint t29a = ( IN ( 9 ) * 14811 + IN ( 23 ) * 7005 + ( 1 < < 13 ) ) > > 14 ;
dctint t29a = ( dctint ) ( IN ( 9 ) * 14811U + IN ( 23 ) * 7005U + ( 1 < < 13 ) ) > > 14 ;
dctint t19a = ( IN ( 25 ) * 15426 - IN ( 7 ) * 5520 + ( 1 < < 13 ) ) > > 14 ;
dctint t19a = ( dctint ) ( IN ( 25 ) * 15426U - IN ( 7 ) * 5520U + ( 1 < < 13 ) ) > > 14 ;
dctint t28a = ( IN ( 25 ) * 5520 + IN ( 7 ) * 15426 + ( 1 < < 13 ) ) > > 14 ;
dctint t28a = ( dctint ) ( IN ( 25 ) * 5520U + IN ( 7 ) * 15426U + ( 1 < < 13 ) ) > > 14 ;
dctint t20a = ( IN ( 5 ) * 3981 - IN ( 27 ) * 15893 + ( 1 < < 13 ) ) > > 14 ;
dctint t20a = ( dctint ) ( IN ( 5 ) * 3981U - IN ( 27 ) * 15893U + ( 1 < < 13 ) ) > > 14 ;
dctint t27a = ( IN ( 5 ) * 15893 + IN ( 27 ) * 3981 + ( 1 < < 13 ) ) > > 14 ;
dctint t27a = ( dctint ) ( IN ( 5 ) * 15893U + IN ( 27 ) * 3981U + ( 1 < < 13 ) ) > > 14 ;
dctint t21a = ( IN ( 21 ) * 14053 - IN ( 11 ) * 8423 + ( 1 < < 13 ) ) > > 14 ;
dctint t21a = ( dctint ) ( IN ( 21 ) * 14053U - IN ( 11 ) * 8423U + ( 1 < < 13 ) ) > > 14 ;
dctint t26a = ( IN ( 21 ) * 8423 + IN ( 11 ) * 14053 + ( 1 < < 13 ) ) > > 14 ;
dctint t26a = ( dctint ) ( IN ( 21 ) * 8423U + IN ( 11 ) * 14053U + ( 1 < < 13 ) ) > > 14 ;
dctint t22a = ( IN ( 13 ) * 9760 - IN ( 19 ) * 13160 + ( 1 < < 13 ) ) > > 14 ;
dctint t22a = ( dctint ) ( IN ( 13 ) * 9760U - IN ( 19 ) * 13160U + ( 1 < < 13 ) ) > > 14 ;
dctint t25a = ( IN ( 13 ) * 13160 + IN ( 19 ) * 9760 + ( 1 < < 13 ) ) > > 14 ;
dctint t25a = ( dctint ) ( IN ( 13 ) * 13160U + IN ( 19 ) * 9760U + ( 1 < < 13 ) ) > > 14 ;
dctint t23a = ( IN ( 29 ) * 16207 - IN ( 3 ) * 2404 + ( 1 < < 13 ) ) > > 14 ;
dctint t23a = ( dctint ) ( IN ( 29 ) * 16207U - IN ( 3 ) * 2404U + ( 1 < < 13 ) ) > > 14 ;
dctint t24a = ( IN ( 29 ) * 2404 + IN ( 3 ) * 16207 + ( 1 < < 13 ) ) > > 14 ;
dctint t24a = ( dctint ) ( IN ( 29 ) * 2404U + IN ( 3 ) * 16207U + ( 1 < < 13 ) ) > > 14 ;
dctint t0 = t0a + t3a ;
dctint t0 = t0a + t3a ;
dctint t1 = t1a + t2a ;
dctint t1 = t1a + t2a ;
@ -1545,20 +1545,20 @@ static av_always_inline void idct32_1d(const dctcoef *in, ptrdiff_t stride,
dctint t30 = t31a - t30a ;
dctint t30 = t31a - t30a ;
dctint t31 = t31a + t30a ;
dctint t31 = t31a + t30a ;
t5a = ( ( t6 - t5 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t5a = ( dctint ) ( ( t6 - t5 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t6a = ( ( t6 + t5 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t6a = ( dctint ) ( ( t6 + t5 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t9a = ( t14 * 6270 - t9 * 15137 + ( 1 < < 13 ) ) > > 14 ;
t9a = ( dctint ) ( t14 * 6270U - t9 * 15137U + ( 1 < < 13 ) ) > > 14 ;
t14a = ( t14 * 15137 + t9 * 6270 + ( 1 < < 13 ) ) > > 14 ;
t14a = ( dctint ) ( t14 * 15137U + t9 * 6270U + ( 1 < < 13 ) ) > > 14 ;
t10a = ( - ( t13 * 15137 + t10 * 6270 ) + ( 1 < < 13 ) ) > > 14 ;
t10a = ( dctint ) ( - ( t13 * 15137U + t10 * 6270U ) + ( 1 < < 13 ) ) > > 14 ;
t13a = ( t13 * 6270 - t10 * 15137 + ( 1 < < 13 ) ) > > 14 ;
t13a = ( dctint ) ( t13 * 6270U - t10 * 15137U + ( 1 < < 13 ) ) > > 14 ;
t17a = ( t30 * 3196 - t17 * 16069 + ( 1 < < 13 ) ) > > 14 ;
t17a = ( dctint ) ( t30 * 3196U - t17 * 16069U + ( 1 < < 13 ) ) > > 14 ;
t30a = ( t30 * 16069 + t17 * 3196 + ( 1 < < 13 ) ) > > 14 ;
t30a = ( dctint ) ( t30 * 16069U + t17 * 3196U + ( 1 < < 13 ) ) > > 14 ;
t18a = ( - ( t29 * 16069 + t18 * 3196 ) + ( 1 < < 13 ) ) > > 14 ;
t18a = ( dctint ) ( - ( t29 * 16069U + t18 * 3196U ) + ( 1 < < 13 ) ) > > 14 ;
t29a = ( t29 * 3196 - t18 * 16069 + ( 1 < < 13 ) ) > > 14 ;
t29a = ( dctint ) ( t29 * 3196U - t18 * 16069U + ( 1 < < 13 ) ) > > 14 ;
t21a = ( t26 * 13623 - t21 * 9102 + ( 1 < < 13 ) ) > > 14 ;
t21a = ( dctint ) ( t26 * 13623U - t21 * 9102U + ( 1 < < 13 ) ) > > 14 ;
t26a = ( t26 * 9102 + t21 * 13623 + ( 1 < < 13 ) ) > > 14 ;
t26a = ( dctint ) ( t26 * 9102U + t21 * 13623U + ( 1 < < 13 ) ) > > 14 ;
t22a = ( - ( t25 * 9102 + t22 * 13623 ) + ( 1 < < 13 ) ) > > 14 ;
t22a = ( dctint ) ( - ( t25 * 9102U + t22 * 13623U ) + ( 1 < < 13 ) ) > > 14 ;
t25a = ( t25 * 13623 - t22 * 9102 + ( 1 < < 13 ) ) > > 14 ;
t25a = ( dctint ) ( t25 * 13623U - t22 * 9102U + ( 1 < < 13 ) ) > > 14 ;
t0a = t0 + t7 ;
t0a = t0 + t7 ;
t1a = t1 + t6a ;
t1a = t1 + t6a ;
@ -1593,18 +1593,18 @@ static av_always_inline void idct32_1d(const dctcoef *in, ptrdiff_t stride,
t30 = t30a + t29a ;
t30 = t30a + t29a ;
t31a = t31 + t28 ;
t31a = t31 + t28 ;
t10a = ( ( t13 - t10 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t10a = ( dctint ) ( ( t13 - t10 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t13a = ( ( t13 + t10 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t13a = ( dctint ) ( ( t13 + t10 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t11 = ( ( t12a - t11a ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t11 = ( dctint ) ( ( t12a - t11a ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t12 = ( ( t12a + t11a ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t12 = ( dctint ) ( ( t12a + t11a ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t18a = ( t29 * 6270 - t18 * 15137 + ( 1 < < 13 ) ) > > 14 ;
t18a = ( dctint ) ( t29 * 6270U - t18 * 15137U + ( 1 < < 13 ) ) > > 14 ;
t29a = ( t29 * 15137 + t18 * 6270 + ( 1 < < 13 ) ) > > 14 ;
t29a = ( dctint ) ( t29 * 15137U + t18 * 6270U + ( 1 < < 13 ) ) > > 14 ;
t19 = ( t28a * 6270 - t19a * 15137 + ( 1 < < 13 ) ) > > 14 ;
t19 = ( dctint ) ( t28a * 6270U - t19a * 15137U + ( 1 < < 13 ) ) > > 14 ;
t28 = ( t28a * 15137 + t19a * 6270 + ( 1 < < 13 ) ) > > 14 ;
t28 = ( dctint ) ( t28a * 15137U + t19a * 6270U + ( 1 < < 13 ) ) > > 14 ;
t20 = ( - ( t27a * 15137 + t20a * 6270 ) + ( 1 < < 13 ) ) > > 14 ;
t20 = ( dctint ) ( - ( t27a * 15137U + t20a * 6270U ) + ( 1 < < 13 ) ) > > 14 ;
t27 = ( t27a * 6270 - t20a * 15137 + ( 1 < < 13 ) ) > > 14 ;
t27 = ( dctint ) ( t27a * 6270U - t20a * 15137U + ( 1 < < 13 ) ) > > 14 ;
t21a = ( - ( t26 * 15137 + t21 * 6270 ) + ( 1 < < 13 ) ) > > 14 ;
t21a = ( dctint ) ( - ( t26 * 15137U + t21 * 6270U ) + ( 1 < < 13 ) ) > > 14 ;
t26a = ( t26 * 6270 - t21 * 15137 + ( 1 < < 13 ) ) > > 14 ;
t26a = ( dctint ) ( t26 * 6270U - t21 * 15137U + ( 1 < < 13 ) ) > > 14 ;
t0 = t0a + t15a ;
t0 = t0a + t15a ;
t1 = t1a + t14 ;
t1 = t1a + t14 ;
@ -1639,14 +1639,14 @@ static av_always_inline void idct32_1d(const dctcoef *in, ptrdiff_t stride,
t30a = t30 + t25 ;
t30a = t30 + t25 ;
t31 = t31a + t24a ;
t31 = t31a + t24a ;
t20 = ( ( t27a - t20a ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t20 = ( dctint ) ( ( t27a - t20a ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t27 = ( ( t27a + t20a ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t27 = ( dctint ) ( ( t27a + t20a ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t21a = ( ( t26 - t21 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t21a = ( dctint ) ( ( t26 - t21 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t26a = ( ( t26 + t21 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t26a = ( dctint ) ( ( t26 + t21 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t22 = ( ( t25a - t22a ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t22 = ( dctint ) ( ( t25a - t22a ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t25 = ( ( t25a + t22a ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t25 = ( dctint ) ( ( t25a + t22a ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t23a = ( ( t24 - t23 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t23a = ( dctint ) ( ( t24 - t23 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
t24a = ( ( t24 + t23 ) * 11585 + ( 1 < < 13 ) ) > > 14 ;
t24a = ( dctint ) ( ( t24 + t23 ) * 11585U + ( 1 < < 13 ) ) > > 14 ;
out [ 0 ] = t0 + t31 ;
out [ 0 ] = t0 + t31 ;
out [ 1 ] = t1 + t30a ;
out [ 1 ] = t1 + t30a ;