improve vp9_idct32x32_34(x1.472)&1024(x1.032)_add_sse2
authorAbo Talib Mahfoodh <ab.mahfoodh@gmail.com>
Tue, 26 Nov 2013 17:26:43 +0000 (12:26 -0500)
committerAbo Talib Mahfoodh <ab.mahfoodh@gmail.com>
Tue, 26 Nov 2013 17:28:26 +0000 (12:28 -0500)
commitf97d91ab673f57e2fa5a44cbee2e1cdec188c43c
tree08a57a5ffea274b4659521b3152b3360ae152a4d
parent5488da280da3eb10e5b39d0f311493cc3946b292
improve vp9_idct32x32_34(x1.472)&1024(x1.032)_add_sse2

vp9_idct32x32_34_add_sse2:
speedup: 1.472
IDCT32_1D_34 and MULTIPLICATION_AND_ADD_2 are optimized
based on the fact that Only upper-left 8x8 has
non-zero values.

vp9_idct32x32_1024_add_sse2:
speedup: 1.032

Tested with: park_joy_420_720p50.y4m

Change-Id: I8670ce547552b48695049de298e2fc46ce28dfbc
vp9/common/x86/vp9_idct_intrin_sse2.c