[x86]: Add aom_highbd_quantize_b_64x64_avx2().
7.3x to 9.7x faster than aom_highbd_quantize_b_64x64_c()
depending on the last nonzero coeff position.
Bug: b/235228922
Change-Id: I7cc00ec6fba5e3fdc484d49d88fdd97d6587fa8c
diff --git a/test/quantize_func_test.cc b/test/quantize_func_test.cc
index 609bf84..6f58898 100644
--- a/test/quantize_func_test.cc
+++ b/test/quantize_func_test.cc
@@ -545,6 +545,8 @@
static_cast<TX_SIZE>(TX_16X16), TYPE_B, AOM_BITS_12),
make_tuple(&aom_highbd_quantize_b_32x32_c, &aom_highbd_quantize_b_32x32_avx2,
static_cast<TX_SIZE>(TX_32X32), TYPE_B, AOM_BITS_12),
+ make_tuple(&aom_highbd_quantize_b_64x64_c, &aom_highbd_quantize_b_64x64_avx2,
+ static_cast<TX_SIZE>(TX_64X64), TYPE_B, AOM_BITS_12),
#if !CONFIG_REALTIME_ONLY
make_tuple(&aom_highbd_quantize_b_adaptive_c,
&aom_highbd_quantize_b_adaptive_avx2,
