Implement sum_sqr_shift() using two passes with no branch inside the loops
authorJean-Marc Valin <jmvalin@jmvalin.ca>
Fri, 19 Feb 2016 02:44:09 +0000 (10:44 +0800)
committerJean-Marc Valin <jmvalin@jmvalin.ca>
Sun, 17 Jul 2016 19:05:54 +0000 (15:05 -0400)
commit7c645606c4825aae7754275a598bbd129a8c6374
treebf2fd9da3b05923f0efcb9008301d19441a40798
parentfe4d91c2e3690ec119577537bf1bd696ad673e86
Implement sum_sqr_shift() using two passes with no branch inside the loops

Slightly slower on x86, about the same speed on ARMv7, should be faster on
DSPs.
silk/sum_sqr_shift.c