cherry-pick from 7241: speed-up matrix44::setconcat when both inputs are
authorreed@google.com <reed@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81>
Thu, 17 Jan 2013 16:01:19 +0000 (16:01 +0000)
committerreed@google.com <reed@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81>
Thu, 17 Jan 2013 16:01:19 +0000 (16:01 +0000)
commit2b16570bda716725e08ad2e92a852f043179c13c
treeb2c761287b8fd61e5d5463e9cd6a0c0530c4a7ed
parent6be0b4c63dfa90fba1acf5ea538faa4d61a1681a
cherry-pick from 7241: speed-up matrix44::setconcat when both inputs are
scale+translate by inlining the zero-assignments, rather than calling sk_bzero.

bench: matrix44_setconcat 2x faster

git-svn-id: http://skia.googlecode.com/svn/trunk@7251 2bbb7eff-a529-9590-31e7-b0007b416f81
src/utils/SkMatrix44.cpp