From: Mike Klein <mtklein@chromium.org>
Date: Tue, 23 May 2017 13:53:36 +0000 (-0400)
Subject: simpler portable round()
X-Git-Tag: accepted/tizen/5.0/unified/20181102.025319~36^2~58
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=e980174d0f3af8e497d9c38851114cc6a3c02e9a;p=platform%2Fupstream%2FlibSkiaSharp.git

simpler portable round()

Calling out lrintf() is probably kind of extreme...
adding half and truncating should be just fine.

Change-Id: Ifb22c1c1a9b764be153b769aad4671ca44311255
Reviewed-on: https://skia-review.googlesource.com/17708
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
---

diff --git a/src/jumper/SkJumper_vectors.h b/src/jumper/SkJumper_vectors.h
index 073b946..b993309 100644
--- a/src/jumper/SkJumper_vectors.h
+++ b/src/jumper/SkJumper_vectors.h
@@ -35,7 +35,7 @@
     SI F   rcp   (F v)          { return 1.0f / v; }
     SI F   rsqrt (F v)          { return 1.0f / sqrtf(v); }
     SI F    sqrt_(F v)          { return sqrtf(v); }
-    SI U32 round (F v, F scale) { return (uint32_t)lrintf(v*scale); }
+    SI U32 round (F v, F scale) { return (uint32_t)(v*scale + 0.5f); }
     SI U16 pack(U32 v)          { return (U16)v; }
     SI U8  pack(U16 v)          { return  (U8)v; }