From: Michael Niedermayer <michaelni@gmx.at>
Date: Mon, 5 Sep 2011 16:42:44 +0000 (+0200)
Subject: swscale: Fix unused asm and enable it
X-Git-Url: https://git.sesse.net/?a=commitdiff_plain;h=27bf599350a869877a1c9304abf9c95991692dd2;p=ffmpeg

swscale: Fix unused asm and enable it
Author of the fix is ronald, the enabling & commit message are mine.

This fixes
	commit 4e3e333a79272944b40695166438359b376d7864
	Author: Ronald S. Bultje <rsbultje@gmail.com>
	Date:   Tue Jul 5 12:49:11 2011 -0700

	    swscale: error dithering for 16/9/10-bit to 8-bit.

	    Based on a somewhat similar idea in FFmpeg's swscale copy.

The Fix was originally commited in: (and i missed it due to the commit message)
	commit 5c391a161a1f71dcdafd3f59b45d80220845f501
	Author: Ronald S. Bultje <rsbultje@gmail.com>
	Date:   Fri Jul 8 14:39:04 2011 -0700

	    swscale: rename uv_off/uv_off2 to uv_off_px/byte.
---

diff --git a/libswscale/x86/swscale_template.c b/libswscale/x86/swscale_template.c
index 6196e985619..cfaf847f938 100644
--- a/libswscale/x86/swscale_template.c
+++ b/libswscale/x86/swscale_template.c
@@ -192,11 +192,6 @@ static void RENAME(yuv2yuvX)(SwsContext *c, const int16_t *lumFilter,
 static av_always_inline void
 dither_8to32(SwsContext *c, const uint8_t *srcDither, int rot)
 {
-int i;
-if(rot) for(i=0; i<8; i++) c->dither32[i] = srcDither[(i+3)&7]<<12;
-else    for(i=0; i<8; i++) c->dither32[i] = srcDither[i&7]<<12;
-return;
-
     if (rot) {
         __asm__ volatile("pxor      %%mm0, %%mm0\n\t"
                          "movq       (%0), %%mm4\n\t"
@@ -213,10 +208,10 @@ return;
                          "punpckhwd %%mm0, %%mm5\n\t"
                          "punpcklwd %%mm0, %%mm6\n\t"
                          "punpckhwd %%mm0, %%mm7\n\t"
-                         "psllw       $12, %%mm4\n\t"
-                         "psllw       $12, %%mm5\n\t"
-                         "psllw       $12, %%mm6\n\t"
-                         "psllw       $12, %%mm7\n\t"
+                         "pslld       $12, %%mm4\n\t"
+                         "pslld       $12, %%mm5\n\t"
+                         "pslld       $12, %%mm6\n\t"
+                         "pslld       $12, %%mm7\n\t"
                          "movq      %%mm4, "DITHER32"+0(%1)\n\t"
                          "movq      %%mm5, "DITHER32"+8(%1)\n\t"
                          "movq      %%mm6, "DITHER32"+16(%1)\n\t"
@@ -235,10 +230,10 @@ return;
                          "punpckhwd %%mm0, %%mm5\n\t"
                          "punpcklwd %%mm0, %%mm6\n\t"
                          "punpckhwd %%mm0, %%mm7\n\t"
-                         "psllw       $12, %%mm4\n\t"
-                         "psllw       $12, %%mm5\n\t"
-                         "psllw       $12, %%mm6\n\t"
-                         "psllw       $12, %%mm7\n\t"
+                         "pslld       $12, %%mm4\n\t"
+                         "pslld       $12, %%mm5\n\t"
+                         "pslld       $12, %%mm6\n\t"
+                         "pslld       $12, %%mm7\n\t"
                          "movq      %%mm4, "DITHER32"+0(%1)\n\t"
                          "movq      %%mm5, "DITHER32"+8(%1)\n\t"
                          "movq      %%mm6, "DITHER32"+16(%1)\n\t"