(previous commit foired)

 . this is a coding style patch which removes all "foo(bar){" constructions
   and most of the tabulations.
 . also, fixed a bug in the default subtitle track.
 . and made a few error messages more explicit, ie. changed "error: %s" to
   "foo error: couldn't initialize bar (%s)"
 . and removed a warning in intf_ctrl.c which isn't used anyway.

plugins/sdl/vout_sdl.c

@@ -331,7 +331,7 @@ static int SDLOpenDisplay( vout_thread_t *p_vout )
         SDL_ShowCursor( 1 );
 
     SDL_WM_SetCaption( VOUT_TITLE , VOUT_TITLE );
-    SDL_EventState(SDL_KEYUP , SDL_IGNORE);	/* ignore keys up */
+    SDL_EventState(SDL_KEYUP , SDL_IGNORE); /* ignore keys up */
 
     if( p_vout->b_need_render )
     {
@@ -382,8 +382,8 @@ static int SDLOpenDisplay( vout_thread_t *p_vout )
 
     p_vout->i_changes |= VOUT_YUV_CHANGE;
 
-    /* Check buffers properties */	
-    p_vout->p_sys->b_must_acquire = 1;		/* always acquire */
+    /* Check buffers properties */
+    p_vout->p_sys->b_must_acquire = 1; /* always acquire */
 
     return( 0 );
 }

src/interface/intf_ctrl.c

@@ -498,8 +498,8 @@ static int SpawnInput( int i_argc, intf_arg_t *p_argv )
     p_main->p_intf->p_input = input_CreateThread( i_method, p_source, i_port, i_vlan_id,
                                                   p_main->p_intf->p_vout, p_main->p_aout,
                                                   NULL );
-    return( INTF_NO_ERROR );
 #endif
+    return( INTF_NO_ERROR );
 }
 
 /*****************************************************************************

src/lpcm_decoder/lpcm_decoder.h

@@ -24,9 +24,9 @@
 typedef struct lpcmdec_s lpcmdec_t;
 
 typedef struct lpcm_sync_info_s {
-    int sample_rate;	/* sample rate in Hz */
-    int frame_size;	/* frame size in bytes */
-    int bit_rate;	/* nominal bit rate in kbps */
+    int sample_rate;    /* sample rate in Hz */
+    int frame_size;     /* frame size in bytes */
+    int bit_rate;       /* nominal bit rate in kbps */
 } lpcm_sync_info_t;
 
 typedef struct lpcm_byte_stream_s {
@@ -57,7 +57,7 @@ struct lpcmdec_s {
      */
 
     /* The bit stream structure handles the PES stream at the bit level */
-    lpcm_bit_stream_t	bit_stream;
+    lpcm_bit_stream_t bit_stream;
 };
 
 

src/lpcm_decoder/lpcm_decoder_thread.c

@@ -2,7 +2,7 @@
  * lpcm_decoder_thread.c: lpcm decoder thread
  *****************************************************************************
  * Copyright (C) 1999, 2000 VideoLAN
- * $Id: lpcm_decoder_thread.c,v 1.7 2000/12/27 18:35:45 massiot Exp $
+ * $Id: lpcm_decoder_thread.c,v 1.8 2001/01/05 14:46:37 sam Exp $
  *
  * Authors:
  *
@@ -66,11 +66,11 @@ static void     EndThread               (lpcmdec_thread_t * p_adec);
 vlc_thread_t lpcmdec_CreateThread (adec_config_t * p_config)
 {
     lpcmdec_thread_t *   p_lpcmdec;
-    intf_DbgMsg ( "LPCM Debug: creating lpcm decoder thread" );
+    intf_DbgMsg ( "lpcm: creating lpcm decoder thread" );
 
     /* Allocate the memory needed to store the thread's structure */
     if ((p_lpcmdec = (lpcmdec_thread_t *)malloc (sizeof(lpcmdec_thread_t))) == NULL) {
-        intf_ErrMsg ( "LPCM Error: not enough memory for lpcmdec_CreateThread() to create the new thread" );
+        intf_ErrMsg ( "lpcm error: cannot create lpcmdec_thread_t" );
         return 0;
     }
 
@@ -92,7 +92,7 @@ vlc_thread_t lpcmdec_CreateThread (adec_config_t * p_config)
 
     /* Spawn the lpcm decoder thread */
     if (vlc_thread_create(&p_lpcmdec->thread_id, "lpcm decoder", (vlc_thread_func_t)RunThread, (void *)p_lpcmdec)) {
-        intf_ErrMsg  ( "LPCM Error: can't spawn lpcm decoder thread" );
+        intf_ErrMsg  ( "lpcm error: cannot spawn thread" );
         free (p_lpcmdec);
         return 0;
     }
@@ -171,38 +171,38 @@ static void RunThread (lpcmdec_thread_t * p_lpcmdec)
    
     while ((!p_lpcmdec->p_fifo->b_die) && (!p_lpcmdec->p_fifo->b_error))
     {
-	    s16 * buffer;
-	    lpcm_sync_info_t sync_info;
+        s16 * buffer;
+        lpcm_sync_info_t sync_info;
 
-	    if (!sync)
+        if (!sync)
         {
             /* have to find a synchro point */
         }
     
         if (DECODER_FIFO_START(*p_lpcmdec->p_fifo)->i_pts)
         {
-	        p_lpcmdec->p_aout_fifo->date[p_lpcmdec->p_aout_fifo->l_end_frame] = DECODER_FIFO_START(*p_lpcmdec->p_fifo)->i_pts;
-	        DECODER_FIFO_START(*p_lpcmdec->p_fifo)->i_pts = 0;
+            p_lpcmdec->p_aout_fifo->date[p_lpcmdec->p_aout_fifo->l_end_frame] = DECODER_FIFO_START(*p_lpcmdec->p_fifo)->i_pts;
+            DECODER_FIFO_START(*p_lpcmdec->p_fifo)->i_pts = 0;
         }
         else
         {
-	        p_lpcmdec->p_aout_fifo->date[p_lpcmdec->p_aout_fifo->l_end_frame] = LAST_MDATE;
+            p_lpcmdec->p_aout_fifo->date[p_lpcmdec->p_aout_fifo->l_end_frame] = LAST_MDATE;
         }
 
-    	p_lpcmdec->p_aout_fifo->l_rate = sync_info.sample_rate;
+        p_lpcmdec->p_aout_fifo->l_rate = sync_info.sample_rate;
 
-	    buffer = ((s16 *)p_lpcmdec->p_aout_fifo->buffer) + (p_lpcmdec->p_aout_fifo->l_end_frame * LPCMDEC_FRAME_SIZE);
+        buffer = ((s16 *)p_lpcmdec->p_aout_fifo->buffer) + (p_lpcmdec->p_aout_fifo->l_end_frame * LPCMDEC_FRAME_SIZE);
 
-	    if (lpcm_decode_frame (&p_lpcmdec->lpcm_decoder, buffer))
+        if (lpcm_decode_frame (&p_lpcmdec->lpcm_decoder, buffer))
         {
-	        sync = 0;
-	        goto bad_frame;
-	    }
-
-	    vlc_mutex_lock (&p_lpcmdec->p_aout_fifo->data_lock);
-	    p_lpcmdec->p_aout_fifo->l_end_frame = (p_lpcmdec->p_aout_fifo->l_end_frame + 1) & AOUT_FIFO_SIZE;
-	    vlc_cond_signal (&p_lpcmdec->p_aout_fifo->data_wait);
-	    vlc_mutex_unlock (&p_lpcmdec->p_aout_fifo->data_lock);
+            sync = 0;
+            goto bad_frame;
+        }
+
+        vlc_mutex_lock (&p_lpcmdec->p_aout_fifo->data_lock);
+        p_lpcmdec->p_aout_fifo->l_end_frame = (p_lpcmdec->p_aout_fifo->l_end_frame + 1) & AOUT_FIFO_SIZE;
+        vlc_cond_signal (&p_lpcmdec->p_aout_fifo->data_wait);
+        vlc_mutex_unlock (&p_lpcmdec->p_aout_fifo->data_lock);
 
         intf_DbgMsg( "LPCM Debug: %x", *buffer );
         bad_frame:

src/lpcm_decoder/lpcm_decoder_thread.h

@@ -2,7 +2,7 @@
  * lpcm_decoder_thread.h : lpcm decoder thread interface
  *****************************************************************************
  * Copyright (C) 1999, 2000 VideoLAN
- * $Id: lpcm_decoder_thread.h,v 1.2 2000/12/21 13:25:51 massiot Exp $
+ * $Id: lpcm_decoder_thread.h,v 1.3 2001/01/05 14:46:37 sam Exp $
  *
  * Authors:
  *
@@ -37,7 +37,7 @@ typedef struct lpcmdec_thread_s
      */
     decoder_fifo_t *    p_fifo;                /* stores the PES stream data */
     data_packet_t *     p_data;
-    int               	sync_ptr;         /* sync ptr from lpcm magic header */
+    int                 sync_ptr;         /* sync ptr from lpcm magic header */
     adec_config_t *     p_config;
 
     /*

src/misc/mtime.c

@@ -94,7 +94,7 @@ mtime_t mdate( void )
 void mwait( mtime_t date )
 {
 #ifdef HAVE_KERNEL_OS_H
-	
+
     mtime_t delay;
     
     delay = date - real_time_clock_usecs();

src/misc/netutils.c

@@ -94,7 +94,7 @@ int BuildInetAddr( struct sockaddr_in *p_sa_in, char *psz_in_addr, int i_port )
         intf_DbgMsg("debug: resolving internet address %s...", psz_in_addr);
         if ( (p_hostent = gethostbyname(psz_in_addr)) == NULL)
         {
-            intf_ErrMsg("error: unknown host %s", psz_in_addr);
+            intf_ErrMsg( "net error: unknown host %s", psz_in_addr );
             return( -1 );
         }
 
@@ -158,8 +158,8 @@ int ReadIfConf(int i_sockfd, if_descr_t* p_ifdescr, char* psz_name)
     }
     else
     {
-        intf_ErrMsg("Cannot read flags for interface %s: %s", psz_name,
-                    strerror(errno));
+        intf_ErrMsg( "net error: cannot read flags for interface %s (%s)",
+                     psz_name, strerror(errno) );
         return -1;
     }
 
@@ -182,8 +182,8 @@ int ReadIfConf(int i_sockfd, if_descr_t* p_ifdescr, char* psz_name)
     }
     else
     {
-        intf_ErrMsg("Cannot read hardware address for interface %s: %s",
-                    psz_name, strerror(errno));
+        intf_ErrMsg( "net error: cannot read hardware address for %s (%s)",
+                     psz_name, strerror(errno) );
         return -1;
     }
 
@@ -197,8 +197,8 @@ int ReadIfConf(int i_sockfd, if_descr_t* p_ifdescr, char* psz_name)
     }
     else
     {
-        intf_ErrMsg("Cannot read network address for interface %s: %s",
-                    psz_name, strerror(errno));
+        intf_ErrMsg( "net error: cannot read network address for %s (%s)",
+                     psz_name, strerror(errno) );
         return -1;
     }
 
@@ -221,8 +221,8 @@ int ReadIfConf(int i_sockfd, if_descr_t* p_ifdescr, char* psz_name)
     }
     else
     {
-        intf_ErrMsg("Cannot read broadcast address for interface %s: %s",
-                    psz_name, strerror(errno));
+        intf_ErrMsg( "net error: cannot read broadcast address for %s (%s)",
+                     psz_name, strerror(errno));
         return -1;
     }
 #endif
@@ -267,8 +267,8 @@ int ReadNetConf(int i_sockfd, net_descr_t* p_net_descr)
         i_rc = ioctl(i_sockfd, SIOCGIFCONF, (byte_t*)&ifc_netconf);
         if( i_rc )
         {
-            intf_ErrMsg("Cannot read network configuration: %s",
-                        strerror(errno));
+            intf_ErrMsg( "net error: cannot read network configuration (%s)",
+                         strerror(errno));
             break;
         }
     }
@@ -298,8 +298,8 @@ int ReadNetConf(int i_sockfd, net_descr_t* p_net_descr)
             /* Read the status of this interface */
             if( ioctl(i_sockfd, SIOCGIFFLAGS, (byte_t *)p_ifr_current_if) < 0 )
             {
-                intf_ErrMsg("Cannot access interface %s: %s",
-                            p_ifr_current_if->ifr_name, strerror(errno));
+                intf_ErrMsg( "net error: cannot access interface %s (%s)",
+                             p_ifr_current_if->ifr_name, strerror(errno) );
                 i_rc = -1;
                 break;
             }

src/misc/playlist.c

@@ -54,14 +54,14 @@ playlist_t * playlist_Create ( void )
     p_playlist->i_index = 0;
     p_playlist->p_list = NULL;
 
-    intf_Msg("Playlist initialized");
+    intf_Msg("Playlist created");
     return( p_playlist );
 }
 
 void playlist_Init( playlist_t * p_playlist, int i_optind )
 {
     int i_list_index = 0;
-    int i_index = 0;
+    int i_index;
     int i_argc = p_main->i_argc;
 
     if( i_optind < i_argc )
@@ -70,11 +70,17 @@ void playlist_Init( playlist_t * p_playlist, int i_optind )
 
         p_playlist->p_list = malloc( i_list_index * sizeof( int ) );
 
-        while( i_argc - i_index > i_optind )
+        for( i_index = 0 ; i_argc - i_index > i_optind ; i_index++ )
         {
-            p_playlist->p_list[ i_index ] =
-                            p_main->ppsz_argv[ i_argc - i_index - 1];
-            i_index++;
+            if( strncmp( p_main->ppsz_argv[ i_argc - i_index - 1], "-", 1 ) )
+            {
+                p_playlist->p_list[ i_index ] =
+                                p_main->ppsz_argv[ i_argc - i_index - 1];
+            }
+            else
+            {
+                p_playlist->p_list[ i_index ] = "/dev/stdin";
+            }
         }
     }
     else

src/misc/plugins.c

@@ -115,6 +115,7 @@ void bank_Init( plugin_bank_t * p_bank )
     
     /* Dummy plugin */
     SEEK_PLUGIN( "dummy" );
+    SEEK_PLUGIN( "null" );
 
 #undef SEEK_PLUGIN
 }
@@ -190,7 +191,7 @@ char * TestPlugin ( plugin_id_t *p_plugin_id, char * psz_name )
         }
 
 #ifndef SYS_BEOS
-        intf_WarnMsg( 1, "Plugin %s failed: %s", psz_plugin, dlerror() );
+        intf_ErrMsg( "plugin error: cannot open %s (%s)", psz_plugin, dlerror() );
 #endif
 
         free( psz_plugin );

src/spu_decoder/spu_decoder.c

@@ -263,7 +263,7 @@ static void RunThread( spudec_thread_t *p_spudec )
                                 p_spu->i_height = p_spu->i_y - ( (i_word << 8)
                                     | GetByte( &p_spudec->bit_stream ) ) + 1;
 
-				i_index += 6;
+                                i_index += 6;
                                 break;
                             case SPU_CMD_SET_OFFSETS:
                                 /* 06xxxxyyyy (byte offsets) */

src/video_decoder/vdec_idctmmx.S

@@ -2,7 +2,7 @@
  * vdec_idctmmx.S : MMX IDCT implementation
  *****************************************************************************
  * Copyright (C) 1999, 2000 VideoLAN
- * $Id: vdec_idctmmx.S,v 1.5 2000/12/21 17:19:52 massiot Exp $
+ * $Id: vdec_idctmmx.S,v 1.6 2001/01/05 14:46:37 sam Exp $
  *
  * Authors:
  *
@@ -166,173 +166,173 @@ vdec_IDCT:
                                 /* with the dequantization in iHuff module */
         pmulhw 8*12(%esi), %mm0         /* V12 */
         movq 8*4(%ecx), %mm1
-        pmulhw 8*4(%esi), %mm1		/* V4 */
+        pmulhw 8*4(%esi), %mm1          /* V4 */
         movq (%ecx), %mm3
-        psraw $1, %mm0			/* t64=t66 */
-        pmulhw (%esi), %mm3		/* V0 */
-        movq 8*8(%ecx), %mm5		/* duplicate V4 */
-        movq %mm1, %mm2			/* added 11/1/96 */
-        pmulhw 8*8(%esi),%mm5		/* V8 */
-        psubsw %mm0, %mm1		/* V16 */
-        pmulhw x5a825a825a825a82, %mm1	/* 23170 ->V18 */
-        paddsw %mm0, %mm2		/* V17 */
-        movq %mm2, %mm0			/* duplicate V17 */
-        psraw $1, %mm2			/* t75=t82 */
-        psraw $2, %mm0			/* t72 */
-        movq %mm3, %mm4			/* duplicate V0 */
-        paddsw %mm5, %mm3		/* V19 */
-        psubsw %mm5, %mm4		/* V20 ;mm5 free */
+        psraw $1, %mm0                  /* t64=t66 */
+        pmulhw (%esi), %mm3             /* V0 */
+        movq 8*8(%ecx), %mm5            /* duplicate V4 */
+        movq %mm1, %mm2                 /* added 11/1/96 */
+        pmulhw 8*8(%esi),%mm5           /* V8 */
+        psubsw %mm0, %mm1               /* V16 */
+        pmulhw x5a825a825a825a82, %mm1  /* 23170 ->V18 */
+        paddsw %mm0, %mm2               /* V17 */
+        movq %mm2, %mm0                 /* duplicate V17 */
+        psraw $1, %mm2                  /* t75=t82 */
+        psraw $2, %mm0                  /* t72 */
+        movq %mm3, %mm4                 /* duplicate V0 */
+        paddsw %mm5, %mm3               /* V19 */
+        psubsw %mm5, %mm4               /* V20 ;mm5 free */
 /* moved from the block below */
         movq 8*10(%ecx), %mm7
-        psraw $1, %mm3			/* t74=t81 */
-        movq %mm3, %mm6			/* duplicate t74=t81 */
-        psraw $2, %mm4			/* t77=t79 */
-        psubsw %mm0, %mm1		/* V21 ; mm0 free */
-        paddsw %mm2, %mm3		/* V22 */
-        movq %mm1, %mm5			/* duplicate V21 */
-        paddsw %mm4, %mm1		/* V23 */
-        movq %mm3, 8*4(%esi)		/* V22 */
-        psubsw %mm5, %mm4		/* V24; mm5 free */
-        movq %mm1, 8*12(%esi)		/* V23 */
-        psubsw %mm2, %mm6		/* V25; mm2 free */
-        movq %mm4, (%esi)		/* V24 */
+        psraw $1, %mm3                  /* t74=t81 */
+        movq %mm3, %mm6                 /* duplicate t74=t81 */
+        psraw $2, %mm4                  /* t77=t79 */
+        psubsw %mm0, %mm1               /* V21 ; mm0 free */
+        paddsw %mm2, %mm3               /* V22 */
+        movq %mm1, %mm5                 /* duplicate V21 */
+        paddsw %mm4, %mm1               /* V23 */
+        movq %mm3, 8*4(%esi)            /* V22 */
+        psubsw %mm5, %mm4               /* V24; mm5 free */
+        movq %mm1, 8*12(%esi)           /* V23 */
+        psubsw %mm2, %mm6               /* V25; mm2 free */
+        movq %mm4, (%esi)               /* V24 */
 /* keep mm6 alive all along the next block */
-        /* movq %mm6, 8*8(%esi) 	V25 */
+        /* movq %mm6, 8*8(%esi)         V25 */
 /* column 0: odd part
  * use V2, V6, V10, V14 to produce V31, V39, V40, V41
  */
 /* moved above: movq 8*10(%ecx), %mm7 */
 
-        pmulhw 8*10(%esi), %mm7		/* V10 */
+        pmulhw 8*10(%esi), %mm7         /* V10 */
         movq 8*6(%ecx), %mm0
-        pmulhw 8*6(%esi), %mm0		/* V6 */
+        pmulhw 8*6(%esi), %mm0          /* V6 */
         movq 8*2(%ecx), %mm5
-        movq %mm7, %mm3			/* duplicate V10 */
-        pmulhw 8*2(%esi), %mm5		/* V2 */
+        movq %mm7, %mm3                 /* duplicate V10 */
+        pmulhw 8*2(%esi), %mm5          /* V2 */
         movq 8*14(%ecx), %mm4
-        psubsw %mm0, %mm7		/* V26 */
-        pmulhw 8*14(%esi), %mm4		/* V14 */
-        paddsw %mm0, %mm3		/* V29 ; free mm0 */
-        movq %mm7, %mm1			/* duplicate V26 */
-        psraw $1, %mm3			/* t91=t94 */
-        pmulhw x539f539f539f539f,%mm7	/* V33 */
-        psraw $1, %mm1			/* t96 */
-        movq %mm5, %mm0			/* duplicate V2 */
-        psraw $2, %mm4			/* t85=t87 */
-        paddsw %mm4,%mm5		/* V27 */
-        psubsw %mm4, %mm0		/* V28 ; free mm4 */
-        movq %mm0, %mm2			/* duplicate V28 */
-        psraw $1, %mm5			/* t90=t93 */
-        pmulhw x4546454645464546,%mm0	/* V35 */
-        psraw $1, %mm2			/* t97 */
-        movq %mm5, %mm4			/* duplicate t90=t93 */
-        psubsw %mm2, %mm1		/* V32 ; free mm2 */
-        pmulhw x61f861f861f861f8,%mm1	/* V36 */
-        psllw $1, %mm7			/* t107 */
-        paddsw %mm3, %mm5		/* V31 */
-        psubsw %mm3, %mm4		/* V30 ; free mm3 */
-        pmulhw x5a825a825a825a82,%mm4	/* V34 */
+        psubsw %mm0, %mm7               /* V26 */
+        pmulhw 8*14(%esi), %mm4         /* V14 */
+        paddsw %mm0, %mm3               /* V29 ; free mm0 */
+        movq %mm7, %mm1                 /* duplicate V26 */
+        psraw $1, %mm3                  /* t91=t94 */
+        pmulhw x539f539f539f539f,%mm7   /* V33 */
+        psraw $1, %mm1                  /* t96 */
+        movq %mm5, %mm0                 /* duplicate V2 */
+        psraw $2, %mm4                  /* t85=t87 */
+        paddsw %mm4,%mm5                /* V27 */
+        psubsw %mm4, %mm0               /* V28 ; free mm4 */
+        movq %mm0, %mm2                 /* duplicate V28 */
+        psraw $1, %mm5                  /* t90=t93 */
+        pmulhw x4546454645464546,%mm0   /* V35 */
+        psraw $1, %mm2                  /* t97 */
+        movq %mm5, %mm4                 /* duplicate t90=t93 */
+        psubsw %mm2, %mm1               /* V32 ; free mm2 */
+        pmulhw x61f861f861f861f8,%mm1   /* V36 */
+        psllw $1, %mm7                  /* t107 */
+        paddsw %mm3, %mm5               /* V31 */
+        psubsw %mm3, %mm4               /* V30 ; free mm3 */
+        pmulhw x5a825a825a825a82,%mm4   /* V34 */
         nop
-        psubsw %mm1, %mm0		/* V38 */
-        psubsw %mm7, %mm1		/* V37 ; free mm7 */
-        psllw $1, %mm1			/* t114 */
+        psubsw %mm1, %mm0               /* V38 */
+        psubsw %mm7, %mm1               /* V37 ; free mm7 */
+        psllw $1, %mm1                  /* t114 */
 /* move from the next block */
-        movq %mm6, %mm3			/* duplicate V25 */
+        movq %mm6, %mm3                 /* duplicate V25 */
 /* move from the next block */
-        movq 8*4(%esi), %mm7		/* V22 */
-        psllw $1, %mm0			/* t110 */
-        psubsw %mm5, %mm0		/* V39 (mm5 needed for next block) */
-        psllw $2, %mm4			/* t112 */
+        movq 8*4(%esi), %mm7            /* V22 */
+        psllw $1, %mm0                  /* t110 */
+        psubsw %mm5, %mm0               /* V39 (mm5 needed for next block) */
+        psllw $2, %mm4                  /* t112 */
 /* moved from the next block */
-        movq 8*12(%esi), %mm2		/* V23 */
-        psubsw %mm0, %mm4		/* V40 */
-        paddsw %mm4, %mm1		/* V41; free mm0 */
+        movq 8*12(%esi), %mm2           /* V23 */
+        psubsw %mm0, %mm4               /* V40 */
+        paddsw %mm4, %mm1               /* V41; free mm0 */
 /* moved from the next block */
-        psllw $1, %mm2			/* t117=t125 */
+        psllw $1, %mm2                  /* t117=t125 */
 /* column 0: output butterfly */
 /* moved above:
- * movq %mm6, %mm3			duplicate V25
- * movq 8*4(%esi), %mm7			V22
- * movq 8*12(%esi), %mm2		V23
- * psllw $1, %mm2			t117=t125
+ * movq %mm6, %mm3                         duplicate V25
+ * movq 8*4(%esi), %mm7                    V22
+ * movq 8*12(%esi), %mm2                   V23
+ * psllw $1, %mm2                          t117=t125
  */
-        psubsw %mm1, %mm6		/* tm6 */
-        paddsw %mm1, %mm3		/* tm8; free mm1 */
-        movq %mm7, %mm1			/* duplicate V22 */
-        paddsw %mm5, %mm7		/* tm0 */
-        movq %mm3, 8*8(%esi)		/* tm8; free mm3 */
-        psubsw %mm5, %mm1		/* tm14; free mm5 */
-        movq %mm6, 8*6(%esi)		/* tm6; free mm6 */
-        movq %mm2, %mm3			/* duplicate t117=t125 */
-        movq (%esi), %mm6		/* V24 */
-        paddsw %mm0, %mm2		/* tm2 */
-        movq %mm7, (%esi)		/* tm0; free mm7 */
-        psubsw %mm0, %mm3		/* tm12; free mm0 */
-        movq %mm1, 8*14(%esi)		/* tm14; free mm1 */
-        psllw $1, %mm6			/* t119=t123 */
-        movq %mm2, 8*2(%esi)		/* tm2; free mm2 */
-        movq %mm6, %mm0			/* duplicate t119=t123 */
-        movq %mm3, 8*12(%esi)		/* tm12; free mm3 */
-        paddsw %mm4, %mm6		/* tm4 */
+        psubsw %mm1, %mm6               /* tm6 */
+        paddsw %mm1, %mm3               /* tm8; free mm1 */
+        movq %mm7, %mm1                 /* duplicate V22 */
+        paddsw %mm5, %mm7               /* tm0 */
+        movq %mm3, 8*8(%esi)            /* tm8; free mm3 */
+        psubsw %mm5, %mm1               /* tm14; free mm5 */
+        movq %mm6, 8*6(%esi)            /* tm6; free mm6 */
+        movq %mm2, %mm3                 /* duplicate t117=t125 */
+        movq (%esi), %mm6               /* V24 */
+        paddsw %mm0, %mm2               /* tm2 */
+        movq %mm7, (%esi)               /* tm0; free mm7 */
+        psubsw %mm0, %mm3               /* tm12; free mm0 */
+        movq %mm1, 8*14(%esi)           /* tm14; free mm1 */
+        psllw $1, %mm6                  /* t119=t123 */
+        movq %mm2, 8*2(%esi)            /* tm2; free mm2 */
+        movq %mm6, %mm0                 /* duplicate t119=t123 */
+        movq %mm3, 8*12(%esi)           /* tm12; free mm3 */
+        paddsw %mm4, %mm6               /* tm4 */
 /* moved from next block */
         movq 8*5(%ecx), %mm1
-        psubsw %mm4, %mm0		/* tm10; free mm4 */
+        psubsw %mm4, %mm0               /* tm10; free mm4 */
 /* moved from next block */
-        pmulhw 8*5(%esi), %mm1		/* V5 */
-        movq %mm6, 8*4(%esi)		/* tm4; free mm6 */
-        movq %mm0, 8*10(%esi)		/* tm10; free mm0 */
+        pmulhw 8*5(%esi), %mm1          /* V5 */
+        movq %mm6, 8*4(%esi)            /* tm4; free mm6 */
+        movq %mm0, 8*10(%esi)           /* tm10; free mm0 */
 /* column 1: even part
  * use V5, V13, V1, V9 to produce V56..V59
  */
 /* moved to prev block:
- *	movq 8*5(%ecx), %mm1
- *	pmulhw 8*5(%esi), %mm1		 V5
+ *        movq 8*5(%ecx), %mm1
+ *        pmulhw 8*5(%esi), %mm1                 V5
  */
         movq 8*13(%ecx), %mm7
-        psllw $1, %mm1			/* t128=t130 */
-        pmulhw 8*13(%esi), %mm7		/* V13 */
-        movq %mm1, %mm2			/* duplicate t128=t130 */
+        psllw $1, %mm1                  /* t128=t130 */
+        pmulhw 8*13(%esi), %mm7         /* V13 */
+        movq %mm1, %mm2                 /* duplicate t128=t130 */
         movq 8(%ecx), %mm3
-        pmulhw 8(%esi), %mm3		/* V1 */
+        pmulhw 8(%esi), %mm3            /* V1 */
         movq 8*9(%ecx), %mm5
-        psubsw %mm7, %mm1		/* V50 */
-        pmulhw 8*9(%esi), %mm5		/* V9 */
-        paddsw %mm7, %mm2		/* V51 */
-        pmulhw x5a825a825a825a82, %mm1	/* 23170 ->V52 */
-        movq %mm2, %mm6			/* duplicate V51 */
-        psraw $1, %mm2			/* t138=t144 */
-        movq %mm3, %mm4			/* duplicate V1 */
-        psraw $2, %mm6			/* t136 */
-        paddsw %mm5, %mm3		/* V53 */
-        psubsw %mm5, %mm4		/* V54 ;mm5 free */
-        movq %mm3, %mm7			/* duplicate V53 */
+        psubsw %mm7, %mm1               /* V50 */
+        pmulhw 8*9(%esi), %mm5          /* V9 */
+        paddsw %mm7, %mm2               /* V51 */
+        pmulhw x5a825a825a825a82, %mm1  /* 23170 ->V52 */
+        movq %mm2, %mm6                 /* duplicate V51 */
+        psraw $1, %mm2                  /* t138=t144 */
+        movq %mm3, %mm4                 /* duplicate V1 */
+        psraw $2, %mm6                  /* t136 */
+        paddsw %mm5, %mm3               /* V53 */
+        psubsw %mm5, %mm4               /* V54 ;mm5 free */
+        movq %mm3, %mm7                 /* duplicate V53 */
 /* moved from next block */
         movq 8*11(%ecx), %mm0
-        psraw $1, %mm4			/* t140=t142 */
-        psubsw %mm6, %mm1		/* V55 ; mm6 free */
-        paddsw %mm2, %mm3		/* V56 */
-        movq %mm4, %mm5			/* duplicate t140=t142 */
-        paddsw %mm1, %mm4		/* V57 */
-        movq %mm3, 8*5(%esi)		/* V56 */
-        psubsw %mm1, %mm5		/* V58; mm1 free */
-        movq %mm4, 8*13(%esi)		/* V57 */
-        psubsw %mm2, %mm7		/* V59; mm2 free */
-        movq %mm5, 8*9(%esi)		/* V58 */
+        psraw $1, %mm4                  /* t140=t142 */
+        psubsw %mm6, %mm1               /* V55 ; mm6 free */
+        paddsw %mm2, %mm3               /* V56 */
+        movq %mm4, %mm5                 /* duplicate t140=t142 */
+        paddsw %mm1, %mm4               /* V57 */
+        movq %mm3, 8*5(%esi)            /* V56 */
+        psubsw %mm1, %mm5               /* V58; mm1 free */
+        movq %mm4, 8*13(%esi)           /* V57 */
+        psubsw %mm2, %mm7               /* V59; mm2 free */
+        movq %mm5, 8*9(%esi)            /* V58 */
 /* keep mm7 alive all along the next block
- *	movq %mm7, 8(%esi)		V59
+ *        movq %mm7, 8(%esi)                V59
  * moved above
- *	movq 8*11(%ecx), %mm0
+ *        movq 8*11(%ecx), %mm0
  */
-        pmulhw 8*11(%esi), %mm0		/* V11 */
+        pmulhw 8*11(%esi), %mm0         /* V11 */
         movq 8*7(%ecx), %mm6
-        pmulhw 8*7(%esi), %mm6		/* V7 */
+        pmulhw 8*7(%esi), %mm6          /* V7 */
         movq 8*15(%ecx), %mm4
-        movq %mm0, %mm3			/* duplicate V11 */
-        pmulhw 8*15(%esi), %mm4		/* V15 */
+        movq %mm0, %mm3                 /* duplicate V11 */
+        pmulhw 8*15(%esi), %mm4         /* V15 */
         movq 8*3(%ecx), %mm5
-        psllw $1, %mm6			/* t146=t152 */
-        pmulhw 8*3(%esi), %mm5		/* V3 */
-        paddsw %mm6, %mm0		/* V63 */
+        psllw $1, %mm6                  /* t146=t152 */
+        pmulhw 8*3(%esi), %mm5          /* V3 */
+        paddsw %mm6, %mm0               /* V63 */
 /* note that V15 computation has a correction step:
  * this is a 'magic' constant that rebiases the results to be closer to the
  * expected result.  this magic constant can be refined to reduce the error
@@ -340,43 +340,43 @@ vdec_IDCT:
  * is actually multiplied by 16
  */
         paddw x0005000200010001, %mm4
-        psubsw %mm6, %mm3		/* V60 ; free mm6 */
-        psraw $1, %mm0			/* t154=t156 */
-        movq %mm3, %mm1			/* duplicate V60 */
-        pmulhw x539f539f539f539f, %mm1	/* V67 */
-        movq %mm5, %mm6			/* duplicate V3 */
-        psraw $2, %mm4			/* t148=t150 */
-        paddsw %mm4, %mm5		/* V61 */
-        psubsw %mm4, %mm6		/* V62 ; free mm4 */
-        movq %mm5, %mm4			/* duplicate V61 */
-        psllw $1, %mm1			/* t169 */
-        paddsw %mm0, %mm5		/* V65 -> result */
-        psubsw %mm0, %mm4		/* V64 ; free mm0 */
-        pmulhw x5a825a825a825a82, %mm4	/* V68 */
-        psraw $1, %mm3			/* t158 */
-        psubsw %mm6, %mm3		/* V66 */
-        movq %mm5, %mm2			/* duplicate V65 */
-        pmulhw x61f861f861f861f8, %mm3	/* V70 */
-        psllw $1, %mm6			/* t165 */
-        pmulhw x4546454645464546, %mm6	/* V69 */
-        psraw $1, %mm2			/* t172 */
+        psubsw %mm6, %mm3               /* V60 ; free mm6 */
+        psraw $1, %mm0                  /* t154=t156 */
+        movq %mm3, %mm1                 /* duplicate V60 */
+        pmulhw x539f539f539f539f, %mm1  /* V67 */
+        movq %mm5, %mm6                 /* duplicate V3 */
+        psraw $2, %mm4                  /* t148=t150 */
+        paddsw %mm4, %mm5               /* V61 */
+        psubsw %mm4, %mm6               /* V62 ; free mm4 */
+        movq %mm5, %mm4                 /* duplicate V61 */
+        psllw $1, %mm1                  /* t169 */
+        paddsw %mm0, %mm5               /* V65 -> result */
+        psubsw %mm0, %mm4               /* V64 ; free mm0 */
+        pmulhw x5a825a825a825a82, %mm4  /* V68 */
+        psraw $1, %mm3                  /* t158 */
+        psubsw %mm6, %mm3               /* V66 */
+        movq %mm5, %mm2                 /* duplicate V65 */
+        pmulhw x61f861f861f861f8, %mm3  /* V70 */
+        psllw $1, %mm6                  /* t165 */
+        pmulhw x4546454645464546, %mm6  /* V69 */
+        psraw $1, %mm2                  /* t172 */
 /* moved from next block */
-        movq 8*5(%esi), %mm0		/* V56 */
-        psllw $1, %mm4			/* t174 */
+        movq 8*5(%esi), %mm0            /* V56 */
+        psllw $1, %mm4                  /* t174 */
 /* moved from next block */
-        psraw $1, %mm0			/* t177=t188 */
+        psraw $1, %mm0                  /* t177=t188 */
         nop
-        psubsw %mm3, %mm6		/* V72 */
-        psubsw %mm1, %mm3		/* V71 ; free mm1 */
-        psubsw %mm2, %mm6		/* V73 ; free mm2 */
+        psubsw %mm3, %mm6               /* V72 */
+        psubsw %mm1, %mm3               /* V71 ; free mm1 */
+        psubsw %mm2, %mm6               /* V73 ; free mm2 */
 /* moved from next block */
-        psraw $1, %mm5			/* t178=t189 */
-        psubsw %mm6, %mm4		/* V74 */
+        psraw $1, %mm5                  /* t178=t189 */
+        psubsw %mm6, %mm4               /* V74 */
 /* moved from next block */
-        movq %mm0, %mm1			/* duplicate t177=t188 */
-        paddsw %mm4, %mm3		/* V75 */
+        movq %mm0, %mm1                 /* duplicate t177=t188 */
+        paddsw %mm4, %mm3               /* V75 */
 /* moved from next block */
-        paddsw %mm5, %mm0		/* tm1 */
+        paddsw %mm5, %mm0               /* tm1 */
 /* location
  *  5 - V56
  * 13 - V57
@@ -388,38 +388,38 @@ vdec_IDCT:
  *  X - V75, mm3
  * free mm0, mm1 & mm2
  * moved above
- *	movq 8*5(%esi), %mm0		V56
- *	psllw $1, %mm0			t177=t188 ! new !!
- *	psllw $1, %mm5			t178=t189 ! new !!
- *	movq %mm0, %mm1			duplicate t177=t188
- *	paddsw %mm5, %mm0		tm1
+ *        movq 8*5(%esi), %mm0             V56
+ *        psllw $1, %mm0                   t177=t188 ! new !!
+ *        psllw $1, %mm5                   t178=t189 ! new !!
+ *        movq %mm0, %mm1                  duplicate t177=t188
+ *        paddsw %mm5, %mm0                tm1
  */
-        movq 8*13(%esi), %mm2		/* V57 */
-        psubsw %mm5, %mm1		/* tm15; free mm5 */
-        movq %mm0, 8(%esi)		/* tm1; free mm0 */
-        psraw $1, %mm7			/* t182=t184 ! new !! */
+        movq 8*13(%esi), %mm2           /* V57 */
+        psubsw %mm5, %mm1               /* tm15; free mm5 */
+        movq %mm0, 8(%esi)              /* tm1; free mm0 */
+        psraw $1, %mm7                  /* t182=t184 ! new !! */
 /* save the store as used directly in the transpose
- *	movq %mm1, 120(%esi)		tm15; free mm1
+ *        movq %mm1, 120(%esi)             tm15; free mm1
  */
-        movq %mm7, %mm5			/* duplicate t182=t184 */
-        psubsw %mm3, %mm7		/* tm7 */
-        paddsw %mm3, %mm5		/* tm9; free mm3 */
-        movq 8*9(%esi), %mm0		/* V58 */
-        movq %mm2, %mm3			/* duplicate V57 */
-        movq %mm7, 8*7(%esi)		/* tm7; free mm7 */
-        psubsw %mm6, %mm3		/* tm13 */
-        paddsw %mm6, %mm2		/* tm3 ; free mm6 */
+        movq %mm7, %mm5                 /* duplicate t182=t184 */
+        psubsw %mm3, %mm7               /* tm7 */
+        paddsw %mm3, %mm5               /* tm9; free mm3 */
+        movq 8*9(%esi), %mm0            /* V58 */
+        movq %mm2, %mm3                 /* duplicate V57 */
+        movq %mm7, 8*7(%esi)            /* tm7; free mm7 */
+        psubsw %mm6, %mm3               /* tm13 */
+        paddsw %mm6, %mm2               /* tm3 ; free mm6 */
 /* moved up from the transpose */
         movq %mm3, %mm7
 /* moved up from the transpose */
         punpcklwd %mm1, %mm3
-        movq %mm0, %mm6			/* duplicate V58 */
-        movq %mm2, 8*3(%esi)		/* tm3; free mm2 */
-        paddsw %mm4, %mm0		/* tm5 */
-        psubsw %mm4, %mm6		/* tm11; free mm4 */
+        movq %mm0, %mm6                 /* duplicate V58 */
+        movq %mm2, 8*3(%esi)            /* tm3; free mm2 */
+        paddsw %mm4, %mm0               /* tm5 */
+        psubsw %mm4, %mm6               /* tm11; free mm4 */
 /* moved up from the transpose */
         punpckhwd %mm1, %mm7
-        movq %mm0, 8*5(%esi)		/* tm5; free mm0 */
+        movq %mm0, 8*5(%esi)            /* tm5; free mm0 */
 /* moved up from the transpose */
         movq %mm5, %mm2
 /* transpose - M4 part
@@ -432,14 +432,14 @@ vdec_IDCT:
  * scheduled before the transpose is done without stores, or use the faster
  * half mmword stores (when possible)
  */
-        movd %mm3, 8*9+4(%esi)		/* MS part of tmt9 */
+        movd %mm3, 8*9+4(%esi)          /* MS part of tmt9 */
         punpcklwd %mm6, %mm5
-        movd %mm7, 8*13+4(%esi)		/* MS part of tmt13 */
+        movd %mm7, 8*13+4(%esi)         /* MS part of tmt13 */
         punpckhwd %mm6, %mm2
-        movd %mm5, 8*9(%esi)		/* LS part of tmt9 */
-        punpckhdq %mm3, %mm5		/* free mm3 */
-        movd %mm2, 8*13(%esi)		/* LS part of tmt13 */
-        punpckhdq %mm7, %mm2		/* free mm7 */
+        movd %mm5, 8*9(%esi)            /* LS part of tmt9 */
+        punpckhdq %mm3, %mm5            /* free mm3 */
+        movd %mm2, 8*13(%esi)           /* LS part of tmt13 */
+        punpckhdq %mm7, %mm2            /* free mm7 */
 /* moved up from the M3 transpose */
         movq 8*8(%esi), %mm0
 /* moved up from the M3 transpose */
@@ -447,19 +447,19 @@ vdec_IDCT:
 /* moved up from the M3 transpose */
         movq %mm0, %mm3
 /* shuffle the rest of the data, and write it with 2 mmword writes */
-        movq %mm5, 8*11(%esi)		/* tmt11 */
+        movq %mm5, 8*11(%esi)           /* tmt11 */
 /* moved up from the M3 transpose */
         punpcklwd %mm1, %mm0
-        movq %mm2, 8*15(%esi)		/* tmt15 */
+        movq %mm2, 8*15(%esi)           /* tmt15 */
 /* moved up from the M3 transpose */
         punpckhwd %mm1, %mm3
 /* transpose - M3 part
  * moved up to previous code section
- *	movq 8*8(%esi), %mm0
- *	movq 8*10(%esi), %mm1
- *	movq %mm0, %mm3
- *	punpcklwd %mm1, %mm0
- *	punpckhwd %mm1, %mm3
+ *        movq 8*8(%esi), %mm0
+ *        movq 8*10(%esi), %mm1
+ *        movq %mm0, %mm3
+ *        punpcklwd %mm1, %mm0
+ *        punpckhwd %mm1, %mm3
  */
         movq 8*12(%esi), %mm6
         movq 8*14(%esi), %mm4
@@ -469,136 +469,136 @@ vdec_IDCT:
         movq %mm0, %mm1
         punpckhdq %mm6, %mm1
         movq %mm3, %mm7
-        punpckhwd %mm4, %mm2		/* free mm4 */
-        punpckldq %mm6, %mm0		/* free mm6 */
+        punpckhwd %mm4, %mm2            /* free mm4 */
+        punpckldq %mm6, %mm0            /* free mm6 */
 /* moved from next block */
-        movq 8*13(%esi), %mm4		/* tmt13 */
+        movq 8*13(%esi), %mm4           /* tmt13 */
         punpckldq %mm2, %mm3
-        punpckhdq %mm2, %mm7		/* free mm2 */
+        punpckhdq %mm2, %mm7            /* free mm2 */
 /* moved from next block */
-        movq %mm3, %mm5			/* duplicate tmt5 */
+        movq %mm3, %mm5                 /* duplicate tmt5 */
 /* column 1: even part (after transpose)
 * moved above
-*	movq %mm3, %mm5			duplicate tmt5
-*	movq 8*13(%esi), %mm4		tmt13
+*        movq %mm3, %mm5                   duplicate tmt5
+*        movq 8*13(%esi), %mm4             tmt13
 */
-        psubsw %mm4, %mm3		/* V134 */
-        pmulhw x5a825a825a825a82, %mm3	/* 23170 ->V136 */
-        movq 8*9(%esi), %mm6		/* tmt9 */
-        paddsw %mm4, %mm5		/* V135 ; mm4 free */
-        movq %mm0, %mm4			/* duplicate tmt1 */
-        paddsw %mm6, %mm0		/* V137 */
-        psubsw %mm6, %mm4		/* V138 ; mm6 free */
-        psllw $2, %mm3			/* t290 */
-        psubsw %mm5, %mm3		/* V139 */
-        movq %mm0, %mm6			/* duplicate V137 */
-        paddsw %mm5, %mm0		/* V140 */
-        movq %mm4, %mm2			/* duplicate V138 */
-        paddsw %mm3, %mm2		/* V141 */
-        psubsw %mm3, %mm4		/* V142 ; mm3 free */
-        movq %mm0, 8*9(%esi)		/* V140 */
-        psubsw %mm5, %mm6		/* V143 ; mm5 free */
+        psubsw %mm4, %mm3               /* V134 */
+        pmulhw x5a825a825a825a82, %mm3  /* 23170 ->V136 */
+        movq 8*9(%esi), %mm6            /* tmt9 */
+        paddsw %mm4, %mm5               /* V135 ; mm4 free */
+        movq %mm0, %mm4                 /* duplicate tmt1 */
+        paddsw %mm6, %mm0               /* V137 */
+        psubsw %mm6, %mm4               /* V138 ; mm6 free */
+        psllw $2, %mm3                  /* t290 */
+        psubsw %mm5, %mm3               /* V139 */
+        movq %mm0, %mm6                 /* duplicate V137 */
+        paddsw %mm5, %mm0               /* V140 */
+        movq %mm4, %mm2                 /* duplicate V138 */
+        paddsw %mm3, %mm2               /* V141 */
+        psubsw %mm3, %mm4               /* V142 ; mm3 free */
+        movq %mm0, 8*9(%esi)            /* V140 */
+        psubsw %mm5, %mm6               /* V143 ; mm5 free */
 /* moved from next block */
-        movq 8*11(%esi), %mm0		/* tmt11 */
-        movq %mm2, 8*13(%esi)		/* V141 */
+        movq 8*11(%esi), %mm0           /* tmt11 */
+        movq %mm2, 8*13(%esi)           /* V141 */
 /* moved from next block */
-        movq %mm0, %mm2			/* duplicate tmt11 */
+        movq %mm0, %mm2                 /* duplicate tmt11 */
 /* column 1: odd part (after transpose) */
 /* moved up to the prev block
- *	movq 8*11(%esi), %mm0		tmt11
- *	movq %mm0, %mm2			duplicate tmt11
+ *        movq 8*11(%esi), %mm0            tmt11
+ *        movq %mm0, %mm2                  duplicate tmt11
  */
-        movq 8*15(%esi), %mm5		/* tmt15 */
-        psubsw %mm7, %mm0		/* V144 */
-        movq %mm0, %mm3			/* duplicate V144 */
-        paddsw %mm7, %mm2		/* V147 ; free mm7 */
-        pmulhw x539f539f539f539f, %mm0	/* 21407-> V151 */
-        movq %mm1, %mm7			/* duplicate tmt3 */
-        paddsw %mm5, %mm7		/* V145 */
-        psubsw %mm5, %mm1		/* V146 ; free mm5 */
-        psubsw %mm1, %mm3		/* V150 */
-        movq %mm7, %mm5			/* duplicate V145 */
-        pmulhw x4546454645464546, %mm1	/* 17734-> V153 */
-        psubsw %mm2, %mm5		/* V148 */
-        pmulhw x61f861f861f861f8, %mm3	/* 25080-> V154 */
-        psllw $2, %mm0			/* t311 */
-        pmulhw x5a825a825a825a82, %mm5	/* 23170-> V152 */
-        paddsw %mm2, %mm7		/* V149 ; free mm2 */
-        psllw $1, %mm1			/* t313 */
-        nop	/* without the nop - freeze here for one clock */
-        movq %mm3, %mm2			/* duplicate V154 */
-        psubsw %mm0, %mm3		/* V155 ; free mm0 */
-        psubsw %mm2, %mm1		/* V156 ; free mm2 */
+        movq 8*15(%esi), %mm5           /* tmt15 */
+        psubsw %mm7, %mm0               /* V144 */
+        movq %mm0, %mm3                 /* duplicate V144 */
+        paddsw %mm7, %mm2               /* V147 ; free mm7 */
+        pmulhw x539f539f539f539f, %mm0  /* 21407-> V151 */
+        movq %mm1, %mm7                 /* duplicate tmt3 */
+        paddsw %mm5, %mm7               /* V145 */
+        psubsw %mm5, %mm1               /* V146 ; free mm5 */
+        psubsw %mm1, %mm3               /* V150 */
+        movq %mm7, %mm5                 /* duplicate V145 */
+        pmulhw x4546454645464546, %mm1  /* 17734-> V153 */
+        psubsw %mm2, %mm5               /* V148 */
+        pmulhw x61f861f861f861f8, %mm3  /* 25080-> V154 */
+        psllw $2, %mm0                  /* t311 */
+        pmulhw x5a825a825a825a82, %mm5  /* 23170-> V152 */
+        paddsw %mm2, %mm7               /* V149 ; free mm2 */
+        psllw $1, %mm1                  /* t313 */
+        nop        /* without the nop - freeze here for one clock */
+        movq %mm3, %mm2                 /* duplicate V154 */
+        psubsw %mm0, %mm3               /* V155 ; free mm0 */
+        psubsw %mm2, %mm1               /* V156 ; free mm2 */
 /* moved from the next block */
-        movq %mm6, %mm2			/* duplicate V143 */
+        movq %mm6, %mm2                 /* duplicate V143 */
 /* moved from the next block */
-        movq 8*13(%esi), %mm0		/* V141 */
-        psllw $1, %mm1			/* t315 */
-        psubsw %mm7, %mm1		/* V157 (keep V149) */
-        psllw $2, %mm5			/* t317 */
-        psubsw %mm1, %mm5		/* V158 */
-        psllw $1, %mm3			/* t319 */
-        paddsw %mm5, %mm3		/* V159 */
+        movq 8*13(%esi), %mm0           /* V141 */
+        psllw $1, %mm1                  /* t315 */
+        psubsw %mm7, %mm1               /* V157 (keep V149) */
+        psllw $2, %mm5                  /* t317 */
+        psubsw %mm1, %mm5               /* V158 */
+        psllw $1, %mm3                  /* t319 */
+        paddsw %mm5, %mm3               /* V159 */
 /* column 1: output butterfly (after transform)
  * moved to the prev block
- *	movq %mm6, %mm2			duplicate V143
- *	movq 8*13(%esi), %mm0		V141
+ *        movq %mm6, %mm2                  duplicate V143
+ *        movq 8*13(%esi), %mm0            V141
  */
-        psubsw %mm3, %mm2		/* V163 */
-        paddsw %mm3, %mm6		/* V164 ; free mm3 */
-        movq %mm4, %mm3			/* duplicate V142 */
-        psubsw %mm5, %mm4		/* V165 ; free mm5 */
-        movq %mm2, scratch7		/* out7 */
+        psubsw %mm3, %mm2               /* V163 */
+        paddsw %mm3, %mm6               /* V164 ; free mm3 */
+        movq %mm4, %mm3                 /* duplicate V142 */
+        psubsw %mm5, %mm4               /* V165 ; free mm5 */
+        movq %mm2, scratch7             /* out7 */
         psraw $4, %mm6
         psraw $4, %mm4
-        paddsw %mm5, %mm3		/* V162 */
-        movq 8*9(%esi), %mm2		/* V140 */
-        movq %mm0, %mm5			/* duplicate V141 */
+        paddsw %mm5, %mm3               /* V162 */
+        movq 8*9(%esi), %mm2            /* V140 */
+        movq %mm0, %mm5                 /* duplicate V141 */
 /* in order not to perculate this line up,
  * we read 72(%esi) very near to this location
  */
-        movq %mm6, 8*9(%esi)		/* out9 */
-        paddsw %mm1, %mm0		/* V161 */
-        movq %mm3, scratch5		/* out5 */
-        psubsw %mm1, %mm5		/* V166 ; free mm1 */
-        movq %mm4, 8*11(%esi)		/* out11 */
+        movq %mm6, 8*9(%esi)            /* out9 */
+        paddsw %mm1, %mm0               /* V161 */
+        movq %mm3, scratch5             /* out5 */
+        psubsw %mm1, %mm5               /* V166 ; free mm1 */
+        movq %mm4, 8*11(%esi)           /* out11 */
         psraw $4, %mm5
-        movq %mm0, scratch3		/* out3 */
-        movq %mm2, %mm4			/* duplicate V140 */
-        movq %mm5, 8*13(%esi)		/* out13 */
-        paddsw %mm7, %mm2		/* V160 */
+        movq %mm0, scratch3             /* out3 */
+        movq %mm2, %mm4                 /* duplicate V140 */
+        movq %mm5, 8*13(%esi)           /* out13 */
+        paddsw %mm7, %mm2               /* V160 */
 /* moved from the next block */
         movq 8(%esi), %mm0
-        psubsw %mm7, %mm4		/* V167 ; free mm7 */
+        psubsw %mm7, %mm4               /* V167 ; free mm7 */
 /* moved from the next block */
         movq 8*3(%esi), %mm7
         psraw $4, %mm4
-        movq %mm2, scratch1		/* out1 */
+        movq %mm2, scratch1             /* out1 */
 /* moved from the next block */
         movq %mm0, %mm1
-        movq %mm4, 8*15(%esi)		/* out15 */
+        movq %mm4, 8*15(%esi)           /* out15 */
 /* moved from the next block */
         punpcklwd %mm7, %mm0
 /* transpose - M2 parts
  * moved up to the prev block
- *	movq 8(%esi), %mm0
- *	movq 8*3(%esi), %mm7
- *	movq %mm0, %mm1
- *	punpcklwd %mm7, %mm0
+ *        movq 8(%esi), %mm0
+ *        movq 8*3(%esi), %mm7
+ *        movq %mm0, %mm1
+ *        punpcklwd %mm7, %mm0
  */
         movq 8*5(%esi), %mm5
         punpckhwd %mm7, %mm1
         movq 8*7(%esi), %mm4
         movq %mm5, %mm3
 /* shuffle the data and write the lower parts of the trasposed in 4 dwords */
-        movd %mm0, 8*8(%esi)		/* LS part of tmt8 */
+        movd %mm0, 8*8(%esi)            /* LS part of tmt8 */
         punpcklwd %mm4, %mm5
-        movd %mm1, 8*12(%esi)		/* LS part of tmt12 */
+        movd %mm1, 8*12(%esi)           /* LS part of tmt12 */
         punpckhwd %mm4, %mm3
-        movd %mm5, 8*8+4(%esi)		/* MS part of tmt8 */
-        punpckhdq %mm5, %mm0		/* tmt10 */
-        movd %mm3, 8*12+4(%esi)		/* MS part of tmt12 */
-        punpckhdq %mm3, %mm1		/* tmt14 */
+        movd %mm5, 8*8+4(%esi)          /* MS part of tmt8 */
+        punpckhdq %mm5, %mm0            /* tmt10 */
+        movd %mm3, 8*12+4(%esi)         /* MS part of tmt12 */
+        punpckhdq %mm3, %mm1            /* tmt14 */
 /* transpose - M1 parts */
         movq (%esi), %mm7
         movq 8*2(%esi), %mm2
@@ -606,203 +606,203 @@ vdec_IDCT:
         movq 8*4(%esi), %mm5
         punpcklwd %mm2, %mm7
         movq 8*6(%esi), %mm4
-        punpckhwd %mm2, %mm6		/* free mm2 */
+        punpckhwd %mm2, %mm6            /* free mm2 */
         movq %mm5, %mm3
         punpcklwd %mm4, %mm5
-        punpckhwd %mm4, %mm3		/* free mm4 */
+        punpckhwd %mm4, %mm3            /* free mm4 */
         movq %mm7, %mm2
         movq %mm6, %mm4
-        punpckldq %mm5, %mm7		/* tmt0 */
-        punpckhdq %mm5, %mm2		/* tmt2 ; free mm5 */
+        punpckldq %mm5, %mm7            /* tmt0 */
+        punpckhdq %mm5, %mm2            /* tmt2 ; free mm5 */
 /* shuffle the rest of the data, and write it with 2 mmword writes */
-        punpckldq %mm3, %mm6		/* tmt4 */
+        punpckldq %mm3, %mm6            /* tmt4 */
 /* moved from next block */
-        movq %mm2, %mm5			/* duplicate tmt2 */
-        punpckhdq %mm3, %mm4		/* tmt6 ; free mm3 */
+        movq %mm2, %mm5                 /* duplicate tmt2 */
+        punpckhdq %mm3, %mm4            /* tmt6 ; free mm3 */
 /* moved from next block */
-        movq %mm0, %mm3			/* duplicate tmt10 */
+        movq %mm0, %mm3                 /* duplicate tmt10 */
 /* column 0: odd part (after transpose)
  *moved up to prev block
- *	movq %mm0, %mm3			duplicate tmt10
- *	movq %mm2, %mm5			duplicate tmt2
+ *        movq %mm0, %mm3                  duplicate tmt10
+ *        movq %mm2, %mm5                  duplicate tmt2
  */
-        psubsw %mm4, %mm0		/* V110 */
-        paddsw %mm4, %mm3		/* V113 ; free mm4 */
-        movq %mm0, %mm4			/* duplicate V110 */
-        paddsw %mm1, %mm2		/* V111 */
-        pmulhw x539f539f539f539f, %mm0	/* 21407-> V117 */
-        psubsw %mm1, %mm5		/* V112 ; free mm1 */
-        psubsw %mm5, %mm4		/* V116 */
-        movq %mm2, %mm1			/* duplicate V111 */
-        pmulhw x4546454645464546, %mm5	/* 17734-> V119 */
-        psubsw %mm3, %mm2		/* V114 */
-        pmulhw x61f861f861f861f8, %mm4	/* 25080-> V120 */
-        paddsw %mm3, %mm1		/* V115 ; free mm3 */
-        pmulhw x5a825a825a825a82, %mm2	/* 23170-> V118 */
-        psllw $2, %mm0			/* t266 */
-        movq %mm1, (%esi)		/* save V115 */
-        psllw $1, %mm5			/* t268 */
-        psubsw %mm4, %mm5		/* V122 */
-        psubsw %mm0, %mm4		/* V121 ; free mm0 */
-        psllw $1, %mm5			/* t270 */
-        psubsw %mm1, %mm5		/* V123 ; free mm1 */
-        psllw $2, %mm2			/* t272 */
-        psubsw %mm5, %mm2		/* V124 (keep V123) */
-        psllw $1, %mm4			/* t274 */
-        movq %mm5, 8*2(%esi)		/* save V123 ; free mm5 */
-        paddsw %mm2, %mm4		/* V125 (keep V124) */
+        psubsw %mm4, %mm0               /* V110 */
+        paddsw %mm4, %mm3               /* V113 ; free mm4 */
+        movq %mm0, %mm4                 /* duplicate V110 */
+        paddsw %mm1, %mm2               /* V111 */
+        pmulhw x539f539f539f539f, %mm0  /* 21407-> V117 */
+        psubsw %mm1, %mm5               /* V112 ; free mm1 */
+        psubsw %mm5, %mm4               /* V116 */
+        movq %mm2, %mm1                 /* duplicate V111 */
+        pmulhw x4546454645464546, %mm5  /* 17734-> V119 */
+        psubsw %mm3, %mm2               /* V114 */
+        pmulhw x61f861f861f861f8, %mm4  /* 25080-> V120 */
+        paddsw %mm3, %mm1               /* V115 ; free mm3 */
+        pmulhw x5a825a825a825a82, %mm2  /* 23170-> V118 */
+        psllw $2, %mm0                  /* t266 */
+        movq %mm1, (%esi)               /* save V115 */
+        psllw $1, %mm5                  /* t268 */
+        psubsw %mm4, %mm5               /* V122 */
+        psubsw %mm0, %mm4               /* V121 ; free mm0 */
+        psllw $1, %mm5                  /* t270 */
+        psubsw %mm1, %mm5               /* V123 ; free mm1 */
+        psllw $2, %mm2                  /* t272 */
+        psubsw %mm5, %mm2               /* V124 (keep V123) */
+        psllw $1, %mm4                  /* t274 */
+        movq %mm5, 8*2(%esi)            /* save V123 ; free mm5 */
+        paddsw %mm2, %mm4               /* V125 (keep V124) */
 /* column 0: even part (after transpose) */
-        movq 8*12(%esi), %mm0		/* tmt12 */
-        movq %mm6, %mm3			/* duplicate tmt4 */
-        psubsw %mm0, %mm6		/* V100 */
-        paddsw %mm0, %mm3		/* V101 ; free mm0 */
-        pmulhw x5a825a825a825a82, %mm6	/* 23170 ->V102 */
-        movq %mm7, %mm5			/* duplicate tmt0 */
-        movq 8*8(%esi), %mm1		/* tmt8 */
-        paddsw %mm1, %mm7		/* V103 */
-        psubsw %mm1, %mm5		/* V104 ; free mm1 */
-        movq %mm7, %mm0			/* duplicate V103 */
-        psllw $2, %mm6			/* t245 */
-        paddsw %mm3, %mm7		/* V106 */
-        movq %mm5, %mm1			/* duplicate V104 */
-        psubsw %mm3, %mm6		/* V105 */
-        psubsw %mm3, %mm0		/* V109; free mm3 */
-        paddsw %mm6, %mm5		/* V107 */
-        psubsw %mm6, %mm1		/* V108 ; free mm6 */
+        movq 8*12(%esi), %mm0           /* tmt12 */
+        movq %mm6, %mm3                 /* duplicate tmt4 */
+        psubsw %mm0, %mm6               /* V100 */
+        paddsw %mm0, %mm3               /* V101 ; free mm0 */
+        pmulhw x5a825a825a825a82, %mm6  /* 23170 ->V102 */
+        movq %mm7, %mm5                 /* duplicate tmt0 */
+        movq 8*8(%esi), %mm1            /* tmt8 */
+        paddsw %mm1, %mm7               /* V103 */
+        psubsw %mm1, %mm5               /* V104 ; free mm1 */
+        movq %mm7, %mm0                 /* duplicate V103 */
+        psllw $2, %mm6                  /* t245 */
+        paddsw %mm3, %mm7               /* V106 */
+        movq %mm5, %mm1                 /* duplicate V104 */
+        psubsw %mm3, %mm6               /* V105 */
+        psubsw %mm3, %mm0               /* V109; free mm3 */
+        paddsw %mm6, %mm5               /* V107 */
+        psubsw %mm6, %mm1               /* V108 ; free mm6 */
 /* column 0: output butterfly (after transform) */
-        movq %mm1, %mm3			/* duplicate V108 */
-        paddsw %mm2, %mm1		/* out4 */
+        movq %mm1, %mm3                 /* duplicate V108 */
+        paddsw %mm2, %mm1               /* out4 */
         psraw $4, %mm1
-        psubsw %mm2, %mm3		/* out10 ; free mm2 */
+        psubsw %mm2, %mm3               /* out10 ; free mm2 */
         psraw $4, %mm3
-        movq %mm0, %mm6			/* duplicate V109 */
-        movq %mm1, 8*4(%esi)		/* out4 ; free mm1 */
-        psubsw %mm4, %mm0		/* out6 */
-        movq %mm3, 8*10(%esi)		/* out10 ; free mm3 */
+        movq %mm0, %mm6                 /* duplicate V109 */
+        movq %mm1, 8*4(%esi)            /* out4 ; free mm1 */
+        psubsw %mm4, %mm0               /* out6 */
+        movq %mm3, 8*10(%esi)           /* out10 ; free mm3 */
         psraw $4, %mm0
-        paddsw %mm4, %mm6		/* out8 ; free mm4 */
-        movq %mm7, %mm1			/* duplicate V106 */
-        movq %mm0, 8*6(%esi)		/* out6 ; free mm0 */
+        paddsw %mm4, %mm6               /* out8 ; free mm4 */
+        movq %mm7, %mm1                 /* duplicate V106 */
+        movq %mm0, 8*6(%esi)            /* out6 ; free mm0 */
         psraw $4, %mm6
-        movq (%esi), %mm4		/* V115 */
-        movq %mm6, 8*8(%esi)		/* out8 ; free mm6 */
-        movq %mm5, %mm2			/* duplicate V107 */
-        movq 8*2(%esi), %mm3		/* V123 */
-        paddsw %mm4, %mm7		/* out0 */
+        movq (%esi), %mm4               /* V115 */
+        movq %mm6, 8*8(%esi)            /* out8 ; free mm6 */
+        movq %mm5, %mm2                 /* duplicate V107 */
+        movq 8*2(%esi), %mm3            /* V123 */
+        paddsw %mm4, %mm7               /* out0 */
 /* moved up from next block */
         movq scratch3, %mm0
         psraw $4, %mm7
 /* moved up from next block */
         movq scratch5, %mm6
-        psubsw %mm4, %mm1		/* out14 ; free mm4 */
-        paddsw %mm3, %mm5		/* out2 */
+        psubsw %mm4, %mm1               /* out14 ; free mm4 */
+        paddsw %mm3, %mm5               /* out2 */
         psraw $4, %mm1
-        movq %mm7, (%esi)		/* out0 ; free mm7 */
+        movq %mm7, (%esi)               /* out0 ; free mm7 */
         psraw $4, %mm5
-        movq %mm1, 8*14(%esi)		/* out14 ; free mm1 */
-        psubsw %mm3, %mm2		/* out12 ; free mm3 */
-        movq %mm5, 8*2(%esi)		/* out2 ; free mm5 */
+        movq %mm1, 8*14(%esi)           /* out14 ; free mm1 */
+        psubsw %mm3, %mm2               /* out12 ; free mm3 */
+        movq %mm5, 8*2(%esi)            /* out2 ; free mm5 */
         psraw $4, %mm2
 /* moved up to the prev block */
         movq scratch7, %mm4
 /* moved up to the prev block */
         psraw $4, %mm0
-        movq %mm2, 8*12(%esi)		/* out12 ; free mm2 */
+        movq %mm2, 8*12(%esi)           /* out12 ; free mm2 */
 /* moved up to the prev block */
         psraw $4, %mm6
 /* move back the data to its correct place
 * moved up to the prev block
- *	movq scratch3, %mm0
- *	movq scratch5, %mm6
- *	movq scratch7, %mm4
- *	psraw $4, %mm0
- *	psraw $4, %mm6
+ *        movq scratch3, %mm0
+ *        movq scratch5, %mm6
+ *        movq scratch7, %mm4
+ *        psraw $4, %mm0
+ *        psraw $4, %mm6
 */
         movq scratch1, %mm1
         psraw $4, %mm4
-        movq %mm0, 8*3(%esi)		/* out3 */
+        movq %mm0, 8*3(%esi)            /* out3 */
         psraw $4, %mm1
-        movq %mm6, 8*5(%esi)		/* out5 */
-        movq %mm4, 8*7(%esi)		/* out7 */
-        movq %mm1, 8(%esi)		/* out1 */
+        movq %mm6, 8*5(%esi)            /* out5 */
+        movq %mm4, 8*7(%esi)            /* out7 */
+        movq %mm1, 8(%esi)              /* out1 */
 /* transpose matrix */
-        movl $8, %ebx			/* ebx is x_size */
-        movl %esi, %edi			/* pointer to the matrix */
+        movl $8, %ebx                   /* ebx is x_size */
+        movl %esi, %edi                 /* pointer to the matrix */
         movl %ebx, %ecx
         sal $2, %ecx
         movl %ebx, %eax
         addl %ebx, %ecx
-        subl $4, %eax			/* eax is inner loop variable */
-        addl %ebx, %ecx			/* ecx is 6*row size */
-        movl %eax, %edx			/* edx is the outer loop variable */
-.L1:	movq (%esi), %mm0		/* first line */
-        movq (%esi,%ebx,4), %mm2	/* third line */
-        movq %mm0, %mm6			/* copy first line */
-        punpcklwd (%esi,%ebx,2), %mm0	/* interleave fist and second lines */
-        movq %mm2, %mm7			/* copy third line */
-        punpcklwd (%esi,%ecx), %mm2	/* interleave third and fourth lines */
-        movq %mm0, %mm4			/* copy first intermediate result */
-        movq (%esi,%ebx,2), %mm1	/* second line */
+        subl $4, %eax                   /* eax is inner loop variable */
+        addl %ebx, %ecx                 /* ecx is 6*row size */
+        movl %eax, %edx                 /* edx is the outer loop variable */
+.L1:        movq (%esi), %mm0           /* first line */
+        movq (%esi,%ebx,4), %mm2        /* third line */
+        movq %mm0, %mm6                 /* copy first line */
+        punpcklwd (%esi,%ebx,2), %mm0   /* interleave fist and second lines */
+        movq %mm2, %mm7                 /* copy third line */
+        punpcklwd (%esi,%ecx), %mm2     /* interleave third and fourth lines */
+        movq %mm0, %mm4                 /* copy first intermediate result */
+        movq (%esi,%ebx,2), %mm1        /* second line */
 /* the next line 'punpcklwd %mm2, %mm0' inverted two pixels. */
 /* punpckldq make printing cleaner */
-        punpckldq %mm2, %mm0		/* interleave to produce result 1 */
-        movq (%esi,%ecx), %mm3		/* fourth line */
-        punpckhdq %mm2, %mm4		/* interleave to produce result 2 */
-        movq %mm0, (%esi)		/* write result 1 */
-        punpckhwd %mm1, %mm6		/* interleave first and second lines */
-        movq %mm4, (%esi,%ebx,2)	/* write result 2 */
-        punpckhwd %mm3, %mm7		/* interleave 3rd and 4th lines */
-        movq %mm6, %mm5			/* copy first intermediate result */
-        punpckldq %mm7, %mm6		/* interleave to produce result 3 */
-        leal (%edi,%ebx,8), %edi	/* point to 4x4 set 4 rows down */
-        punpckhdq %mm7, %mm5		/* interleave to produce result 4 */
-        movq %mm6, (%esi,%ebx,4)	/* write result 3 */
-        movq %mm5, (%esi,%ecx)		/* write result 4 */
+        punpckldq %mm2, %mm0            /* interleave to produce result 1 */
+        movq (%esi,%ecx), %mm3          /* fourth line */
+        punpckhdq %mm2, %mm4            /* interleave to produce result 2 */
+        movq %mm0, (%esi)               /* write result 1 */
+        punpckhwd %mm1, %mm6            /* interleave first and second lines */
+        movq %mm4, (%esi,%ebx,2)        /* write result 2 */
+        punpckhwd %mm3, %mm7            /* interleave 3rd and 4th lines */
+        movq %mm6, %mm5                 /* copy first intermediate result */
+        punpckldq %mm7, %mm6            /* interleave to produce result 3 */
+        leal (%edi,%ebx,8), %edi        /* point to 4x4 set 4 rows down */
+        punpckhdq %mm7, %mm5            /* interleave to produce result 4 */
+        movq %mm6, (%esi,%ebx,4)        /* write result 3 */
+        movq %mm5, (%esi,%ecx)          /* write result 4 */
         /* check to see if number of rows left is zero */
         cmpl $0, %edx
         /* last time through you are done and ready to exit */
         je .L3
-.L2:	movq 8(%esi), %mm0		/* first line */
-        movq 8(%esi,%ebx,4), %mm2	/* third line */
-        movq %mm0, %mm6			/* copy first line */
-        punpcklwd 8(%esi,%ebx,2), %mm0	/* interleave first and second lines */
-        movq %mm2, %mm7			/* copy third line */
-        punpcklwd 8(%esi,%ecx), %mm2	/* interleave 3rd and 4th lines */
-        movq %mm0, %mm4			/* copy first intermediate */
-        movq (%edi), %mm1		/* first line */
-        punpckldq %mm2, %mm0		/* interleave to produce 1st result */
-        movq (%edi,%ebx,4), %mm3	/* third line */
-        punpckhdq %mm2, %mm4		/* interleave to produce 2nd result */
-        punpckhwd 8(%esi,%ebx,2), %mm6	/* interleave 1st and 2nd lines */
-        movq %mm1, %mm2			/* copy first line */
-        punpckhwd 8(%esi,%ecx), %mm7	/* interleave 3rd and 4th lines */
-        movq %mm6, %mm5			/* copy first intermediate */
-        movq %mm0, (%edi)		/* write result 1 */
-        punpckhdq %mm7, %mm5		/* produce third result */
-        punpcklwd (%edi,%ebx,2), %mm1	/* interleave 1st and 2nd lines */
-        movq %mm3, %mm0			/* copy third line */
-        punpckhwd (%edi,%ebx,2), %mm2	/* interleave 1st and 2nd lines */
-        movq %mm4, (%edi,%ebx,2)	/* write result 2 */
-        punpckldq %mm7, %mm6		/* produce fourth result */
-        punpcklwd (%edi,%ecx), %mm3	/* interleave 3rd and 4th lines */
-        movq %mm1, %mm4			/* copy first intermediate */
-        movq %mm6, (%edi,%ebx,4)	/* write result 3 */
+.L2:        movq 8(%esi), %mm0          /* first line */
+        movq 8(%esi,%ebx,4), %mm2       /* third line */
+        movq %mm0, %mm6                 /* copy first line */
+        punpcklwd 8(%esi,%ebx,2), %mm0  /* interleave first and second lines */
+        movq %mm2, %mm7                 /* copy third line */
+        punpcklwd 8(%esi,%ecx), %mm2    /* interleave 3rd and 4th lines */
+        movq %mm0, %mm4                 /* copy first intermediate */
+        movq (%edi), %mm1               /* first line */
+        punpckldq %mm2, %mm0            /* interleave to produce 1st result */
+        movq (%edi,%ebx,4), %mm3        /* third line */
+        punpckhdq %mm2, %mm4            /* interleave to produce 2nd result */
+        punpckhwd 8(%esi,%ebx,2), %mm6  /* interleave 1st and 2nd lines */
+        movq %mm1, %mm2                 /* copy first line */
+        punpckhwd 8(%esi,%ecx), %mm7    /* interleave 3rd and 4th lines */
+        movq %mm6, %mm5                 /* copy first intermediate */
+        movq %mm0, (%edi)               /* write result 1 */
+        punpckhdq %mm7, %mm5            /* produce third result */
+        punpcklwd (%edi,%ebx,2), %mm1   /* interleave 1st and 2nd lines */
+        movq %mm3, %mm0                 /* copy third line */
+        punpckhwd (%edi,%ebx,2), %mm2   /* interleave 1st and 2nd lines */
+        movq %mm4, (%edi,%ebx,2)        /* write result 2 */
+        punpckldq %mm7, %mm6            /* produce fourth result */
+        punpcklwd (%edi,%ecx), %mm3     /* interleave 3rd and 4th lines */
+        movq %mm1, %mm4                 /* copy first intermediate */
+        movq %mm6, (%edi,%ebx,4)        /* write result 3 */
         punpckldq %mm3, %mm1
-        punpckhwd (%edi,%ecx), %mm0	/* interleave 3rd and 4th lines */
-        movq %mm2, %mm6			/* copy second intermediate */
-        movq %mm5, (%edi,%ecx)		/* write result 4 */
-        punpckhdq %mm3, %mm4		/* produce second result */
-        movq %mm1, 8(%esi)		/* write result 5 */
-        punpckldq %mm0, %mm2		/* produce third result */
-        movq %mm4, 8(%esi,%ebx,2)	/* write result 6 */
-        punpckhdq %mm0, %mm6		/* produce fourth result */
-        movq %mm2, 8(%esi,%ebx,4)	/* write result 7 */
-        movq %mm6, 8(%esi,%ecx)		/* write result 8 */
+        punpckhwd (%edi,%ecx), %mm0     /* interleave 3rd and 4th lines */
+        movq %mm2, %mm6                 /* copy second intermediate */
+        movq %mm5, (%edi,%ecx)          /* write result 4 */
+        punpckhdq %mm3, %mm4            /* produce second result */
+        movq %mm1, 8(%esi)              /* write result 5 */
+        punpckldq %mm0, %mm2            /* produce third result */
+        movq %mm4, 8(%esi,%ebx,2)       /* write result 6 */
+        punpckhdq %mm0, %mm6            /* produce fourth result */
+        movq %mm2, 8(%esi,%ebx,4)       /* write result 7 */
+        movq %mm6, 8(%esi,%ecx)         /* write result 8 */
         /* increment %esi to point to next 4x4 block in same row */
         addl $8, %esi
         /* increment %edi to point to nxt 4x4 block below current */
         leal (%edi,%ebx,8), %edi
-        sub $4, %eax			/* decrement inner loop var */
+        sub $4, %eax                    /* decrement inner loop var */
         jnz .L2
         /* %edi points to start of second row in block just finished */
         sal $1, %edx
@@ -810,13 +810,13 @@ vdec_IDCT:
         subl %edx, %esi
         /* subtract the number of bytes in last row */
         /* now we point to spot where row=col */
-        subl $8, %edx			/* sub 4 from row number */
+        subl $8, %edx                   /* sub 4 from row number */
         sarl $1, %edx
         mov %esi, %edi
         mov %edx, %eax
         /* reset x_size to outer loop variable to start new row */
         jmp .L1
-.L3:	emms
+.L3:        emms
         popl %edi
         popl %esi
         popl %edx

src/video_decoder/vdec_motion.c

@@ -2,7 +2,7 @@
  * vdec_motion.c : motion compensation routines
  *****************************************************************************
  * Copyright (C) 1999, 2000 VideoLAN
- * $Id: vdec_motion.c,v 1.33 2000/12/22 13:04:45 sam Exp $
+ * $Id: vdec_motion.c,v 1.34 2001/01/05 14:46:37 sam Exp $
  *
  * Authors: Christophe Massiot <massiot@via.ecp.fr>
  *          Jean-Marc Dressler <polux@via.ecp.fr>
@@ -55,71 +55,71 @@
 #include "video_parser.h"
 #include "video_fifo.h"
 
-#define __MotionComponents(width,height)		\
-void MotionComponent_x_y_copy_##width##_##height ();	\
-void MotionComponent_X_y_copy_##width##_##height ();	\
-void MotionComponent_x_Y_copy_##width##_##height ();	\
-void MotionComponent_X_Y_copy_##width##_##height ();	\
-void MotionComponent_x_y_avg_##width##_##height ();	\
-void MotionComponent_X_y_avg_##width##_##height ();	\
-void MotionComponent_x_Y_avg_##width##_##height ();	\
+#define __MotionComponents(width,height)                \
+void MotionComponent_x_y_copy_##width##_##height ();    \
+void MotionComponent_X_y_copy_##width##_##height ();    \
+void MotionComponent_x_Y_copy_##width##_##height ();    \
+void MotionComponent_X_Y_copy_##width##_##height ();    \
+void MotionComponent_x_y_avg_##width##_##height ();     \
+void MotionComponent_X_y_avg_##width##_##height ();     \
+void MotionComponent_x_Y_avg_##width##_##height ();     \
 void MotionComponent_X_Y_avg_##width##_##height ();
 
-__MotionComponents (16,16)	/* 444, 422, 420 */
-__MotionComponents (16,8)	/* 444, 422, 420 */
-__MotionComponents (8,8)	/* 422, 420 */
-__MotionComponents (8,4)	/* 420 */
+__MotionComponents (16,16)        /* 444, 422, 420 */
+__MotionComponents (16,8)        /* 444, 422, 420 */
+__MotionComponents (8,8)        /* 422, 420 */
+__MotionComponents (8,4)        /* 420 */
 #if 0
-__MotionComponents (8,16)	/* 422 */
+__MotionComponents (8,16)        /* 422 */
 #endif
 
-#define ___callTheRightOne(width,height)				     \
-    if ((i_width == width) && (i_height == height))			     \
-    {									     \
-	if (!b_average)							     \
-	{								     \
-	    switch (i_select)						     \
-	    {								     \
-	    case 0:							     \
-		MotionComponent_x_y_copy_##width##_##height (p_src, p_dest,  \
-							     i_stride);	     \
-		break;							     \
-	    case 1:							     \
-		MotionComponent_X_y_copy_##width##_##height (p_src, p_dest,  \
-							     i_stride);	     \
-		break;							     \
-	    case 2:							     \
-		MotionComponent_x_Y_copy_##width##_##height (p_src, p_dest,  \
-							     i_stride);      \
-		break;							     \
-	    case 3:							     \
-		MotionComponent_X_Y_copy_##width##_##height (p_src, p_dest,  \
-							     i_stride);	     \
-		break;							     \
-	    }								     \
-	}								     \
-	else								     \
-	{								     \
-	    switch (i_select)						     \
-	    {								     \
-	    case 0:							     \
-		MotionComponent_x_y_avg_##width##_##height (p_src, p_dest,   \
-							    i_stride);	     \
-		break;							     \
-	    case 1:							     \
-		MotionComponent_X_y_avg_##width##_##height (p_src, p_dest,   \
-							    i_stride);	     \
-		break;							     \
-	    case 2:							     \
-		MotionComponent_x_Y_avg_##width##_##height (p_src, p_dest,   \
-							    i_stride);       \
-		break;							     \
-	    case 3:							     \
-		MotionComponent_X_Y_avg_##width##_##height (p_src, p_dest,   \
-							    i_stride);	     \
-		break;							     \
-	    }								     \
-	}								     \
+#define ___callTheRightOne(width,height)                                     \
+    if ((i_width == width) && (i_height == height))                          \
+    {                                                                        \
+        if (!b_average)                                                      \
+        {                                                                    \
+            switch (i_select)                                                \
+            {                                                                \
+            case 0:                                                          \
+                MotionComponent_x_y_copy_##width##_##height (p_src, p_dest,  \
+                                                             i_stride);      \
+                break;                                                       \
+            case 1:                                                          \
+                MotionComponent_X_y_copy_##width##_##height (p_src, p_dest,  \
+                                                             i_stride);      \
+                break;                                                       \
+            case 2:                                                          \
+                MotionComponent_x_Y_copy_##width##_##height (p_src, p_dest,  \
+                                                             i_stride);      \
+                break;                                                       \
+            case 3:                                                          \
+                MotionComponent_X_Y_copy_##width##_##height (p_src, p_dest,  \
+                                                             i_stride);      \
+                break;                                                       \
+            }                                                                \
+        }                                                                    \
+        else                                                                 \
+        {                                                                    \
+            switch (i_select)                                                \
+            {                                                                \
+            case 0:                                                          \
+                MotionComponent_x_y_avg_##width##_##height (p_src, p_dest,   \
+                                                            i_stride);       \
+                break;                                                       \
+            case 1:                                                          \
+                MotionComponent_X_y_avg_##width##_##height (p_src, p_dest,   \
+                                                            i_stride);       \
+                break;                                                       \
+            case 2:                                                          \
+                MotionComponent_x_Y_avg_##width##_##height (p_src, p_dest,   \
+                                                            i_stride);       \
+                break;                                                       \
+            case 3:                                                          \
+                MotionComponent_X_Y_avg_##width##_##height (p_src, p_dest,   \
+                                                            i_stride);       \
+                break;                                                       \
+            }                                                                \
+        }                                                                    \
     }
 
 /*****************************************************************************
@@ -136,12 +136,12 @@ static __inline__ void MotionComponent(
                     boolean_t b_average     /* (explicit) averaging of several
                                              * predictions */ )
 {
-___callTheRightOne (16,16)
-___callTheRightOne (16,8)
-___callTheRightOne (8,8)
-___callTheRightOne (8,4)
+    ___callTheRightOne (16,16)
+    ___callTheRightOne (16,8)
+    ___callTheRightOne (8,8)
+    ___callTheRightOne (8,4)
 #if 0
-___callTheRightOne (8,16)
+    ___callTheRightOne (8,16)
 #endif
 }
 
@@ -263,7 +263,7 @@ static __inline__ void Motion422(
                        + (p_mb->i_l_x + (i_mv_x >> 1))
                        + (p_mb->i_motion_l_y + i_offset
                           + b_source_field)
-		       * p_mb->p_picture->i_width
+                       * p_mb->p_picture->i_width
                        + (i_mv_y >> 1) * p_mb->i_l_stride,
                      /* destination */
                      p_mb->p_picture->p_y
@@ -671,7 +671,7 @@ void vdec_MotionFrameField444( macroblock_t * p_mb )
     MOTION( p_mb, p_mb->p_forward, 0, 0,                                \
             p_mb->pppi_motion_vectors[0][0][0],                         \
             p_mb->pppi_motion_vectors[0][0][1],                         \
-            /* XXX?? XXX?? >> 1 ? */                                        \
+            /* XXX?? XXX?? >> 1 ? */                                    \
             p_mb->i_l_stride << 1, p_mb->i_c_stride << 1, 8, 0, 0 );    \
                                                                         \
     /* predict and add to top field from bottom field */                \
@@ -683,7 +683,7 @@ void vdec_MotionFrameField444( macroblock_t * p_mb )
     MOTION( p_mb, p_mb->p_forward, 1, 1,                                \
             p_mb->pppi_motion_vectors[0][0][0],                         \
             p_mb->pppi_motion_vectors[0][0][1],                         \
-            /* XXX?? XXX?? >> 1 ? */                                        \
+            /* XXX?? XXX?? >> 1 ? */                                    \
             p_mb->i_l_stride << 1, p_mb->i_c_stride << 1, 8, 0, 0 );    \
                                                                         \
     /* predict and add to bottom field from top field */                \

src/video_decoder/vdec_motion_inner.c

@@ -2,7 +2,7 @@
  * vdec_motion_inner.c : motion compensation inner routines
  *****************************************************************************
  * Copyright (C) 1999, 2000 VideoLAN
- * $Id: vdec_motion_inner.c,v 1.10 2000/12/21 17:19:52 massiot Exp $
+ * $Id: vdec_motion_inner.c,v 1.11 2001/01/05 14:46:37 sam Exp $
  *
  * Authors: Christophe Massiot <massiot@via.ecp.fr>
  *          Jean-Marc Dressler <polux@via.ecp.fr>
@@ -55,190 +55,190 @@
 #include "video_parser.h"
 #include "video_fifo.h"
 
-#define __MotionComponent_x_y_copy(width,height)			\
-void MotionComponent_x_y_copy_##width##_##height(yuv_data_t * p_src,	\
-						 yuv_data_t * p_dest,	\
-						 int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    p_dest[i_x] = p_src[i_x];					\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_x_y_copy(width,height)                        \
+void MotionComponent_x_y_copy_##width##_##height(yuv_data_t * p_src,    \
+                                                 yuv_data_t * p_dest,   \
+                                                 int i_stride)          \
+{                                                                       \
+    int i_x, i_y;                                                       \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            p_dest[i_x] = p_src[i_x];                                   \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponent_X_y_copy(width,height)			\
-void MotionComponent_X_y_copy_##width##_##height(yuv_data_t * p_src,	\
-						 yuv_data_t * p_dest,	\
-						 int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    p_dest[i_x] = (unsigned int)(p_src[i_x]			\
-					 + p_src[i_x + 1]		\
-					 + 1) >> 1;			\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_X_y_copy(width,height)                        \
+void MotionComponent_X_y_copy_##width##_##height(yuv_data_t * p_src,    \
+                                                 yuv_data_t * p_dest,   \
+                                                 int i_stride)          \
+{                                                                       \
+    int i_x, i_y;                                                       \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            p_dest[i_x] = (unsigned int)(p_src[i_x]                     \
+                                         + p_src[i_x + 1]               \
+                                         + 1) >> 1;                     \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponent_x_Y_copy(width,height)			\
-void MotionComponent_x_Y_copy_##width##_##height(yuv_data_t * p_src,	\
-						 yuv_data_t * p_dest,	\
-						 int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    p_dest[i_x] = (unsigned int)(p_src[i_x]			\
-					 + p_src[i_x + i_stride]	\
-					 + 1) >> 1;			\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_x_Y_copy(width,height)                        \
+void MotionComponent_x_Y_copy_##width##_##height(yuv_data_t * p_src,    \
+                                                 yuv_data_t * p_dest,   \
+                                                 int i_stride)          \
+{                                                                       \
+    int i_x, i_y;                                                       \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            p_dest[i_x] = (unsigned int)(p_src[i_x]                     \
+                                         + p_src[i_x + i_stride]        \
+                                         + 1) >> 1;                     \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponent_X_Y_copy(width,height)			\
-void MotionComponent_X_Y_copy_##width##_##height(yuv_data_t * p_src,	\
-						 yuv_data_t * p_dest,	\
-						 int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    p_dest[i_x] = (unsigned int)(p_src[i_x]			\
-					 + p_src[i_x + 1]		\
-					 + p_src[i_x + i_stride]	\
-					 + p_src[i_x + i_stride + 1]	\
-					 + 2) >> 2;			\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_X_Y_copy(width,height)                        \
+void MotionComponent_X_Y_copy_##width##_##height(yuv_data_t * p_src,    \
+                                                 yuv_data_t * p_dest,   \
+                                                 int i_stride)          \
+{                                                                       \
+    int i_x, i_y;                                                       \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            p_dest[i_x] = (unsigned int)(p_src[i_x]                     \
+                                         + p_src[i_x + 1]               \
+                                         + p_src[i_x + i_stride]        \
+                                         + p_src[i_x + i_stride + 1]    \
+                                         + 2) >> 2;                     \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponent_x_y_avg(width,height)				\
-void MotionComponent_x_y_avg_##width##_##height(yuv_data_t * p_src,	\
-						yuv_data_t * p_dest,	\
-						int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-    unsigned int i_dummy;						\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    i_dummy = p_dest[i_x] + p_src[i_x];				\
-	    p_dest[i_x] = (i_dummy + 1) >> 1;  				\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_x_y_avg(width,height)                         \
+void MotionComponent_x_y_avg_##width##_##height(yuv_data_t * p_src,     \
+                                                yuv_data_t * p_dest,    \
+                                                int i_stride)           \
+{                                                                       \
+    int i_x, i_y;                                                       \
+    unsigned int i_dummy;                                               \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            i_dummy = p_dest[i_x] + p_src[i_x];                         \
+            p_dest[i_x] = (i_dummy + 1) >> 1;                           \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponent_X_y_avg(width,height)				\
-void MotionComponent_X_y_avg_##width##_##height(yuv_data_t * p_src,	\
-						yuv_data_t * p_dest,	\
-						int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-    unsigned int i_dummy;						\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    i_dummy = p_dest[i_x] + ((unsigned int)(p_src[i_x]		\
-						    + p_src[i_x + 1]	\
-						    + 1) >> 1);		\
-	    p_dest[i_x] = (i_dummy + 1) >> 1;  				\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_X_y_avg(width,height)                         \
+void MotionComponent_X_y_avg_##width##_##height(yuv_data_t * p_src,     \
+                                                yuv_data_t * p_dest,    \
+                                                int i_stride)           \
+{                                                                       \
+    int i_x, i_y;                                                       \
+    unsigned int i_dummy;                                               \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            i_dummy = p_dest[i_x] + ((unsigned int)(p_src[i_x]          \
+                                                    + p_src[i_x + 1]    \
+                                                    + 1) >> 1);         \
+            p_dest[i_x] = (i_dummy + 1) >> 1;                           \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponent_x_Y_avg(width,height)				\
-void MotionComponent_x_Y_avg_##width##_##height(yuv_data_t * p_src,	\
-						yuv_data_t * p_dest,	\
-						int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-    unsigned int i_dummy;						\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    i_dummy =							\
-		p_dest[i_x] + ((unsigned int)(p_src[i_x]		\
-					      + p_src[i_x + i_stride]	\
-					      + 1) >> 1);		\
-	    p_dest[i_x] = (i_dummy + 1) >> 1;  				\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_x_Y_avg(width,height)                         \
+void MotionComponent_x_Y_avg_##width##_##height(yuv_data_t * p_src,     \
+                                                yuv_data_t * p_dest,    \
+                                                int i_stride)           \
+{                                                                       \
+    int i_x, i_y;                                                       \
+    unsigned int i_dummy;                                               \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            i_dummy =                                                   \
+                p_dest[i_x] + ((unsigned int)(p_src[i_x]                \
+                                              + p_src[i_x + i_stride]   \
+                                              + 1) >> 1);               \
+            p_dest[i_x] = (i_dummy + 1) >> 1;                           \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponent_X_Y_avg(width,height)				\
-void MotionComponent_X_Y_avg_##width##_##height(yuv_data_t * p_src,	\
-						yuv_data_t * p_dest,	\
-						int i_stride)		\
-{									\
-    int i_x, i_y;      							\
-    unsigned int i_dummy;						\
-									\
-    for( i_y = 0; i_y < height; i_y ++ )				\
-    {									\
-	for( i_x = 0; i_x < width; i_x++ )     				\
-	{								\
-	    i_dummy =							\
-		p_dest[i_x] + ((unsigned int)(p_src[i_x]		\
-					    + p_src[i_x + 1]		\
-					    + p_src[i_x + i_stride]	\
-					    + p_src[i_x + i_stride + 1]	\
-					    + 2) >> 2);                 \
-	    p_dest[i_x] = (i_dummy + 1) >> 1;  				\
-	}								\
-	p_dest += i_stride;						\
-	p_src += i_stride;						\
-    }									\
+#define __MotionComponent_X_Y_avg(width,height)                         \
+void MotionComponent_X_Y_avg_##width##_##height(yuv_data_t * p_src,     \
+                                                yuv_data_t * p_dest,    \
+                                                int i_stride)           \
+{                                                                       \
+    int i_x, i_y;                                                       \
+    unsigned int i_dummy;                                               \
+                                                                        \
+    for( i_y = 0; i_y < height; i_y ++ )                                \
+    {                                                                   \
+        for( i_x = 0; i_x < width; i_x++ )                              \
+        {                                                               \
+            i_dummy =                                                   \
+                p_dest[i_x] + ((unsigned int)(p_src[i_x]                \
+                                            + p_src[i_x + 1]            \
+                                            + p_src[i_x + i_stride]     \
+                                            + p_src[i_x + i_stride + 1] \
+                                            + 2) >> 2);                 \
+            p_dest[i_x] = (i_dummy + 1) >> 1;                           \
+        }                                                               \
+        p_dest += i_stride;                                             \
+        p_src += i_stride;                                              \
+    }                                                                   \
 }
 
-#define __MotionComponents(width,height)	\
-__MotionComponent_x_y_copy(width,height)	\
-__MotionComponent_X_y_copy(width,height)	\
-__MotionComponent_x_Y_copy(width,height)	\
-__MotionComponent_X_Y_copy(width,height)	\
-__MotionComponent_x_y_avg(width,height)		\
-__MotionComponent_X_y_avg(width,height)		\
-__MotionComponent_x_Y_avg(width,height)		\
+#define __MotionComponents(width,height)        \
+__MotionComponent_x_y_copy(width,height)        \
+__MotionComponent_X_y_copy(width,height)        \
+__MotionComponent_x_Y_copy(width,height)        \
+__MotionComponent_X_Y_copy(width,height)        \
+__MotionComponent_x_y_avg(width,height)         \
+__MotionComponent_X_y_avg(width,height)         \
+__MotionComponent_x_Y_avg(width,height)         \
 __MotionComponent_X_Y_avg(width,height)
 
-__MotionComponents (16,16)	/* 444, 422, 420 */
-__MotionComponents (16,8)	/* 444, 422, 420 */
-__MotionComponents (8,8)	/* 422, 420 */
-__MotionComponents (8,4)	/* 420 */
+__MotionComponents (16,16)      /* 444, 422, 420 */
+__MotionComponents (16,8)       /* 444, 422, 420 */
+__MotionComponents (8,8)        /* 422, 420 */
+__MotionComponents (8,4)        /* 420 */
 #if 0
-__MotionComponents (8,16)	/* 422 */
+__MotionComponents (8,16)       /* 422 */
 #endif

src/video_decoder/vdec_motion_inner_mmx.c

@@ -3,7 +3,7 @@
  *                           MMX
  *****************************************************************************
  * Copyright (C) 1999, 2000 VideoLAN
- * $Id: vdec_motion_inner_mmx.c,v 1.5 2000/12/21 17:19:52 massiot Exp $
+ * $Id: vdec_motion_inner_mmx.c,v 1.6 2001/01/05 14:46:37 sam Exp $
  *
  * Authors: Christophe Massiot <massiot@via.ecp.fr>, largerly inspired by the
  *          work done by the livid project <http://www.linuxvideo.org/>
@@ -487,10 +487,10 @@ __MotionComponent_X_y_avg(width,height)                                     \
 __MotionComponent_x_Y_avg(width,height)                                     \
 __MotionComponent_X_Y_avg(width,height)
 
-__MotionComponents (16,16)        /* 444, 422, 420 */
-__MotionComponents (16,8)        /* 444, 422, 420 */
+__MotionComponents (16,16)      /* 444, 422, 420 */
+__MotionComponents (16,8)       /* 444, 422, 420 */
 __MotionComponents (8,8)        /* 422, 420 */
 __MotionComponents (8,4)        /* 420 */
 #if 0
-__MotionComponents (8,16)        /* 422 */
+__MotionComponents (8,16)       /* 422 */
 #endif

src/video_output/video_output.c

@@ -104,7 +104,8 @@ vout_thread_t * vout_CreateThread   ( char *psz_display, int i_root_window,
     p_vout = (vout_thread_t *) malloc( sizeof(vout_thread_t) );
     if( p_vout == NULL )
     {
-        intf_ErrMsg( "vout error: %s", strerror(ENOMEM) );
+        intf_ErrMsg( "vout error: vout thread creation returned %s",
+                     strerror(ENOMEM) );
         return( NULL );
     }
 
@@ -460,7 +461,8 @@ subpicture_t *vout_CreateSubPicture( vout_thread_t *p_vout, int i_type,
             p_free_subpic->i_type   =  EMPTY_SUBPICTURE;
             p_free_subpic->i_status =  FREE_SUBPICTURE;
             p_free_subpic =            NULL;
-            intf_ErrMsg("spu warning: %s", strerror( ENOMEM ) );
+            intf_ErrMsg( "vout error: spu allocation returned %s",
+                         strerror( ENOMEM ) );
         }
 
 #ifdef DEBUG_VOUT
@@ -698,7 +700,8 @@ picture_t *vout_CreatePicture( vout_thread_t *p_vout, int i_type,
             p_free_picture->i_type   =  EMPTY_PICTURE;
             p_free_picture->i_status =  FREE_PICTURE;
             p_free_picture =            NULL;
-            intf_ErrMsg( "vout warning: %s", strerror( ENOMEM ) );
+            intf_ErrMsg( "vout error: picture allocation returned %s",
+                         strerror( ENOMEM ) );
         }
 
 #ifdef DEBUG_VOUT
@@ -929,7 +932,7 @@ static int InitThread( vout_thread_t *p_vout )
         /* Initialize convertion tables and functions */
         if( vout_InitYUV( p_vout ) )
         {
-            intf_ErrMsg("error: can't allocate YUV translation tables");
+            intf_ErrMsg("vout error: can't allocate YUV translation tables");
             return( 1 );
         }
     }
@@ -1142,7 +1145,6 @@ static void RunThread( vout_thread_t *p_vout)
                 
             } else {
                 /* no splash screen ! */
-                intf_ErrMsgImm("End of splash screen");  
                 p_vout->init_display_date=0;
             }
         }
@@ -2042,7 +2044,7 @@ static int Manage( vout_thread_t *p_vout )
     {
         if( vout_ResetYUV( p_vout ) )
         {
-            intf_ErrMsg("error: can't rebuild convertion tables");
+            intf_ErrMsg( "vout error: can't rebuild conversion tables" );
             return( 1 );
         }
     }
@@ -2058,7 +2060,7 @@ static int Manage( vout_thread_t *p_vout )
     {
         /* Some changes were not acknowledged by p_vout->p_sys_manage or this
          * function, it means they should not be authorized */
-        intf_ErrMsg( "error: unauthorized changes in the video output thread" );
+        intf_ErrMsg( "vout error: unauthorized changes in the vout thread" );
         return( 1 );
     }
 
@@ -2127,6 +2129,6 @@ static int Align( vout_thread_t *p_vout, int *pi_x, int *pi_y,
 static void     SetPalette        ( p_vout_thread_t p_vout, u16 *red,
                                     u16 *green, u16 *blue, u16 *transp )
 {
-    intf_ErrMsg( "SetPalette: method does not support palette changing" );
+    intf_ErrMsg( "vout error: method does not support palette changing" );
 }
 

src/video_output/video_text.c

@@ -228,7 +228,8 @@ vout_font_t *vout_LoadFont( const char *psz_name )
     p_font = malloc( sizeof( vout_font_t ) );
     if( p_font == NULL )
     {
-        intf_ErrMsg("vout error: %s", strerror(ENOMEM));
+        intf_ErrMsg( "vout error: cannot allocate vout_font_t (%s)",
+                     strerror(ENOMEM) );
         close( i_file );
         return( NULL );
     }
@@ -245,7 +246,7 @@ vout_font_t *vout_LoadFont( const char *psz_name )
         /* Read font header - two bytes indicate the font properties */
         if( read( i_file, pi_buffer, 2 ) != 2)
         {
-            intf_ErrMsg("error: unexpected end of file '%s'", psz_name );
+            intf_ErrMsg( "vout error: unexpected end of file '%s'", psz_name );
             free( p_font );
             close( i_file );
             return( NULL );
@@ -265,7 +266,8 @@ vout_font_t *vout_LoadFont( const char *psz_name )
         p_font->p_data = malloc( 2 * 256 * pi_buffer[1] );
         if( p_font->p_data == NULL )
         {
-            intf_ErrMsg("error: %s", strerror(ENOMEM));
+            intf_ErrMsg( "vout error: cannot allocate font space (%s)",
+                         strerror(ENOMEM) );
             free( p_font );
             close( i_file );
             return( NULL );
@@ -274,7 +276,7 @@ vout_font_t *vout_LoadFont( const char *psz_name )
         /* Copy raw data */
         if( read( i_file, p_font->p_data, 256 * pi_buffer[1] ) != 256 * pi_buffer[1] )
         {
-            intf_ErrMsg("error: unexpected end of file '%s'", psz_name );
+            intf_ErrMsg("vout error: unexpected end of file '%s'", psz_name );
             free( p_font->p_data );
             free( p_font );
             close( i_file );
@@ -300,7 +302,7 @@ vout_font_t *vout_LoadFont( const char *psz_name )
 
         break;
     default:
-        intf_ErrMsg("error: file '%s' has an unknown format", psz_name );
+        intf_ErrMsg("vout error: file '%s' has an unknown format", psz_name );
         free( p_font );
         close( i_file );
         return( NULL );