Rename subframe_size to a name that is not specific to some specific use

of ff_acelp_convolve_circ()

Originally committed as revision 14894 to svn://svn.ffmpeg.org/ffmpeg/trunk

libavcodec/acelp_filters.c

@@ -85,22 +85,22 @@ void ff_acelp_convolve_circ(
         int16_t* fc_out,
         const int16_t* fc_in,
         const int16_t* filter,
-        int subframe_size)
+        int len)
 {
     int i, k;
 
-    memset(fc_out, 0, subframe_size * sizeof(int16_t));
+    memset(fc_out, 0, len * sizeof(int16_t));
 
     /* Since there are few pulses over an entire subframe (i.e. almost
        all fc_in[i] are zero) it is faster to loop over fc_in first. */
-    for(i=0; i<subframe_size; i++)
+    for(i=0; i<len; i++)
     {
         if(fc_in[i])
         {
             for(k=0; k<i; k++)
-                fc_out[k] += (fc_in[i] * filter[subframe_size + k - i]) >> 15;
+                fc_out[k] += (fc_in[i] * filter[len + k - i]) >> 15;
 
-            for(k=i; k<subframe_size; k++)
+            for(k=i; k<len; k++)
                 fc_out[k] += (fc_in[i] * filter[k - i]) >> 15;
         }
     }

libavcodec/acelp_filters.h

@@ -75,7 +75,7 @@ void ff_acelp_convolve_circ(
         int16_t* fc_out,
         const int16_t* fc_in,
         const int16_t* filter,
-        int subframe_size);
+        int len);
 
 /**
  * LP synthesis filter.