Copy macroblock data to a buffer before encoding it

I got this idea from Pascal (Thanks). Before encoding a macroblock,
copy it to a 16x16 buffer, and then read source data from there
instead. This will help keep the source data in cache, and help
with the performance.

Change-Id: Id05f4cb601299150511d59dcba0ae62c49b5b757

vp8/encoder/firstpass.c

@@ -577,6 +577,9 @@ void vp8_first_pass(VP8_COMP *cpi)
             xd->dst.v_buffer = new_yv12->v_buffer + recon_uvoffset;
             xd->left_available = (mb_col != 0);
 
+            //Copy current mb to a buffer
+            RECON_INVOKE(&xd->rtcd->recon, copy16x16)(x->src.y_buffer, x->src.y_stride, x->thismb, 16);
+
             // do intra 16x16 prediction
             this_error = vp8_encode_intra(cpi, x, use_dc_pred);