libvpx: Pull from upstream

source/libvpx/vpx_dsp/mips/convolve8_horiz_dspr2.c

 /*
  *  Copyright (c) 2013 The WebM project authors. All Rights Reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS.  All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
 #include <assert.h>
 #include <stdio.h>
 
 #include "./vpx_dsp_rtcd.h"
-#include "vpx_dsp/mips/vpx_common_dspr2.h"
+#include "vpx_dsp/mips/convolve_common_dspr2.h"
 #include "vpx_dsp/vpx_dsp_common.h"
 #include "vpx_dsp/vpx_filter.h"
 #include "vpx_ports/mem.h"
 
 #if HAVE_DSPR2
 static void convolve_horiz_4_dspr2(const uint8_t *src,
                                    int32_t src_stride,
                                    uint8_t *dst,
                                    int32_t dst_stride,
                                    const int16_t *filter_x0,
@@ skipping 808 matching lines @@
     src_ptr += src_stride;
     dst_ptr += dst_stride;
   }
 }
 
 void vpx_convolve8_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
                                uint8_t *dst, ptrdiff_t dst_stride,
                                const int16_t *filter_x, int x_step_q4,
                                const int16_t *filter_y, int y_step_q4,
                                int w, int h) {
-  if (((const int32_t *)filter_x)[1] == 0x800000) {
-    vpx_convolve_copy(src, src_stride,
-                      dst, dst_stride,
-                      filter_x, x_step_q4,
-                      filter_y, y_step_q4,
-                      w, h);
-  } else if (((const int32_t *)filter_x)[0] == 0) {
+  assert(x_step_q4 == 16);
+  assert(((const int32_t *)filter_x)[1] != 0x800000);
+
+  if (((const int32_t *)filter_x)[0] == 0) {
     vpx_convolve2_horiz_dspr2(src, src_stride,
                               dst, dst_stride,
                               filter_x, x_step_q4,
                               filter_y, y_step_q4,
                               w, h);
   } else {
-    if (16 == x_step_q4) {
-      uint32_t pos = 38;
+    uint32_t pos = 38;
 
-      prefetch_load((const uint8_t *)filter_x);
-      src -= 3;
+    prefetch_load((const uint8_t *)filter_x);
+    src -= 3;
 
-      /* bit positon for extract from acc */
-      __asm__ __volatile__ (
-        "wrdsp      %[pos],     1           \n\t"
-        :
-        : [pos] "r" (pos)
-      );
+    /* bit positon for extract from acc */
+    __asm__ __volatile__ (
+      "wrdsp      %[pos],     1           \n\t"
+      :
+      : [pos] "r" (pos)
+    );
 
-      /* prefetch data to cache memory */
-      prefetch_load(src);
-      prefetch_load(src + 32);
-      prefetch_store(dst);
+    /* prefetch data to cache memory */
+    prefetch_load(src);
+    prefetch_load(src + 32);
+    prefetch_store(dst);
 
-      switch (w) {
-        case 4:
-          convolve_horiz_4_dspr2(src, (int32_t)src_stride,
-                                 dst, (int32_t)dst_stride,
-                                 filter_x, (int32_t)h);
-          break;
-        case 8:
-          convolve_horiz_8_dspr2(src, (int32_t)src_stride,
-                                 dst, (int32_t)dst_stride,
-                                 filter_x, (int32_t)h);
-          break;
-        case 16:
-          convolve_horiz_16_dspr2(src, (int32_t)src_stride,
-                                  dst, (int32_t)dst_stride,
-                                  filter_x, (int32_t)h, 1);
-          break;
-        case 32:
-          convolve_horiz_16_dspr2(src, (int32_t)src_stride,
-                                  dst, (int32_t)dst_stride,
-                                  filter_x, (int32_t)h, 2);
-          break;
-        case 64:
-          prefetch_load(src + 64);
-          prefetch_store(dst + 32);
+    switch (w) {
+      case 4:
+        convolve_horiz_4_dspr2(src, (int32_t)src_stride,
+                               dst, (int32_t)dst_stride,
+                               filter_x, (int32_t)h);
+        break;
+      case 8:
+        convolve_horiz_8_dspr2(src, (int32_t)src_stride,
+                               dst, (int32_t)dst_stride,
+                               filter_x, (int32_t)h);
+        break;
+      case 16:
+        convolve_horiz_16_dspr2(src, (int32_t)src_stride,
+                                dst, (int32_t)dst_stride,
+                                filter_x, (int32_t)h, 1);
+        break;
+      case 32:
+        convolve_horiz_16_dspr2(src, (int32_t)src_stride,
+                                dst, (int32_t)dst_stride,
+                                filter_x, (int32_t)h, 2);
+        break;
+      case 64:
+        prefetch_load(src + 64);
+        prefetch_store(dst + 32);
 
-          convolve_horiz_64_dspr2(src, (int32_t)src_stride,
-                                  dst, (int32_t)dst_stride,
-                                  filter_x, (int32_t)h);
-          break;
-        default:
-          vpx_convolve8_horiz_c(src + 3, src_stride,
-                                dst, dst_stride,
-                                filter_x, x_step_q4,
-                                filter_y, y_step_q4,
-                                w, h);
-          break;
-      }
-    } else {
-      vpx_convolve8_horiz_c(src, src_stride,
-                            dst, dst_stride,
-                            filter_x, x_step_q4,
-                            filter_y, y_step_q4,
-                            w, h);
+        convolve_horiz_64_dspr2(src, (int32_t)src_stride,
+                                dst, (int32_t)dst_stride,
+                                filter_x, (int32_t)h);
+        break;
+      default:
+        vpx_convolve8_horiz_c(src + 3, src_stride,
+                              dst, dst_stride,
+                              filter_x, x_step_q4,
+                              filter_y, y_step_q4,
+                              w, h);
+        break;
     }
   }
 }
 #endif