update v2.6 benchmark, test=develop (#3663)

update v2.6 benchmark

docs/benchmark/benchmark.md

@@ -18,7 +18,7 @@
 
 * 测试机器(android ndk ndk-r17c)
    *  骁龙855
-      * xiaomi mi9, snapdragon 855 
+      * xiaomi mi9, snapdragon 855 (enable sdot instruction)
       * 4xA76(1@2.84GHz + 3@2.4GHz) + 4xA55@1.78GHz
 
    *  骁龙845
@@ -33,7 +33,7 @@
       * HUAWEI Mate10
  
 * 测试说明
-    * branch: release/v2.3.0
+    * branch: release/v2.6.0
     * warmup=10, repeats=30，统计平均时间，单位是ms
     * 当线程数为1时，```DeviceInfo::Global().SetRunMode```设置LITE_POWER_HIGH，否者设置LITE_POWER_NO_BIND
     * 模型的输入图像的维度是{1, 3, 224, 224}，输入图像的每一位数值是1
@@ -48,75 +48,75 @@
 骁龙855|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 
-mobilenet_v1 |33.27 |19.52 |11.14 |31.72 |18.76 |10.24 |
-mobilenet_v2 |29.08 |15.79 |9.25 |25.89 |14.17 |8.38 |
-shufflenet_v2 |4.40 |3.09 |2.30 |4.28 |3.02 |2.35 |
-squeezenet_v1.1 |19.96 |12.61 |8.76 |18.25 |11.46 |7.97 |
-mnasnet |21.00 |12.54 |7.28 |19.65 |11.65 |6.96 |
+mobilenet_v1 |35.11 |20.67 |11.83 |30.56 |18.59 |10.44 |
+mobilenet_v2 |26.36 |15.83 |9.29 |21.64 |13.25 |7.95 |
+shufflenet_v2 |4.56 |3.14 |2.35 |4.07 |2.89 |2.28 |
+squeezenet_v1.1 |21.27 |13.55 |8.49 |18.05 |11.51 |7.83 |
+mnasnet |21.40 |13.18 |7.63 |18.84 |11.40 |6.80 |
 
 
 骁龙845|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 
-mobilenet_v1 |66.36 |35.97 |19.45 |62.66 |33.87 |17.85 |
-mobilenet_v2 |45.86 |25.53 |14.6 |41.58 |23.24 |13.39 |
-shufflenet_v2 |7.58 |4.89 |3.41 |7.44 |4.91 |3.58 |
-squeezenet_v1.1 |37.15 |22.74 |13.51 |34.69 |21.27 |12.74 |
-mnasnet |40.09 |21.73 |11.91 |38.19 |21.02 |12.11 |
+mobilenet_v1 |65.56 |37.17 |19.65 |63.23 |32.98 |17.68 |
+mobilenet_v2 |45.89 |25.20 |14.39 |41.03 |22.94 |12.98 |
+shufflenet_v2 |7.31 |4.66 |3.27 |7.08 |4.71 |3.41 |
+squeezenet_v1.1 |36.98 |22.53 |13.45 |34.27 |20.96 |12.60 |
+mnasnet |39.85 |23.64 |12.25 |37.81 |20.70 |11.81 |
 
 
 骁龙835|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 
-mobilenet_v1 |96.98 |53.92 |32.24 |89.31 |48.02 |27.58 |
-mobilenet_v2 |67.72 |37.66 |23.82 |60.10 |34.36 |21.05 |
-shufflenet_v2 |10.72 |6.62 |4.63 |10.10 |6.44 |4.63 |
-squeezenet_v1.1 |53.89 |33.28 |20.73 |50.83 |32.31 |19.51 |
-mnasnet |59.55 |33.53 |20.32 |56.21 |31.58 |19.06 |
+mobilenet_v1 |92.77 |51.56 |30.14 |87.46 |48.02 |26.42 |
+mobilenet_v2 |65.78 |36.52 |22.34 |58.31 |33.04 |19.87 |
+shufflenet_v2 |10.39 |6.26 |4.46 |9.72 |6.19 |4.41 |
+squeezenet_v1.1 |53.59 |33.16 |20.13 |51.56 |31.81 |19.10 |
+mnasnet |57.44 |32.62 |19.47 |54.99 |30.69 |17.98 |
 
 #### caffe model
 
 骁龙855|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 |
-mobilenet_v1 |33.36 |19.45 |11.26 |31.63 |18.74 |10.31 |
-mobilenet_v2 |31.63 |19.21 |11.61 |28.34 |17.14 |10.16 |
-shufflenet_v2 |4.46 |3.08 |2.32 |4.26 |2.98 |2.35 |
+mobilenet_v1 |32.38 |18.65 |10.69 |30.75 |18.11 |9.88 |
+mobilenet_v2 |29.45 |17.86 |10.81 |26.61 |16.26 |9.67 |
+shufflenet_v2 |5.04 |3.14 |2.20 |4.09 |2.85 |2.25 |
 
 
 骁龙845|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 |
-mobilenet_v1 |66.32 |35.83 |19.56 |62.52 |33.79 |17.91 |
-mobilenet_v2 |58.46 |32.69 |18.56 |53.72 |29.86 |16.80 |
-shufflenet_v2 |7.65 |4.82 |3.46 |7.55 |4.97 |3.62 |
+mobilenet_v1 |65.26 |35.19 |19.11 |61.42 |33.15 |17.48 |
+mobilenet_v2 |55.59 |31.31 |17.68 |51.54 |29.69 |16.00 |
+shufflenet_v2 |7.42 |4.73 |3.33 |7.18 |4.75 |3.39 |
 
 
 骁龙835|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 |
-mobilenet_v1 |95.38 |54.09 |32.03 |95.05 |48.33 |27.54 |
-mobilenet_v2 |88.46 |48.98 |30.23 |79.28 |44.64 |27.10 |
-shufflenet_v2 |10.07 |6.51 |4.61 |10.31 |6.50 |4.66 |
+mobilenet_v1 |95.38 |52.16 |30.37 |92.10 |46.71 |26.31 |
+mobilenet_v2 |82.89 |45.49 |28.14 |74.91 |41.88 |25.25 |
+shufflenet_v2 |10.25 |6.36 |4.42 |9.68 |6.20 |4.42 |
 
 #### int8量化模型测试数据
 
 骁龙855|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 |
-mobilenet_v1 |36.80 |21.58 |11.12 | 14.01 |8.13 |4.32 |
-mobilenet_v2 |28.72 |19.08 |12.49 | 17.24 |11.55 |7.82 |
+mobilenet_v1 |37.18 |21.71 |11.16 | 14.41 |8.34 |4.37 |
+mobilenet_v2 |27.95 |16.57 |8.97 | 13.68 |8.16 |4.67 |
 
 
 骁龙835|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 |
-mobilenet_v1 |60.76 |32.25 |16.66 |56.57 |29.84 |15.24 |
-mobilenet_v2 |49.38 |31.10 |22.07 |47.52 |28.18 |19.24 |
+mobilenet_v1 |61.63 |32.60 |16.49 |57.36 |29.74 |15.50 |
+mobilenet_v2 |47.13 |25.62 |13.56 |41.87 |22.42 |11.72 |
 
 
 麒麟970|armv7 | armv7 |  armv7 |armv8 | armv8 |armv8 
 ----| ---- | ---- | ---- | ----  |----  |----
 threads num|1 |2 |4 |1 |2 |4 |
-mobilenet_v1 |65.95 |34.39 |18.68 |60.86 |30.98 |16.31 |
-mobilenet_v2 |68.87 |39.39 |24.43 |65.57 |37.31 |20.87 |
+mobilenet_v1 |63.13 |32.63 |16.85 |58.92 |29.96 |15.42 |
+mobilenet_v2 |48.60 |25.43 |13.76 |43.06 |22.10 |12.09 |