VLIB是TI提供的针对C6x优化过的视觉库，下载地址：http://software-dl.ti.com/libs/vlib/latest/index_FDS.html
提供40多个核心函数，主要实现以下功能：
- Background Modeling & Subtraction
- Object Feature Extraction
- Tracking & Recognition
- Low-level Pixel Processing
- HoughLinesDemo
- Regression
以下是（1）HoughLinesDemo的测试数据（暂时Load Program失败）
以下是（2）Regression的测试数据（测试平台6657evm）
[C66xx_0]
+---------------------------------------+
| TESTING: VLIB_afast12_detectCorners |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_afast12_detectCorners testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 76800 | 29.49 | 29.52 | 0.00 | 29.49 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 2876,
| | | | | | | | realistic data, sort
| 1 | PASS | 19200 | 43.85 | 43.83 | 0.00 | 43.85 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 891,
| | | | | | | | realistic data, sort
| 2 | PASS | 76800 | 3.70 | 3.69 | 0.00 | 3.70 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 0, flat
| | | | | | | | image, sort
| 3 | PASS | 19200 | 3.72 | 3.68 | 0.00 | 3.72 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 0, flat
| | | | | | | | image, sort
| 4 | PASS | 76800 | 198.54 | 198.54 | 0.00 | 198.54 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 9996,
| | | | | | | | checkerboard image, sort
| 5 | PASS | 19200 | 146.10 | 146.10 | 0.00 | 146.10 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 3350,
| | | | | | | | checkerboard image, sort
| 6 | PASS | 76800 | 21.04 | 21.07 | 0.00 | 21.04 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 2876,
| | | | | | | | realistic data, no sort
| 7 | PASS | 19200 | 35.91 | 36.00 | 0.00 | 35.91 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 891,
| | | | | | | | realistic data, no sort
| 8 | PASS | 76800 | 3.64 | 3.63 | 0.00 | 3.64 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 0, flat
| | | | | | | | image, no sort
| 9 | PASS | 19200 | 3.63 | 3.56 | 0.00 | 3.63 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 0, flat
| | | | | | | | image, no sort
| 10 | PASS | 76800 | 28.98 | 28.98 | 0.00 | 28.98 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 9996,
| | | | | | | | checkerboard image, no sort
| 11 | PASS | 19200 | 39.36 | 39.28 | 0.00 | 39.36 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 3350,
| | | | | | | | checkerboard image, no sort
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 12
Range:
Min: 3.63 cycles/pt Avg: 46.50 cycles/pt
Max: 198.54 cycles/pt Med: 28.98 cycles/pt
Num pt = number of input pixels (img_width*img_height); Performance is data dependent.
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 376 bytes
+---------------------------------------+
| TESTING: VLIB_afast9_detectCorners |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_afast9_detectCorners testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 16384 | 10.23 | 111.31 | 0.00 | 10.23 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width= 134, height= 134, n_features found= 1060, constant
| | | | | | | | data
| 1 | PASS | 71136 | 10.03 | 86.02 | 0.00 | 10.03 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width= 320, height= 240, n_features found= 6337, realistic
| | | | | | | | data
| 2 | PASS | 16416 | 10.28 | 123.32 | 0.00 | 10.28 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width= 160, height= 120, n_features found= 4215, realistic
| | | | | | | | data
| 3 | PASS | 71136 | 10.03 | 56.52 | 0.00 | 10.03 | Constant generated input | Opt and NatC results compared |
| | | | | | | | width= 320, height= 240, n_features found= 0, flat image
| 4 | PASS | 16416 | 10.31 | 56.68 | 0.00 | 10.31 | Constant generated input | Opt and NatC results compared |
| | | | | | | | width= 160, height= 120, n_features found= 0, flat image
| 5 | PASS | 71136 | 10.05 | 97.88 | 0.00 | 10.05 | Custom generated input | Opt and NatC results compared |
| | | | | | | | width= 320, height= 240, n_features found= 17901,
| | | | | | | | checkerboard image
| 6 | PASS | 16416 | 10.55 | 98.43 | 0.00 | 10.55 | Custom generated input | Opt and NatC results compared |
| | | | | | | | width= 160, height= 120, n_features found= 4104,
| | | | | | | | checkerboard image
| 7 | PASS | 71136 | 10.51 | 86.53 | 0.00 | 10.51 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width= 320, height= 240, n_features found= 6337, realistic
| | | | | | | | data
| 8 | PASS | 16416 | 10.26 | 71.78 | 0.00 | 10.26 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width= 160, height= 120, n_features found= 41, realistic
| | | | | | | | data
| 9 | PASS | 71136 | 10.06 | 56.51 | 0.00 | 10.06 | Constant generated input | Opt and NatC results compared |
| | | | | | | | width= 320, height= 240, n_features found= 0, flat image
| 10 | PASS | 16416 | 10.32 | 56.68 | 0.00 | 10.32 | Constant generated input | Opt and NatC results compared |
| | | | | | | | width= 160, height= 120, n_features found= 0, flat image
| 11 | PASS | 71136 | 10.04 | 56.51 | 0.00 | 10.04 | Custom generated input | Opt and NatC results compared |
| | | | | | | | width= 320, height= 240, n_features found= 0, checkerboard
| | | | | | | | image
| 12 | PASS | 16416 | 10.31 | 98.04 | 0.00 | 10.31 | Custom generated input | Opt and NatC results compared |
| | | | | | | | width= 160, height= 120, n_features found= 4104,
| | | | | | | | checkerboard image
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 13
Formula:
Cycles: 10.34385*N - 131.53547*M + 14898
Where: N = ((width-6)&0xFFF0) * (height-6), M=height-6
Range:
Min: 10.03 cycles/pt Avg: 10.23 cycles/pt
Max: 10.55 cycles/pt Med: 10.23 cycles/pt
Num pt = ((width-6)&0xFFF0) * (height-6)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 124 bytes
+---------------------------------------+
| TESTING: VLIB_aFast_nonmaxSuppression |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_aFast_nonmaxSuppression testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 2872 | 29.68 | 120.67 | 0.00 | 29.68 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 2876, output features= 1290, realistic data
| 1 | PASS | 888 | 35.13 | 127.32 | 0.00 | 35.13 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 891, output features= 609, realistic data
| 2 | PASS | 208 | 55.61 | 144.00 | 0.00 | 55.61 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 210, output features= 157, fits in 32k
| 3 | PASS | 176 | 61.61 | 146.03 | 0.00 | 61.61 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 180, output features= 132, fits in 32k
| 4 | PASS | 120 | 74.56 | 150.83 | 0.00 | 74.56 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 120, output features= 90, fits in 32k
| 5 | PASS | 1288 | 37.41 | 132.10 | 0.00 | 37.41 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 1292, output features= 1288, realistic data,
| | | | | | | | already suppressed
| 6 | PASS | 608 | 40.26 | 134.01 | 0.00 | 40.26 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 612, output features= 608, realistic data,
| | | | | | | | already suppressed
| 7 | PASS | 208 | 57.68 | 145.82 | 0.00 | 57.68 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 210, output features= 208, fits in 32k,
| | | | | | | | already suppressed
| 8 | PASS | 176 | 62.86 | 148.91 | 0.00 | 62.86 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 180, output features= 176, fits in 32k,
| | | | | | | | already suppressed
| 9 | PASS | 120 | 75.49 | 153.42 | 0.00 | 75.49 | Statically generated input | Opt and NatC results compared |
| | | | | | | | inNumScores = 120, output features= 120, fits in 32k,
| | | | | | | | already suppressed
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 28.23980*N + 6439
Where: N = inNumScores
Range:
Min: 29.68 cycles/pt Avg: 53.03 cycles/pt
Max: 75.49 cycles/pt Med: 55.61 cycles/pt
Num pt = inNumScores
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_bhattacharyaDistance_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_bhattacharyaDistance_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 512 | 20.95 | 65.80 | 0.00 | 20.95 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | N=512
| 1 | PASS | 32 | 84.66 | 192.50 | 0.00 | 84.66 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | N=32
| 2 | PASS | 64 | 50.20 | 154.11 | 0.00 | 50.20 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=64
| 3 | PASS | 128 | 32.98 | 142.40 | 0.00 | 32.98 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=128
| 4 | PASS | 256 | 26.52 | 136.54 | 0.00 | 26.52 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=256
| 5 | PASS | 1024 | 18.02 | 132.78 | 0.00 | 18.02 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=1024
| 6 | PASS | 4096 | 16.75 | 131.94 | 0.00 | 16.75 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 16.18650*N + 2255
Where: N = N
Range:
Min: 16.75 cycles/pt Avg: 35.72 cycles/pt
Max: 84.66 cycles/pt Med: 20.95 cycles/pt
Num pt = N
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_bhattacharyaDistance_U32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_bhattacharyaDistance_U32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 512 | 144.55 | 143.74 | 0.00 | 144.55 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | N=512
| 1 | PASS | 32 | 171.00 | 175.06 | 0.00 | 171.00 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | N=32
| 2 | PASS | 64 | 154.50 | 156.53 | 0.00 | 154.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=64
| 3 | PASS | 128 | 148.17 | 149.19 | 0.00 | 148.17 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=128
| 4 | PASS | 256 | 145.05 | 145.63 | 0.00 | 145.05 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=256
| 5 | PASS | 1024 | 143.30 | 142.93 | 0.00 | 143.30 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=1024
| 6 | PASS | 4096 | 142.57 | 142.67 | 0.00 | 142.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | N=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 142.37362*N + 857
Where: N = N
Range:
Min: 142.57 cycles/pt Avg: 149.88 cycles/pt
Max: 171.00 cycles/pt Med: 144.55 cycles/pt
Num pt = N
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_binarySkeleton |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_binarySkeleton testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4032 | 3.17 | 3.43 | 0.00 | 3.17 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width=32, height=16, num_st=3, iters=3
| 1 | PASS | 8640 | 2.30 | 2.42 | 0.00 | 2.30 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width=32, height=32, num_st=3, iters=3
| 2 | PASS | 28800 | 1.80 | 1.90 | 0.00 | 1.80 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width=64, height=32, num_st=3, iters=5
| 3 | PASS | 2688 | 3.71 | 3.92 | 0.00 | 3.71 | Statically generated input | Opt and NatC results compared |
| | | | | | | | width=32, height=16, num_st=2, iters=3
| 4 | PASS | 304640 | 1.64 | 1.64 | 0.00 | 1.64 | Constant generated input | Opt and NatC results compared |
| | | | | | | | width=320, height=240, num_st=2, iters=2
| 5 | PASS | 380800 | 1.61 | 1.61 | 0.00 | 1.61 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | width=320, height=240, num_st=1, iters=5
| 6 | PASS | 4894720 | 1.75 | 1.76 | 0.00 | 1.75 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | width=640, height=480, num_st=2, iters=8
| 7 | PASS | 5506560 | 1.75 | 1.75 | 0.00 | 1.75 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | width=640, height=480, num_st=3, iters=6
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 1.75362*N - 10912
Where: N = width * (height - 2) * num_st * iters
Range:
Min: 1.61 cycles/pt Avg: 2.22 cycles/pt
Max: 3.71 cycles/pt Med: 1.75 cycles/pt
Num pt = width * (height - 2) * num_st * iters
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 80 bytes
+---------------------------------------+
| TESTING: VLIB_blobAnalysis |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_blobAnalysis testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
num of blobs, scratch buffer size (in bytes) for
CCL, Create Blob list, Blob analysis
| 0 | PASS | 640 | 117.26 | 308.77 | 0.00 | 117.26 | 7 blobs, 26776 bytes, 224 bytes, 1464 bytes
| 1 | PASS | 64000 | 22.83 | 126.55 | 0.00 | 22.83 | 80 blobs, 2021688 bytes, 2560 bytes, 8012 bytes
| 2 | PASS | 8192 | 27.95 | 30.57 | 0.00 | 27.95 | 13 blobs, 255528 bytes, 416 bytes, 23312 bytes
| 3 | PASS | 8192 | 27.98 | 30.70 | 0.00 | 27.98 | 13 blobs, 255528 bytes, 416 bytes, 23312 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 4
Range:
Min: 22.83 cycles/pt Avg: 49.01 cycles/pt
Max: 117.26 cycles/pt Med: 27.95 cycles/pt
Num pt = imageWidth * imageHeight
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 108 bytes
+---------------------------------------+
| TESTING: VLIB_blockMedian |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_blockMedian testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 128 | 96.03 | 53.04 | 0.00 | 96.03 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=11, imageHeight=6, blockWidth=4,
| | | | | | | | blockHeight=4, stepSizeHorz=2, stepSizeVert=2, out size=8
| 1 | PASS | 3849120 | 2.85 | 7.81 | 0.00 | 2.85 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=640, imageHeight=480, blockWidth=66,
| | | | | | | | blockHeight=60, stepSizeHorz=16, stepSizeVert=16, out size=972
| 2 | PASS | 1546380 | 2.25 | 7.50 | 0.00 | 2.25 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=640, imageHeight=480, blockWidth=121,
| | | | | | | | blockHeight=71, stepSizeHorz=35, stepSizeVert=35, out size=180
| 3 | PASS | 307200 | 2.78 | 8.40 | 0.00 | 2.78 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=640, imageHeight=480, blockWidth=640,
| | | | | | | | blockHeight=480, stepSizeHorz=100, stepSizeVert=100, out size=1
| 4 | PASS | 204800 | 2.86 | 8.40 | 0.00 | 2.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=640, imageHeight=480, blockWidth=320,
| | | | | | | | blockHeight=320, stepSizeHorz=320, stepSizeVert=320, out size=2
| 5 | PASS | 19076 | 3.84 | 9.04 | 0.00 | 3.84 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=633, imageHeight=477, blockWidth=251,
| | | | | | | | blockHeight=19, stepSizeHorz=311, stepSizeVert=311, out size=4
| 6 | PASS | 387904 | 5.53 | 10.00 | 0.00 | 5.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=633, imageHeight=477, blockWidth=32,
| | | | | | | | blockHeight=19, stepSizeHorz=21, stepSizeVert=21, out size=638
| 7 | PASS | 9216 | 7.17 | 11.42 | 0.00 | 7.17 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=203, imageHeight=101, blockWidth=32,
| | | | | | | | blockHeight=32, stepSizeHorz=64, stepSizeVert=32, out size=9
| 8 | PASS | 9744 | 3.99 | 9.02 | 0.00 | 3.99 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=203, imageHeight=101, blockWidth=203,
| | | | | | | | blockHeight=16, stepSizeHorz=64, stepSizeVert=32, out size=3
| 9 | PASS | 16000 | 3.11 | 8.57 | 0.00 | 3.11 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=2000, imageHeight=10, blockWidth=2000,
| | | | | | | | blockHeight=8, stepSizeHorz=200, stepSizeVert=32, out size=1
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 2.76691*N + 71705
Where: N = ((imageHeight - blockHeight)/stepSizeVert + 1) * ((imageWidth - blockWidth)/stepSizeHorz + 1) * blockHeight*blockWidth
Range:
Min: 2.25 cycles/pt Avg: 13.04 cycles/pt
Max: 96.03 cycles/pt Med: 3.11 cycles/pt
Num pt = ((imageHeight - blockHeight)/stepSizeVert + 1) * ((imageWidth - blockWidth)/stepSizeHorz + 1) * blockHeight*blockWidth; Cycles/pt get smaller as the blockWidth gets larger.
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 88 bytes
+---------------------------------------+
| TESTING: VLIB_blockStatistics |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_blockStatistics testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4096 | 23.42 | 56.89 | 0.00 | 23.42 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 64, imageHeight= 64, statBlockWidth= 1,
| | | | | | | | statBlockHeight= 1, mode = 0
| 1 | PASS | 4096 | 51.76 | 56.76 | 0.00 | 51.76 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 64, imageHeight= 64, statBlockWidth= 1,
| | | | | | | | statBlockHeight= 1, mode = 1
| 2 | PASS | 4096 | 4.25 | 11.91 | 0.00 | 4.25 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 64, imageHeight= 64, statBlockWidth= 5,
| | | | | | | | statBlockHeight= 5, mode = 0
| 3 | PASS | 4096 | 7.93 | 12.18 | 0.00 | 7.93 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 64, imageHeight= 64, statBlockWidth= 5,
| | | | | | | | statBlockHeight= 5, mode = 1
| 4 | PASS | 4096 | 4.43 | 12.84 | 0.00 | 4.43 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 64, imageHeight= 64, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 5, mode = 0
| 5 | PASS | 4096 | 3.05 | 8.37 | 0.00 | 3.05 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 64, imageHeight= 64, statBlockWidth= 5,
| | | | | | | | statBlockHeight= 11, mode = 0
| 6 | PASS | 16384 | 8.47 | 21.51 | 0.00 | 8.47 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 128, imageHeight= 128, statBlockWidth= 1,
| | | | | | | | statBlockHeight= 3, mode = 0
| 7 | PASS | 32768 | 14.61 | 44.38 | 0.00 | 14.61 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 128, imageHeight= 256, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 1, mode = 0
| 8 | PASS | 32768 | 2.29 | 8.60 | 0.00 | 2.29 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 128, imageHeight= 256, statBlockWidth= 1,
| | | | | | | | statBlockHeight= 32, mode = 0
| 9 | PASS | 16384 | 27.45 | 8.59 | 0.00 | 27.45 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 128, imageHeight= 128, statBlockWidth= 1,
| | | | | | | | statBlockHeight= 32, mode = 1
| 10 | PASS | 32768 | 13.04 | 40.52 | 0.00 | 13.04 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 128, imageHeight= 256, statBlockWidth= 32,
| | | | | | | | statBlockHeight= 1, mode = 0
| 11 | PASS | 16384 | 4.86 | 40.44 | 0.00 | 4.86 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 128, imageHeight= 128, statBlockWidth= 32,
| | | | | | | | statBlockHeight= 1, mode = 1
| 12 | PASS | 76800 | 2.47 | 9.21 | 0.00 | 2.47 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 21,
| | | | | | | | statBlockHeight= 11, mode = 0
| 13 | PASS | 76800 | 3.70 | 9.20 | 0.00 | 3.70 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 21,
| | | | | | | | statBlockHeight= 11, mode = 1
| 14 | PASS | 77361 | 6.21 | 18.09 | 0.00 | 6.21 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 321, imageHeight= 241, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 3, mode = 0
| 15 | PASS | 77924 | 5.79 | 17.62 | 0.00 | 5.79 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 322, imageHeight= 242, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 3, mode = 0
| 16 | PASS | 78489 | 5.82 | 17.78 | 0.00 | 5.82 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 323, imageHeight= 243, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 3, mode = 0
| 17 | PASS | 79056 | 5.85 | 17.76 | 0.00 | 5.85 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 324, imageHeight= 244, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 3, mode = 0
| 18 | PASS | 76800 | 5.75 | 17.70 | 0.00 | 5.75 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 3, mode = 0
| 19 | PASS | 76800 | 12.43 | 17.74 | 0.00 | 12.43 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 3,
| | | | | | | | statBlockHeight= 3, mode = 1
| 20 | PASS | 76800 | 2.63 | 9.39 | 0.00 | 2.63 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 1,
| | | | | | | | statBlockHeight= 18, mode = 0
| 21 | PASS | 76800 | 12.96 | 38.58 | 0.00 | 12.96 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 18,
| | | | | | | | statBlockHeight= 1, mode = 0
| 22 | PASS | 76800 | 2.29 | 8.95 | 0.00 | 2.29 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 16,
| | | | | | | | statBlockHeight= 16, mode = 0
| 23 | PASS | 76800 | 4.27 | 8.97 | 0.00 | 4.27 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, statBlockWidth= 16,
| | | | | | | | statBlockHeight= 16, mode = 1
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 24
Range:
Min: 2.29 cycles/pt Avg: 9.82 cycles/pt
Max: 51.76 cycles/pt Med: 5.79 cycles/pt
Num pt = imageWidth*imageHeight
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 56 bytes
+---------------------------------------+
| TESTING: VLIB_Canny_Edge_Detection |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_Canny_Edge_Detection testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4096 | 30.14 | 102.04 | 0.00 | 30.14 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=64
| 1 | PASS | 480 | 33.85 | 71.30 | 0.00 | 33.85 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=15
| 2 | PASS | 216 | 53.51 | 75.99 | 0.00 | 53.51 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=18, height=12
| 3 | PASS | 360 | 36.17 | 65.60 | 0.00 | 36.17 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=30, height=12
| 4 | PASS | 3600 | 35.94 | 106.06 | 0.00 | 35.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=36, height=100
| 5 | PASS | 4180 | 36.04 | 107.29 | 0.00 | 36.04 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=38, height=110
| 6 | PASS | 19200 | 38.63 | 114.90 | 0.00 | 38.63 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120
| 7 | PASS | 39042 | 41.81 | 120.27 | 0.00 | 41.81 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=162, height=241
| 8 | PASS | 76800 | 58.30 | 137.49 | 0.00 | 58.30 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240
| 9 | PASS | 307200 | 65.16 | 143.96 | 0.00 | 65.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, height=480
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Range:
Min: 30.14 cycles/pt Avg: 42.95 cycles/pt
Max: 65.16 cycles/pt Med: 36.17 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 140 bytes
+---------------------------------------+
| TESTING: VLIB_Connected_Components_Labeling |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_Connected_Components_Labeling testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 256 | 93.37 | 0.00 | 0.00 | 93.37 | Statically generated input | Opt results compared to static
| | | | | | | | results | width=32, height=8, minBlobArea=7,
| | | | | | | | connected8Flag=1, numCCs=2
| 1 | PASS | 512 | 85.71 | 0.00 | 0.00 | 85.71 | Statically generated input | Opt results compared to static
| | | | | | | | results | width=64, height=8, minBlobArea=7,
| | | | | | | | connected8Flag=0, numCCs=8
| 2 | PASS | 512 | 75.88 | 0.00 | 0.00 | 75.88 | Statically generated input | Opt results compared to static
| | | | | | | | results | width=64, height=8, minBlobArea=7,
| | | | | | | | connected8Flag=1, numCCs=7
| 3 | PASS | 64000 | 13.15 | 0.00 | 0.00 | 13.15 | Statically generated input | Opt results compared to static
| | | | | | | | results | width=256, height=250, minBlobArea=7,
| | | | | | | | connected8Flag=1, numCCs=78
| 4 | PASS | 8192 | 104.93 | 0.00 | 0.00 | 104.93 | Statically generated input | Opt results compared to static
| | | | | | | | results | width=128, height=64, minBlobArea=7,
| | | | | | | | connected8Flag=1, numCCs=1
| 5 | PASS | 76800 | 235.96 | 0.00 | 0.00 | 235.96 | Randomly generated input | Opt results compared to static
| | | | | | | | results | width=320, height=240, minBlobArea=7,
| | | | | | | | connected8Flag=1, numCCs=18
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Range:
Min: 13.15 cycles/pt Avg: 101.50 cycles/pt
Max: 235.96 cycles/pt Med: 85.71 cycles/pt
Num pt = CELING(width,32) * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 300 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYint_to_HSLpl |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYint_to_HSLpl testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 391.91 | 241.66 | 0.00 | 391.91 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 512 | 115.25 | 113.97 | 0.00 | 115.25 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, pitch=32, height=16, num_pts=512
| 2 | PASS | 2688 | 101.27 | 105.90 | 0.00 | 101.27 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 3 | PASS | 7040 | 98.70 | 104.53 | 0.00 | 98.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=55, num_pts=7040
| 4 | PASS | 76800 | 98.69 | 104.67 | 0.00 | 98.69 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 5 | PASS | 307200 | 99.42 | 104.68 | 0.00 | 99.42 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 6 | PASS | 32 | 377.16 | 235.38 | 0.00 | 377.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=20, height=2, num_pts=32
| 7 | PASS | 2688 | 100.68 | 106.86 | 0.00 | 100.68 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=84, height=42, num_pts=2688
| 8 | PASS | 7040 | 98.67 | 105.29 | 0.00 | 98.67 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=136, height=55, num_pts=7040
| 9 | PASS | 76800 | 97.61 | 104.62 | 0.00 | 97.61 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=324, height=240, num_pts=76800
| 10 | PASS | 307200 | 97.82 | 104.76 | 0.00 | 97.82 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=644, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 11
Range:
Min: 97.61 cycles/pt Avg: 152.47 cycles/pt
Max: 391.91 cycles/pt Med: 98.70 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYint_to_LABpl |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYint_to_LABpl testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 6558.50 | 5723.91 | 0.00 | 6558.50 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 512 | 6024.64 | 5277.42 | 0.00 | 6024.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, pitch=32, height=16, num_pts=512
| 2 | PASS | 2688 | 5927.80 | 5188.16 | 0.00 | 5927.80 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 3 | PASS | 7040 | 5872.00 | 5133.73 | 0.00 | 5872.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=55, num_pts=7040
| 4 | PASS | 32 | 6309.88 | 5501.00 | 0.00 | 6309.88 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=20, height=2, num_pts=32
| 5 | PASS | 2688 | 5920.68 | 5181.16 | 0.00 | 5920.68 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=84, height=42, num_pts=2688
| 6 | PASS | 7040 | 5900.05 | 5160.99 | 0.00 | 5900.05 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=136, height=55, num_pts=7040
| 7 | PASS | 76800 | 5898.14 | 5159.19 | 0.00 | 5898.14 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 8 | PASS | 76800 | 5890.23 | 5151.66 | 0.00 | 5890.23 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=324, height=240, num_pts=76800
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 9
Range:
Min: 5872.00 cycles/pt Avg: 6033.55 cycles/pt
Max: 6558.50 cycles/pt Med: 5900.05 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 428 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYint_to_LABpl_LUT |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYint_to_LABpl_LUT testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 586.88 | 589.00 | 0.00 | 586.88 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 512 | 395.81 | 409.11 | 0.00 | 395.81 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, pitch=32, height=16, num_pts=512
| 2 | PASS | 2688 | 80.60 | 88.36 | 0.00 | 80.60 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 3 | PASS | 7040 | 73.03 | 80.79 | 0.00 | 73.03 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=55, num_pts=7040
| 4 | PASS | 76800 | 70.90 | 77.18 | 0.00 | 70.90 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 5 | PASS | 307200 | 70.63 | 77.71 | 0.00 | 70.63 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 6 | PASS | 32 | 554.50 | 571.38 | 0.00 | 554.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=22, height=2, num_pts=32
| 7 | PASS | 2604 | 81.26 | 89.25 | 0.00 | 81.26 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=62, pitch=82, height=42, num_pts=2604
| 8 | PASS | 7040 | 72.90 | 81.04 | 0.00 | 72.90 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=136, height=55, num_pts=7040
| 9 | PASS | 76800 | 69.28 | 77.43 | 0.00 | 69.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=324, height=240, num_pts=76800
| 10 | PASS | 307200 | 68.98 | 77.53 | 0.00 | 68.98 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=644, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 11
Range:
Min: 68.98 cycles/pt Avg: 193.16 cycles/pt
Max: 586.88 cycles/pt Med: 72.90 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 104 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYint_to_RGBpl |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYint_to_RGBpl testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 62.53 | 59.28 | 0.00 | 62.53 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 512 | 10.89 | 16.40 | 0.00 | 10.89 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, pitch=32, height=16, num_pts=512
| 2 | PASS | 2688 | 8.00 | 13.22 | 0.00 | 8.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 3 | PASS | 7040 | 7.56 | 13.34 | 0.00 | 7.56 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=55, num_pts=7040
| 4 | PASS | 76800 | 7.91 | 13.46 | 0.00 | 7.91 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 5 | PASS | 307200 | 8.92 | 14.20 | 0.00 | 8.92 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 6 | PASS | 32 | 62.53 | 60.78 | 0.00 | 62.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=20, height=2, num_pts=32
| 7 | PASS | 2688 | 8.94 | 14.07 | 0.00 | 8.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=84, height=42, num_pts=2688
| 8 | PASS | 7040 | 7.47 | 13.55 | 0.00 | 7.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=136, height=55, num_pts=7040
| 9 | PASS | 76800 | 7.86 | 13.40 | 0.00 | 7.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=324, height=240, num_pts=76800
| 10 | PASS | 307200 | 8.80 | 14.04 | 0.00 | 8.80 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=644, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 11
Formula:
Cycles: 9.84714*N - 663.62969*M + 13935
Where: N = width * height; M=height
Range:
Min: 7.47 cycles/pt Avg: 18.31 cycles/pt
Max: 62.53 cycles/pt Med: 8.00 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 36 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYint_to_YUV420pl |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYint_to_YUV420pl testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 37.94 | 45.44 | 0.00 | 37.94 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 2688 | 5.27 | 7.22 | 0.00 | 5.27 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 2 | PASS | 6912 | 4.57 | 7.31 | 0.00 | 4.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=54, num_pts=6912
| 3 | PASS | 76800 | 4.55 | 7.37 | 0.00 | 4.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 4 | PASS | 307200 | 5.11 | 7.86 | 0.00 | 5.11 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 5 | PASS | 32 | 37.00 | 45.63 | 0.00 | 37.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=20, height=2, num_pts=32
| 6 | PASS | 2688 | 5.79 | 8.06 | 0.00 | 5.79 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=84, height=42, num_pts=2688
| 7 | PASS | 6912 | 4.78 | 7.42 | 0.00 | 4.78 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=136, height=54, num_pts=6912
| 8 | PASS | 76800 | 5.13 | 7.35 | 0.00 | 5.13 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=324, height=240, num_pts=76800
| 9 | PASS | 307200 | 5.68 | 7.91 | 0.00 | 5.68 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=644, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 5.96245*N - 383.48895*M + 8737
Where: N = width * height; M = height
Range:
Min: 4.55 cycles/pt Avg: 11.58 cycles/pt
Max: 37.94 cycles/pt Med: 5.13 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYint_to_YUV422pl |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYint_to_YUV422pl testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 38.25 | 55.22 | 0.00 | 38.25 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 2688 | 5.19 | 8.14 | 0.00 | 5.19 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 2 | PASS | 7040 | 4.94 | 8.36 | 0.00 | 4.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=55, num_pts=7040
| 3 | PASS | 76800 | 5.52 | 9.15 | 0.00 | 5.52 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 4 | PASS | 307200 | 6.25 | 9.64 | 0.00 | 6.25 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 5 | PASS | 32 | 50.03 | 36.50 | 0.00 | 50.03 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=20, height=2, num_pts=32
| 6 | PASS | 2688 | 6.09 | 8.78 | 0.00 | 6.09 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=84, height=42, num_pts=2688
| 7 | PASS | 7040 | 5.72 | 8.30 | 0.00 | 5.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=136, height=55, num_pts=7040
| 8 | PASS | 76800 | 5.94 | 9.16 | 0.00 | 5.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=324, height=240, num_pts=76800
| 9 | PASS | 307200 | 6.44 | 9.70 | 0.00 | 6.44 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=644, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: pitch == width
Test vectors run: 5
Formula:
Cycles: 6.26499*N - 14486
Where: N = width * height
Range:
Min: 4.94 cycles/pt Avg: 12.03 cycles/pt
Max: 38.25 cycles/pt Med: 5.19 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: pitch > width
Test vectors run: 5
Formula:
Cycles: 6.96773*N - 352.21212*M + 7718
Where: N = width * height; M = height
Range:
Min: 5.72 cycles/pt Avg: 14.84 cycles/pt
Max: 50.03 cycles/pt Med: 5.94 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYint_to_YUV444pl |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYint_to_YUV444pl testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 38.72 | 38.81 | 0.00 | 38.72 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 2688 | 11.73 | 10.97 | 0.00 | 11.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 2 | PASS | 4400 | 10.61 | 10.65 | 0.00 | 10.61 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=80, pitch=80, height=55, num_pts=4400
| 3 | PASS | 76800 | 12.39 | 11.71 | 0.00 | 12.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 4 | PASS | 307200 | 13.15 | 12.49 | 0.00 | 13.15 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 5 | PASS | 32 | 38.69 | 37.34 | 0.00 | 38.69 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=20, height=2, num_pts=32
| 6 | PASS | 2688 | 12.30 | 12.42 | 0.00 | 12.30 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=84, height=42, num_pts=2688
| 7 | PASS | 4400 | 10.61 | 10.92 | 0.00 | 10.61 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=80, pitch=84, height=55, num_pts=4400
| 8 | PASS | 76800 | 12.33 | 12.40 | 0.00 | 12.33 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=324, height=240, num_pts=76800
| 9 | PASS | 307200 | 13.10 | 13.11 | 0.00 | 13.10 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=644, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 13.89292*N - 516.51440*M + 9858
Where: N = width * height; M = height
Range:
Min: 10.61 cycles/pt Avg: 17.36 cycles/pt
Max: 38.72 cycles/pt Med: 12.33 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYpl_to_YUVint |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYpl_to_YUVint testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 53.72 | 44.06 | 0.00 | 53.72 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 2688 | 6.35 | 7.39 | 0.00 | 6.35 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 2 | PASS | 7040 | 6.14 | 7.78 | 0.00 | 6.14 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=55, num_pts=7040
| 3 | PASS | 76800 | 6.56 | 7.59 | 0.00 | 6.56 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 4 | PASS | 307200 | 7.22 | 8.18 | 0.00 | 7.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 5 | PASS | 32 | 58.34 | 45.47 | 0.00 | 58.34 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=32, height=2, num_pts=32
| 6 | PASS | 2688 | 6.88 | 8.35 | 0.00 | 6.88 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=80, height=42, num_pts=2688
| 7 | PASS | 7040 | 6.41 | 7.89 | 0.00 | 6.41 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=144, height=55, num_pts=7040
| 8 | PASS | 76800 | 6.32 | 8.11 | 0.00 | 6.32 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=336, height=240, num_pts=76800
| 9 | PASS | 307200 | 6.86 | 8.69 | 0.00 | 6.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=656, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: pitch == width
Test vectors run: 5
Formula:
Cycles: 7.23121*N - 12709
Where: N = width * height
Range:
Min: 6.14 cycles/pt Avg: 16.00 cycles/pt
Max: 53.72 cycles/pt Med: 6.35 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: pitch > width
Test vectors run: 5
Formula:
Cycles: 7.42535*N - 380.60763*M + 9410
Where: N = width * height; M = height
Range:
Min: 6.32 cycles/pt Avg: 16.96 cycles/pt
Max: 58.34 cycles/pt Med: 6.41 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_convertUYVYsemipl_to_YUVpl |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_convertUYVYsemipl_to_YUVpl testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 35.06 | 35.16 | 0.00 | 35.06 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=16, height=2, num_pts=32
| 1 | PASS | 2688 | 2.31 | 5.15 | 0.00 | 2.31 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=64, height=42, num_pts=2688
| 2 | PASS | 6912 | 2.18 | 5.63 | 0.00 | 2.18 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=128, height=54, num_pts=6912
| 3 | PASS | 76800 | 2.20 | 5.72 | 0.00 | 2.20 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=320, height=240, num_pts=76800
| 4 | PASS | 307200 | 2.61 | 5.99 | 0.00 | 2.61 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=640, height=480, num_pts=307200
| 5 | PASS | 32 | 29.50 | 35.16 | 0.00 | 29.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, pitch=24, height=2, num_pts=32
| 6 | PASS | 2688 | 3.40 | 5.03 | 0.00 | 3.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, pitch=72, height=42, num_pts=2688
| 7 | PASS | 6912 | 2.60 | 5.78 | 0.00 | 2.60 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, pitch=136, height=54, num_pts=6912
| 8 | PASS | 76800 | 2.56 | 5.77 | 0.00 | 2.56 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, pitch=328, height=240, num_pts=76800
| 9 | PASS | 307200 | 2.95 | 6.04 | 0.00 | 2.95 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=640, pitch=648, height=480, num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: pitch == width
Test vectors run: 5
Formula:
Cycles: 2.61871*N - 7215
Where: N = width * height
Range:
Min: 2.18 cycles/pt Avg: 8.87 cycles/pt
Max: 35.06 cycles/pt Med: 2.20 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: pitch > width
Test vectors run: 5
Formula:
Cycles: 3.35406*N - 273.22763*M + 6877
Where: N = width * height; M = height
Range:
Min: 2.56 cycles/pt Avg: 8.20 cycles/pt
Max: 29.50 cycles/pt Med: 2.60 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_coOccurrenceMatrix |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_coOccurrenceMatrix testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 121476 | 7.17 | 32.14 | 0.00 | 7.17 | Statically generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 191, imageHeight= 159, outWidth= 4, outHeight= 4
| | | | | | | | mode =3
| 1 | PASS | 121476 | 2.43 | 32.14 | 0.00 | 2.43 | Statically generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 191, imageHeight= 159, outWidth= 4, outHeight= 4
| | | | | | | | mode =1
| 2 | PASS | 121476 | 4.09 | 32.08 | 0.00 | 4.09 | Statically generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 191, imageHeight= 159, outWidth= 4, outHeight= 4
| | | | | | | | mode =2
| 3 | PASS | 4096 | 9.28 | 35.88 | 0.00 | 9.28 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 64, imageHeight= 64, outWidth= 4, outHeight= 4
| | | | | | | | mode =3
| 4 | PASS | 230400 | 7.39 | 31.95 | 0.00 | 7.39 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 32, outHeight=
| | | | | | | | 32 mode =3
| 5 | PASS | 230400 | 2.54 | 31.89 | 0.00 | 2.54 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 32, outHeight=
| | | | | | | | 32 mode =1
| 6 | PASS | 230400 | 7.70 | 33.60 | 0.00 | 7.70 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 15, outHeight=
| | | | | | | | 15 mode =3
| 7 | PASS | 230400 | 2.61 | 33.61 | 0.00 | 2.61 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 8, outHeight= 8
| | | | | | | | mode =1
| 8 | PASS | 4096 | 54.65 | 86.22 | 0.00 | 54.65 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 64, imageHeight= 64, outWidth= 256, outHeight=
| | | | | | | | 256 mode =3
| 9 | PASS | 4096 | 610.57 | 86.02 | 0.00 | 610.57 | Constant generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 64, imageHeight= 64, outWidth= 256, outHeight=
| | | | | | | | 256 mode =2
| 10 | PASS | 153600 | 4.56 | 33.82 | 0.00 | 4.56 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 16, outHeight=
| | | | | | | | 16 mode =1
| 11 | PASS | 230400 | 7.72 | 33.57 | 0.00 | 7.72 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 8, outHeight= 8
| | | | | | | | mode =3
| 12 | PASS | 230400 | 4.49 | 33.61 | 0.00 | 4.49 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 32,
| | | | | | | | outHeight= 32 mode =1
| 13 | PASS | 230400 | 7.77 | 33.60 | 0.00 | 7.77 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 320, imageHeight= 240, outWidth= 32,
| | | | | | | | outHeight= 32 mode =3
| 14 | PASS | 57600 | 5.91 | 24.82 | 0.00 | 5.91 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | imageWidth= 160, imageHeight= 120, outWidth= 32,
| | | | | | | | outHeight= 32 mode =3
| 15 | PASS | 12288 | 5.20 | 17.90 | 0.00 | 5.20 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | imageWidth= 64, imageHeight= 64, outWidth= 32, outHeight= 32
| | | | | | | | mode =3
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: Scratch used; internal LUT not required
Test vectors run: 5
Range:
Min: 2.43 cycles/pt Avg: 3.33 cycles/pt
Max: 4.56 cycles/pt Med: 2.54 cycles/pt
Num pt = imageWidth*imageHeight*numOffsets
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: Scratch used; internal LUT required
Test vectors run: 2
Range:
Min: 4.09 cycles/pt Avg: 307.33 cycles/pt
Max: 610.57 cycles/pt Med: 4.09 cycles/pt
Num pt = imageWidth*imageHeight*numOffsets
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 3: Scratch not used; internal LUT required
Test vectors run: 9
Range:
Min: 5.20 cycles/pt Avg: 12.53 cycles/pt
Max: 54.65 cycles/pt Med: 7.39 cycles/pt
Num pt = imageWidth*imageHeight*numOffsets
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 364 bytes
+---------------------------------------+
| TESTING: VLIB_dilate_bin_cross |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_dilate_bin_cross testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1920 | 0.94 | 1.04 | 0.00 | 0.94 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 1 | PASS | 2880 | 0.80 | 0.85 | 0.00 | 0.80 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=96, in_height=32, cols=2880, pitch=96
| 2 | PASS | 3840 | 0.69 | 0.75 | 0.00 | 0.69 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=128, in_height=32, cols=3840, pitch=128
| 3 | PASS | 9408 | 0.56 | 0.57 | 0.00 | 0.56 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=192, in_height=51, cols=9408, pitch=192
| 4 | PASS | 29696 | 0.49 | 0.53 | 0.00 | 0.49 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=256, in_height=118, cols=29696, pitch=256
| 5 | PASS | 76160 | 0.49 | 0.49 | 0.00 | 0.49 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=320, in_height=240, cols=76160, pitch=320
| 6 | PASS | 305920 | 0.45 | 0.45 | 0.00 | 0.45 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=640, in_height=480, cols=305920,
| | | | | | | | pitch=640
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 0.44544*N + 1371
Where: N = cols = in_width * (in_height - 2)
Range:
Min: 0.45 cycles/pt Avg: 0.63 cycles/pt
Max: 0.94 cycles/pt Med: 0.49 cycles/pt
Num pt = cols = in_width * (in_height - 2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_dilate_bin_mask |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_dilate_bin_mask testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1920 | 1.27 | 1.78 | 0.00 | 1.27 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 1 | PASS | 1920 | 1.27 | 1.80 | 0.00 | 1.27 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 2 | PASS | 2880 | 1.00 | 1.55 | 0.00 | 1.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=96, in_height=32, cols=2880, pitch=96
| 3 | PASS | 3840 | 0.88 | 1.41 | 0.00 | 0.88 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=128, in_height=32, cols=3840, pitch=128
| 4 | PASS | 9408 | 0.70 | 1.19 | 0.00 | 0.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=192, in_height=51, cols=9408, pitch=192
| 5 | PASS | 29696 | 0.55 | 1.10 | 0.00 | 0.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=256, in_height=118, cols=29696, pitch=256
| 6 | PASS | 76160 | 0.54 | 1.06 | 0.00 | 0.54 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=320, in_height=240, cols=76160, pitch=320
| 7 | PASS | 305920 | 0.48 | 1.05 | 0.00 | 0.48 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=640, in_height=480, cols=305920,
| | | | | | | | pitch=640
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 0.47966*N + 1965
Where: N = cols = in_width * (in_height - 2)
Range:
Min: 0.48 cycles/pt Avg: 0.84 cycles/pt
Max: 1.27 cycles/pt Med: 0.70 cycles/pt
Num pt = cols = in_width * (in_height - 2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_dilate_bin_square |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_dilate_bin_square testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1920 | 0.98 | 1.12 | 0.00 | 0.98 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 1 | PASS | 2880 | 0.84 | 0.90 | 0.00 | 0.84 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=96, in_height=32, cols=2880, pitch=96
| 2 | PASS | 3840 | 0.73 | 0.81 | 0.00 | 0.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=128, in_height=32, cols=3840, pitch=128
| 3 | PASS | 9408 | 0.58 | 0.62 | 0.00 | 0.58 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=192, in_height=51, cols=9408, pitch=192
| 4 | PASS | 29696 | 0.52 | 0.55 | 0.00 | 0.52 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=256, in_height=118, cols=29696, pitch=256
| 5 | PASS | 76160 | 0.50 | 0.54 | 0.00 | 0.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=320, in_height=240, cols=76160, pitch=320
| 6 | PASS | 305920 | 0.47 | 0.50 | 0.00 | 0.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=640, in_height=480, cols=305920,
| | | | | | | | pitch=640
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 0.46574*N + 1390
Where: N = cols = in_width * (in_height - 2)
Range:
Min: 0.47 cycles/pt Avg: 0.66 cycles/pt
Max: 0.98 cycles/pt Med: 0.52 cycles/pt
Num pt = cols = in_width * (in_height - 2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_disparity_SAD16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_disparity_SAD16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 162 | 297.63 | 224.86 | 0.00 | 297.63 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 1 | PASS | 162 | 300.30 | 229.86 | 0.00 | 300.30 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 2 | PASS | 162 | 299.30 | 232.44 | 0.00 | 299.30 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 3 | PASS | 162 | 299.96 | 230.90 | 0.00 | 299.96 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 4 | PASS | 1431 | 76.92 | 69.76 | 0.00 | 76.92 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=52, height=4, windowSize=5, numDisp=9
| 5 | PASS | 6027 | 99.47 | 89.99 | 0.00 | 99.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=40, height=8, windowSize=5, numDisp=21
| 6 | PASS | 19866 | 109.16 | 100.23 | 0.00 | 109.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=31, height=8, windowSize=5, numDisp=86
| 7 | PASS | 37410 | 122.50 | 108.91 | 0.00 | 122.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=27, height=16, windowSize=9, numDisp=86
| 8 | PASS | 307840 | 44.16 | 39.97 | 0.00 | 44.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=183, height=14, windowSize=11, numDisp=128
| 9 | PASS | 302016 | 52.29 | 46.71 | 0.00 | 52.29 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=119, height=14, windowSize=11, numDisp=192
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 29.64242*N + 2722.66015*M - 12206
Where: N = padWidth*(outHeight-1)*(disparitySearchRange); M = (outHeight-1)*(disparitySearchRange)
Range:
Min: 44.16 cycles/pt Avg: 170.17 cycles/pt
Max: 300.30 cycles/pt Med: 109.16 cycles/pt
Num pt = padWidth*(outHeight-1)*(disparitySearchRange)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_disparity_SAD8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_disparity_SAD8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 162 | 318.52 | 243.04 | 0.00 | 318.52 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 1 | PASS | 162 | 306.78 | 235.01 | 0.00 | 306.78 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 2 | PASS | 162 | 317.48 | 238.40 | 0.00 | 317.48 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 3 | PASS | 162 | 314.37 | 241.34 | 0.00 | 314.37 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 4 | PASS | 1539 | 62.51 | 51.69 | 0.00 | 62.51 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=52, height=4, windowSize=5, numDisp=9
| 5 | PASS | 6027 | 80.87 | 67.55 | 0.00 | 80.87 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=40, height=8, windowSize=5, numDisp=21
| 6 | PASS | 19866 | 110.45 | 86.93 | 0.00 | 110.45 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=31, height=8, windowSize=5, numDisp=86
| 7 | PASS | 42570 | 110.54 | 84.60 | 0.00 | 110.54 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=27, height=16, windowSize=9, numDisp=86
| 8 | PASS | 307840 | 32.00 | 32.87 | 0.00 | 32.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=183, height=14, windowSize=11, numDisp=128
| 9 | PASS | 302016 | 41.92 | 40.20 | 0.00 | 41.92 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=119, height=14, windowSize=11, numDisp=192
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 14.88903*N + 3242.30224*M - 35337
Where: N = padWidth*(outHeight-1)*(disparitySearchRange); M = (outHeight-1)*(disparitySearchRange)
Range:
Min: 32.00 cycles/pt Avg: 169.54 cycles/pt
Max: 318.52 cycles/pt Med: 110.45 cycles/pt
Num pt = padWidth*(outHeight-1)*(disparitySearchRange)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_disparity_SAD_firstRow16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_disparity_SAD_firstRow16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 81 | 273.80 | 248.88 | 0.00 | 273.80 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 1 | PASS | 81 | 266.19 | 251.06 | 0.00 | 266.19 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 2 | PASS | 81 | 263.06 | 241.23 | 0.00 | 263.06 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 3 | PASS | 81 | 263.05 | 239.77 | 0.00 | 263.05 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 4 | PASS | 477 | 96.20 | 135.49 | 0.00 | 96.20 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=52, height=4, windowSize=5, numDisp=9
| 5 | PASS | 861 | 109.33 | 146.97 | 0.00 | 109.33 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=40, height=8, windowSize=5, numDisp=21
| 6 | PASS | 2838 | 154.14 | 188.43 | 0.00 | 154.14 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=31, height=8, windowSize=5, numDisp=86
| 7 | PASS | 2494 | 244.10 | 273.34 | 0.00 | 244.10 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=27, height=16, windowSize=9, numDisp=86
| 8 | PASS | 23680 | 117.40 | 195.34 | 0.00 | 117.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=183, height=14, windowSize=11, numDisp=128
| 9 | PASS | 23232 | 134.11 | 208.45 | 0.00 | 134.11 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=119, height=14, windowSize=11, numDisp=192
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 100.75276*N + 3886.89013*M - 43166
Where: N = padWidth*(disparitySearchRange); M = disparitySearchRange
Range:
Min: 96.20 cycles/pt Avg: 192.14 cycles/pt
Max: 273.80 cycles/pt Med: 154.14 cycles/pt
Num pt = padWidth*(disparitySearchRange)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_disparity_SAD_firstRow8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_disparity_SAD_firstRow8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 81 | 228.40 | 205.12 | 0.00 | 228.40 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 1 | PASS | 81 | 229.58 | 201.67 | 0.00 | 229.58 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 2 | PASS | 81 | 225.06 | 206.53 | 0.00 | 225.06 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 3 | PASS | 81 | 221.21 | 204.09 | 0.00 | 221.21 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=6, height=3, windowSize=3, numDisp=9
| 4 | PASS | 513 | 59.97 | 98.46 | 0.00 | 59.97 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=52, height=4, windowSize=5, numDisp=9
| 5 | PASS | 861 | 69.74 | 115.40 | 0.00 | 69.74 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=40, height=8, windowSize=5, numDisp=21
| 6 | PASS | 2838 | 105.72 | 141.71 | 0.00 | 105.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=31, height=8, windowSize=5, numDisp=86
| 7 | PASS | 2838 | 154.87 | 180.98 | 0.00 | 154.87 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=27, height=16, windowSize=9, numDisp=86
| 8 | PASS | 23680 | 74.08 | 156.17 | 0.00 | 74.08 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=183, height=14, windowSize=11, numDisp=128
| 9 | PASS | 23232 | 90.63 | 166.80 | 0.00 | 90.63 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=119, height=14, windowSize=11, numDisp=192
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Formula:
Cycles: 61.36349*N + 3270.45654*M - 39263
Where: N = padWidth*(disparitySearchRange); M = disparitySearchRange
Range:
Min: 59.97 cycles/pt Avg: 145.93 cycles/pt
Max: 229.58 cycles/pt Med: 105.72 cycles/pt
Num pt = padWidth*(disparitySearchRange)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_erode_bin_cross |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_erode_bin_cross testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1920 | 1.06 | 1.05 | 0.00 | 1.06 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 1 | PASS | 2880 | 0.86 | 0.85 | 0.00 | 0.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=96, in_height=32, cols=2880, pitch=96
| 2 | PASS | 3840 | 0.73 | 0.74 | 0.00 | 0.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=128, in_height=32, cols=3840, pitch=128
| 3 | PASS | 9408 | 0.57 | 0.57 | 0.00 | 0.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=192, in_height=51, cols=9408, pitch=192
| 4 | PASS | 29696 | 0.51 | 0.50 | 0.00 | 0.51 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=256, in_height=118, cols=29696, pitch=256
| 5 | PASS | 76160 | 0.49 | 0.49 | 0.00 | 0.49 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=320, in_height=240, cols=76160, pitch=320
| 6 | PASS | 305920 | 0.45 | 0.45 | 0.00 | 0.45 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=640, in_height=480, cols=305920,
| | | | | | | | pitch=640
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 0.44493*N + 1644
Where: N = cols = in_width * (in_height - 2)
Range:
Min: 0.45 cycles/pt Avg: 0.67 cycles/pt
Max: 1.06 cycles/pt Med: 0.51 cycles/pt
Num pt = cols = in_width * (in_height - 2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_erode_bin_mask |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_erode_bin_mask testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1920 | 1.22 | 1.94 | 0.00 | 1.22 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 1 | PASS | 1920 | 1.22 | 1.91 | 0.00 | 1.22 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 2 | PASS | 2880 | 1.00 | 1.73 | 0.00 | 1.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=96, in_height=32, cols=2880, pitch=96
| 3 | PASS | 3840 | 0.87 | 1.59 | 0.00 | 0.87 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=128, in_height=32, cols=3840, pitch=128
| 4 | PASS | 9408 | 0.70 | 1.20 | 0.00 | 0.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=192, in_height=51, cols=9408, pitch=192
| 5 | PASS | 29696 | 0.61 | 1.08 | 0.00 | 0.61 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=256, in_height=118, cols=29696, pitch=256
| 6 | PASS | 76160 | 0.57 | 1.04 | 0.00 | 0.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=320, in_height=240, cols=76160, pitch=320
| 7 | PASS | 305920 | 0.51 | 1.02 | 0.00 | 0.51 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=640, in_height=480, cols=305920,
| | | | | | | | pitch=640
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 0.50565*N + 2101
Where: N = cols = in_width * (in_height - 2)
Range:
Min: 0.51 cycles/pt Avg: 0.84 cycles/pt
Max: 1.22 cycles/pt Med: 0.70 cycles/pt
Num pt = cols = in_width * (in_height - 2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_erode_bin_singlePixel |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_erode_bin_singlePixel testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1920 | 0.93 | 1.24 | 0.00 | 0.93 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 1 | PASS | 2880 | 0.79 | 0.89 | 0.00 | 0.79 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=96, in_height=32, cols=2880, pitch=96
| 2 | PASS | 3840 | 0.71 | 0.78 | 0.00 | 0.71 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=128, in_height=32, cols=3840, pitch=128
| 3 | PASS | 9408 | 0.58 | 0.61 | 0.00 | 0.58 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=192, in_height=51, cols=9408, pitch=192
| 4 | PASS | 29696 | 0.55 | 0.55 | 0.00 | 0.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=256, in_height=118, cols=29696, pitch=256
| 5 | PASS | 76160 | 0.51 | 0.55 | 0.00 | 0.51 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=320, in_height=240, cols=76160, pitch=320
| 6 | PASS | 305920 | 0.47 | 0.49 | 0.00 | 0.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=640, in_height=480, cols=305920,
| | | | | | | | pitch=640
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 0.46888*N + 1498
Where: N = cols = in_width * (in_height - 2)
Range:
Min: 0.47 cycles/pt Avg: 0.65 cycles/pt
Max: 0.93 cycles/pt Med: 0.55 cycles/pt
Num pt = cols = in_width * (in_height - 2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_erode_bin_square |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_erode_bin_square testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1920 | 0.98 | 1.12 | 0.00 | 0.98 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | in_width=64, in_height=32, cols=1920, pitch=64
| 1 | PASS | 2880 | 0.84 | 0.90 | 0.00 | 0.84 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=96, in_height=32, cols=2880, pitch=96
| 2 | PASS | 3840 | 0.73 | 0.82 | 0.00 | 0.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=128, in_height=32, cols=3840, pitch=128
| 3 | PASS | 9408 | 0.64 | 0.62 | 0.00 | 0.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=192, in_height=51, cols=9408, pitch=192
| 4 | PASS | 29696 | 0.52 | 0.56 | 0.00 | 0.52 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=256, in_height=118, cols=29696, pitch=256
| 5 | PASS | 76160 | 0.51 | 0.55 | 0.00 | 0.51 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=320, in_height=240, cols=76160, pitch=320
| 6 | PASS | 305920 | 0.47 | 0.50 | 0.00 | 0.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | in_width=640, in_height=480, cols=305920,
| | | | | | | | pitch=640
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 0.46339*N + 1585
Where: N = cols = in_width * (in_height - 2)
Range:
Min: 0.47 cycles/pt Avg: 0.67 cycles/pt
Max: 0.98 cycles/pt Med: 0.52 cycles/pt
Num pt = cols = in_width * (in_height - 2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_extract8bitBackgroundS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_extract8bitBackgroundS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 26.72 | 33.97 | 0.00 | 26.72 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 16.22 | 20.13 | 0.00 | 16.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 128 | 8.66 | 13.06 | 0.00 | 8.66 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=128
| 3 | PASS | 512 | 5.26 | 7.06 | 0.00 | 5.26 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 4.69 | 6.12 | 0.00 | 4.69 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 4.39 | 5.36 | 0.00 | 4.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 4.23123*N + 619
Where: N = pixelCount
Range:
Min: 4.39 cycles/pt Avg: 10.99 cycles/pt
Max: 26.72 cycles/pt Med: 5.26 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_extractLumaFromUYUV |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_extractLumaFromUYUV testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 40.88 | 29.88 | 0.00 | 40.88 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inPitch=16, inRows=2, num_pts=32
| 1 | PASS | 2688 | 4.79 | 5.80 | 0.00 | 4.79 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inPitch=64, inRows=42, num_pts=2688
| 2 | PASS | 7040 | 4.35 | 5.57 | 0.00 | 4.35 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inPitch=128, inRows=55, num_pts=7040
| 3 | PASS | 76800 | 4.38 | 5.69 | 0.00 | 4.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inPitch=320, inRows=240, num_pts=76800
| 4 | PASS | 307200 | 4.89 | 6.05 | 0.00 | 4.89 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inPitch=640, inRows=480,
| | | | | | | | num_pts=307200
| 5 | PASS | 32 | 45.22 | 29.75 | 0.00 | 45.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inPitch=20, inRows=2, num_pts=32
| 6 | PASS | 2688 | 5.77 | 6.74 | 0.00 | 5.77 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inPitch=84, inRows=42, num_pts=2688
| 7 | PASS | 7040 | 4.60 | 5.85 | 0.00 | 4.60 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inPitch=136, inRows=55, num_pts=7040
| 8 | PASS | 76800 | 4.44 | 5.66 | 0.00 | 4.44 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inPitch=324, inRows=240, num_pts=76800
| 9 | PASS | 307200 | 4.86 | 5.94 | 0.00 | 4.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inPitch=644, inRows=480,
| | | | | | | | num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: inPitch == inCols
Test vectors run: 5
Formula:
Cycles: 4.89807*N - 8803
Where: N = inCols * inRows
Range:
Min: 4.35 cycles/pt Avg: 11.86 cycles/pt
Max: 40.88 cycles/pt Med: 4.38 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: inPitch > inCols
Test vectors run: 5
Formula:
Cycles: 5.29170*N - 295.98699*M + 8264
Where: N = inCols * inRows; M = inRows
Range:
Min: 4.44 cycles/pt Avg: 12.98 cycles/pt
Max: 45.22 cycles/pt Med: 4.60 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_extractLumaFromYUYV |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_extractLumaFromYUYV testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 36.19 | 29.88 | 0.00 | 36.19 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inPitch=16, inRows=2, num_pts=32
| 1 | PASS | 2688 | 4.49 | 5.81 | 0.00 | 4.49 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inPitch=64, inRows=42, num_pts=2688
| 2 | PASS | 7040 | 4.27 | 5.65 | 0.00 | 4.27 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inPitch=128, inRows=55, num_pts=7040
| 3 | PASS | 76800 | 4.38 | 5.70 | 0.00 | 4.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inPitch=320, inRows=240, num_pts=76800
| 4 | PASS | 307200 | 4.89 | 6.05 | 0.00 | 4.89 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inPitch=640, inRows=480,
| | | | | | | | num_pts=307200
| 5 | PASS | 32 | 36.31 | 29.75 | 0.00 | 36.31 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inPitch=20, inRows=2, num_pts=32
| 6 | PASS | 2688 | 5.64 | 6.76 | 0.00 | 5.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inPitch=84, inRows=42, num_pts=2688
| 7 | PASS | 7040 | 4.55 | 5.79 | 0.00 | 4.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inPitch=136, inRows=55, num_pts=7040
| 8 | PASS | 76800 | 4.43 | 5.65 | 0.00 | 4.43 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inPitch=324, inRows=240, num_pts=76800
| 9 | PASS | 307200 | 4.86 | 5.95 | 0.00 | 4.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inPitch=644, inRows=480,
| | | | | | | | num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: inPitch == inCols
Test vectors run: 5
Formula:
Cycles: 4.89994*N - 9248
Where: N = inCols * inRows
Range:
Min: 4.27 cycles/pt Avg: 10.84 cycles/pt
Max: 36.19 cycles/pt Med: 4.38 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: inPitch > inCols
Test vectors run: 5
Formula:
Cycles: 5.29710*N - 300.43356*M + 8067
Where: N = inCols * inRows; M = inRows
Range:
Min: 4.43 cycles/pt Avg: 11.16 cycles/pt
Max: 36.31 cycles/pt Med: 4.55 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_gauss5x5PyramidKernel_16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_gauss5x5PyramidKernel_16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4 | 364.25 | 497.75 | 0.00 | 364.25 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=5, N=4, M=1
| 1 | PASS | 8 | 202.63 | 285.63 | 0.00 | 202.63 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=8, N=8, M=3
| 2 | PASS | 168 | 30.29 | 78.57 | 0.00 | 30.29 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=16, N=168, M=7
| 3 | PASS | 840 | 23.39 | 64.06 | 0.00 | 23.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=64, N=840, M=31
| 4 | PASS | 1736 | 22.01 | 62.47 | 0.00 | 22.01 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=128, N=1736, M=63
| 5 | PASS | 840 | 21.38 | 63.35 | 0.00 | 21.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=32, N=840, M=15
| 6 | PASS | 1800 | 20.08 | 60.22 | 0.00 | 20.08 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=64, N=1800, M=31
| 7 | PASS | 3720 | 19.84 | 59.21 | 0.00 | 19.84 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=128, N=3720, M=63
| 8 | PASS | 7080 | 19.72 | 58.39 | 0.00 | 19.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=240, N=7080, M=119
| 9 | PASS | 18408 | 18.73 | 55.96 | 0.00 | 18.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=316, height=240, N=18408, M=119
| 10 | PASS | 75208 | 19.32 | 55.90 | 0.00 | 19.32 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=636, height=480, N=75208, M=239
| 11 | PASS | 18328 | 22.85 | 60.02 | 0.00 | 22.85 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=236, height=320, N=18328, M=159
| 12 | PASS | 75048 | 20.28 | 56.56 | 0.00 | 20.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=476, height=640, N=75048, M=319
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 13
Formula:
Cycles: 18.55331*N + 363.90161*M - 9472
Where: N = output pixels = (width-4) * (height-3) / 4; M = output lines = (height-3) / 2
Range:
Min: 18.73 cycles/pt Avg: 61.90 cycles/pt
Max: 364.25 cycles/pt Med: 20.28 cycles/pt
Num pt = output pixels = (width-4) * (height-3) / 4
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_gauss5x5PyramidKernel_8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_gauss5x5PyramidKernel_8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4 | 519.75 | 453.25 | 0.00 | 519.75 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=5, N=4, M=1
| 1 | PASS | 168 | 20.85 | 70.37 | 0.00 | 20.85 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=16, N=168, M=7
| 2 | PASS | 840 | 11.04 | 59.34 | 0.00 | 11.04 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=64, N=840, M=31
| 3 | PASS | 1736 | 9.73 | 58.06 | 0.00 | 9.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=128, N=1736, M=63
| 4 | PASS | 840 | 10.82 | 57.12 | 0.00 | 10.82 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=32, N=840, M=15
| 5 | PASS | 1800 | 9.16 | 54.95 | 0.00 | 9.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=64, N=1800, M=31
| 6 | PASS | 3720 | 8.58 | 54.19 | 0.00 | 8.58 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=128, N=3720, M=63
| 7 | PASS | 7080 | 8.16 | 53.89 | 0.00 | 8.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=124, height=240, N=7080, M=119
| 8 | PASS | 18408 | 7.96 | 51.90 | 0.00 | 7.96 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=316, height=240, N=18408, M=119
| 9 | PASS | 75208 | 13.63 | 51.47 | 0.00 | 13.63 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=636, height=480, N=75208, M=239
| 10 | PASS | 18328 | 9.85 | 54.10 | 0.00 | 9.85 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=236, height=320, N=18328, M=159
| 11 | PASS | 75048 | 15.10 | 53.25 | 0.00 | 15.10 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=476, height=640, N=75048, M=319
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 12
Formula:
Cycles: 18.71414*N - 1316.66870*M + 28886
Where: N = output pixels = (width-4) * (height-3) / 4; M = output columns = (width-4) / 2
Range:
Min: 7.96 cycles/pt Avg: 53.72 cycles/pt
Max: 519.75 cycles/pt Med: 9.85 cycles/pt
Num pt = output pixels = (width-4) * (height-3) / 4
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_goodFeaturestoTrack |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_goodFeaturestoTrack testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 2048 | 23.66 | 65.81 | 0.00 | 23.66 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32, patch=7, thresh=14000,
| | | | | | | | mindist=10, good_points_number=23
| 1 | PASS | 2048 | 29.83 | 72.17 | 0.00 | 29.83 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32, patch=7, thresh=14000,
| | | | | | | | mindist=10, good_points_number=20
| 2 | PASS | 2048 | 24.01 | 70.30 | 0.00 | 24.01 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32, patch=7, thresh=14000,
| | | | | | | | mindist=7, good_points_number=20
| 3 | PASS | 2048 | 25.93 | 70.53 | 0.00 | 25.93 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32, patch=5, thresh=14000,
| | | | | | | | mindist=10, good_points_number=27
| 4 | PASS | 2048 | 38.00 | 93.10 | 0.00 | 38.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32, patch=3, thresh=14000,
| | | | | | | | mindist=10, good_points_number=48
| 5 | PASS | 24320 | 22.27 | 71.59 | 0.00 | 22.27 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=7, thresh=14000,
| | | | | | | | mindist=10, good_points_number=50
| 6 | PASS | 24320 | 28.36 | 80.27 | 0.00 | 28.36 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=5, thresh=14000,
| | | | | | | | mindist=10, good_points_number=50
| 7 | PASS | 24320 | 55.24 | 114.09 | 0.00 | 55.24 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=3, thresh=14000,
| | | | | | | | mindist=10, good_points_number=50
| 8 | PASS | 24320 | 70.33 | 117.90 | 0.00 | 70.33 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=7, thresh=250,
| | | | | | | | mindist=10, good_points_number=50
| 9 | PASS | 24320 | 72.90 | 123.05 | 0.00 | 72.90 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=5, thresh=250,
| | | | | | | | mindist=10, good_points_number=50
| 10 | PASS | 24320 | 76.04 | 133.11 | 0.00 | 76.04 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=3, thresh=250,
| | | | | | | | mindist=10, good_points_number=50
| 11 | PASS | 24320 | 15.22 | 64.69 | 0.00 | 15.22 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=7, thresh=14000,
| | | | | | | | mindist=10, good_points_number=0
| 12 | PASS | 24320 | 15.22 | 67.21 | 0.00 | 15.22 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=5, thresh=14000,
| | | | | | | | mindist=10, good_points_number=0
| 13 | PASS | 24320 | 15.87 | 74.84 | 0.00 | 15.87 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=3, thresh=14000,
| | | | | | | | mindist=10, good_points_number=0
| 14 | PASS | 24320 | 170.39 | 219.52 | 0.00 | 170.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=7, thresh=14000,
| | | | | | | | mindist=10, good_points_number=50
| 15 | PASS | 24320 | 775.91 | 827.65 | 0.00 | 775.91 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=5, thresh=14000,
| | | | | | | | mindist=10, good_points_number=50
| 16 | PASS | 24320 | 5710.80 | 5769.48 | 0.00 | 5710.80 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=152, patch=3, thresh=14000,
| | | | | | | | mindist=10, good_points_number=50
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 17
Range:
Min: 15.22 cycles/pt Avg: 421.76 cycles/pt
Max: 5710.80 cycles/pt Med: 28.36 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 176 bytes
+---------------------------------------+
| TESTING: VLIB_gradientH5x5PyramidKernel_8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_gradientH5x5PyramidKernel_8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4 | 405.50 | 501.50 | 0.00 | 405.50 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=5, num_output_pt=4
| 1 | PASS | 4 | 413.00 | 573.50 | 0.00 | 413.00 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=5, num_output_pt=4
| 2 | PASS | 392 | 27.03 | 62.21 | 0.00 | 27.03 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=32, num_output_pt=392
| 3 | PASS | 4484 | 17.73 | 53.37 | 0.00 | 17.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=156, height=121, num_output_pt=4484
| 4 | PASS | 18880 | 16.00 | 51.83 | 0.00 | 16.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=324, height=240, num_output_pt=18880
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 14.54763*N + 219.08050*M + 1482
Where: N = (width-4)/2) * ((height-3)/2); M = (height-3)/2
Range:
Min: 16.00 cycles/pt Avg: 175.85 cycles/pt
Max: 413.00 cycles/pt Med: 17.73 cycles/pt
Num pt = (width-4)/2) * ((height-3)/2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_gradientV5x5PyramidKernel_8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_gradientV5x5PyramidKernel_8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4 | 337.00 | 412.25 | 0.00 | 337.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=5, num_output_pt=4
| 1 | PASS | 4 | 337.00 | 410.75 | 0.00 | 337.00 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=5, num_output_pt=4
| 2 | PASS | 392 | 22.69 | 52.55 | 0.00 | 22.69 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=60, height=32, num_output_pt=392
| 3 | PASS | 4484 | 14.23 | 43.85 | 0.00 | 14.23 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=156, height=121, num_output_pt=4484
| 4 | PASS | 18880 | 12.97 | 42.01 | 0.00 | 12.97 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=324, height=240, num_output_pt=18880
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 11.91547*N + 156.49969*M + 1377
Where: N = (width-4)/2) * ((height-3)/2); M = (height-3)/2
Range:
Min: 12.97 cycles/pt Avg: 144.78 cycles/pt
Max: 337.00 cycles/pt Med: 14.23 cycles/pt
Num pt = (width-4)/2) * ((height-3)/2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_grayscale_morphology |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
GRAYSCALE_MORPH_DILATE testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 34596 | 0.51 | 11.01 | 0.00 | 0.51 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 3 | Mask with all ones
| 1 | PASS | 90000 | 0.38 | 7.42 | 0.00 | 0.38 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 5 | Mask with random
| | | | | | | | ones
| 2 | PASS | 55800 | 0.70 | 10.34 | 0.00 | 0.70 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 5 | Mask with all ones
| 3 | PASS | 178200 | 0.46 | 7.09 | 0.00 | 0.46 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 11 | Mask with random
| | | | | | | | ones
| 4 | PASS | 48384 | 1.28 | 29.98 | 0.00 | 1.28 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones :: Special Optmization as (blk_h-se_h+1)%se_h=0
| 5 | PASS | 48384 | 1.38 | 14.32 | 0.00 | 1.38 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 3, se_h= 1 | Mask with random
| | | | | | | | ones
| 6 | PASS | 8627300 | 0.32 | 4.85 | 0.00 | 0.32 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 11, se_h= 11 | Mask with all
| | | | | | | | ones
| 7 | PASS | 681156 | 0.40 | 10.93 | 0.00 | 0.40 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with random
| | | | | | | | ones
| 8 | PASS | 681156 | 0.40 | 10.93 | 0.00 | 0.40 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with all
| | | | | | | | ones
| 9 | PASS | 5645640 | 0.34 | 4.92 | 0.00 | 0.34 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 322, blk_h= 241, se_w= 11, se_h= 7 | Mask with random
| | | | | | | | ones
| 10 | PASS | 948480 | 0.46 | 26.11 | 0.00 | 0.46 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones
| 11 | PASS | 973440 | 0.23 | 26.11 | 0.00 | 0.23 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 246, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
| 12 | PASS | 228480 | 1.27 | 29.81 | 0.00 | 1.27 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones
| 13 | PASS | 230400 | 1.16 | 29.81 | 0.00 | 1.16 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 242, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 14
Range:
Min: 0.23 cycles/pt Avg: 0.66 cycles/pt
Max: 1.38 cycles/pt Med: 0.46 cycles/pt
Num pt = ((blk_w - se_w + 1) * (blk_h - se_h + 1) * se_w * se_h)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 80 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
GRAYSCALE_MORPH_ERODE testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 34596 | 0.54 | 11.03 | 0.00 | 0.54 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 3 | Mask with all ones
| 1 | PASS | 90000 | 0.33 | 7.41 | 0.00 | 0.33 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 5 | Mask with random
| | | | | | | | ones
| 2 | PASS | 55800 | 0.71 | 10.35 | 0.00 | 0.71 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 5 | Mask with all ones
| 3 | PASS | 178200 | 0.46 | 7.08 | 0.00 | 0.46 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 11 | Mask with random
| | | | | | | | ones
| 4 | PASS | 48384 | 1.29 | 29.99 | 0.00 | 1.29 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones :: Special Optmization as (blk_h-se_h+1)%se_h=0
| 5 | PASS | 48384 | 1.39 | 14.31 | 0.00 | 1.39 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 3, se_h= 1 | Mask with random
| | | | | | | | ones
| 6 | PASS | 8627300 | 0.32 | 4.85 | 0.00 | 0.32 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 11, se_h= 11 | Mask with all
| | | | | | | | ones
| 7 | PASS | 681156 | 0.40 | 10.93 | 0.00 | 0.40 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with random
| | | | | | | | ones
| 8 | PASS | 681156 | 0.40 | 10.93 | 0.00 | 0.40 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with all
| | | | | | | | ones
| 9 | PASS | 5645640 | 0.34 | 4.92 | 0.00 | 0.34 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 322, blk_h= 241, se_w= 11, se_h= 7 | Mask with random
| | | | | | | | ones
| 10 | PASS | 948480 | 0.46 | 26.11 | 0.00 | 0.46 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones
| 11 | PASS | 973440 | 0.23 | 26.11 | 0.00 | 0.23 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 246, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
| 12 | PASS | 228480 | 1.27 | 29.81 | 0.00 | 1.27 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones
| 13 | PASS | 230400 | 1.17 | 29.81 | 0.00 | 1.17 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 242, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 14
Range:
Min: 0.23 cycles/pt Avg: 0.66 cycles/pt
Max: 1.39 cycles/pt Med: 0.46 cycles/pt
Num pt = ((blk_w - se_w + 1) * (blk_h - se_h + 1) * se_w * se_h)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 108 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
GRAYSCALE_MORPH_OPEN testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 34596 | 0.84 | 21.16 | 0.00 | 0.84 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 3 | Mask with all ones
| 1 | PASS | 90000 | 0.62 | 13.81 | 0.00 | 0.62 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 5 | Mask with random
| | | | | | | | ones
| 2 | PASS | 55800 | 1.31 | 19.60 | 0.00 | 1.31 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 5 | Mask with all ones
| 3 | PASS | 178200 | 0.76 | 12.44 | 0.00 | 0.76 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 11 | Mask with random
| | | | | | | | ones
| 4 | PASS | 48384 | 2.40 | 58.82 | 0.00 | 2.40 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones :: Special Optmization as (blk_h-se_h+1)%se_h=0
| 5 | PASS | 48384 | 2.77 | 27.70 | 0.00 | 2.77 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 3, se_h= 1 | Mask with random
| | | | | | | | ones
| 6 | PASS | 8627300 | 0.62 | 9.33 | 0.00 | 0.62 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 11, se_h= 11 | Mask with all
| | | | | | | | ones
| 7 | PASS | 681156 | 0.73 | 21.62 | 0.00 | 0.73 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with random
| | | | | | | | ones
| 8 | PASS | 681156 | 0.73 | 21.62 | 0.00 | 0.73 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with all
| | | | | | | | ones
| 9 | PASS | 5645640 | 0.65 | 9.54 | 0.00 | 0.65 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 322, blk_h= 241, se_w= 11, se_h= 7 | Mask with random
| | | | | | | | ones
| 10 | PASS | 948480 | 0.86 | 50.77 | 0.00 | 0.86 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones
| 11 | PASS | 973440 | 0.57 | 50.80 | 0.00 | 0.57 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 246, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
| 12 | PASS | 228480 | 2.82 | 59.07 | 0.00 | 2.82 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones
| 13 | PASS | 230400 | 2.36 | 59.03 | 0.00 | 2.36 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 242, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 14
Range:
Min: 0.57 cycles/pt Avg: 1.29 cycles/pt
Max: 2.82 cycles/pt Med: 0.76 cycles/pt
Num pt = ((blk_w - se_w + 1) * (blk_h - se_h + 1) * se_w * se_h)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 108 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
GRAYSCALE_MORPH_CLOSE testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 34596 | 0.85 | 21.15 | 0.00 | 0.85 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 3 | Mask with all ones
| 1 | PASS | 90000 | 0.63 | 13.81 | 0.00 | 0.63 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 5 | Mask with random
| | | | | | | | ones
| 2 | PASS | 55800 | 1.31 | 19.59 | 0.00 | 1.31 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 5 | Mask with all ones
| 3 | PASS | 178200 | 0.77 | 12.45 | 0.00 | 0.77 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 11 | Mask with random
| | | | | | | | ones
| 4 | PASS | 48384 | 2.41 | 58.82 | 0.00 | 2.41 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones :: Special Optmization as (blk_h-se_h+1)%se_h=0
| 5 | PASS | 48384 | 2.77 | 27.69 | 0.00 | 2.77 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 3, se_h= 1 | Mask with random
| | | | | | | | ones
| 6 | PASS | 8627300 | 0.62 | 9.33 | 0.00 | 0.62 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 11, se_h= 11 | Mask with all
| | | | | | | | ones
| 7 | PASS | 681156 | 0.73 | 21.62 | 0.00 | 0.73 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with random
| | | | | | | | ones
| 8 | PASS | 681156 | 0.73 | 21.62 | 0.00 | 0.73 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with all
| | | | | | | | ones
| 9 | PASS | 5645640 | 0.65 | 9.54 | 0.00 | 0.65 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 322, blk_h= 241, se_w= 11, se_h= 7 | Mask with random
| | | | | | | | ones
| 10 | PASS | 948480 | 0.86 | 50.77 | 0.00 | 0.86 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones
| 11 | PASS | 973440 | 0.58 | 50.80 | 0.00 | 0.58 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 246, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
| 12 | PASS | 228480 | 2.82 | 59.07 | 0.00 | 2.82 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones
| 13 | PASS | 230400 | 2.36 | 59.02 | 0.00 | 2.36 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 242, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 14
Range:
Min: 0.58 cycles/pt Avg: 1.29 cycles/pt
Max: 2.82 cycles/pt Med: 0.77 cycles/pt
Num pt = ((blk_w - se_w + 1) * (blk_h - se_h + 1) * se_w * se_h)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 108 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
GRAYSCALE_MORPH_TOPHAT testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 34596 | 0.93 | 21.44 | 0.00 | 0.93 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 3 | Mask with all ones
| 1 | PASS | 90000 | 0.63 | 13.90 | 0.00 | 0.63 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 5 | Mask with random
| | | | | | | | ones
| 2 | PASS | 55800 | 1.34 | 19.75 | 0.00 | 1.34 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 5 | Mask with all ones
| 3 | PASS | 178200 | 0.78 | 12.48 | 0.00 | 0.78 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 11 | Mask with random
| | | | | | | | ones
| 4 | PASS | 48384 | 2.79 | 63.57 | 0.00 | 2.79 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones :: Special Optmization as (blk_h-se_h+1)%se_h=0
| 5 | PASS | 48384 | 2.56 | 28.91 | 0.00 | 2.56 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 3, se_h= 1 | Mask with random
| | | | | | | | ones
| 6 | PASS | 8627300 | 0.62 | 9.35 | 0.00 | 0.62 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 11, se_h= 11 | Mask with all
| | | | | | | | ones
| 7 | PASS | 681156 | 0.75 | 21.93 | 0.00 | 0.75 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with random
| | | | | | | | ones
| 8 | PASS | 681156 | 0.75 | 21.93 | 0.00 | 0.75 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with all
| | | | | | | | ones
| 9 | PASS | 5645640 | 0.65 | 9.58 | 0.00 | 0.65 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 322, blk_h= 241, se_w= 11, se_h= 7 | Mask with random
| | | | | | | | ones
| 10 | PASS | 948480 | 0.92 | 51.00 | 0.00 | 0.92 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones
| 11 | PASS | 973440 | 0.70 | 51.02 | 0.00 | 0.70 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 246, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
| 12 | PASS | 228480 | 2.71 | 60.02 | 0.00 | 2.71 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones
| 13 | PASS | 230400 | 2.63 | 60.01 | 0.00 | 2.63 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 242, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 14
Range:
Min: 0.62 cycles/pt Avg: 1.34 cycles/pt
Max: 2.79 cycles/pt Med: 0.78 cycles/pt
Num pt = ((blk_w - se_w + 1) * (blk_h - se_h + 1) * se_w * se_h)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 108 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
GRAYSCALE_MORPH_BOTHAT testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 34596 | 0.93 | 21.46 | 0.00 | 0.93 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 3 | Mask with all ones
| 1 | PASS | 90000 | 0.64 | 13.91 | 0.00 | 0.64 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 5 | Mask with random
| | | | | | | | ones
| 2 | PASS | 55800 | 1.33 | 19.77 | 0.00 | 1.33 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 5 | Mask with all ones
| 3 | PASS | 178200 | 0.78 | 12.48 | 0.00 | 0.78 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 11 | Mask with random
| | | | | | | | ones
| 4 | PASS | 48384 | 2.78 | 63.65 | 0.00 | 2.78 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones :: Special Optmization as (blk_h-se_h+1)%se_h=0
| 5 | PASS | 48384 | 2.55 | 28.91 | 0.00 | 2.55 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 3, se_h= 1 | Mask with random
| | | | | | | | ones
| 6 | PASS | 8627300 | 0.62 | 9.35 | 0.00 | 0.62 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 11, se_h= 11 | Mask with all
| | | | | | | | ones
| 7 | PASS | 681156 | 0.75 | 21.92 | 0.00 | 0.75 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with random
| | | | | | | | ones
| 8 | PASS | 681156 | 0.75 | 21.92 | 0.00 | 0.75 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with all
| | | | | | | | ones
| 9 | PASS | 5645640 | 0.65 | 9.58 | 0.00 | 0.65 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 322, blk_h= 241, se_w= 11, se_h= 7 | Mask with random
| | | | | | | | ones
| 10 | PASS | 948480 | 0.92 | 50.99 | 0.00 | 0.92 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones
| 11 | PASS | 973440 | 0.70 | 51.02 | 0.00 | 0.70 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 246, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
| 12 | PASS | 228480 | 2.71 | 60.02 | 0.00 | 2.71 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones
| 13 | PASS | 230400 | 2.64 | 60.01 | 0.00 | 2.64 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 242, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 14
Range:
Min: 0.62 cycles/pt Avg: 1.34 cycles/pt
Max: 2.78 cycles/pt Med: 0.78 cycles/pt
Num pt = ((blk_w - se_w + 1) * (blk_h - se_h + 1) * se_w * se_h)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 108 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
GRAYSCALE_MORPH_GRADIENT testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 34596 | 0.71 | 22.24 | 0.00 | 0.71 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 3 | Mask with all ones
| 1 | PASS | 90000 | 0.46 | 14.91 | 0.00 | 0.46 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 5 | Mask with random
| | | | | | | | ones
| 2 | PASS | 55800 | 0.87 | 20.84 | 0.00 | 0.87 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 3, se_h= 5 | Mask with all ones
| 3 | PASS | 178200 | 0.60 | 14.20 | 0.00 | 0.60 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 64, blk_h= 64, se_w= 5, se_h= 11 | Mask with random
| | | | | | | | ones
| 4 | PASS | 48384 | 1.44 | 60.51 | 0.00 | 1.44 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones :: Special Optmization as (blk_h-se_h+1)%se_h=0
| 5 | PASS | 48384 | 1.63 | 29.13 | 0.00 | 1.63 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 128, blk_h= 128, se_w= 3, se_h= 1 | Mask with random
| | | | | | | | ones
| 6 | PASS | 8627300 | 0.45 | 9.72 | 0.00 | 0.45 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 11, se_h= 11 | Mask with all
| | | | | | | | ones
| 7 | PASS | 681156 | 0.56 | 22.25 | 0.00 | 0.56 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with random
| | | | | | | | ones
| 8 | PASS | 681156 | 0.56 | 22.25 | 0.00 | 0.56 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 3, se_h= 3 | Mask with all
| | | | | | | | ones
| 9 | PASS | 5645640 | 0.47 | 9.88 | 0.00 | 0.47 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 322, blk_h= 241, se_w= 11, se_h= 7 | Mask with random
| | | | | | | | ones
| 10 | PASS | 948480 | 0.42 | 52.45 | 0.00 | 0.42 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones
| 11 | PASS | 973440 | 0.24 | 52.46 | 0.00 | 0.24 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 246, se_w= 1, se_h= 13 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
| 12 | PASS | 228480 | 1.17 | 60.72 | 0.00 | 1.17 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 240, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones
| 13 | PASS | 230400 | 1.31 | 60.69 | 0.00 | 1.31 | Randomly generated input | Opt and NatC results compared |
| | | | | | | | blk_w= 320, blk_h= 242, se_w= 1, se_h= 3 | Mask with all
| | | | | | | | ones : Special Optmization as (blk_h-se_h+1)%se_h=0
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 14
Range:
Min: 0.24 cycles/pt Avg: 0.78 cycles/pt
Max: 1.63 cycles/pt Med: 0.56 cycles/pt
Num pt = ((blk_w - se_w + 1) * (blk_h - se_h + 1) * se_w * se_h)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 160 bytes
+---------------------------------------+
| TESTING: VLIB_hammingDistance |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_hammingDistance testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 128 | 13.16 | 27.09 | 0.00 | 13.16 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | mode=0, xSize=32, ySize=4
| 1 | PASS | 128 | 12.50 | 33.44 | 0.00 | 12.50 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | mode=1, xSize=32, ySize=4
| 2 | PASS | 1024 | 3.53 | 22.08 | 0.00 | 3.53 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | mode=0, xSize=256, ySize=4
| 3 | PASS | 1024 | 3.34 | 21.27 | 0.00 | 3.34 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | mode=1, xSize=256, ySize=4
| 4 | PASS | 1024 | 3.93 | 27.77 | 0.00 | 3.93 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=0, xSize=128, ySize=8
| 5 | PASS | 896 | 3.55 | 27.45 | 0.00 | 3.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=1, xSize=128, ySize=7
| 6 | PASS | 1024 | 5.03 | 27.86 | 0.00 | 5.03 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=0, xSize=64, ySize=16
| 7 | PASS | 256 | 7.21 | 30.42 | 0.00 | 7.21 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=1, xSize=64, ySize=4
| 8 | PASS | 256 | 9.02 | 31.49 | 0.00 | 9.02 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=0, xSize=32, ySize=8
| 9 | PASS | 256 | 7.94 | 31.09 | 0.00 | 7.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=1, xSize=16, ySize=16
| 10 | PASS | 256 | 10.23 | 33.15 | 0.00 | 10.23 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=0, xSize=8, ySize=32
| 11 | PASS | 256 | 11.29 | 34.37 | 0.00 | 11.29 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | mode=1, xSize=4, ySize=64
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: mode=0, One-to-one
Test vectors run: 6
Formula:
Cycles: 2.75073*N + 33.70095*M + 1127
Where: N = xSize * ySize; M = ySize
Range:
Min: 3.53 cycles/pt Avg: 7.48 cycles/pt
Max: 13.16 cycles/pt Med: 5.03 cycles/pt
Num pt = xSize * ySize
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: mode=1, Many-to-one
Test vectors run: 6
Formula:
Cycles: 2.02868*N + 17.33717*M + 1256
Where: N = xSize * ySize; M = ySize
Range:
Min: 3.34 cycles/pt Avg: 7.64 cycles/pt
Max: 12.50 cycles/pt Med: 7.21 cycles/pt
Num pt = xSize * ySize
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_harrisScore_7x7 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_harrisScore_7x7 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4096 | 27.31 | 175.10 | 0.00 | 27.31 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=64, k=1310,
| 1 | PASS | 3072 | 27.29 | 167.78 | 0.00 | 27.29 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=48, height=64, k=1310,
| 2 | PASS | 2048 | 28.15 | 150.83 | 0.00 | 28.15 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=64, k=1310,
| 3 | PASS | 1024 | 31.22 | 130.52 | 0.00 | 31.22 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=32, k=1310,
| 4 | PASS | 76800 | 26.63 | 221.30 | 0.00 | 26.63 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, k=1310,
| 5 | PASS | 76800 | 26.60 | 221.44 | 0.00 | 26.60 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, k=1600,
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 26.57071*N + 3315
Where: N = width * height
Range:
Min: 26.60 cycles/pt Avg: 27.87 cycles/pt
Max: 31.22 cycles/pt Med: 27.29 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 20 bytes
+---------------------------------------+
| TESTING: VLIB_harrisScore_7x7_S32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_harrisScore_7x7_S32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 4096 | 27.17 | 171.20 | 0.00 | 27.17 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=64, k=1310,
| 1 | PASS | 3072 | 27.63 | 163.16 | 0.00 | 27.63 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=48, height=64, k=1310,
| 2 | PASS | 2048 | 27.85 | 147.46 | 0.00 | 27.85 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=64, k=1310,
| 3 | PASS | 1024 | 31.80 | 126.88 | 0.00 | 31.80 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=32, k=1310,
| 4 | PASS | 76800 | 27.69 | 215.57 | 0.00 | 27.69 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, k=1310,
| 5 | PASS | 76800 | 27.68 | 215.70 | 0.00 | 27.68 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, k=1600,
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 27.67538*N + 622
Where: N = width * height
Range:
Min: 27.17 cycles/pt Avg: 28.30 cycles/pt
Max: 31.80 cycles/pt Med: 27.68 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 20 bytes
+---------------------------------------+
| TESTING: VLIB_histogram_1D_Init_U16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_histogram_1D_Init_U16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 235 | 11.19 | 11.95 | 0.00 | 11.19 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=10, maxSize=250
| 1 | PASS | 20 | 43.90 | 40.60 | 0.00 | 43.90 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=25
| 2 | PASS | 251 | 10.86 | 11.52 | 0.00 | 10.86 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=3, maxSize=256
| 3 | PASS | 71 | 18.55 | 18.76 | 0.00 | 18.55 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=13, maxSize=240
| 4 | PASS | 511 | 9.41 | 10.30 | 0.00 | 9.41 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=7, maxSize=512
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 7.97294*N + 736
Where: N = binEdges[numBins-1] - binEdges[0] + 1
Range:
Min: 9.41 cycles/pt Avg: 18.78 cycles/pt
Max: 43.90 cycles/pt Med: 10.86 cycles/pt
Num pt = binEdges[numBins-1] - binEdges[0] + 1
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_histogram_1D_Init_U8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_histogram_1D_Init_U8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 235 | 11.47 | 12.55 | 0.00 | 11.47 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=10, maxSize=250
| 1 | PASS | 20 | 48.80 | 50.75 | 0.00 | 48.80 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=25
| 2 | PASS | 251 | 11.25 | 12.45 | 0.00 | 11.25 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=3, maxSize=256
| 3 | PASS | 71 | 19.49 | 20.76 | 0.00 | 19.49 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=13, maxSize=240
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 4
Formula:
Cycles: 8.0*N + 816
Where: N = binEdges[numBins-1] - binEdges[0] + 1
Range:
Min: 11.25 cycles/pt Avg: 22.75 cycles/pt
Max: 48.80 cycles/pt Med: 11.47 cycles/pt
Num pt = binEdges[numBins-1] - binEdges[0] + 1
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_histogram_1D_U16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_histogram_1D_U16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 100 | 25.71 | 24.49 | 0.00 | 25.71 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=10, maxSize=254, numX=100
| 1 | PASS | 64 | 25.34 | 23.33 | 0.00 | 25.34 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=25, numX=64
| 2 | PASS | 512 | 9.86 | 8.43 | 0.00 | 9.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=3, maxSize=253, numX=512
| 3 | PASS | 4096 | 6.58 | 6.36 | 0.00 | 6.58 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=13, maxSize=200, numX=4096
| 4 | PASS | 4096 | 6.75 | 6.59 | 0.00 | 6.75 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=7, maxSize=512, numX=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 6.26233*N + 1662
Where: N = numX
Range:
Min: 6.58 cycles/pt Avg: 14.85 cycles/pt
Max: 25.71 cycles/pt Med: 6.75 cycles/pt
Num pt = numX
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_histogram_1D_U8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_histogram_1D_U8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 100 | 18.53 | 19.83 | 0.00 | 18.53 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=10, maxSize=254, numX=100
| 1 | PASS | 64 | 21.84 | 20.81 | 0.00 | 21.84 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=25, numX=64
| 2 | PASS | 512 | 6.38 | 6.60 | 0.00 | 6.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=253, numX=512
| 3 | PASS | 4096 | 3.87 | 4.16 | 0.00 | 3.87 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=14, maxSize=200, numX=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 4
Formula:
Cycles: 3.54076*N + 1371
Where: N = numX
Range:
Min: 3.87 cycles/pt Avg: 12.66 cycles/pt
Max: 21.84 cycles/pt Med: 6.38 cycles/pt
Num pt = numX
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_histogram_equal_8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_histogram_equal_8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 16384 | 11.08 | 11.67 | 0.00 | 11.08 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | n=16384
| 1 | PASS | 64 | 81.13 | 78.95 | 0.00 | 81.13 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | n=64
| 2 | PASS | 1024 | 15.28 | 15.58 | 0.00 | 15.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | n=1024
| 3 | PASS | 4096 | 11.88 | 12.48 | 0.00 | 11.88 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | n=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 4
Formula:
Cycles: 10.80045*N + 4510
Where: N = n
Range:
Min: 11.08 cycles/pt Avg: 29.84 cycles/pt
Max: 81.13 cycles/pt Med: 11.88 cycles/pt
Num pt = n
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 260 bytes
+---------------------------------------+
| TESTING: VLIB_histogram_equal_8_f32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_histogram_equal_8_f32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 16384 | 42.78 | 46.26 | 0.00 | 42.78 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | n=16384
| 1 | PASS | 64 | 147.16 | 140.59 | 0.00 | 147.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | n=64
| 2 | PASS | 1024 | 48.74 | 52.02 | 0.00 | 48.74 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | n=1024
| 3 | PASS | 4096 | 44.06 | 47.34 | 0.00 | 44.06 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | n=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 4
Formula:
Cycles: 42.37407*N + 6696
Where: N = n
Range:
Min: 42.78 cycles/pt Avg: 70.68 cycles/pt
Max: 147.16 cycles/pt Med: 44.06 cycles/pt
Num pt = n
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 1068 bytes
+---------------------------------------+
| TESTING: VLIB_histogram_nD_U16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_histogram_nD_U16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 100 | 26.76 | 23.59 | 0.00 | 26.76 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numX=25, dimX=4, maxSize=254, outsize=48
| 1 | PASS | 48 | 123094.15 | 767.75 | 0.00 | 123094.15 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=4, dimX=12, maxSize=25, outsize=725760
| 2 | PASS | 128 | 21.65 | 20.31 | 0.00 | 21.65 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=32, dimX=4, maxSize=253, outsize=48
| 3 | PASS | 256 | 14.15 | 13.52 | 0.00 | 14.15 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=64, dimX=4, maxSize=253, outsize=48
| 4 | PASS | 2048 | 6.90 | 8.02 | 0.00 | 6.90 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=512, dimX=4, maxSize=253, outsize=48
| 5 | PASS | 12288 | 24.05 | 22.61 | 0.00 | 24.05 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=4096, dimX=3, maxSize=200, outsize=24800
| 6 | PASS | 8192 | 8.74 | 9.69 | 0.00 | 8.74 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=4096, dimX=2, maxSize=512, outsize=180
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 5.61566*N + 5.69578*M + 8.13938*L + 1000
Where: N = numX*dimX; M = numX; L = outsize (product of num of bins in each dim)
Range:
Min: 6.90 cycles/pt Avg: 17599.49 cycles/pt
Max: 123094.15 cycles/pt Med: 14.15 cycles/pt
Num pt = numX*dimX
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_homographyDecomposition |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_homographyDecomposition testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 200 | 34851.62 | 37055.74 | 0.00 | 34851.62 | Statically generated input | Opt and NatC results compared
| | | | | | | | to ref | "Num pt" value is actually the number of times the
| | | | | | | | test was called, so that the value shown is the average cycle count per call
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 1
Range:
Min: 34851.62 cycles/pt Avg: 34851.62 cycles/pt
Max: 34851.62 cycles/pt Med: 34851.62 cycles/pt
Num pt = 1 homography matrix
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 1156 bytes
+---------------------------------------+
| TESTING: VLIB_houghLineFromList |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_houghLineFromList testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1023945 | 2.25 | 2.73 | 0.00 | 2.25 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | listSize=3835, thetaRange=267, rhoMaxLength=267
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 1
Formula:
Equation could not be determined.
Range:
Min: 2.25 cycles/pt Avg: 2.25 cycles/pt
Max: 2.25 cycles/pt Med: 2.25 cycles/pt
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 72 bytes
+---------------------------------------+
| TESTING: VLIB_hysteresisThresholding |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_hysteresisThresholding testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 2916 | 12.52 | 12.40 | 0.00 | 12.52 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=54, height=54
| 1 | PASS | 110 | 22.95 | 22.74 | 0.00 | 22.95 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | width=22, height=5
| 2 | PASS | 16 | 133.38 | 122.94 | 0.00 | 133.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=2
| 3 | PASS | 40 | 56.72 | 55.17 | 0.00 | 56.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=20, height=2
| 4 | PASS | 2340 | 21.68 | 21.21 | 0.00 | 21.68 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=26, height=90
| 5 | PASS | 2800 | 21.45 | 23.00 | 0.00 | 21.45 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=28, height=100
| 6 | PASS | 16500 | 20.33 | 19.89 | 0.00 | 20.33 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=150, height=110
| 7 | PASS | 35112 | 20.22 | 20.05 | 0.00 | 20.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=152, height=231
| 8 | PASS | 71300 | 21.22 | 20.99 | 0.00 | 21.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=310, height=230
| 9 | PASS | 296100 | 24.34 | 24.26 | 0.00 | 24.34 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=630, height=470
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 10
Range:
Min: 12.52 cycles/pt Avg: 35.48 cycles/pt
Max: 133.38 cycles/pt Med: 21.45 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_imagePyramid16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_imagePyramid16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 42 | 57.26 | 51.07 | 0.00 | 57.26 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inRows=8, num_output_pt=42
| 1 | PASS | 84 | 32.99 | 34.61 | 0.00 | 32.99 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | inCols=32, inRows=8, num_output_pt=84
| 2 | PASS | 42 | 57.26 | 51.79 | 0.00 | 57.26 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inRows=8, num_output_pt=42
| 3 | PASS | 1008 | 14.16 | 15.85 | 0.00 | 14.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inRows=48, num_output_pt=1008
| 4 | PASS | 1911 | 12.99 | 16.22 | 0.00 | 12.99 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=104, inRows=56, num_output_pt=1911
| 5 | PASS | 2352 | 12.24 | 15.58 | 0.00 | 12.24 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=112, inRows=64, num_output_pt=2352
| 6 | PASS | 16800 | 12.34 | 15.04 | 0.00 | 12.34 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inRows=160, num_output_pt=16800
| 7 | PASS | 100800 | 14.00 | 17.13 | 0.00 | 14.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inRows=480, num_output_pt=100800
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 13.99567*N - 3505
Where: N = inCols * inRows * 21 / 64
Range:
Min: 12.24 cycles/pt Avg: 26.65 cycles/pt
Max: 57.26 cycles/pt Med: 14.00 cycles/pt
Num pt = inCols * inRows * 21 / 64
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_imagePyramid8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_imagePyramid8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 42 | 42.52 | 58.48 | 0.00 | 42.52 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inRows=8, num_output_pt=42
| 1 | PASS | 84 | 22.99 | 34.74 | 0.00 | 22.99 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | inCols=32, inRows=8, num_output_pt=84
| 2 | PASS | 1008 | 6.76 | 15.37 | 0.00 | 6.76 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inRows=48, num_output_pt=1008
| 3 | PASS | 1911 | 5.19 | 13.46 | 0.00 | 5.19 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=104, inRows=56, num_output_pt=1911
| 4 | PASS | 2352 | 5.22 | 13.37 | 0.00 | 5.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=112, inRows=64, num_output_pt=2352
| 5 | PASS | 16800 | 4.42 | 13.71 | 0.00 | 4.42 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inRows=160, num_output_pt=16800
| 6 | PASS | 100800 | 4.18 | 13.05 | 0.00 | 4.18 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inRows=480, num_output_pt=100800
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 4.16395*N + 2382
Where: N = inCols * inRows * 21 / 64
Range:
Min: 4.18 cycles/pt Avg: 13.04 cycles/pt
Max: 42.52 cycles/pt Med: 5.19 cycles/pt
Num pt = inCols * inRows * 21 / 64
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 64 bytes
+---------------------------------------+
| TESTING: VLIB_image_rescale |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_image_rescale testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 6336 | 22.75 | 29.11 | 0.00 | 22.75 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=176, inRows=144, num_output_pt=6336, mode=2
| 1 | PASS | 6336 | 22.32 | 29.17 | 0.00 | 22.32 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=176, inRows=144, num_output_pt=6336, mode=2
| 2 | PASS | 2048 | 20.53 | 29.52 | 0.00 | 20.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inRows=64, num_output_pt=2048, mode=2
| 3 | PASS | 8192 | 21.26 | 29.03 | 0.00 | 21.26 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=256, inRows=128, num_output_pt=8192, mode=2
| 4 | PASS | 4096 | 24.60 | 24.46 | 0.00 | 24.60 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inRows=128, num_output_pt=4096, mode=1
| 5 | PASS | 3200 | 31.98 | 30.83 | 0.00 | 31.98 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inRows=160, num_output_pt=3200, mode=1
| 6 | PASS | 21504 | 17.97 | 18.00 | 0.00 | 17.97 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=112, inRows=48, num_output_pt=21504, mode=1
| 7 | PASS | 19200 | 18.94 | 19.03 | 0.00 | 18.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=160, inRows=120, num_output_pt=19200,
| | | | | | | | mode=1
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: Luma scaling using nFactor
Test vectors run: 4
Formula:
Cycles: 15.98813*N + 89.95793*M + 39215
Where: N = out width * out height; M = out height
Range:
Min: 17.97 cycles/pt Avg: 23.37 cycles/pt
Max: 31.98 cycles/pt Med: 18.94 cycles/pt
Num pt = out width * out height
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: YUYV or UYVY down-scaling by 2
Test vectors run: 4
Formula:
Cycles: 18.98820*N + 483.63916*M - 12322
Where: N = out width * out height; M = out height
Range:
Min: 20.53 cycles/pt Avg: 21.72 cycles/pt
Max: 22.75 cycles/pt Med: 21.26 cycles/pt
Num pt = out width * out height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 7676 bytes
+---------------------------------------+
| TESTING: VLIB_initMeanWithLumaS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_initMeanWithLumaS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 31.00 | 32.91 | 0.00 | 31.00 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 18.63 | 17.02 | 0.00 | 18.63 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 128 | 11.15 | 11.47 | 0.00 | 11.15 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=128
| 3 | PASS | 512 | 6.44 | 6.45 | 0.00 | 6.44 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 5.66 | 5.76 | 0.00 | 5.66 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 5.15 | 5.13 | 0.00 | 5.15 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 4.94385*N + 805
Where: N = pixelCount
Range:
Min: 5.15 cycles/pt Avg: 13.00 cycles/pt
Max: 31.00 cycles/pt Med: 6.44 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_initMeanWithLumaS32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_initMeanWithLumaS32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 28.97 | 29.44 | 0.00 | 28.97 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 17.59 | 15.27 | 0.00 | 17.59 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 128 | 11.80 | 11.73 | 0.00 | 11.80 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=128
| 3 | PASS | 512 | 8.40 | 8.44 | 0.00 | 8.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 7.89 | 7.87 | 0.00 | 7.89 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 7.47 | 7.51 | 0.00 | 7.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 7.31914*N + 615
Where: N = pixelCount
Range:
Min: 7.47 cycles/pt Avg: 13.69 cycles/pt
Max: 28.97 cycles/pt Med: 8.40 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_initUYVYint_to_LABpl_LUT |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_initUYVYint_to_LABpl_LUT testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 39304 | 623.32 | 614.25 | 0.00 | 623.32 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | d=4, lut_size=39304 bytes
| 1 | PASS | 287496 | 626.90 | 617.80 | 0.00 | 626.90 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | d=3, lut_size=287496 bytes
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 2
Range:
Min: 623.32 cycles/pt Avg: 625.11 cycles/pt
Max: 626.90 cycles/pt Med: 623.32 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 372 bytes
+---------------------------------------+
| TESTING: VLIB_initVarWithConstS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_initVarWithConstS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 21.81 | 15.22 | 0.00 | 21.81 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 10.97 | 8.11 | 0.00 | 10.97 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 128 | 5.55 | 5.48 | 0.00 | 5.55 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=128
| 3 | PASS | 512 | 3.53 | 3.09 | 0.00 | 3.53 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 3.12 | 2.94 | 0.00 | 3.12 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 2.85 | 2.94 | 0.00 | 2.85 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 2.72912*N + 467
Where: N = pixelCount
Range:
Min: 2.85 cycles/pt Avg: 7.97 cycles/pt
Max: 21.81 cycles/pt Med: 3.53 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_initVarWithConstS32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_initVarWithConstS32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 14.09 | 15.63 | 0.00 | 14.09 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 9.95 | 12.22 | 0.00 | 9.95 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 128 | 8.15 | 10.19 | 0.00 | 8.15 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=128
| 3 | PASS | 512 | 6.12 | 6.03 | 0.00 | 6.12 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 5.83 | 5.71 | 0.00 | 5.83 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 5.57 | 5.55 | 0.00 | 5.57 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 5.50019*N + 309
Where: N = pixelCount
Range:
Min: 5.57 cycles/pt Avg: 8.29 cycles/pt
Max: 14.09 cycles/pt Med: 6.12 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_insertLumaIntoYUYV |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_insertLumaIntoYUYV testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 37.78 | 33.69 | 0.00 | 37.78 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inPitch=16, inRows=2, num_pts=32
| 1 | PASS | 2688 | 4.71 | 5.27 | 0.00 | 4.71 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inPitch=64, inRows=42, num_pts=2688
| 2 | PASS | 7040 | 4.55 | 5.10 | 0.00 | 4.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inPitch=128, inRows=55, num_pts=7040
| 3 | PASS | 76800 | 4.57 | 5.18 | 0.00 | 4.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inPitch=320, inRows=240, num_pts=76800
| 4 | PASS | 307200 | 5.38 | 5.86 | 0.00 | 5.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inPitch=640, inRows=480,
| | | | | | | | num_pts=307200
| 5 | PASS | 32 | 42.44 | 33.50 | 0.00 | 42.44 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inPitch=20, inRows=2, num_pts=32
| 6 | PASS | 2688 | 5.87 | 5.84 | 0.00 | 5.87 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inPitch=84, inRows=42, num_pts=2688
| 7 | PASS | 7040 | 4.83 | 5.12 | 0.00 | 4.83 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=128, inPitch=136, inRows=55, num_pts=7040
| 8 | PASS | 76800 | 4.57 | 5.16 | 0.00 | 4.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inPitch=324, inRows=240, num_pts=76800
| 9 | PASS | 307200 | 5.35 | 5.88 | 0.00 | 5.35 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inPitch=644, inRows=480,
| | | | | | | | num_pts=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: inPitch == inCols
Test vectors run: 5
Formula:
Cycles: 5.39202*N - 14535
Where: N = inCols * inRows
Range:
Min: 4.55 cycles/pt Avg: 11.40 cycles/pt
Max: 37.78 cycles/pt Med: 4.57 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: inPitch > inCols
Test vectors run: 5
Formula:
Cycles: 6.14133*N - 537.21228*M + 13481
Where: N = inCols * inRows; M = inRows
Range:
Min: 4.57 cycles/pt Avg: 12.61 cycles/pt
Max: 42.44 cycles/pt Med: 4.83 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_integralImage16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_integralImage16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 53.84 | 37.39 | 0.00 | 53.84 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inRows=4, num_pt=64
| 1 | PASS | 128 | 31.89 | 31.49 | 0.00 | 31.89 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | inCols=32, inRows=4, num_pt=128
| 2 | PASS | 3072 | 8.24 | 10.21 | 0.00 | 8.24 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inRows=48, num_pt=3072
| 3 | PASS | 5720 | 8.53 | 11.40 | 0.00 | 8.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=104, inRows=55, num_pt=5720
| 4 | PASS | 6944 | 8.91 | 11.30 | 0.00 | 8.91 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=112, inRows=62, num_pt=6944
| 5 | PASS | 51200 | 7.86 | 11.85 | 0.00 | 7.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inRows=160, num_pt=51200
| 6 | PASS | 307200 | 10.76 | 13.54 | 0.00 | 10.76 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inRows=480, num_pt=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 10.78083*N - 26359
Where: N = inCols * inRows
Range:
Min: 7.86 cycles/pt Avg: 18.58 cycles/pt
Max: 53.84 cycles/pt Med: 8.53 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_integralImage8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_integralImage8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 38.11 | 35.30 | 0.00 | 38.11 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | inCols=16, inRows=4, num_pt=64
| 1 | PASS | 128 | 26.66 | 22.51 | 0.00 | 26.66 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | inCols=32, inRows=4, num_pt=128
| 2 | PASS | 3072 | 8.77 | 9.58 | 0.00 | 8.77 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=64, inRows=48, num_pt=3072
| 3 | PASS | 5720 | 8.20 | 10.27 | 0.00 | 8.20 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=104, inRows=55, num_pt=5720
| 4 | PASS | 6944 | 8.47 | 9.85 | 0.00 | 8.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=112, inRows=62, num_pt=6944
| 5 | PASS | 51200 | 8.69 | 9.81 | 0.00 | 8.69 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=320, inRows=160, num_pt=51200
| 6 | PASS | 307200 | 10.82 | 11.96 | 0.00 | 10.82 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | inCols=640, inRows=480, num_pt=307200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 10.83898*N - 21654
Where: N = inCols * inRows
Range:
Min: 8.20 cycles/pt Avg: 15.67 cycles/pt
Max: 38.11 cycles/pt Med: 8.69 cycles/pt
Num pt = inCols * inRows
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_2x4_Correct |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_2x4_Correct testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 50 | 4134.12 | 3614.92 | 0.00 | 4134.12 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=50
| 1 | PASS | 20 | 4105.50 | 3571.00 | 0.00 | 4105.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 4232.48 | 3788.88 | 0.00 | 4232.48 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 4207.74 | 3772.90 | 0.00 | 4207.74 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 4034.91 | 3655.43 | 0.00 | 4034.91 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 4020.54516*N + 10122
Where: N = number of measurements
Range:
Min: 4034.91 cycles/pt Avg: 4142.95 cycles/pt
Max: 4232.48 cycles/pt Med: 4105.50 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 56 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_2x4_Correct_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_2x4_Correct_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 5249.45 | 5756.55 | 0.00 | 5249.45 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 5468.20 | 5649.30 | 0.00 | 5468.20 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 5223.04 | 5723.80 | 0.00 | 5223.04 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 5503.36 | 5625.30 | 0.00 | 5503.36 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 5640.10 | 6435.32 | 0.00 | 5640.10 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 5666.48242*N - 17208
Where: N = number of measurements
Range:
Min: 5223.04 cycles/pt Avg: 5416.83 cycles/pt
Max: 5640.10 cycles/pt Med: 5249.45 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 100 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_2x4_Correct_S16_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_2x4_Correct_S16_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 6132.05 | 6308.30 | 0.00 | 6132.05 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 6462.70 | 6258.40 | 0.00 | 6462.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 6153.40 | 6473.56 | 0.00 | 6153.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 6472.66 | 6238.54 | 0.00 | 6472.66 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 6304.99 | 6332.52 | 0.00 | 6304.99 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 6306.40673*N + 90
Where: N = number of measurements
Range:
Min: 6132.05 cycles/pt Avg: 6305.16 cycles/pt
Max: 6472.66 cycles/pt Med: 6153.40 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 112 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_2x4_Predict |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_2x4_Predict testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 50 | 1445.20 | 1718.28 | 0.00 | 1445.20 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=50
| 1 | PASS | 20 | 1557.00 | 1792.70 | 0.00 | 1557.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 1589.24 | 1617.52 | 0.00 | 1589.24 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 1453.88 | 1697.72 | 0.00 | 1453.88 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 1394.00 | 1732.92 | 0.00 | 1394.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 1383.05480*N + 7140
Where: N = number of measurements
Range:
Min: 1394.00 cycles/pt Avg: 1487.86 cycles/pt
Max: 1589.24 cycles/pt Med: 1445.20 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_2x4_Predict_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_2x4_Predict_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 2708.95 | 2358.15 | 0.00 | 2708.95 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 2379.10 | 2363.00 | 0.00 | 2379.10 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 2619.60 | 2497.08 | 0.00 | 2619.60 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 2364.94 | 2402.00 | 0.00 | 2364.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 2658.21 | 2234.96 | 0.00 | 2658.21 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 2669.94116*N - 9113
Where: N = number of measurements
Range:
Min: 2364.94 cycles/pt Avg: 2546.16 cycles/pt
Max: 2708.95 cycles/pt Med: 2379.10 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 28 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_2x4_Predict_S16_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_2x4_Predict_S16_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 2449.00 | 2714.60 | 0.00 | 2449.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 2268.00 | 3060.30 | 0.00 | 2268.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 2319.00 | 2679.08 | 0.00 | 2319.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 2268.00 | 3061.92 | 0.00 | 2268.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 2731.02 | 2429.60 | 0.00 | 2731.02 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 2767.71118*N - 25675
Where: N = number of measurements
Range:
Min: 2268.00 cycles/pt Avg: 2407.00 cycles/pt
Max: 2731.02 cycles/pt Med: 2268.00 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 28 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_4x6_Correct |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_4x6_Correct testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 8954.25 | 8342.05 | 0.00 | 8954.25 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 8738.40 | 8195.00 | 0.00 | 8738.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 8691.16 | 8235.56 | 0.00 | 8691.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 8746.74 | 8258.54 | 0.00 | 8746.74 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 8324.25 | 8305.54 | 0.00 | 8324.25 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 8283.1875*N + 27508
Where: N = number of measurements
Range:
Min: 8324.25 cycles/pt Avg: 8690.96 cycles/pt
Max: 8954.25 cycles/pt Med: 8691.16 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 180 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_4x6_Correct_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_4x6_Correct_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 16189.80 | 16444.35 | 0.00 | 16189.80 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 16151.70 | 16824.30 | 0.00 | 16151.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 16163.24 | 16177.00 | 0.00 | 16163.24 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 16103.76 | 16149.96 | 0.00 | 16103.76 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 17037.54 | 16060.80 | 0.00 | 17037.54 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 17118.15039*N - 55576
Where: N = number of measurements
Range:
Min: 16103.76 cycles/pt Avg: 16329.21 cycles/pt
Max: 17037.54 cycles/pt Med: 16151.70 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 532 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_4x6_Correct_S16_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_4x6_Correct_S16_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 16043.80 | 17640.40 | 0.00 | 16043.80 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 15991.50 | 17191.60 | 0.00 | 15991.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 15815.40 | 17147.68 | 0.00 | 15815.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 15947.94 | 16991.08 | 0.00 | 15947.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 16136.06 | 16980.82 | 0.00 | 16136.06 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 16154.39355*N - 12949
Where: N = number of measurements
Range:
Min: 15815.40 cycles/pt Avg: 15986.94 cycles/pt
Max: 16136.06 cycles/pt Med: 15947.94 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 484 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_4x6_Predict |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_4x6_Predict testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 3461.90 | 3501.25 | 0.00 | 3461.90 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 3401.90 | 3494.80 | 0.00 | 3401.90 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 3311.60 | 3350.08 | 0.00 | 3311.60 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 3387.14 | 3475.24 | 0.00 | 3387.14 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 3372.27 | 3417.74 | 0.00 | 3372.27 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 3371.33178*N + 380
Where: N = number of measurements
Range:
Min: 3311.60 cycles/pt Avg: 3386.96 cycles/pt
Max: 3461.90 cycles/pt Med: 3372.27 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 72 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_4x6_Predict_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_4x6_Predict_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 4756.80 | 4475.30 | 0.00 | 4756.80 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 4676.50 | 4269.60 | 0.00 | 4676.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 4676.80 | 4454.28 | 0.00 | 4676.80 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 4673.56 | 4369.44 | 0.00 | 4673.56 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 4702.67 | 5103.35 | 0.00 | 4702.67 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 4703.40673*N - 764
Where: N = number of measurements
Range:
Min: 4673.56 cycles/pt Avg: 4697.27 cycles/pt
Max: 4756.80 cycles/pt Med: 4676.50 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 36 bytes
+---------------------------------------+
| TESTING: VLIB_kalmanFilter_4x6_Predict_S16_F32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_kalmanFilter_4x6_Predict_S16_F32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 40 | 4325.15 | 4470.75 | 0.00 | 4325.15 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=40
| 1 | PASS | 20 | 4641.30 | 4580.60 | 0.00 | 4641.30 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=20
| 2 | PASS | 75 | 4520.96 | 4613.04 | 0.00 | 4520.96 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=75
| 3 | PASS | 100 | 4608.06 | 4696.10 | 0.00 | 4608.06 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=100
| 4 | PASS | 600 | 4286.15 | 4671.85 | 0.00 | 4286.15 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | num_measurements=600
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 4265.07666*N + 15212
Where: N = number of measurements
Range:
Min: 4286.15 cycles/pt Avg: 4476.32 cycles/pt
Max: 4641.30 cycles/pt Med: 4325.15 cycles/pt
Num pt = number of measurements
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 36 bytes
+---------------------------------------+
| TESTING: VLIB_L1DistanceS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_L1DistanceS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 8 | 109.13 | 101.50 | 0.00 | 109.13 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | d=8
| 1 | PASS | 32 | 32.53 | 31.41 | 0.00 | 32.53 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | d=32
| 2 | PASS | 64 | 18.89 | 17.39 | 0.00 | 18.89 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | d=64
| 3 | PASS | 128 | 11.41 | 12.16 | 0.00 | 11.41 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | d=128
| 4 | PASS | 256 | 7.68 | 9.55 | 0.00 | 7.68 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | d=256
| 5 | PASS | 1024 | 4.98 | 7.65 | 0.00 | 4.98 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | d=1024
| 6 | PASS | 4096 | 4.29 | 7.15 | 0.00 | 4.29 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | d=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 4.07079*N + 914
Where: N = d
Range:
Min: 4.29 cycles/pt Avg: 26.99 cycles/pt
Max: 109.13 cycles/pt Med: 7.68 cycles/pt
Num pt = d
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_L2_norm_f32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_L2_norm_f32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 10 | 118.80 | 136.20 | 0.00 | 118.80 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | len=10
| 1 | PASS | 32 | 42.19 | 47.63 | 0.00 | 42.19 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | len=32
| 2 | PASS | 64 | 24.09 | 26.81 | 0.00 | 24.09 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=64
| 3 | PASS | 128 | 15.05 | 16.41 | 0.00 | 15.05 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=128
| 4 | PASS | 256 | 10.52 | 11.20 | 0.00 | 10.52 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=256
| 5 | PASS | 1024 | 7.19 | 7.36 | 0.00 | 7.19 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=1024
| 6 | PASS | 4096 | 6.47 | 6.37 | 0.00 | 6.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 6.19134*N + 1119
Where: N = len
Range:
Min: 6.47 cycles/pt Avg: 32.04 cycles/pt
Max: 118.80 cycles/pt Med: 10.52 cycles/pt
Num pt = len
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_L2_norm_i16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_L2_norm_i16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 10 | 100.10 | 80.70 | 0.00 | 100.10 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | len=10
| 1 | PASS | 32 | 31.47 | 24.47 | 0.00 | 31.47 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | len=32
| 2 | PASS | 64 | 18.17 | 14.67 | 0.00 | 18.17 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=64
| 3 | PASS | 128 | 10.54 | 8.79 | 0.00 | 10.54 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=128
| 4 | PASS | 256 | 6.72 | 5.96 | 0.00 | 6.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=256
| 5 | PASS | 1024 | 3.84 | 3.67 | 0.00 | 3.84 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=1024
| 6 | PASS | 4096 | 3.17 | 3.27 | 0.00 | 3.17 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 2.93922*N + 955
Where: N = len
Range:
Min: 3.17 cycles/pt Avg: 24.86 cycles/pt
Max: 100.10 cycles/pt Med: 6.72 cycles/pt
Num pt = len
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_L2_norm_i32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_L2_norm_i32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 10 | 103.10 | 87.50 | 0.00 | 103.10 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | len=10
| 1 | PASS | 32 | 37.09 | 32.22 | 0.00 | 37.09 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | len=32
| 2 | PASS | 64 | 21.45 | 18.55 | 0.00 | 21.45 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=64
| 3 | PASS | 128 | 13.63 | 12.41 | 0.00 | 13.63 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=128
| 4 | PASS | 256 | 9.72 | 9.11 | 0.00 | 9.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=256
| 5 | PASS | 1024 | 6.85 | 6.70 | 0.00 | 6.85 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=1024
| 6 | PASS | 4096 | 6.25 | 6.09 | 0.00 | 6.25 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 6.01415*N + 961
Where: N = len
Range:
Min: 6.25 cycles/pt Avg: 28.30 cycles/pt
Max: 103.10 cycles/pt Med: 9.72 cycles/pt
Num pt = len
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_L2_norm_i8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_L2_norm_i8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 10 | 91.10 | 88.50 | 0.00 | 91.10 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | len=10
| 1 | PASS | 32 | 29.22 | 28.41 | 0.00 | 29.22 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | len=32
| 2 | PASS | 64 | 14.70 | 14.30 | 0.00 | 14.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=64
| 3 | PASS | 128 | 8.57 | 8.13 | 0.00 | 8.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=128
| 4 | PASS | 256 | 5.01 | 4.79 | 0.00 | 5.01 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=256
| 5 | PASS | 1024 | 2.34 | 2.32 | 0.00 | 2.34 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=1024
| 6 | PASS | 4096 | 1.85 | 1.77 | 0.00 | 1.85 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | len=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 1.63848*N + 853
Where: N = len
Range:
Min: 1.85 cycles/pt Avg: 21.83 cycles/pt
Max: 91.10 cycles/pt Med: 5.01 cycles/pt
Num pt = len
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_lbp |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_lbp testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 256 | 39.23 | 43.74 | 0.00 | 39.23 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=16, imageHeight=16, out_size=59
| 1 | PASS | 256 | 39.29 | 43.74 | 0.00 | 39.29 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=16, imageHeight=16, out_size=58
| 2 | PASS | 91 | 74.63 | 53.10 | 0.00 | 74.63 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=13, imageHeight=7, out_size=59
| 3 | PASS | 198 | 47.39 | 45.44 | 0.00 | 47.39 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=11, imageHeight=18, out_size=58
| 4 | PASS | 432 | 29.32 | 42.07 | 0.00 | 29.32 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=24, imageHeight=18, out_size=59
| 5 | PASS | 2048 | 19.99 | 42.87 | 0.00 | 19.99 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=64, imageHeight=32, out_size=255
| 6 | PASS | 6144 | 17.62 | 43.22 | 0.00 | 17.62 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=128, imageHeight=48, out_size=256
| 7 | PASS | 25600 | 16.86 | 43.94 | 0.00 | 16.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | imageWidth=320, imageHeight=80, out_size=256
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 16.52757*N + 43.11733*M + 5107
Where: N = imageWidth * imageHeight; M = imageHeight
Range:
Min: 16.86 cycles/pt Avg: 35.54 cycles/pt
Max: 74.63 cycles/pt Med: 29.32 cycles/pt
Num pt = imageWidth * imageHeight
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 64 bytes
+---------------------------------------+
| TESTING: VLIB_legendreMoments |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_legendreMoments testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 19600 | 2.22 | 1.93 | 0.00 | 2.22 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | ImH=20, Order=6
| 1 | PASS | 19600 | 2.25 | 1.93 | 0.00 | 2.25 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=20, Order=6
| 2 | PASS | 313600 | 0.86 | 0.60 | 0.00 | 0.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=80, Order=6
| 3 | PASS | 802816 | 0.76 | 0.50 | 0.00 | 0.76 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=128, Order=6
| 4 | PASS | 3211264 | 0.93 | 0.70 | 0.00 | 0.93 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=256, Order=6
| 5 | PASS | 2822400 | 0.81 | 0.55 | 0.00 | 0.81 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=80, Order=20
| 6 | PASS | 7225344 | 0.71 | 0.45 | 0.00 | 0.71 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=128, Order=20
| 7 | PASS | 28901376 | 0.88 | 0.66 | 0.00 | 0.88 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=256, Order=20
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 0.87950*N - 155175
Where: N = (Order+1)*(Order+1)*ImH*ImH
Range:
Min: 0.71 cycles/pt Avg: 1.18 cycles/pt
Max: 2.25 cycles/pt Med: 0.86 cycles/pt
Num pt = (Order+1)*(Order+1)*ImH*ImH
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_legendreMoments_Init |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_legendreMoments_Init testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 100 | 41.90 | 39.22 | 0.00 | 41.90 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | ImH=20, Order=6
| 1 | PASS | 100 | 42.20 | 39.22 | 0.00 | 42.20 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=20, Order=6
| 2 | PASS | 400 | 23.09 | 23.33 | 0.00 | 23.09 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=80, Order=6
| 3 | PASS | 640 | 20.94 | 20.58 | 0.00 | 20.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=128, Order=6
| 4 | PASS | 1280 | 19.08 | 18.90 | 0.00 | 19.08 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=256, Order=6
| 5 | PASS | 1520 | 16.42 | 16.44 | 0.00 | 16.42 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=80, Order=20
| 6 | PASS | 2432 | 15.88 | 15.69 | 0.00 | 15.88 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=128, Order=20
| 7 | PASS | 4864 | 15.33 | 15.28 | 0.00 | 15.33 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | ImH=256, Order=20
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 14.00733*N + 15.60297*M + 2464
Where: N = (Order-1)*ImH; M = ImH
Range:
Min: 15.33 cycles/pt Avg: 24.35 cycles/pt
Max: 42.20 cycles/pt Med: 19.08 cycles/pt
Num pt = (Order-1)*ImH
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 100 bytes
+---------------------------------------+
| TESTING: VLIB_mixtureOfGaussiansS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_mixtureOfGaussiansS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 320 | 280.32 | 275.58 | 0.00 | 280.32 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | im_size=32, num_images=10
| 1 | PASS | 512 | 168.19 | 167.82 | 0.00 | 168.19 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=64, num_images=8
| 2 | PASS | 288 | 128.64 | 127.80 | 0.00 | 128.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=96, num_images=3
| 3 | PASS | 600 | 115.02 | 111.59 | 0.00 | 115.02 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=120, num_images=5
| 4 | PASS | 960 | 76.43 | 75.81 | 0.00 | 76.43 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=320, num_images=3
| 5 | PASS | 2560 | 66.11 | 65.72 | 0.00 | 66.11 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=640, num_images=4
| 6 | PASS | 3840 | 60.93 | 60.31 | 0.00 | 60.93 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=1280, num_images=3
| 7 | PASS | 64000 | 147.86 | 149.46 | 0.00 | 147.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=32000, num_images=2
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 150.37290*N + 28535.91601*M - 231551
Where: N = im_size*num_images; M = num_images
Range:
Min: 60.93 cycles/pt Avg: 130.44 cycles/pt
Max: 280.32 cycles/pt Med: 115.02 cycles/pt
Num pt = im_size*num_images
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 160 bytes
+---------------------------------------+
| TESTING: VLIB_mixtureOfGaussiansS32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_mixtureOfGaussiansS32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 320 | 434.93 | 537.32 | 0.00 | 434.93 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | im_size=32, num_images=10
| 1 | PASS | 512 | 255.62 | 315.73 | 0.00 | 255.62 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=64, num_images=8
| 2 | PASS | 288 | 199.66 | 239.78 | 0.00 | 199.66 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=96, num_images=3
| 3 | PASS | 600 | 174.93 | 207.18 | 0.00 | 174.93 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=120, num_images=5
| 4 | PASS | 960 | 119.20 | 129.82 | 0.00 | 119.20 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=320, num_images=3
| 5 | PASS | 2560 | 103.09 | 108.63 | 0.00 | 103.09 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=640, num_images=4
| 6 | PASS | 3840 | 99.81 | 102.11 | 0.00 | 99.81 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=1280, num_images=3
| 7 | PASS | 64000 | 249.40 | 248.86 | 0.00 | 249.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | im_size=32000, num_images=2
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 253.81118*N + 47751.35546*M - 400947
Where: N = im_size*num_images; M = num_images
Range:
Min: 99.81 cycles/pt Avg: 204.58 cycles/pt
Max: 434.93 cycles/pt Med: 174.93 cycles/pt
Num pt = im_size*num_images
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 172 bytes
+---------------------------------------+
| TESTING: VLIB_nonMaxSuppress_3x3_S16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_nonMaxSuppress_3x3_S16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 16 | 67.88 | 107.81 | 0.00 | 67.88 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=4, out_size=16
| 1 | PASS | 320 | 8.99 | 66.17 | 0.00 | 8.99 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=20, height=18, out_size=320
| 2 | PASS | 2944 | 5.70 | 67.94 | 0.00 | 5.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48, out_size=2944
| 3 | PASS | 2944 | 5.70 | 68.05 | 0.00 | 5.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48, out_size=2944
| 4 | PASS | 18880 | 5.67 | 68.99 | 0.00 | 5.67 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120, out_size=18880
| 5 | PASS | 76160 | 5.58 | 69.64 | 0.00 | 5.58 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, out_size=76160
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 5.56711*N + 927
Where: N = width * (height-2)
Range:
Min: 5.58 cycles/pt Avg: 16.59 cycles/pt
Max: 67.88 cycles/pt Med: 5.70 cycles/pt
Num pt = width * (height-2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_nonMaxSuppress_5x5_S16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_nonMaxSuppress_5x5_S16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 8 | 168.38 | 158.25 | 0.00 | 168.38 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=5, out_size=8
| 1 | PASS | 280 | 9.92 | 75.43 | 0.00 | 9.92 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=20, height=18, out_size=280
| 2 | PASS | 2816 | 6.00 | 83.95 | 0.00 | 6.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48, out_size=2816
| 3 | PASS | 2816 | 6.00 | 84.01 | 0.00 | 6.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48, out_size=2816
| 4 | PASS | 18560 | 5.74 | 86.69 | 0.00 | 5.74 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120, out_size=18560
| 5 | PASS | 75520 | 5.62 | 87.73 | 0.00 | 5.62 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, out_size=75520
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 5.60925*N + 1386
Where: N = width * (height-4)
Range:
Min: 5.62 cycles/pt Avg: 33.61 cycles/pt
Max: 168.38 cycles/pt Med: 6.00 cycles/pt
Num pt = width * (height-4)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_nonMaxSuppress_7x7_S16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_nonMaxSuppress_7x7_S16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 12 | 120.67 | 162.08 | 0.00 | 120.67 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=12, height=7, out_size=12
| 1 | PASS | 240 | 12.89 | 106.75 | 0.00 | 12.89 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=20, height=18, out_size=240
| 2 | PASS | 2688 | 7.01 | 132.27 | 0.00 | 7.01 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48, out_size=2688
| 3 | PASS | 2688 | 6.99 | 132.49 | 0.00 | 6.99 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48, out_size=2688
| 4 | PASS | 18240 | 6.67 | 139.46 | 0.00 | 6.67 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120, out_size=18240
| 5 | PASS | 74880 | 6.54 | 142.27 | 0.00 | 6.54 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, out_size=74880
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 6.52518*N + 1579
Where: N = width * (height-6)
Range:
Min: 6.54 cycles/pt Avg: 26.79 cycles/pt
Max: 120.67 cycles/pt Med: 6.99 cycles/pt
Num pt = width * (height-6)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_nonMaxSuppress_U16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_nonMaxSuppress_U16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 256 | 36.06 | 54.99 | 0.00 | 36.06 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=7, threshold=8, in
| | | | | | | | size=256
| 1 | PASS | 256 | 36.20 | 57.43 | 0.00 | 36.20 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=7, threshold=8, in
| | | | | | | | size=256
| 2 | PASS | 360 | 29.51 | 48.26 | 0.00 | 29.51 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=20, h=18, filterWidth=7, threshold=80, in
| | | | | | | | size=360
| 3 | PASS | 3072 | 15.46 | 60.09 | 0.00 | 15.46 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=7, threshold=80, in
| | | | | | | | size=3072
| 4 | PASS | 3072 | 15.28 | 60.09 | 0.00 | 15.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=7, threshold=12000, in
| | | | | | | | size=3072
| 5 | PASS | 19200 | 15.38 | 65.44 | 0.00 | 15.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=160, h=120, filterWidth=7, threshold=80, in
| | | | | | | | size=19200
| 6 | PASS | 76800 | 32.53 | 76.79 | 0.00 | 32.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=320, h=240, filterWidth=7, threshold=80, in
| | | | | | | | size=76800
| 7 | PASS | 256 | 35.85 | 63.67 | 0.00 | 35.85 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=5, threshold=8, in
| | | | | | | | size=256
| 8 | PASS | 256 | 36.13 | 63.77 | 0.00 | 36.13 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=5, threshold=8, in
| | | | | | | | size=256
| 9 | PASS | 360 | 29.92 | 61.61 | 0.00 | 29.92 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=20, h=18, filterWidth=5, threshold=80, in
| | | | | | | | size=360
| 10 | PASS | 3072 | 16.22 | 62.95 | 0.00 | 16.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=5, threshold=80, in
| | | | | | | | size=3072
| 11 | PASS | 3072 | 16.02 | 63.19 | 0.00 | 16.02 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=5, threshold=12000, in
| | | | | | | | size=3072
| 12 | PASS | 19200 | 15.14 | 67.74 | 0.00 | 15.14 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=160, h=120, filterWidth=5, threshold=80, in
| | | | | | | | size=19200
| 13 | PASS | 76800 | 30.13 | 81.26 | 0.00 | 30.13 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=320, h=240, filterWidth=5, threshold=80, in
| | | | | | | | size=76800
| 14 | PASS | 256 | 30.30 | 76.48 | 0.00 | 30.30 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=3, threshold=8, in
| | | | | | | | size=256
| 15 | PASS | 256 | 30.46 | 76.65 | 0.00 | 30.46 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=3, threshold=8, in
| | | | | | | | size=256
| 16 | PASS | 360 | 27.06 | 76.04 | 0.00 | 27.06 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=20, h=18, filterWidth=3, threshold=80, in
| | | | | | | | size=360
| 17 | PASS | 3072 | 15.13 | 75.15 | 0.00 | 15.13 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=3, threshold=80, in
| | | | | | | | size=3072
| 18 | PASS | 3072 | 14.97 | 75.13 | 0.00 | 14.97 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=3, threshold=12000, in
| | | | | | | | size=3072
| 19 | PASS | 19200 | 15.04 | 75.88 | 0.00 | 15.04 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=160, h=120, filterWidth=3, threshold=80, in
| | | | | | | | size=19200
| 20 | PASS | 76800 | 35.36 | 91.90 | 0.00 | 35.36 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=320, h=240, filterWidth=3, threshold=80, in
| | | | | | | | size=76800
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: filterWidth = 3
Test vectors run: 7
Formula:
Cycles: 35.31177*N - 73664
Where: N = w * h
Range:
Min: 14.97 cycles/pt Avg: 24.05 cycles/pt
Max: 35.36 cycles/pt Med: 15.13 cycles/pt
Num pt = w * h
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: filterWidth = 5
Test vectors run: 7
Formula:
Cycles: 30.06113*N - 51964
Where: N = w * h
Range:
Min: 15.14 cycles/pt Avg: 25.63 cycles/pt
Max: 36.13 cycles/pt Med: 16.22 cycles/pt
Num pt = w * h
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 3: filterWidth = 7
Test vectors run: 7
Formula:
Cycles: 32.47440*N - 61192
Where: N = w * h
Range:
Min: 15.28 cycles/pt Avg: 25.77 cycles/pt
Max: 36.20 cycles/pt Med: 15.46 cycles/pt
Num pt = w * h
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 120 bytes
+---------------------------------------+
| TESTING: VLIB_nonMaxSuppress_U32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_nonMaxSuppress_U32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 256 | 42.50 | 59.48 | 0.00 | 42.50 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=7, threshold=8, in
| | | | | | | | size=256
| 1 | PASS | 256 | 41.37 | 59.22 | 0.00 | 41.37 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=7, threshold=8, in
| | | | | | | | size=256
| 2 | PASS | 360 | 34.48 | 51.19 | 0.00 | 34.48 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=20, h=18, filterWidth=7, threshold=80, in
| | | | | | | | size=360
| 3 | PASS | 3072 | 21.73 | 57.27 | 0.00 | 21.73 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=7, threshold=80, in
| | | | | | | | size=3072
| 4 | PASS | 3072 | 21.64 | 57.47 | 0.00 | 21.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=7, threshold=12000, in
| | | | | | | | size=3072
| 5 | PASS | 19200 | 21.55 | 61.33 | 0.00 | 21.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=160, h=120, filterWidth=7, threshold=80, in
| | | | | | | | size=19200
| 6 | PASS | 76800 | 48.72 | 89.84 | 0.00 | 48.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=320, h=240, filterWidth=7, threshold=80, in
| | | | | | | | size=76800
| 7 | PASS | 256 | 37.41 | 70.35 | 0.00 | 37.41 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=5, threshold=8, in
| | | | | | | | size=256
| 8 | PASS | 256 | 37.38 | 69.45 | 0.00 | 37.38 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=5, threshold=8, in
| | | | | | | | size=256
| 9 | PASS | 360 | 30.59 | 63.49 | 0.00 | 30.59 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=20, h=18, filterWidth=5, threshold=80, in
| | | | | | | | size=360
| 10 | PASS | 3072 | 21.28 | 61.04 | 0.00 | 21.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=5, threshold=80, in
| | | | | | | | size=3072
| 11 | PASS | 3072 | 21.04 | 61.09 | 0.00 | 21.04 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=5, threshold=12000, in
| | | | | | | | size=3072
| 12 | PASS | 19200 | 20.82 | 65.87 | 0.00 | 20.82 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=160, h=120, filterWidth=5, threshold=80, in
| | | | | | | | size=19200
| 13 | PASS | 76800 | 45.78 | 94.40 | 0.00 | 45.78 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=320, h=240, filterWidth=5, threshold=80, in
| | | | | | | | size=76800
| 14 | PASS | 256 | 30.29 | 85.70 | 0.00 | 30.29 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=3, threshold=8, in
| | | | | | | | size=256
| 15 | PASS | 256 | 29.63 | 86.92 | 0.00 | 29.63 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | w=16, h=16, filterWidth=3, threshold=8, in
| | | | | | | | size=256
| 16 | PASS | 360 | 27.27 | 81.58 | 0.00 | 27.27 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=20, h=18, filterWidth=3, threshold=80, in
| | | | | | | | size=360
| 17 | PASS | 3072 | 19.45 | 78.77 | 0.00 | 19.45 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=3, threshold=80, in
| | | | | | | | size=3072
| 18 | PASS | 3072 | 19.55 | 78.63 | 0.00 | 19.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=64, h=48, filterWidth=3, threshold=12000, in
| | | | | | | | size=3072
| 19 | PASS | 19200 | 19.02 | 79.38 | 0.00 | 19.02 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=160, h=120, filterWidth=3, threshold=80, in
| | | | | | | | size=19200
| 20 | PASS | 76800 | 43.17 | 108.66 | 0.00 | 43.17 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | w=320, h=240, filterWidth=3, threshold=80, in
| | | | | | | | size=76800
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: filterWidth = 3
Test vectors run: 7
Formula:
Cycles: 43.12415*N - 88150
Where: N = w * h
Range:
Min: 19.02 cycles/pt Avg: 26.91 cycles/pt
Max: 43.17 cycles/pt Med: 19.55 cycles/pt
Num pt = w * h
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: filterWidth = 5
Test vectors run: 7
Formula:
Cycles: 45.72509*N - 90705
Where: N = w * h
Range:
Min: 20.82 cycles/pt Avg: 30.61 cycles/pt
Max: 45.78 cycles/pt Med: 21.28 cycles/pt
Num pt = w * h
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 3: filterWidth = 7
Test vectors run: 7
Formula:
Cycles: 48.65757*N - 98592
Where: N = w * h
Range:
Min: 21.55 cycles/pt Avg: 33.14 cycles/pt
Max: 48.72 cycles/pt Med: 21.73 cycles/pt
Num pt = w * h
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 128 bytes
+---------------------------------------+
| TESTING: VLIB_normalFlow_16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_normalFlow_16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 41.47 | 57.98 | 0.00 | 41.47 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=8, numEls=64
| 1 | PASS | 768 | 16.48 | 34.64 | 0.00 | 16.48 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=24, numEls=768
| 2 | PASS | 1040 | 18.47 | 36.18 | 0.00 | 18.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=40, height=26, numEls=1040
| 3 | PASS | 4992 | 16.56 | 34.85 | 0.00 | 16.56 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=96, height=52, numEls=4992
| 4 | PASS | 8192 | 26.53 | 72.23 | 0.00 | 26.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, height=64, numEls=8192
| 5 | PASS | 19200 | 14.70 | 35.66 | 0.00 | 14.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120, numEls=19200
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 15.44717*N + 14608
Where: N = numEls
Range:
Min: 14.70 cycles/pt Avg: 22.37 cycles/pt
Max: 41.47 cycles/pt Med: 16.56 cycles/pt
Num pt = numEls
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_ORB_bestFeaturesToFront |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_ORB_bestFeaturesToFront testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1292 | 119.08 | 533.28 | 0.00 | 119.08 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 1292, desired output features= 512, actual
| | | | | | | | output n_features= 512, realistic data
| 1 | PASS | 612 | 122.98 | 490.75 | 0.00 | 122.98 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 612, desired output features= 303, actual
| | | | | | | | output n_features= 303, realistic data
| 2 | PASS | 1292 | 221.67 | 331.00 | 0.00 | 221.67 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 1292, desired output features= 646,
| | | | | | | | actual output n_features= 646, inversly pre-sorted
| 3 | PASS | 1292 | 108.33 | 306.08 | 0.00 | 108.33 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 1292, desired output features= 646,
| | | | | | | | actual output n_features= 646, pre-sorted
| 4 | PASS | 200 | 172.00 | 295.93 | 0.00 | 172.00 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 200, desired output features= 100,
| | | | | | | | actual output n_features= 100, inversly pre-sorted
| 5 | PASS | 100 | 310.57 | 295.72 | 0.00 | 310.57 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 100, desired output features= 50, actual
| | | | | | | | output n_features= 50, inversly pre-sorted
| 6 | PASS | 200 | 141.82 | 270.24 | 0.00 | 141.82 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 200, desired output features= 100,
| | | | | | | | actual output n_features= 100, pre-sorted
| 7 | PASS | 100 | 182.48 | 271.21 | 0.00 | 182.48 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 100, desired output features= 50, actual
| | | | | | | | output n_features= 50, pre-sorted
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Range:
Min: 108.33 cycles/pt Avg: 172.36 cycles/pt
Max: 310.57 cycles/pt Med: 141.82 cycles/pt
Num pt = number of input features;
Min (best case) is when input is already sorted.
Max (worst case) is where input is sorted in reverse order.
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 64 bytes
+---------------------------------------+
| TESTING: VLIB_ORB_computeOrientation |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_ORB_computeOrientation testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 545 | 527.46 | 1632.50 | 0.00 | 527.46 | Statically generated input | Opt and NatC results compared |
| | | | | | | | numFeatures= 545, realistic data
| 1 | PASS | 1236 | 563.79 | 1660.22 | 0.00 | 563.79 | Statically generated input | Opt and NatC results compared |
| | | | | | | | numFeatures= 1236, realistic data
| 2 | PASS | 749 | 551.87 | 1660.67 | 0.00 | 551.87 | Statically generated input | Opt and NatC results compared |
| | | | | | | | numFeatures= 749, realistic data
| 3 | PASS | 272 | 559.88 | 1649.61 | 0.00 | 559.88 | Statically generated input | Opt and NatC results compared |
| | | | | | | | numFeatures= 272, realistic data
| 4 | PASS | 262 | 618.75 | 1683.56 | 0.00 | 618.75 | Statically generated input | Opt and NatC results compared |
| | | | | | | | numFeatures= 262, realistic data
| 5 | PASS | 76 | 743.05 | 1716.84 | 0.00 | 743.05 | Statically generated input | Opt and NatC results compared |
| | | | | | | | numFeatures= 76, realistic data
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 550.28216*N + 6774
Where: N = numFeatures
Range:
Min: 527.46 cycles/pt Avg: 594.13 cycles/pt
Max: 743.05 cycles/pt Med: 559.88 cycles/pt
Num pt = numFeatures
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 284 bytes
+---------------------------------------+
| TESTING: VLIB_ORB_computeRBrief |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_ORB_computeRBrief testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1121 | 2305.20 | 2816.45 | 0.00 | 2305.20 | Statically generated input | Opt and NatC results compared |
| | | | | | | | output->n_features= 1121, realistic data
| 1 | PASS | 463 | 2091.43 | 2612.67 | 0.00 | 2091.43 | Statically generated input | Opt and NatC results compared |
| | | | | | | | output->n_features= 463, realistic data
| 2 | PASS | 668 | 2411.65 | 2937.61 | 0.00 | 2411.65 | Statically generated input | Opt and NatC results compared |
| | | | | | | | output->n_features= 668, realistic data
| 3 | PASS | 232 | 2366.03 | 2878.15 | 0.00 | 2366.03 | Statically generated input | Opt and NatC results compared |
| | | | | | | | output->n_features= 232, realistic data
| 4 | PASS | 208 | 2827.00 | 3363.88 | 0.00 | 2827.00 | Statically generated input | Opt and NatC results compared |
| | | | | | | | output->n_features= 208, realistic data
| 5 | PASS | 42 | 3815.33 | 4261.71 | 0.00 | 3815.33 | Statically generated input | Opt and NatC results compared |
| | | | | | | | output->n_features= 42, realistic data
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Range:
Min: 2091.43 cycles/pt Avg: 2636.11 cycles/pt
Max: 3815.33 cycles/pt Med: 2366.03 cycles/pt
Num pt = Total Number of Features(numFeatures)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 56 bytes
+---------------------------------------+
| TESTING: VLIB_ORB_getHarrisScore |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_ORB_getHarrisScore testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1292 | 135.39 | 134.31 | 0.00 | 135.39 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 1292, width= 320, height= 240
| 1 | PASS | 612 | 120.36 | 114.52 | 0.00 | 120.36 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 612, width= 160, height= 120
| 2 | PASS | 100 | 129.50 | 113.23 | 0.00 | 129.50 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 100, width= 160, height= 120
| 3 | PASS | 50 | 107.02 | 109.78 | 0.00 | 107.02 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 50, width= 80, height= 60
| 4 | PASS | 40 | 111.50 | 120.63 | 0.00 | 111.50 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 40, width= 80, height= 60
| 5 | PASS | 30 | 113.97 | 122.73 | 0.00 | 113.97 | Statically generated input | Opt and NatC results compared |
| | | | | | | | input n_features= 30, width= 80, height= 60
| 6 | PASS | 200 | 131.34 | 132.83 | 0.00 | 131.34 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 200, width= 320, height= 240
| 7 | PASS | 100 | 138.56 | 135.79 | 0.00 | 138.56 | Sequentially generated input | Opt and NatC results compared
| | | | | | | | | input n_features= 100, width= 320, height= 240
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 133.99118*N - 1238
Where: N = output->n_features
Range:
Min: 107.02 cycles/pt Avg: 123.45 cycles/pt
Max: 138.56 cycles/pt Med: 120.36 cycles/pt
Num pt = output->n_features
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_originalfast12_detectCorners |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_originalfast12_detectCorners testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 76800 | 58.98 | 58.97 | 0.00 | 58.98 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 2876,
| | | | | | | | realistic data
| 1 | PASS | 19200 | 67.03 | 66.91 | 0.00 | 67.03 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 891,
| | | | | | | | realistic data
| 2 | PASS | 76800 | 35.21 | 35.20 | 0.00 | 35.21 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 0, flat
| | | | | | | | image
| 3 | PASS | 19200 | 29.77 | 29.69 | 0.00 | 29.77 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 0, flat
| | | | | | | | image
| 4 | PASS | 76800 | 48.05 | 48.15 | 0.00 | 48.05 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 10000,
| | | | | | | | checkerboard image
| 5 | PASS | 19200 | 65.05 | 65.18 | 0.00 | 65.05 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 3350,
| | | | | | | | checkerboard image
| 6 | PASS | 76800 | 59.33 | 59.37 | 0.00 | 59.33 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 2876,
| | | | | | | | realistic data
| 7 | PASS | 19200 | 66.96 | 66.89 | 0.00 | 66.96 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 891,
| | | | | | | | realistic data
| 8 | PASS | 76800 | 35.20 | 35.17 | 0.00 | 35.20 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 0, flat
| | | | | | | | image
| 9 | PASS | 19200 | 29.78 | 29.67 | 0.00 | 29.78 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 0, flat
| | | | | | | | image
| 10 | PASS | 76800 | 48.05 | 48.16 | 0.00 | 48.05 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 10000,
| | | | | | | | checkerboard image
| 11 | PASS | 19200 | 64.68 | 64.79 | 0.00 | 64.68 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 3350,
| | | | | | | | checkerboard image
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 12
Range:
Min: 29.77 cycles/pt Avg: 50.67 cycles/pt
Max: 67.03 cycles/pt Med: 48.05 cycles/pt
Num pt = number of input pixels (img_width*img_height); Performance is data dependent.
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 100 bytes
+---------------------------------------+
| TESTING: VLIB_originalfast12_score |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_originalfast12_score testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 2876 | 1061.95 | 1066.38 | 0.00 | 1061.95 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 2876, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 1 | PASS | 891 | 1043.57 | 1047.02 | 0.00 | 1043.57 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 891, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 2 | PASS | 306 | 1056.26 | 1062.55 | 0.00 | 1056.26 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 306, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 3 | PASS | 63 | 1119.71 | 1124.48 | 0.00 | 1119.71 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 63, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 4 | PASS | 2876 | 421.43 | 423.96 | 0.00 | 421.43 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 2876, fastDifferenceThreshold= 20, flat image
| 5 | PASS | 891 | 417.58 | 418.35 | 0.00 | 417.58 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 891, fastDifferenceThreshold= 20, flat image
| 6 | PASS | 306 | 426.63 | 435.42 | 0.00 | 426.63 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 306, fastDifferenceThreshold= 20, flat image
| 7 | PASS | 63 | 476.46 | 465.05 | 0.00 | 476.46 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 63, fastDifferenceThreshold= 20, flat image
| 8 | PASS | 2876 | 750.59 | 752.62 | 0.00 | 750.59 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 2876, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 9 | PASS | 891 | 758.54 | 758.75 | 0.00 | 758.54 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 891, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 10 | PASS | 306 | 784.05 | 790.90 | 0.00 | 784.05 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 306, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 11 | PASS | 63 | 860.54 | 862.16 | 0.00 | 860.54 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 63, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 12 | PASS | 2876 | 421.61 | 424.68 | 0.00 | 421.61 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 2876, fastDifferenceThreshold= 50, flat image
| 13 | PASS | 891 | 418.58 | 418.75 | 0.00 | 418.58 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 891, fastDifferenceThreshold= 50, flat image
| 14 | PASS | 306 | 429.96 | 433.61 | 0.00 | 429.96 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 306, fastDifferenceThreshold= 50, flat image
| 15 | PASS | 63 | 474.84 | 472.76 | 0.00 | 474.84 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_12= 63, fastDifferenceThreshold= 50, flat image
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 16
Range:
Min: 417.58 cycles/pt Avg: 682.65 cycles/pt
Max: 1119.71 cycles/pt Med: 476.46 cycles/pt
Num pt = n_features_12; Performance is data dependent.
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 92 bytes
+---------------------------------------+
| TESTING: VLIB_originalfast9_detectCorners |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_originalfast9_detectCorners testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 76800 | 61.88 | 61.78 | 0.00 | 61.88 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 6029,
| | | | | | | | realistic data
| 1 | PASS | 19200 | 74.22 | 74.06 | 0.00 | 74.22 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 2478,
| | | | | | | | realistic data
| 2 | PASS | 76800 | 35.23 | 35.19 | 0.00 | 35.23 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 0, flat
| | | | | | | | image
| 3 | PASS | 19200 | 29.77 | 29.76 | 0.00 | 29.77 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 0, flat
| | | | | | | | image
| 4 | PASS | 76800 | 45.21 | 45.19 | 0.00 | 45.21 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 10000,
| | | | | | | | checkerboard image
| 5 | PASS | 19200 | 60.96 | 60.90 | 0.00 | 60.96 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 3350,
| | | | | | | | checkerboard image
| 6 | PASS | 76800 | 62.30 | 62.33 | 0.00 | 62.30 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 6029,
| | | | | | | | realistic data
| 7 | PASS | 19200 | 74.21 | 74.06 | 0.00 | 74.21 | Statically generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 2478,
| | | | | | | | realistic data
| 8 | PASS | 76800 | 35.23 | 35.19 | 0.00 | 35.23 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 0, flat
| | | | | | | | image
| 9 | PASS | 19200 | 29.77 | 29.76 | 0.00 | 29.77 | Constant generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 0, flat
| | | | | | | | image
| 10 | PASS | 76800 | 45.19 | 45.19 | 0.00 | 45.19 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 320, img_height= 240, n_features found= 10000,
| | | | | | | | checkerboard image
| 11 | PASS | 19200 | 60.55 | 60.59 | 0.00 | 60.55 | Custom generated input | Opt and NatC results compared |
| | | | | | | | img_width= 160, img_height= 120, n_features found= 3350,
| | | | | | | | checkerboard image
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 12
Range:
Min: 29.77 cycles/pt Avg: 51.21 cycles/pt
Max: 74.22 cycles/pt Med: 45.21 cycles/pt
Num pt = number of input pixels (img_width*img_height); Performance is data dependent.
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 100 bytes
+---------------------------------------+
| TESTING: VLIB_originalfast9_score |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_originalfast9_score testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 6024 | 55.37 | 277.84 | 0.00 | 55.37 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 6024, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 1 | PASS | 2472 | 49.64 | 271.58 | 0.00 | 49.64 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 2472, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 2 | PASS | 896 | 53.77 | 273.93 | 0.00 | 53.77 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 896, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 3 | PASS | 176 | 74.30 | 292.49 | 0.00 | 74.30 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 176, fastDifferenceThreshold= 20, realistic
| | | | | | | | data
| 4 | PASS | 6024 | 55.27 | 280.14 | 0.00 | 55.27 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 6024, fastDifferenceThreshold= 20, flat image
| 5 | PASS | 2472 | 49.42 | 274.78 | 0.00 | 49.42 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 2472, fastDifferenceThreshold= 20, flat image
| 6 | PASS | 896 | 53.28 | 277.53 | 0.00 | 53.28 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 896, fastDifferenceThreshold= 20, flat image
| 7 | PASS | 176 | 74.64 | 296.67 | 0.00 | 74.64 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 176, fastDifferenceThreshold= 20, flat image
| 8 | PASS | 6024 | 55.31 | 277.49 | 0.00 | 55.31 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 6024, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 9 | PASS | 2472 | 49.26 | 271.19 | 0.00 | 49.26 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 2472, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 10 | PASS | 896 | 54.01 | 274.90 | 0.00 | 54.01 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 896, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 11 | PASS | 176 | 74.30 | 292.49 | 0.00 | 74.30 | Statically generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 176, fastDifferenceThreshold= 50, realistic
| | | | | | | | data
| 12 | PASS | 6024 | 55.34 | 279.98 | 0.00 | 55.34 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 6024, fastDifferenceThreshold= 50, flat image
| 13 | PASS | 2472 | 49.90 | 273.67 | 0.00 | 49.90 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 2472, fastDifferenceThreshold= 50, flat image
| 14 | PASS | 896 | 53.38 | 276.87 | 0.00 | 53.38 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 896, fastDifferenceThreshold= 50, flat image
| 15 | PASS | 176 | 74.64 | 298.95 | 0.00 | 74.64 | Constant generated input | Opt and NatC results compared |
| | | | | | | | n_features_9= 176, fastDifferenceThreshold= 50, flat image
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 16
Formula:
Cycles: 55.01145*N - 2362
Where: N=numFeatures
Range:
Min: 49.26 cycles/pt Avg: 58.24 cycles/pt
Max: 74.64 cycles/pt Med: 54.01 cycles/pt
Num pt = numFeatures
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 124 bytes
+---------------------------------------+
| TESTING: VLIB_packMask32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_packMask32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 18.05 | 15.19 | 0.00 | 18.05 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 1 | PASS | 128 | 10.56 | 9.43 | 0.00 | 10.56 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=128
| 2 | PASS | 256 | 5.60 | 6.83 | 0.00 | 5.60 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=256
| 3 | PASS | 480 | 3.81 | 5.77 | 0.00 | 3.81 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=480
| 4 | PASS | 640 | 3.20 | 5.48 | 0.00 | 3.20 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=640
| 5 | PASS | 1280 | 2.50 | 5.30 | 0.00 | 2.50 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1280
| 6 | PASS | 2600 | 2.16 | 4.56 | 0.00 | 2.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=2600
| 7 | PASS | 4800 | 1.94 | 4.49 | 0.00 | 1.94 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4800
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 1.73421*N + 1023
Where: N = pixelCount
Range:
Min: 1.94 cycles/pt Avg: 5.98 cycles/pt
Max: 18.05 cycles/pt Med: 3.20 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_recursiveFilterHoriz1stOrderS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_recursiveFilterHoriz1stOrderS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 64.42 | 56.86 | 0.00 | 64.42 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=8
| 1 | PASS | 512 | 35.72 | 44.37 | 0.00 | 35.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=16
| 2 | PASS | 512 | 35.09 | 45.31 | 0.00 | 35.09 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, height=32
| 3 | PASS | 2048 | 29.74 | 40.61 | 0.00 | 29.74 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32
| 4 | PASS | 3072 | 29.27 | 41.48 | 0.00 | 29.27 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=48, height=64
| 5 | PASS | 3072 | 29.22 | 40.30 | 0.00 | 29.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48
| 6 | PASS | 19200 | 28.85 | 40.24 | 0.00 | 28.85 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120
| 7 | PASS | 76800 | 29.22 | 39.66 | 0.00 | 29.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 29.55120*N - 131.23123*M + 5127
Where: N = width * height; M = height
Range:
Min: 28.85 cycles/pt Avg: 35.19 cycles/pt
Max: 64.42 cycles/pt Med: 29.27 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 64 bytes
+---------------------------------------+
| TESTING: VLIB_recursiveFilterHoriz1stOrder |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_recursiveFilterHoriz1stOrder testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 41.28 | 62.30 | 0.00 | 41.28 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=8
| 1 | PASS | 512 | 11.24 | 45.96 | 0.00 | 11.24 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=16
| 2 | PASS | 512 | 10.39 | 45.49 | 0.00 | 10.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, height=32
| 3 | PASS | 2048 | 7.58 | 44.04 | 0.00 | 7.58 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32
| 4 | PASS | 3072 | 6.00 | 42.45 | 0.00 | 6.00 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=48, height=64
| 5 | PASS | 3072 | 7.15 | 44.21 | 0.00 | 7.15 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48
| 6 | PASS | 19200 | 5.81 | 42.72 | 0.00 | 5.81 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120
| 7 | PASS | 76800 | 5.57 | 42.67 | 0.00 | 5.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 5.46344*N + 24.85923*M + 2478
Where: N = width * height; M = height
Range:
Min: 5.57 cycles/pt Avg: 11.88 cycles/pt
Max: 41.28 cycles/pt Med: 7.15 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 44 bytes
+---------------------------------------+
| TESTING: VLIB_recursiveFilterVert1stOrderS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_recursiveFilterVert1stOrderS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 38.94 | 62.58 | 0.00 | 38.94 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=8
| 1 | PASS | 64 | 39.41 | 61.36 | 0.00 | 39.41 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=8
| 2 | PASS | 512 | 11.66 | 46.16 | 0.00 | 11.66 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=16
| 3 | PASS | 512 | 11.77 | 45.33 | 0.00 | 11.77 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, height=32
| 4 | PASS | 2048 | 5.91 | 41.48 | 0.00 | 5.91 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32
| 5 | PASS | 3072 | 6.32 | 41.53 | 0.00 | 6.32 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=48, height=64
| 6 | PASS | 3072 | 5.64 | 41.02 | 0.00 | 5.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48
| 7 | PASS | 19200 | 5.42 | 39.37 | 0.00 | 5.42 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120
| 8 | PASS | 76800 | 6.75 | 41.80 | 0.00 | 6.75 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 9
Formula:
Cycles: 7.66170*N - 332.44354*M + 7823
Where: N = width * height; M = height
Range:
Min: 5.42 cycles/pt Avg: 14.65 cycles/pt
Max: 39.41 cycles/pt Med: 6.32 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_recursiveFilterVert1stOrder |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_recursiveFilterVert1stOrder testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 35.20 | 56.39 | 0.00 | 35.20 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=8
| 1 | PASS | 512 | 9.37 | 46.33 | 0.00 | 9.37 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=16
| 2 | PASS | 512 | 9.23 | 46.04 | 0.00 | 9.23 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, height=32
| 3 | PASS | 2048 | 6.44 | 44.39 | 0.00 | 6.44 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=32
| 4 | PASS | 3072 | 6.01 | 44.01 | 0.00 | 6.01 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=48, height=64
| 5 | PASS | 3072 | 6.06 | 43.10 | 0.00 | 6.06 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=48
| 6 | PASS | 19200 | 4.55 | 41.81 | 0.00 | 4.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=160, height=120
| 7 | PASS | 76800 | 4.58 | 41.78 | 0.00 | 4.58 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 4.54902*N - 6.55885*M + 3187
Where: N = width * height; M = height
Range:
Min: 4.55 cycles/pt Avg: 10.18 cycles/pt
Max: 35.20 cycles/pt Med: 6.06 cycles/pt
Num pt = width * height
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_simplex_3D |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_simplex_3D testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1 | 29643.00 | 29551.00 | 0.00 | 29643.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | MaxIteration=200, EPSILON=1342
| 1 | PASS | 1 | 24765.00 | 24819.00 | 0.00 | 24765.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | MaxIteration=200, EPSILON=134217
| 2 | PASS | 1 | 14145.00 | 14192.00 | 0.00 | 14145.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | MaxIteration=40, EPSILON=1342
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 3
Range:
Min: 14145.00 cycles/pt Avg: 22851.00 cycles/pt
Max: 29643.00 cycles/pt Med: 14145.00 cycles/pt
Num pt = Performance number includes iterations of cost function
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 80 bytes
+---------------------------------------+
| TESTING: VLIB_simplex |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_simplex testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 1 | 42590.00 | 39399.00 | 0.00 | 42590.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | N=3, MaxIteration=200, EPSILON=1342
| 1 | PASS | 1 | 35749.00 | 33141.00 | 0.00 | 35749.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | N=3, MaxIteration=200, EPSILON=134217
| 2 | PASS | 1 | 20006.00 | 18629.00 | 0.00 | 20006.00 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | N=3, MaxIteration=40, EPSILON=1342
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 3
Range:
Min: 20006.00 cycles/pt Avg: 32781.67 cycles/pt
Max: 42590.00 cycles/pt Med: 20006.00 cycles/pt
Num pt = Performance number includes iterations of cost function
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 104 bytes
+---------------------------------------+
| TESTING: VLIB_subtractBackgroundS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_subtractBackgroundS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 56.06 | 44.91 | 0.00 | 56.06 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=32
| 1 | PASS | 64 | 29.22 | 26.72 | 0.00 | 29.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=64
| 2 | PASS | 128 | 17.66 | 17.89 | 0.00 | 17.66 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=128
| 3 | PASS | 512 | 9.86 | 12.41 | 0.00 | 9.86 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=512
| 4 | PASS | 1024 | 8.65 | 11.34 | 0.00 | 8.65 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=1024
| 5 | PASS | 4096 | 7.72 | 10.46 | 0.00 | 7.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 7.38024*N + 1374
Where: N = PixelCount
Range:
Min: 7.72 cycles/pt Avg: 21.53 cycles/pt
Max: 56.06 cycles/pt Med: 9.86 cycles/pt
Num pt = PixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_subtractBackgroundS32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_subtractBackgroundS32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 58.34 | 53.03 | 0.00 | 58.34 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=32
| 1 | PASS | 64 | 35.42 | 34.08 | 0.00 | 35.42 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=64
| 2 | PASS | 128 | 23.95 | 25.23 | 0.00 | 23.95 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=128
| 3 | PASS | 512 | 16.18 | 19.39 | 0.00 | 16.18 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=512
| 4 | PASS | 1024 | 14.83 | 18.18 | 0.00 | 14.83 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=1024
| 5 | PASS | 4096 | 14.72 | 18.18 | 0.00 | 14.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | PixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 14.40792*N + 1099
Where: N = PixelCount
Range:
Min: 14.72 cycles/pt Avg: 27.24 cycles/pt
Max: 58.34 cycles/pt Med: 16.18 cycles/pt
Num pt = PixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_trackFeaturesLucasKanade_7x7 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_trackFeaturesLucasKanade_7x7 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 80 | 434.56 | 1086.65 | 0.00 | 434.56 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=10 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
| 1 | PASS | 80 | 326.48 | 1085.75 | 0.00 | 326.48 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=10 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=0(1=external,0=internal)
| 2 | PASS | 53 | 459.81 | 1379.36 | 0.00 | 459.81 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=10 |
| | | | | | | | earlyExit=1(1=en,0=dis) | gradCalculation=0(1=external,0=internal)
| 3 | PASS | 53 | 613.34 | 1374.38 | 0.00 | 613.34 | Custom generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=10 |
| | | | | | | | earlyExit=1(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
| 4 | PASS | 100 | 230.62 | 1204.86 | 0.00 | 230.62 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=2 | max_iters=50 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=0(1=external,0=internal)
| 5 | PASS | 100 | 272.27 | 1199.18 | 0.00 | 272.27 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=2 | max_iters=50 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
| 6 | PASS | 23 | 1016.04 | 2087.04 | 0.00 | 1016.04 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=2 | max_iters=50 |
| | | | | | | | earlyExit=1(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
| 7 | PASS | 8 | 2698.50 | 5688.88 | 0.00 | 2698.50 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=50 |
| | | | | | | | earlyExit=1(1=en,0=dis) | gradCalculation=0(1=external,0=internal)
| 8 | PASS | 8 | 2176.88 | 3372.88 | 0.00 | 2176.88 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=20 |
| | | | | | | | earlyExit=1(1=en,0=dis) | gradCalculation=0(1=external,0=internal)
| 9 | PASS | 1600 | 118.48 | 627.53 | 0.00 | 118.48 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=200 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=0(1=external,0=internal)
| 10 | PASS | 160 | 337.11 | 891.59 | 0.00 | 337.11 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=20 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
| 11 | PASS | 160 | 312.37 | 891.48 | 0.00 | 312.37 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=20 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
| 12 | PASS | 160 | 313.38 | 895.09 | 0.00 | 313.38 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=20 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
| 13 | PASS | 160 | 320.39 | 885.94 | 0.00 | 320.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | nfeatures=8 | max_iters=20 |
| | | | | | | | earlyExit=0(1=en,0=dis) | gradCalculation=1(1=external,0=internal)
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 1: gradient is externally calculated and results passed to function
Test vectors run: 8
Formula:
Cycles: 159.14889*N + 1569.69018*M + 12371
Where: N = nfeatures * num_iters; M = nfeatures
Range:
Min: 272.27 cycles/pt Avg: 452.43 cycles/pt
Max: 1016.04 cycles/pt Med: 320.39 cycles/pt
Num pt = nfeatures * num_iters
---------------------------------------------------------------------------------------------------------------------------------------------
Mode 2: gradient is internally calculated
Test vectors run: 6
Formula:
Cycles: 106.96007*N + 1004.96215*M + 10356
Where: N = nfeatures * num_iters; M = nfeatures
Range:
Min: 118.48 cycles/pt Avg: 1001.79 cycles/pt
Max: 2698.50 cycles/pt Med: 326.48 cycles/pt
Num pt = nfeatures * num_iters
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 252 bytes
+---------------------------------------+
| TESTING: VLIB_unpackMask32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_unpackMask32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 64 | 15.41 | 19.67 | 0.00 | 15.41 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 1 | PASS | 128 | 7.89 | 11.34 | 0.00 | 7.89 | Constant generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=128
| 2 | PASS | 256 | 4.91 | 7.14 | 0.00 | 4.91 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=256
| 3 | PASS | 480 | 2.97 | 5.21 | 0.00 | 2.97 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=480
| 4 | PASS | 640 | 2.46 | 4.89 | 0.00 | 2.46 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=640
| 5 | PASS | 1280 | 2.28 | 3.98 | 0.00 | 2.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1280
| 6 | PASS | 2600 | 1.79 | 3.67 | 0.00 | 1.79 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=2600
| 7 | PASS | 4800 | 1.75 | 3.43 | 0.00 | 1.75 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4800
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 1.56846*N + 768
Where: N = pixelCount
Range:
Min: 1.75 cycles/pt Avg: 4.93 cycles/pt
Max: 15.41 cycles/pt Med: 2.46 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_updateEWRMeanS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_updateEWRMeanS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 41.13 | 41.25 | 0.00 | 41.13 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 23.33 | 26.94 | 0.00 | 23.33 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 72 | 22.54 | 27.99 | 0.00 | 22.54 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=72
| 3 | PASS | 512 | 7.12 | 12.72 | 0.00 | 7.12 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 5.93 | 11.01 | 0.00 | 5.93 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 5.28 | 10.19 | 0.00 | 5.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 4.99506*N + 1136
Where: N = pixelCount
Range:
Min: 5.28 cycles/pt Avg: 17.55 cycles/pt
Max: 41.13 cycles/pt Med: 7.12 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_updateEWRMeanS32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_updateEWRMeanS32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 44.69 | 60.88 | 0.00 | 44.69 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 26.22 | 56.38 | 0.00 | 26.22 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 72 | 25.18 | 54.18 | 0.00 | 25.18 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=72
| 3 | PASS | 512 | 10.55 | 40.87 | 0.00 | 10.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 8.56 | 40.20 | 0.00 | 8.56 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 7.92 | 40.42 | 0.00 | 7.92 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 7.61075*N + 1230
Where: N = pixelCount
Range:
Min: 7.92 cycles/pt Avg: 20.52 cycles/pt
Max: 44.69 cycles/pt Med: 10.55 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_updateEWRVarianceS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_updateEWRVarianceS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 62.13 | 45.72 | 0.00 | 62.13 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 34.28 | 30.39 | 0.00 | 34.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 72 | 31.06 | 27.68 | 0.00 | 31.06 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=72
| 3 | PASS | 512 | 12.23 | 15.01 | 0.00 | 12.23 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 9.77 | 12.53 | 0.00 | 9.77 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 8.24 | 11.86 | 0.00 | 8.24 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 7.80547*N + 1861
Where: N = pixelCount
Range:
Min: 8.24 cycles/pt Avg: 26.28 cycles/pt
Max: 62.13 cycles/pt Med: 12.23 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_updateEWRVarianceS32 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_updateEWRVarianceS32 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 63.78 | 64.13 | 0.00 | 63.78 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 37.75 | 64.17 | 0.00 | 37.75 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 72 | 35.63 | 62.63 | 0.00 | 35.63 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=72
| 3 | PASS | 512 | 16.47 | 49.80 | 0.00 | 16.47 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 14.59 | 47.14 | 0.00 | 14.59 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 13.37 | 50.56 | 0.00 | 13.37 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 12.96732*N + 1657
Where: N = pixelCount
Range:
Min: 13.37 cycles/pt Avg: 30.27 cycles/pt
Max: 63.78 cycles/pt Med: 16.47 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_updateUWRMeanS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_updateUWRMeanS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 68.84 | 86.44 | 0.00 | 68.84 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 34.53 | 49.52 | 0.00 | 34.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 72 | 33.57 | 42.40 | 0.00 | 33.57 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=72
| 3 | PASS | 512 | 12.08 | 16.39 | 0.00 | 12.08 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 10.51 | 14.51 | 0.00 | 10.51 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 9.42 | 12.91 | 0.00 | 9.42 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 8.99146*N + 1701
Where: N = pixelCount
Range:
Min: 9.42 cycles/pt Avg: 28.16 cycles/pt
Max: 68.84 cycles/pt Med: 12.08 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_updateUWRVarianceS16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_updateUWRVarianceS16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 32 | 112.47 | 83.59 | 0.00 | 112.47 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=32
| 1 | PASS | 64 | 64.09 | 68.58 | 0.00 | 64.09 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=64
| 2 | PASS | 72 | 55.67 | 65.50 | 0.00 | 55.67 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=72
| 3 | PASS | 512 | 19.70 | 45.11 | 0.00 | 19.70 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=512
| 4 | PASS | 1024 | 17.39 | 43.39 | 0.00 | 17.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=1024
| 5 | PASS | 4096 | 16.10 | 42.97 | 0.00 | 16.10 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | pixelCount=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 6
Formula:
Cycles: 15.37891*N + 2722
Where: N = pixelCount
Range:
Min: 16.10 cycles/pt Avg: 47.57 cycles/pt
Max: 112.47 cycles/pt Med: 19.70 cycles/pt
Num pt = pixelCount
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 20 bytes
+---------------------------------------+
| TESTING: VLIB_weightedHistogram_1D_U16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_weightedHistogram_1D_U16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 100 | 26.67 | 25.24 | 0.00 | 26.67 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=10, maxSize=254, numX=100
| 1 | PASS | 64 | 27.42 | 23.95 | 0.00 | 27.42 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=25, numX=64
| 2 | PASS | 512 | 11.30 | 11.28 | 0.00 | 11.30 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=3, maxSize=253, numX=512
| 3 | PASS | 4096 | 8.99 | 6.70 | 0.00 | 8.99 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=13, maxSize=200, numX=4096
| 4 | PASS | 4096 | 9.32 | 8.00 | 0.00 | 9.32 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=7, maxSize=512, numX=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 5
Formula:
Cycles: 8.80880*N + 1423
Where: N = numX
Range:
Min: 8.99 cycles/pt Avg: 16.74 cycles/pt
Max: 27.42 cycles/pt Med: 9.32 cycles/pt
Num pt = numX
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_weightedHistogram_1D_U8 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_weightedHistogram_1D_U8 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 100 | 19.92 | 21.79 | 0.00 | 19.92 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numBins=10, maxSize=254, numX=100
| 1 | PASS | 64 | 22.36 | 26.67 | 0.00 | 22.36 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=25, numX=64
| 2 | PASS | 512 | 8.48 | 9.42 | 0.00 | 8.48 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=4, maxSize=253, numX=512
| 3 | PASS | 4096 | 5.90 | 6.93 | 0.00 | 5.90 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numBins=14, maxSize=200, numX=4096
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 4
Formula:
Cycles: 5.58259*N + 1324
Where: N = numX
Range:
Min: 5.90 cycles/pt Avg: 14.17 cycles/pt
Max: 22.36 cycles/pt Med: 8.48 cycles/pt
Num pt = numX
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_weightedHistogram_nD_U16 |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_weightedHistogram_nD_U16 testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 100 | 28.47 | 27.99 | 0.00 | 28.47 | Statically generated input | Opt results compared to NatC
| | | | | | | | results | numX=25, dimX=4, maxSize=254, outsize=48
| 1 | PASS | 48 | 769.90 | 777.38 | 0.00 | 769.90 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=4, dimX=12, maxSize=25, outsize=725760
| 2 | PASS | 128 | 23.18 | 24.09 | 0.00 | 23.18 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=32, dimX=4, maxSize=253, outsize=48
| 3 | PASS | 256 | 14.71 | 15.77 | 0.00 | 14.71 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=64, dimX=4, maxSize=253, outsize=48
| 4 | PASS | 2048 | 7.48 | 8.52 | 0.00 | 7.48 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=512, dimX=4, maxSize=253, outsize=48
| 5 | PASS | 12288 | 25.60 | 24.20 | 0.00 | 25.60 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=4096, dimX=3, maxSize=200, outsize=24800
| 6 | PASS | 8192 | 10.03 | 11.13 | 0.00 | 10.03 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | numX=4096, dimX=2, maxSize=512, outsize=180
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 7
Formula:
Cycles: 53.07368*N - 81.72695*M + 0.06674*L - 13400
Where: N = numX*dimX; M = numX; L = outsize (product of num of bins in each dim)
Range:
Min: 7.48 cycles/pt Avg: 125.62 cycles/pt
Max: 769.90 cycles/pt Med: 14.71 cycles/pt
Num pt = numX*dimX
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_xyGradientsAndMagnitude |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_xyGradientsAndMagnitude testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 16 | 88.19 | 81.63 | 0.00 | 88.19 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=4, out size=16
| 1 | PASS | 16 | 88.19 | 81.63 | 0.00 | 88.19 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=4, out size=16
| 2 | PASS | 96 | 24.36 | 22.03 | 0.00 | 24.36 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, height=8, out size=96
| 3 | PASS | 448 | 12.72 | 18.09 | 0.00 | 12.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=16, out size=448
| 4 | PASS | 1856 | 9.68 | 17.82 | 0.00 | 9.68 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=31, out size=1856
| 5 | PASS | 7936 | 10.39 | 17.76 | 0.00 | 10.39 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, height=64, out size=7936
| 6 | PASS | 76160 | 11.64 | 18.38 | 0.00 | 11.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, out size=76160
| 7 | PASS | 75768 | 11.28 | 18.75 | 0.00 | 11.28 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=328, height=233, out size=75768
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 11.46981*N - 1143
Where: N = width * (height-2)
Range:
Min: 9.68 cycles/pt Avg: 32.06 cycles/pt
Max: 88.19 cycles/pt Med: 11.64 cycles/pt
Num pt = width * (height-2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_xyGradients |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_xyGradients testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 16 | 66.19 | 63.75 | 0.00 | 66.19 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=4, out size=16
| 1 | PASS | 16 | 66.19 | 63.75 | 0.00 | 66.19 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=8, height=4, out size=16
| 2 | PASS | 96 | 17.40 | 19.25 | 0.00 | 17.40 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=16, height=8, out size=96
| 3 | PASS | 448 | 8.61 | 14.63 | 0.00 | 8.61 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=32, height=16, out size=448
| 4 | PASS | 1856 | 6.64 | 13.22 | 0.00 | 6.64 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=64, height=31, out size=1856
| 5 | PASS | 7936 | 6.53 | 12.97 | 0.00 | 6.53 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=128, height=64, out size=7936
| 6 | PASS | 76160 | 7.44 | 13.71 | 0.00 | 7.44 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=320, height=240, out size=76160
| 7 | PASS | 75768 | 8.87 | 13.23 | 0.00 | 8.87 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | width=328, height=233, out size=75768
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 8.16935*N - 1907
Where: N = width * (height-2)
Range:
Min: 6.53 cycles/pt Avg: 23.48 cycles/pt
Max: 66.19 cycles/pt Med: 8.61 cycles/pt
Num pt = width * (height-2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 16 bytes
+---------------------------------------+
| TESTING: VLIB_xyGradients_Magnitude_Orientations |
+---------------------------------------+
---------------------------------------------------------------------------------------------------------------------------------------------
VLIB_xyGradients_Magnitude_Orientations testing starts.
---------------------------------------------------------------------------------------------------------------------------------------------
| No | Status | Num pt | Opt Compute | NatC Compute |I/O | Opt Comp || IO | Description
| | | | cyc/pt | cyc/pt | cyc/pt | Total cyc/pt |
---------------------------------------------------------------------------------------------------------------------------------------------
| 0 | PASS | 12 | 1808.17 | 2299.25 | 0.00 | 1808.17 | Sequentially generated input | Opt results compared to NatC
| | | | | | | | results | Width=8, Height=4, bins=9, out size=12
| 1 | PASS | 128 | 288.98 | 541.38 | 0.00 | 288.98 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | Width=10, Height=18, bins=9, out size=128
| 2 | PASS | 84 | 377.87 | 636.27 | 0.00 | 377.87 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | Width=16, Height=8, bins=9, out size=84
| 3 | PASS | 420 | 176.06 | 413.54 | 0.00 | 176.06 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | Width=32, Height=16, bins=9, out size=420
| 4 | PASS | 1798 | 136.72 | 369.58 | 0.00 | 136.72 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | Width=64, Height=31, bins=9, out size=1798
| 5 | PASS | 7812 | 136.55 | 368.61 | 0.00 | 136.55 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | Width=128, Height=64, bins=12, out size=7812
| 6 | PASS | 75684 | 125.16 | 357.04 | 0.00 | 125.16 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | Width=320, Height=240, bins=9, out size=75684
| 7 | PASS | 304964 | 124.96 | 356.95 | 0.00 | 124.96 | Randomly generated input | Opt results compared to NatC
| | | | | | | | results | Width=640, Height=480, bins=9, out size=304964
---------------------------------------------------------------------------------------------------------------------------------------------
Test vectors run: 8
Formula:
Cycles: 124.85874*N + 31477
Where: N = (Width-2) * (Height-2)
Range:
Min: 124.96 cycles/pt Avg: 396.81 cycles/pt
Max: 1808.17 cycles/pt Med: 136.72 cycles/pt
Num pt = (Width-2) * (Height-2)
---------------------------------------------------------------------------------------------------------------------------------------------
Stack usage: 848 bytes