仿照imglib 中11x11的卷积 写了一段19X19的卷积
优化后结果如下：
;*----------------------------------------------------------------------------*
;* SOFTWARE PIPELINE INFORMATION
;*
;* Loop source line : 72
;* Loop opening brace source line : 73
;* Loop closing brace source line : 159
;* Known Minimum Trip Count : 19
;* Known Maximum Trip Count : 19
;* Known Max Trip Count Factor : 19
;* Loop Carried Dependency Bound(^) : 4
;* Unpartitioned Resource Bound : 12
;* Partitioned Resource Bound(*) : 12
;* Resource Partition:
;* A-side B-side
;* .L units 0 0
;* .S units 3 2
;* .D units 3 3
;* .M units 7 4
;* .X cross paths 8 8
;* .T address paths 6 6
;* Long read paths 0 0
;* Long write paths 0 0
;* Logical ops (.LS) 7 13 (.L or .S unit)
;* Addition ops (.LSD) 21 16 (.L or .S or .D unit)
;* Bound(.L .S .LS) 5 8
;* Bound(.L .S .D .LS .LSD) 12* 12*
;*
;* Searching for software pipeline schedule at ...
;* ii = 12 Schedule found with 3 iterations in parallel
;* Done
;*
;* Loop will be splooped
;* Collapsed epilog stages : 0
;* Collapsed prolog stages : 0
;* Minimum required memory pad : 0 bytes
;*
;* Minimum safe trip count : 1
开发新手，不知道获取这个信息后，应该着手从什么方向进一步进行优化。
;* Bound(.L .S .D .LS .LSD) 12* 12*
;* Bound(.L .S .LS) 5 8
这种信息具体是指什么，TI的文档里面好多看不太明白