- xlopt
-
Enable kernel optimizations based on the specified value:
-
0: No kernel optimizations.
-
1: Computation of heap size, generation of guidance based on LLVM IR analysis, and insertion of loop pragmas.
-
2: The same optimizations as 1 with the addition of loop peeling for unrolled loops, and automatic inlining.
xlopt=2 -
- Xxloptstr
-
Option string to enable or disable optimizations in XLOpt level 1,2.
- -xlinline-threshold=T: set inlining threshold to T (default T = 5000, only for optlevel=2).
- -annotate-pragma: insertion of loop unrolling, pipelining, and flattening pragmas (default = true)
- -align-global-array=0,1 if set to 1, global arrays are aligned on 2 times the width of the physical memory bank. This automatic alignment optimizes load/store performance. Load/Store ports bitwidth is twice the width of a physical bank and aligned on even index physical bank. This option automatically aligns the data to the even index bank address.
Xxloptstr=-annotate-pragma