Description
Specifies a target performance value for the transaction interval (target_ti) of loops; where a transaction is a complete set of
loop iterations (tripcount) and the interval is the time between when the first transaction
starts and the second transaction starts. The PERFORMANCE
pragma does not guarantee the specified value will be achieved, and so it is only a target.
- Target Transaction Interval (
target_ti) - Specifies the number of clock cycles between successive starts of the loop. In other words, the clock cycles from the start of the first transaction of a loop, or nested loop, and the start of the next transaction of the loop.
The transaction interval is the initiation interval (II) of the loop times
the number of iterations, or tripcount: target_ti = II *
loop tripcount. Conversely, target_ti
= FreqHz / Operations per second.
For example, assuming an image processing function that processes a single
frame per invocation with a throughput goal of 60 fps, then the target throughput for the
function is 60 invocations per second. If the clock frequency is 180 MHz, then target_ti is 180M/60, or 3 million clock cycles per function
invocation.
Vitis HLS always tries to achieve the
specified performance target in the design. When the PERFORMANCE pragma is specified, the
tool automatically applies pragmas or directives such as PIPELINE, UNROLL, or
ARRAY_PARTITION to achieve the target_ti.
Syntax
Place the pragma within the boundary a loop, or the outer loop of a loop nest.
#pragma HLS performance target_ti=<value>Where:
-
target_ti=<value> - Specifies a target transaction interval defined as the number of clock
cycles for the function, loop, or region of code to complete an iteration. The
<value> can be specified as an integer, floating point, or constant expression
that is resolved by the tool as an integer. Note: A warning will be returned if truncation occurs.
Example 1
Function func is specified to have target
transaction interval of 4 clock cycles.
int func(char x, char a, char b, char c) {
#pragma HLS performance target_ti=4
char y;
y = x*a+b+c;
return y
}