Description
syn.directive.performance applies to loops and loop nests, and requires a known
loop tripcount to determine the performance. If your loop has a variable tripcount then you
must also specify syn.directive.tripcount.The syn.directive.performance lets you
specify a high-level constraint, target_ti or target_tl. The directive allows you to define the number of
clock cycles between successive starts of a loop. The directive lets the tool infer
lower-level UNROLL, PIPELINE, ARRAY_PARTITION, and INLINE directives for achieving the
desired result. The syn.directive.performance does not
guarantee the specified value will be achieved, and so it is only a target.
The target_ti is the interval between
successive starts of the loop. It is also the interval between the start of the first
iteration of the loop, and the next start of the first iteration of the loop. In the
following code example, a target_ti=T means the target
interval for the start of loop L2 between two consecutive
iterations of L1 is 100 cycles.
const int T = 100;
L1: for (int i=0; i<N; i++)
L2: for (int j=0; j<M; j++){
#pragma HLS PERFORMANCE target_ti=T
...
}The target_tl is the interval between start of the loop and end of the
loop, or between the start of the first iteration of the loop and the completion of the last
iteration of the loop. In the preceding code example a target_tl=T means
the target completion of loop L2 for a single iteration of
L1 should be 100 cycles.
syn.directive.inline applies automatically to functions inside any pipelined
loop with II=1 to improve throughput. If you apply the PERFORMANCE pragma, the tool also
triggers auto-inline optimization. If you apply a directive that infers a pipeline with
II=1, the tool also triggers auto-inline optimization. You can disable this for specific
functions by using syn.directive.inline off.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, assume an image processing function processes a single frame
per invocation with a throughput goal of 60 fps. 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.
Syntax
Specify the directive for a labeled loop.
syn.directive.performance=<location> [Options]Where:
<location> specifies the loop in the format function/loop_label.
Options
-
target_ti=<value> - Specifies a target transaction interval that is the number of clock
cycles for the loop to complete an iteration. The transaction interval refers to the
number of clock cycles from the first transaction of a loop, or nested loop to the start
of the next transaction of the loop. The <value> can be an integer, floating
point, or constant expression that is resolved by the tool as an integer. Note: The tool returns a warning if truncation occurs.
-
target_tl=<value> -
Specifies a target latency that is the number of clock cycles for the loop to complete all iterations. The transaction latency is the interval between the start of the first iteration of the loop, and the completion of the last iteration of the loop. The <value> can be an integer, floating point, or constant expression that is resolved by the tool as an integer.
- unit=[sec | cycle]
- Specifies the unit associated with the
target_tiortarget_tlvalues. The unit can either be seconds, or clock cycles. When the unit is specified as seconds, you can specify it with a value to indicate nanoseconds (ns), picoseconds (ps) or microseconds (us).
Example 1
The loop labeled loop_1 has a target
transaction interval of 4 clock cycles:
syn.directive.performance=loop_1 target_ti=4