From bbb5abeeb03183de09d495422d8b06c8e4ddcb21 Mon Sep 17 00:00:00 2001
From: Yaman Umuroglu <maltanar@gmail.com>
Date: Thu, 3 Sep 2020 23:26:24 +0200
Subject: [PATCH] [Analysis] clarify critical_path_cycles
---
src/finn/analysis/fpgadataflow/dataflow_performance.py | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/finn/analysis/fpgadataflow/dataflow_performance.py b/src/finn/analysis/fpgadataflow/dataflow_performance.py
index 6f19254d6..e678630ae 100644
--- a/src/finn/analysis/fpgadataflow/dataflow_performance.py
+++ b/src/finn/analysis/fpgadataflow/dataflow_performance.py
@@ -32,6 +32,9 @@ from finn.util.fpgadataflow import is_fpgadataflow_node
def dataflow_performance(model):
"""Extract key performance indicators from given model with dataflow nodes.
+ Note that the latency (critical path) analysis is very pessimistic, it
+ assumes no overlap between executions and simply sums the expected cycles
+ for each node along the critical path.
Preconditions:
- model consists of fpgadataflow nodes
@@ -41,7 +44,7 @@ def dataflow_performance(model):
Returns:
- max_cycles : number of cycles for slowest node
- max_cycles_node_name : name of slowest node
- - critical_path_cycles : total expected latency from input to output
+ - critical_path_cycles : pessimistic expected latency from input to output
"""
latency_at_node_output = {}
max_cycles = 0
--
GitLab