From 8bf6e713d5e9985886df581b2b00c57752d18022 Mon Sep 17 00:00:00 2001
From: Dominic Symes <dominic.symes@arm.com>
Date: Thu, 7 Sep 2023 11:10:59 +0100
Subject: Main Conformance: Add local_bound to RFFT2D

Also express the fp32_t acc_min_normal calculation
without a double negation.

Signed-off-by: Dominic Symes <dominic.symes@arm.com>
Change-Id: I34ba274264bc511e60f0e01601d5d0a3f983e594
---
 chapters/introduction.adoc | 4 ++--
 tosa.xml                   | 8 ++++++++
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/chapters/introduction.adoc b/chapters/introduction.adoc
index d415e24..c8cfc30 100644
--- a/chapters/introduction.adoc
+++ b/chapters/introduction.adoc
@@ -336,10 +336,10 @@ fp64_t out_err_sum = 0.0;
 fp64_t out_err_sumsq = 0.0;
 fp64_t acc_prec;       // 1<<(M+1) where M is the number of mantissa bits
 fp64_t acc_min_normal; // accumulator minimum normal greater than zero
-fp64_t two_m63 = -1.0/(fp64)((int64_t)-1<<63);         // pow(2,  -63)
+fp64_t two_m42 = 1.0/static_cast<fp64_t>((int64_t)1<<42);         // pow(2,  -42)
 switch (acc_t) {
     case fp32_t: acc_prec       = static_cast<fp64_t>(1<<24);     // pow(2,   24)
-                 acc_min_normal = two_m63 * two_m63;   // pow(2, -126)
+                 acc_min_normal = two_m42 * two_m42 * two_m42;    // pow(2, -126)
                  break;
     case fp16_t: acc_prec       = static_cast<fp64_t>(1<<11);     // pow(2,   11)
                  acc_min_normal = 1.0/static_cast<fp64_t>(1<<14); // pow(2,  -14)
diff --git a/tosa.xml b/tosa.xml
index 8cf9fce..cf9b3c3 100644
--- a/tosa.xml
+++ b/tosa.xml
@@ -568,6 +568,14 @@
             <description>Real part of the complex output</description>
             <rank min="3" max="3"/>
           </argument>
+          <argument category="attribute" name="local_bound" type="tensor_t" shape="-" tensor-element-type="bool_t" optional="true">
+            <description>
+                This optional attribute affects the floating-point compliance error bound.
+                The default of false allows for direct and transform based, fast convolution algorithms.
+                Only set to true if direct dot-product calculation precision is required.
+            </description>
+            <rank min="0" max="0"/>
+          </argument>
           <argument category="output" name="output_imag" type="tensor_t" shape="[N,H,W/2 + 1]" tensor-element-type="in_out_t">
             <description>Imaginary part of the complex output.</description>
             <rank min="3" max="3"/>
-- 
cgit v1.2.1