aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorAndrew Waterman <andrew@sifive.com>2019-05-06 23:51:52 -0700
committerAndrew Waterman <andrew@sifive.com>2019-05-06 23:52:09 -0700
commit8772ea6f0f0e7136c21981d75ad25cd799a83c33 (patch)
treeeae8ea15a316473a0261dac79536d9b9c3688db6
parentf02a75082ca87fed504792553dcb9d3c88b32404 (diff)
downloadriscv-isa-manual-8772ea6f0f0e7136c21981d75ad25cd799a83c33.zip
riscv-isa-manual-8772ea6f0f0e7136c21981d75ad25cd799a83c33.tar.gz
riscv-isa-manual-8772ea6f0f0e7136c21981d75ad25cd799a83c33.tar.bz2
Add note about FLE vs. BGE inconsistency
-rw-r--r--src/f.tex7
1 files changed, 7 insertions, 0 deletions
diff --git a/src/f.tex b/src/f.tex
index 76786c9..7680347 100644
--- a/src/f.tex
+++ b/src/f.tex
@@ -765,6 +765,13 @@ FCMP & S & src2 & src1 & EQ/LT/LE & dest & OP-FP \\
\end{tabular}
\end{center}
+\begin{commentary}
+The F extension provides a $\leq$ comparison, whereas the base ISA provides
+a $\geq$ branch comparison. Because $\leq$ can be synthesized from $\geq$ and
+vice-versa, there is no performance implication to this inconsistency, but it
+is nevertheless an unfortunate incongruity in the ISA.
+\end{commentary}
+
\section{Single-Precision Floating-Point Classify Instruction}
The FCLASS.S instruction examines the value in floating-point register {\em