aboutsummaryrefslogtreecommitdiff
path: root/src/q.tex
diff options
context:
space:
mode:
authords2horner <ds2horner@gmail.com>2017-07-01 20:42:25 -0400
committerAndrew Waterman <andrew@sifive.com>2017-07-01 22:08:54 -0700
commit6c372b544b4bed13455d711bc0ca1316ff7a7fa4 (patch)
tree50c9bb4f941aeafad6901f4da9b99324c14579f3 /src/q.tex
parent2634695db445868f429978fd2b2cbaace69aeec3 (diff)
downloadriscv-isa-manual-6c372b544b4bed13455d711bc0ca1316ff7a7fa4.zip
riscv-isa-manual-6c372b544b4bed13455d711bc0ca1316ff7a7fa4.tar.gz
riscv-isa-manual-6c372b544b4bed13455d711bc0ca1316ff7a7fa4.tar.bz2
Clarify Float Compare description
Modeled after SLT description: Make register compare order explicit (rs1 to rs2) Make Boolean result explicit (1 written if true, zero otherwise).
Diffstat (limited to 'src/q.tex')
-rw-r--r--src/q.tex6
1 files changed, 3 insertions, 3 deletions
diff --git a/src/q.tex b/src/q.tex
index 9a54234..17a5bcc 100644
--- a/src/q.tex
+++ b/src/q.tex
@@ -242,9 +242,9 @@ RV128 supports FMV.X.Q and FMV.Q.X in the Q extension.
\section{Quad-Precision Floating-Point Compare Instructions}
-Floating-point compare instructions perform the specified comparison (equal,
-less than, or less than or equal) between floating-point registers {\em rs1}
-and {\em rs2} and record the Boolean result in integer register {\em rd}.
+The quad-precision floating-point compare instructions are
+defined analogously to their double-precision counterparts, but operate on
+quad-precision operands.
\vspace{-0.2in}
\begin{center}