
<HTML>

<HEAD>
<TITLE>timesoftfloat</TITLE>
</HEAD>

<BODY>

<H1>Berkeley TestFloat Release 3c: <CODE>timesoftfloat</CODE></H1>

<P>
John R. Hauser<BR>
2017 February 10<BR>
</P>


<H2>Overview</H2>

<P>
The <CODE>timesoftfloat</CODE> program provides a simple way to evaluate the
speed of the floating-point operations of the Berkeley SoftFloat library.
Program <CODE>timesoftfloat</CODE> is included with the Berkeley TestFloat
package, a small collection of programs for testing that an implementation of
floating-point conforms to the IEEE Standard for Binary Floating-Point
Arithmetic.
Although <CODE>timesoftfloat</CODE> does not test floating-point correctness
like the other TestFloat programs, nevertheless <CODE>timesoftfloat</CODE> is a
partner to TestFloat&rsquo;s <CODE>testsoftfloat</CODE> program.
For more about TestFloat generally and <CODE>testsoftfloat</CODE> specifically,
see file
<A HREF="TestFloat-general.html"><NOBR><CODE>TestFloat-general.html</CODE></NOBR></A>.
</P>

<P>
Ordinarily, <CODE>timesoftfloat</CODE> will measure a function&rsquo;s speed
separately for each of the five rounding modes defined by the IEEE
Floating-Point Standard, one after the other, plus possibly a sixth mode,
<I>round to odd</I> (depending on the options selected when
<CODE>timesoftfloat</CODE> was compiled).
If an operation is not supposed to require rounding, it will by default be
timed only with the rounding mode set to <CODE>near_even</CODE> (nearest/even).
In the same way, if an operation is affected by the way in which underflow
tininess is detected, <CODE>timesoftfloat</CODE> times the function with
tininess detected both before rounding and after rounding.
For <NOBR>80-bit</NOBR> double-extended-precision operations affected by
rounding precision control, <CODE>timesoftfloat</CODE> also times the function
for each of the three rounding precision modes, one after the other.
Evaluation of a function can be limited to a single rounding mode, a single
tininess mode, and/or a single rounding precision with appropriate command-line
options.
</P>

<P>
For each function and mode evaluated, <CODE>timesoftfloat</CODE> reports the
measured speed of the function in Mop/s, or &ldquo;millions of operations per
second&rdquo;.
The speeds reported by <CODE>timesoftfloat</CODE> may be affected somewhat by
other software executing at the same time as <CODE>timesoftfloat</CODE>.
Be aware also that the exact execution time of any SoftFloat function depends
partly on the values of arguments and the state of the processor&rsquo;s caches
at the time the function is called.
Your actual experience with SoftFloat may differ from the speeds reported by
<CODE>timesoftfloat</CODE> for all these reasons.
</P>

<P>
Note that the remainder operations for larger formats (<CODE>f64_rem</CODE>,
<CODE>extF80_rem</CODE>, and <CODE>f128_rem</CODE>) can be markedly slower than
other operations, particularly for double-extended-precision
(<CODE>extF80_rem</CODE>) and quadruple precision (<CODE>f128_rem</CODE>).
This is inherent to the remainder operation itself and is not a failing of the
SoftFloat implementation.
</P>


<H2>Command Syntax</H2>

<P>
The <CODE>timesoftfloat</CODE> program is executed as a command with this
syntax:
<BLOCKQUOTE>
<PRE>
timesoftfloat [&lt;<I>option</I>&gt;...] &lt;<I>function</I>&gt;
</PRE>
</BLOCKQUOTE>
Square brackets (<CODE>[ ]</CODE>) denote optional arguments,
<CODE>&lt;<I>option</I>&gt;</CODE> is a supported option, and
<CODE>&lt;<I>function</I>&gt;</CODE> is the name of either a testable function
or a function set.
The available options and function sets are documented below.
If <CODE>timesoftfloat</CODE> is executed without any arguments, a summary of
usage is written.
</P>


<H2>Options</H2>

<P>
The <CODE>timesoftfloat</CODE> program accepts several command options.
If mutually contradictory options are given, the last one has priority.
</P>

<H3><CODE>-help</CODE></H3>

<P>
The <CODE>-help</CODE> option causes a summary of program usage to be written,
after which the program exits.
</P>

<H3><CODE>-precision32, -precision64, -precision80</CODE></H3>

<P>
For <NOBR>80-bit</NOBR> double-extended-precision funcions affected by
rounding precision control, the <CODE>-precision32</CODE> option restricts
timing of an operation to only the cases in which the rounding precision is
<NOBR>32 bits</NOBR>, equivalent to <NOBR>32-bit</NOBR> single-precision.
Other rounding precision choices are not timed.
Likewise, <CODE>-precision64</CODE> fixes the rounding precision to
<NOBR>64 bits</NOBR>, equivalent to <NOBR>64-bit</NOBR> double-precision;
and <CODE>-precision80</CODE> fixes the rounding precision to the full
<NOBR>80 bits</NOBR> of the double-extended-precision format.
All these options are ignored for operations not affected by rounding precision
control.
</P>

<H3><CODE>-rnear_even, -rnear_maxMag, -rminMag, -rmin, -rmax, -rodd</CODE></H3>

<P>
The <CODE>-rnear_even</CODE> option restricts timing of an operation to only
the cases in which the rounding mode is nearest/even.
Other rounding mode choices are not timed.
Likewise, <CODE>-rnear_maxMag</CODE> forces rounding to nearest/maximum
magnitude (nearest-away), <CODE>-rminMag</CODE> forces rounding to minimum
magnitude (toward zero), <CODE>-rmin</CODE> forces rounding to minimum (down,
toward negative infinity), <CODE>-rmax</CODE> forces rounding to maximum (up,
toward positive infinity), and <CODE>-rodd</CODE>, if supported, forces
rounding to odd.
These options are ignored for operations that are exact and thus do not round.
</P>

<H3><CODE>-tininessbefore, -tininessafter</CODE></H3>

<P>
The <CODE>-tininessbefore</CODE> option restricts timing of an operation to
only the cases in which tininess on underflow is detected before rounding.
Likewise, <CODE>-tininessafter</CODE> restricts measurement to only the cases
in which tininess on underflow is detected after rounding.
</P>

<H3><CODE>-notexact, -exact</CODE></H3>

<P>
For functions that round to an integer (conversions to integer types and the
<CODE>roundToInt</CODE> functions), the <CODE>-notexact</CODE> option restricts
timing of an operation to only the cases for which the
<CODE><I>exact</I></CODE> operand (specifying whether the <I>inexact</I>
exception flag may be raised) is <CODE>false</CODE>.
Likewise, the <CODE>-exact</CODE> option restricts measurement to only the
cases for which the <CODE><I>exact</I></CODE> operand is <CODE>true</CODE>.
</P>


<H2>Function Sets</H2>

<P>
Just as <CODE>timesoftfloat</CODE> can time a function for all five or six
rounding modes in sequence, multiple functions can be timed with a single
execution of <CODE>timesoftfloat</CODE>.
Three sets are recognized:
<CODE>-all1</CODE>, <CODE>-all2</CODE>, and <CODE>-all</CODE>.
The set <CODE>-all1</CODE> is all one-operand operations, <CODE>-all2</CODE> is
all two-operand operations, and <CODE>-all</CODE> is obviously all operations.
A function set is used in place of a function name in the
<CODE>timesoftfloat</CODE> command line, such as
<BLOCKQUOTE>
<PRE>
timesoftfloat [&lt;<I>option</I>&gt;...] -all1
</PRE>
</BLOCKQUOTE>
</P>

<P>
For the purpose of deciding the number of operands of an operation, any
<CODE><I>roundingMode</I></CODE> and <CODE><I>exact</I></CODE> arguments are
ignored.
(Such arguments specify the rounding mode and whether the <I>inexact</I>
exception flag may be raised, respectively.)
Thus, functions that convert to integer type and the <CODE>roundToInt</CODE>
functions are included in the set of one-operand operations timed by
<CODE>-all1</CODE>.
</P>


</BODY>