<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=us-ascii" http-equiv=Content-Type>
<META name=GENERATOR content="MSHTML 11.00.9600.19130"></HEAD>
<BODY>
<DIV dir=ltr align=left><SPAN class=166360815-17092018><FONT color=#0000ff
size=2 face=Arial>Makes sense .. thanks.</FONT></SPAN></DIV>
<DIV dir=ltr align=left><SPAN class=166360815-17092018><FONT color=#0000ff
size=2 face=Arial></FONT></SPAN> </DIV>
<DIV dir=ltr align=left><SPAN class=166360815-17092018><FONT color=#0000ff
size=2 face=Arial>Mike</FONT></SPAN></DIV><BR>
<BLOCKQUOTE
style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px solid; MARGIN-RIGHT: 0px"
dir=ltr>
<DIV lang=en-us class=OutlookMessageHeader dir=ltr align=left>
<HR tabIndex=-1>
<FONT size=2 face=Tahoma><B>From:</B> Eric Schwarz
[mailto:eschwarz@us.ibm.com] <BR><B>Sent:</B> 17 September 2018
13:26<BR><B>To:</B> Mike Cowlishaw<BR><B>Cc:</B> cfp-interest@oakapple.net;
'Jim Thomas'<BR><B>Subject:</B> RE: [Cfp-interest] Not trigger
sNaN<BR></FONT><BR></DIV>
<DIV></DIV>
<P><FONT size=2>Mike,</FONT><BR><BR><FONT size=2>I don't remember defining the
sameQuantum operation. </FONT><BR><FONT size=2>I think it was to compare
exponents or as a compare units instruction </FONT><BR><FONT size=2>and used
it to determine whether the result had the expected quantum. </FONT><BR><FONT
size=2>This was before IBM architecture came up with a flag for quantum
exceptions and</FONT><BR><FONT size=2>before we figured out we could test the
most significant digit for non-zero.</FONT><BR><FONT size=2>So if we are just
testing a result for correct units, we don't want to trigger
exceptions,</FONT><BR><FONT size=2>though we may want to force it into the
correct units.</FONT><BR><FONT size=2>An example is we have a table of data to
output that should have the units of 10^-2, we</FONT><BR><FONT size=2>test
each data item before converting the significand to BCD and outputting
the</FONT><BR><FONT size=2>data. Probably figured SNaNs would have already
been seen by this point and</FONT><BR><FONT size=2>not a point in the program
where we want to take any exceptions.</FONT><BR><FONT
size=2><BR>Regards,<BR>Eric<BR><BR><BR>eschwarz@us.ibm.com</FONT><BR><BR><TT><FONT
size=2>"Mike Cowlishaw" <mfc@speleotrove.com> wrote on 09/17/2018
06:47:19 AM:<BR><BR>> From: "Mike Cowlishaw"
<mfc@speleotrove.com></FONT></TT><BR><TT><FONT size=2>> To: "'Jim
Thomas'" <jaswthomas@sbcglobal.net></FONT></TT><BR><TT><FONT size=2>>
Cc: <cfp-interest@oakapple.net>, "'Eric Schwarz'"
<eschwarz@us.ibm.com></FONT></TT><BR><TT><FONT size=2>> Date:
09/17/2018 06:47 AM</FONT></TT><BR><TT><FONT size=2>> Subject: RE:
[Cfp-interest] Not trigger sNaN</FONT></TT><BR><TT><FONT size=2>> <BR>>
I believe it was, but I do not recall the reasoning. Eric Schwartz
might<BR>> remember; cc:ing him ...<BR>> <BR>> Mike <BR>> <BR>>
> -----Original Message-----<BR>> > From: Jim Thomas [<A
href="mailto:jaswthomas@sbcglobal.net">mailto:jaswthomas@sbcglobal.net</A>]
<BR>> > Sent: 16 September 2018 16:40<BR>> > To: Mike
Cowlishaw<BR>> > Cc: cfp-interest@oakapple.net<BR>> > Subject: Re:
[Cfp-interest] Not trigger sNaN<BR>> > <BR>> > Mike,<BR>> >
<BR>> > 754 seems to say sameQuantum(sNaN, qNaN) is true with no
<BR>> > exception. Just double checking that it is as intended.<BR>>
> <BR>> > Jim<BR>> > <BR>> > > On Aug 28, 2018, at
7:16 PM, Fred J. Tydeman <BR>> > <tydeman@tybor.com>
wrote:<BR>> > > <BR>> > > On Tue, 28 Aug 2018 13:13:45 -0500
Rajan Bhakta wrote:<BR>> > >> <BR>> > >>
*AI*: Fred: See which other functions have the need to not
<BR>> > >> trigger signaling nans but are functions (need to be
macros).<BR>> > > <BR>> > > 754 Operations that shall not
trigger an sNaN (signal invalid):<BR>> > > <BR>> > > "*" are
functions in C<BR>> > > <BR>> > > <BR>> > ...<BR>>
> > sameQuantum() = samequantum()*<BR>> > > <BR>> >
<BR>> <BR></FONT></TT><BR></P></BLOCKQUOTE></BODY></HTML>