View Issue Details

Jump to NotesJump to History
IDProjectCategoryView StatusDate SubmittedLast Update
00000681003.1(2008)/Issue 7System Interfacespublic2009-06-26 15:492013-04-16 13:06
Reportermsbrown Assigned Toajosey  
PrioritynormalSeverityCommentTypeError
Status ClosedResolutionAccepted As Marked 
NameMark Brown
OrganizationIBM
User Reference
Sectiontgamma
Page Number2105
Line Number66605
Interp StatusApproved
Final Accepted Text0000068:0000125
Summary0000068: tgamma range error overflow text
DescriptionCOMMENT Enhancement Request Number 6
 mtk.manpages:xxxxxxxxx Bug in XSHd5.1 tgamma (rdvk# 4)
 {tgamma} Thu, 24 Jul 2008 07:04:12 +0100 (BST)
 _____________________________________________________________________________

 In Section 7.12.8.4 of the C99 spec, it says:

 A range error may occur if the magnitude of x is too large or too small.

 XSH specifies the overflow condition, but does not specify the underflow
 condition. It should also specify the underflow condition.
Desired ActionAt line 66605, append the following text:

 ==
 If the correct value would cause underflow, a range error shall occur
 and tgamma(), tgammaf(), and tgammal() shall return 0.
 ==

 (A point about which I'm unsure mathematically is whether the last
 piece should read:

     ... shall return 0, with
     the same sign as the correct value of the function.

 I'll leave that to the math experts.)

 ==========

 At line 66626, append the following text:

 ==
 Range Error The value underflows.
 If the integer expression (math_errhandling & MATH_ERRNO) is non-zero,
 then errno shall be set to [ERANGE]. If the integer expression
 (math_errhandling & MATH_ERREXCEPT) is non-zero, then the underflow
 floating-point exception shall be raised.
Tagstc1-2008

Relationships

related to 0000348 Closedajosey 1003.1(2008)/Issue 7 XRAT margin code descriptions for MX and MXX 
related to 0001382 Closed 1003.1(2016/18)/Issue7+TC2 Ambiguous statement in asin() and other math functions 

Activities

msbrown

2009-06-26 15:49

manager   bugnote:0000125

Last edited: 2009-10-09 16:33

 Interpretation response
------------------------
The standard does not speak to this issue, and as such
no conformance distinction can be made between alternative
implementations based on this. This is being referred to the
sponsor.

Rationale:
-------------
None.

Notes to the Editor (not part of this interpretation):
-------------------------------------------------------
This is a cross volume change
All page and line numbers are for draft 5.1.

Changes to XBD ...

At page 9 line 246 replace the description of the MX code with:

MX IEC 60559 Floating-Point
     The functionality described is optional. The functionality
     described is mandated by the ISO C standard only for
     implementations that define __STDC_IEC_559__.

and add an MXX code after it:

MXX IEC 60559 Floating-Point Extension
     The functionality described is part of the IEC 60559
     Floating-Point option, but is an extension to the ISO C standard.

(The "part of" here is because the functionality provided by both
MX and MXX is available when __STDC_IEC_559__ is defined. Effectively
the two together form a single option.)

Note that the above changes include removal of the second paragraph of
the MX description:

    Where applicable, functions are marked with the MX margin legend
    in the SYNOPSIS section. Where additional semantics apply to a
    function, the material is identified by use of the MX margin legend.

(This is because there are no functions whose SYNOPSIS has MX shading,
so there is no need to distinguish between MX markers in the SYNOPSIS
and those elsewhere.)

Changes to XSH ...

At P593 L20443 (asin), P595 L20501 (asinh), P599 L20603 (atan),
P605 L20752 (atanh), P790 L26409 (expm1), P1249 L41188 (log1p),
P1958 L62217 (sin), P1960 L62286 (sinh), P2074 L65711 (tan) and
P2076 L65793 (tanh) change:

    [MX] If x is subnormal, a range error may occur and x should be
    returned.[/MX]

to (substituting the appropriate function names for asin(), asinf()
and asinl()):

    [MX] If x is subnormal, a range error may occur[/MX] [MXX] and x
    should be returned.[/MXX]
    [MX] If x is not returned, asin(), asinf(), and asinl() shall
    return an implementation-defined value no greater in magnitude
    than DBL_MIN, FLT_MIN, and LDBL_MIN, respectively.[/MX]

At P601 L20660 (atan2) change:

    [MX] If the result underflows, a range error may occur and y/x
    should be returned.[/MX]

to:

    If the correct value would cause underflow, a range error may
    occur, and atan2(), atan2f(), and atan2l() shall return an
    implementation-defined value no greater in magnitude than DBL_MIN,
    FLT_MIN, and LDBL_MIN, respectively. [MXX] If IEC 60559
    Floating-Point is supported, y/x should be returned.[/MXX]

At P765 L25519 (erf) change:

    [MX] If x is subnormal, a range error may occur, and 2 * x/sqrt(pi)
    should be returned.[/MX]

to:

    If the correct value would cause underflow, a range error may
    occur, and erf(), erff(), and erfl() shall return an
    implementation-defined value no greater in magnitude than
    DBL_MIN, FLT_MIN, and LDBL_MIN, respectively. [MXX] If IEC
    60559 Floating-Point is supported, 2 * x/sqrt(pi) should be
    returned.[/MXX]

At P767 L25583 (erfc) change:

    If the correct value would cause underflow and is not
    representable, a range error may occur and [MX] either 0.0 (if
    representable), or[/MX] an implementation-defined value shall be
    returned.

to:

    If the correct value would cause underflow, [MXX]and is not
    representable[/MXX], a range error may occur, and erfc(), erfcf(),
    and erfcl() shall return [MXX] 0.0, or [/MXX] (if IEC 60559
    Floating-Point is not supported) an implementation-defined value
    no greater in magnitude than DBL_MIN, FLT_MIN, and LDBL_MIN,
    respectively.

At P767 L25589 (erfc) change MX shading to MXX for:

    If the correct value would cause underflow and is representable,
    a range error may occur and the correct value shall be returned.

At P786 L26270 (exp), P788 L26347 (exp2), P870 L28890 (fmod),
P1210 L39939 (ldexp), P1518 L49006 (pow) and P2074 L65707 (tan)
change:

    If the correct value would cause underflow, and is not
    representable, a range error may occur, and [MX] either 0.0
    (if supported), or[/MX] an implementation-defined value shall be
    returned.

to (substituting the appropriate function names for exp(), expf() and
expl()):

    If the correct value would cause underflow, [MXX]and is not
    representable[/MXX], a range error may occur, and exp(), expf(),
    and expl() shall return [MXX] 0.0, or [/MXX] (if IEC 60559
    Floating-Point is not supported) an implementation-defined value
    no greater in magnitude than DBL_MIN, FLT_MIN, and LDBL_MIN,
    respectively.

At P1795 L57424 (scalbln) change:

    If the correct value would cause underflow, and is not
    representable, a range error may occur, and [MX] either 0.0
    (if supported), or[/MX] an implementation-defined value shall be
    returned.

to:

    If the correct value would cause underflow, [MXX]and is not
    representable[/MXX], a range error may occur, and scalbln(),
    scalblnf(), scalblnl(), scalbn(), scalbnf(), and scalbnl()
    shall return [MXX] 0.0, or [/MXX] (if IEC 60559 Floating-Point
    is not supported) an implementation-defined value no greater in
    magnitude than DBL_MIN, FLT_MIN, LDBL_MIN, DBL_MIN, FLT_MIN, and
    LDBL_MIN, respectively.

At P786 L26276 (exp), P788 L26353 (exp2), P870 L28899 (fmod),
P1210 L39943 (ldexp), P1519 L49029 (pow), P1795 L57429 (scalbln)
and P2074 L65714 (tan) change MX shading to MXX for:

    If the correct value would cause underflow, and is representable,
    a range error may occur and the correct value shall be returned.

At P818 L27345 (fdim) change:

    If x-y is positive and underflows, a range error may occur, and
    either (x-y) (if representable), [XSI] or 0.0 (if supported),[/XSI]
    or an implementation-defined value shall be returned.

to:

    If the correct value would cause underflow, a range error may
    occur, and fdim(), fdimf(), and fdiml() shall return [MXX] the
    correct value, or [/MXX] (if IEC 60559 Floating-Point is not
    supported) an implementation-defined value no greater in magnitude
    than DBL_MIN, FLT_MIN, and LDBL_MIN, respectively.

At P819 L27364 (fdim) delete from APPLICATION USAGE:

    On implementations supporting IEEE Std 754-1985, x-y cannot
    underflow, and hence the 0.0 return value is shaded as an
    extension for implementations supporting the XSI option rather
    than an MX extension.

At P1098 L36650 (hypot) change MX shading to MXX for:

    If both arguments are subnormal and the correct result is subnormal,
    a range error may occur and the correct result is returned.

and change "is returned" to "shall be returned".

At P1196 L39586 (j0) add MXX shading to:

    If x is NaN, a NaN shall be returned.

At P1367 L44844 (nextafter) change:

    [MX] If x!=y and the correct function value is subnormal, zero, or
    underflows, a range error shall occur, and either the correct
    function value (if representable) or 0.0 shall be returned.[/MX]

to:

    [MX] If x!=y and the correct function value is subnormal, zero, or
    underflows, a range error shall occur, and[/MX] [MXX] the correct
    function value (if representable) or[/MXX] [MX] 0.0 shall be
    returned.[/MX]

At P2105 L66604 (tgamma) add:

    If the correct value would cause underflow, [MXX]and is not
    representable[/MXX], a range error may occur, and tgamma(),
    tgammaf(), and tgammal() shall return [MXX] 0.0, or [/MXX] (if
    IEC 60559 Floating-Point is not supported) an implementation-
    defined value no greater in magnitude than DBL_MIN, FLT_MIN,
    and LDBL_MIN, respectively.

    [MXX] If the correct value would cause underflow, and is
    representable, a range error may occur and the correct value
    shall be returned.[/MXX]

At P2106 L66636 (tgamma) add:

    Range Error The result underflows.

    If the integer expression (math_errhandling & MATH_ERRNO) is non-zero,
    then errno shall be set to [ERANGE]. If the integer expression
    (math_errhandling & MATH_ERREXCEPT) is non-zero, then the underflow
    floating-point exception shall be raised.

At P2274 L71531 (y0) add MXX shading to:

    If x is NaN, a NaN shall be returned.


(Target for TC1)

Issue History

Date Modified Username Field Change
2009-06-26 15:49 msbrown New Issue
2009-06-26 15:49 msbrown Status New => Under Review
2009-06-26 15:49 msbrown Assigned To => ajosey
2009-06-26 15:49 msbrown Name => Mark Brown
2009-06-26 15:49 msbrown Organization => IBM
2009-06-26 15:49 msbrown Section => tgamma
2009-06-26 15:49 msbrown Page Number => 2105
2009-06-26 15:49 msbrown Line Number => 66605
2009-06-26 15:49 msbrown Note Added: 0000125
2009-06-26 15:49 msbrown Status Under Review => Interpretation Required
2009-06-26 15:49 msbrown Resolution Open => Accepted As Marked
2009-06-26 15:50 msbrown Final Accepted Text => 0000068:0000125
2009-08-11 16:29 Don Cragun Interp Status => Pending
2009-09-17 15:41 nick Interp Status Pending => Proposed
2009-10-09 16:33 ajosey Note Edited: 0000125
2009-10-09 16:33 ajosey Interp Status Proposed => Approved
2010-09-20 09:12 geoffclare Tag Attached: tc1-2008
2010-11-05 17:11 geoffclare Relationship added related to 0000348
2013-04-16 13:06 ajosey Status Interpretation Required => Closed
2020-12-17 17:12 nick Relationship added related to 0001382