This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Fix FLT_EPSILON docs


On Fri, Feb 17, 2012 at 6:48 PM, Joseph S. Myers
<joseph@codesourcery.com> wrote:
> Bug 5077 points out that the documentation of FLT_EPSILON is inaccurate.
> I've applied this patch to correct it. ?Feel free to improve it further if
> you can think of even better wording.
>
> diff --git a/ChangeLog b/ChangeLog
> index 6a9b16b..32cbb65 100644
> --- a/ChangeLog
> +++ b/ChangeLog
> @@ -1,3 +1,9 @@
> +2012-02-17 ?Joseph Myers ?<joseph@codesourcery.com>
> +
> + ? ? ? [BZ #5077]
> + ? ? ? * manual/lang.texi (FLT_EPSILON): Avoid description depending on
> + ? ? ? rounding modes.
> +
> ?2012-02-17 ?Fabrice Bauzac ?<fabrice.bauzac@wanadoo.fr>
>
> ? ? ? ?[BZ #6907]
> diff --git a/NEWS b/NEWS
> index 59e07ea..3b97e5c 100644
> --- a/NEWS
> +++ b/NEWS
> @@ -9,9 +9,9 @@ Version 2.16
>
> ?* The following bugs are resolved with this release:
>
> - ?174, 350, 411, 3335, 4026, 4822, 5805, 6884, 6907, 9902, 10140, 10210,
> - ?11494, 12047, 13525, 13526, 13527, 13528, 13529, 13530, 13531, 13532,
> - ?13533, 13547, 13551, 13552, 13553, 13555, 13559, 13583, 13618
> + ?174, 350, 411, 3335, 4026, 4822, 5077, 5805, 6884, 6907, 9902, 10140,
> + ?10210, 11494, 12047, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
> + ?13532, 13533, 13547, 13551, 13552, 13553, 13555, 13559, 13583, 13618
>
> ?* ISO C11 support:
>
> diff --git a/manual/lang.texi b/manual/lang.texi
> index 82c5e96..9bc9871 100644
> --- a/manual/lang.texi
> +++ b/manual/lang.texi
> @@ -1202,8 +1202,8 @@ macro's value is the same as the type it describes.
> ?@comment ISO
> ?@item FLT_EPSILON
>
> -This is the minimum positive floating point number of type @code{float}
> -such that @code{1.0 + FLT_EPSILON != 1.0} is true. ?It's supposed to
> +This is the difference between 1 and the smallest floating point
> +number of type @code{float} that is greater than 1. ?It's supposed to
> ?be no greater than @code{1E-5}.

This looks good to me.

I noticed the report mentioned that this problem also existed for
DBL_EPSILON and LDBL_EPSILON but I did not find such references.

Therefore it looks like your patch catches all the instance of this bug.

Cheers,
Carlos.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]