From: Kbwms AT aol DOT com Message-ID: <57.21830678.2c83b892@aol.com> Date: Sun, 31 Aug 2003 16:46:10 EDT Subject: Re: and To: djgpp-workers AT delorie DOT com MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="part1_57.21830678.2c83b892_boundary" X-Mailer: 8.0 for Windows sub 6015 Reply-To: djgpp-workers AT delorie DOT com --part1_57.21830678.2c83b892_boundary Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit In a message dated 8/31/2003 2:24:16 PM Eastern Standard Time, eliz AT elta DOT co DOT il writes: > Please allow me a couple of comments: > > >typedef int __int32_t; > > I think we don't need this typedef in math.h. If it is required to > compile libm.a, it should be moved into fdlibm.h, which is private to > that library. > Great idea. Good to be rid of it. That means that the following must be moved to fdlibm.h, too. Also, I can uncomment their definition for HUGE_VAL. typedef int __int32_t; typedef unsigned int __uint32_t; union __dmath { __uint32_t i[2]; double d; }; extern const union __dmath __infinity; //#define HUGE_VAL (__infinity.d) --part1_57.21830678.2c83b892_boundary Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable In a message dated 8/31/2003 2:24:16 PM Eastern Standard= Time, eliz AT elta DOT co DOT il writes:

Please allow me a couple of comments:

>typedef int __int32_t;

I think we don't need this typedef in math.h.  If it is required to
compile libm.a, it should be moved into fdlibm.h, which is private to
that library.


Great idea.  Good to be rid of it.  That means that the following=20= must be moved to fdlibm.h, too.  Also, I can uncomment their definition= for HUGE_VAL.

typedef int __int32_t;
typedef unsigned int __uint32_t;

union __dmath
{
  __uint32_t i[2];
  double d;
};

extern const union __dmath __infinity;
//#define HUGE_VAL (__infinity.d)


 --part1_57.21830678.2c83b892_boundary--