Date:	Thu, 10 Jun 1999 10:44:13 +0300 (IDT)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	Scott Rohde <rohde AT prairienet DOT org>
cc:	djgpp AT delorie DOT com
Subject:	Re: Linking in the math library produces peculiar results.
In-Reply-To:	<_3K63.2422$xg.109@firefly>
Message-ID:	<Pine.SUN.3.91.990610104327.17697A-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp AT delorie DOT com
X-Mailing-List:	djgpp AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Mon, 7 Jun 1999, Scott Rohde wrote:

> The following test program, which uses the sine function, works as
> expected when compiled without the -lm option (I assume that means it
> is using a definition for sin(double) in libc.a).  But with -lm, I get
> very peculiar results: for the integral (but not int!) values of x
> between x = 29.0 and x = 75.0, the value returned by the sine function
> for one of the four angles having x for a reference angle is off in
> all but eleven cases.

Thanks for a detailed and comprehensive report!

This problem is due to a subtle bug in the emulator that is triggered
by the elaborate argument-reduction algorithm in libm.a.  I have fixed
the bug, and the fixed version will be part of the next DJGPP release.
If you need the fixed version now, mail me privately and I will send
you the corrected emulator.

> Whether or
> not I link in floating point emulation with the -lemu switch makes no
> difference in the program results.

This is normal: the emulator and libemu.a share the same code (and the
same bugs ;-).

In addition, libemu.a from v2.02 didn't include the modified
npxsetup.o module which sets things up so that the emulation is
handled by the linked-in functions instead of loading emu387.dxe.  So
it's possible that your program linked with -lemu was still loading
the emulator at run time.

- Raw text -