X-Authentication-Warning:	delorie.com: mail set sender to djgpp-bounces using -f
X-Recipient:	djgpp AT delorie DOT com
Message-ID:	<5A4FA4F9.4000601@gmx.de>
Date:	Fri, 05 Jan 2018 17:16:57 +0100
From:	"Juan Manuel Guerrero (juan DOT guerrero AT gmx DOT de) [via djgpp AT delorie DOT com]" <djgpp AT delorie DOT com>
User-Agent:	Mozilla/5.0 (X11; U; Linux x86_64; de; rv:1.9.2.13) Gecko/20101206 SUSE/3.1.7 Thunderbird/3.1.7
MIME-Version:	1.0
To:	djgpp AT delorie DOT com
Subject:	Re: Fixing various bugs in frexp.S
References:	<5A4946EB DOT 3090500 AT gmx DOT de>
In-Reply-To:	<5A4946EB.3090500@gmx.de>
X-Provags-ID:	V03:K0:p4/crrzLlsJZzh+v5bHStiw4NUWK9uN6ZeEgde+EZicGqXEiPhX
	5wJSfwlJxun+tGheBlyqwiOHVWWsvr+64eUe24PIfPBQet7CkAwBe6zMyDLRbYHPdVJc2s5
	sLWWGr9MMojilgGlX2msSZaRXB8M9AEj4hM/WA4vpqXvc2e/83BnfxPmZrRmOoT92nuugz2
	faaWQhu+m5fno32UfIfqg==
X-UI-Out-Filterresults:	notjunk:1;V01:K0:aBfmka/aghE=:NQb4uxXmjdEusTX0waNnvN
	NvKs+7a6wOFa5cdvkgt+hLxx+Hd6HDmkUwabN7Hxj9UMJIdOB72sOUrLggXDIuzq1KArnl2Sr
	bNddkiEpjQm8VsNdtQBc8D0eVCvfQSuNe3RxSqJjjT39wSF8hNzcRxdPI66zALxgT6AB4CqNB
	riLrsmWtM+HWxU2iSxi3mX9UK5obkPP4cSH69AyAwRp1gqNgOt41JjYLtM3y7IzVDKNbhvZ3N
	fDxePrcIRaQZbXmnwT1J+wHHvFkCtlxdAvyzpOWlWjXyEDt2EAv7UYd+K07fnF/Z2v0u0L/aC
	2wKRPYpd+nNUXxR7e0MHQexMuxen0K0J8blOUzRKRWAQTsb5G0BWVR9FmbXH5YCcTAuHsrPSA
	kEiEbw9dcscbbVqqaMgtRgercUFz7t8oKo7c8AnB7QDjZdILtqUL36TjUXAi2jDm/T+qQ30N5
	iJaRJf6A2k9X8fF6nCdxqTwIwHAgqk5Fwj6KbpieGoQ9nJO+RHQn2US/hAQxKEySVQuG3d4Fv
	B4mDPxd92SAmSRPtvoLl/S6Snt1y5hufWzINrs+mqxqyCemcXpZ6QIyqjuwx83lWQgjRnPrbH
	baMp2bifcpMneuaCNfDjKRhBUrs/znrKcZ7NZWBc1huMKIn8EggtzMiDxUcFHq9XWZNPAiMub
	EHJ063RtOqBxxLBbhe2mt1cI1rJ42PwBXujVdQMEQxFj19txFkz6MqAcV0xuR4QjIUfMp6KSr
	d2JGjPn/QKYQ66CCGtB5OtKpJ6DEB8jRYgOJxQIYgGMCnS+38TF+HdQjarYna9Xcvn2DifcQq
	6znwtmLoSm0p+b0SFkyNn66mP/MPs8NdaD6P73g7muBPw/VRFA=
Reply-To:	djgpp AT delorie DOT com

Am 31.12.2017 21:22, schrieb Juan Manuel Guerrero (juan DOT guerrero AT gmx DOT de) [via djgpp AT delorie DOT com]:
> The frexp function provided by libc.a has a couple of bugs:
> - it sets errno to EDOM if the argument is either infinity or NaN
> - it produces the wrong value if the argument is -0
> - it returns NaN if the argument is infinity
> - for NaN and infinity arguments no sign checks are done thus the
> results are always positive
>
> All these issues are serious deviations from the ansi/posix behavior.
> With the patch below I have tried to fix these issues. If I do not get
> any serious objection I will commite the patch in a couple of days.

OFYI, I commited the changes.

Regards,
Juan M. Guerrero

- Raw text -