[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
BOUNCE rtl@rtlinux.org: Approval required: Non-member submissionfrom [Suessmilch Bernd <Bernd.Suessmilch@SWAROVSKI.COM>] (fwd)
- To: rtl@rtlinux.org
- Subject: BOUNCE rtl@rtlinux.org: Approval required: Non-member submissionfrom [Suessmilch Bernd <Bernd.Suessmilch@SWAROVSKI.COM>] (fwd)
- From: Der Herr Hofrat <der.herr@hofr.at>
- Date: Mon, 28 May 2001 07:06:29 +0200 (CEST)
>From owner-rtl Mon May 28 00:03:18 2001
Received: from lan212095014004.ibk.telering.at (lan212095014004.ibk.telering.at [212.95.14.4])
by hq.fsmlabs.com (8.11.2/8.11.2) with SMTP id f4S63Hs17865
for <rtl@fsmlabs.com>; Mon, 28 May 2001 00:03:17 -0600
Received: from no.name.available by lan212095014004.ibk.telering.at
via smtpd (for hq.fsmlabs.com [209.155.42.197]) with SMTP; 28 May 2001 05:58:10 UT
Received: from 193.188.217.34 by dsw80.wattens.swarovski.com (InterScan E-Mail VirusWall NT); Mon, 28 May 2001 07:58:06 +0200 (W. Europe Daylight Time)
Received: from dsw29.wattens.swarovski.com by [193.164.238.249]
via smtpd (for [193.164.238.240]) with SMTP; 28 May 2001 05:58:07 UT
Received: by dsw29.wattens.swarovski.com with Internet Mail Service (5.5.2650.21)
id <28N2ZCW4>; Mon, 28 May 2001 07:58:06 +0200
Message-ID: <853DD5D67AF0D411A3300000D1DD5C3B260E62@dsw89.wattens.swarovski.com>
From: Suessmilch Bernd <Bernd.Suessmilch@SWAROVSKI.COM>
To: "'rtl@fsmlabs.com'" <rtl@fsmlabs.com>,
"'hgull@mail.ukans.edu'"
<hgull@mail.ukans.edu>
Subject: RE: [rtl] uresolved symbol __udivdi3 and __umoddi3
Date: Mon, 28 May 2001 07:58:06 +0200
X-Mailer: Internet Mail Service (5.5.2650.21)
Hello Havard,
I think the simplest solution for your problem is not posted so far:
just tell the compiler to use the inline code of the math-functions.
You can do this by adding the "-O2" flag to your CFLAGS.
The inline math-functions are faster than the one in libm, but less
error checking is performed. You have to check the returned values
for "NaN" to detect errors.
A further drawback of the inline functions is that gdb has difficulties
to identify the right source lines when debugging.
Bernd
* - * - * - * - * - * - * - * - * - * -
SWAROVSKI & CO
Dipl.-Inf. Bernd Suessmilch
Elektronik - Entwicklung 2
Tel. ++43 - +5224 - 500 DW 3671
Fax ++43 - +5224 - 571 90
Mailto: Bernd.Suessmilch@swarovski.com