From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.30) id 1B3hqH-00034O-5z for qemu-devel@nongnu.org; Wed, 17 Mar 2004 15:40:45 -0500 Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.30) id 1B3hpi-0002nd-0X for qemu-devel@nongnu.org; Wed, 17 Mar 2004 15:40:42 -0500 Received: from [213.191.74.84] (helo=mx2.ngi.de) by monty-python.gnu.org with esmtp (Exim 4.30) id 1B3hph-0002mb-3Y for qemu-devel@nongnu.org; Wed, 17 Mar 2004 15:40:09 -0500 Date: Wed, 17 Mar 2004 19:12:17 +0100 From: Richard Zidlicky Message-ID: <20040317181217.GC2562@linux-m68k.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Subject: [Qemu-devel] [patch] m68k updates #3 Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Fabrice Bellard , Qemu ML Hi, this is another patch where I could use some advice. m68k has hw support for "long double" so I thought I could easilly improve performance a little bit and get better precision by using them. Appended patch shows how I was trying to do this, but there must be something else I have missed because regression tests dont work.. Anyway, the change might be usefull if someone else attempts to do the same for some other arch. Richard diff -urN qemu-cvs/target-i386/cpu.h qemu-cvs-new/target-i386/cpu.h --- qemu-cvs/target-i386/cpu.h Thu Feb 26 00:14:19 2004 +++ qemu-cvs-new/target-i386/cpu.h Wed Mar 17 18:39:40 2004 @@ -252,7 +252,8 @@ CC_OP_NB, }; -#ifdef __i386__ +/* should work for m68k as well but needs more work somewhere */ +#if defined(__i386__) //|| defined (__mc68000) #define USE_X86LDOUBLE #endif diff -urN qemu-cvs/target-i386/exec.h qemu-cvs-new/target-i386/exec.h --- qemu-cvs/target-i386/exec.h Thu Feb 26 00:26:33 2004 +++ qemu-cvs-new/target-i386/exec.h Wed Mar 17 18:39:39 2004 @@ -440,7 +440,7 @@ /* XXX: same endianness assumed */ -#ifdef CONFIG_USER_ONLY +#if defined(CONFIG_USER_ONLY) && defined (__i386__) static inline CPU86_LDouble helper_fldt(uint8_t *ptr) {