From mint-bounce@lists.fishpool.fi Sat Jan 26 16:39:41 2008 Message-ID: <479BA786.10403@freesbee.fr> Date: Sat, 26 Jan 2008 22:35:02 +0100 From: =?ISO-8859-1?Q?Vincent_Rivi=E8re?= User-Agent: Thunderbird 2.0.0.9 (Windows/20071031) MIME-Version: 1.0 To: mint Subject: Re: [MiNT] Alignment References: <4798E197.6050500@freesbee.fr> In-Reply-To: <4798E197.6050500@freesbee.fr> Content-Type: text/plain; charset=ISO-8859-1; format=flowed X-ecartis-version: Ecartis v1.0.0 Sender: mint-bounce@lists.fishpool.fi Errors-to: mint-bounce@lists.fishpool.fi X-original-sender: vincent.riviere@freesbee.fr Precedence: bulk List-help: List-unsubscribe: List-Id: X-List-ID: List-subscribe: List-owner: List-post: Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by mail.sparemint.org id m0QLdZTo005970 > Question 1: On a standard TT or Falcon with 68030, is it faster to do a > move.l to an address multiple of 4 than to an address not multiple of 4 ? Thank you for your answers. So the conclusion is: Aligning the longs on multiples of 4 is faster only inside the FastRAM. There are 2 cases: 1) Standard ST and Falcon 030 Aligning the longs on multiples of 4 is useless, because these machines have only ST-RAM, which is always wired on a 16-bit data bus. The biggest useful (and required) alignment is 2 bytes for words and longs (and maybe FPU floats ?) 2) Standard TT and accelerator cards for ST and Falcon with FastRAM The FastRAM is wired on a 32-bit data bus, and can only be accessed by the CPU. Thus it can work at full-speed when the longs are aligned on a 4-byte boundary. Does anyone agree ? -- Vincent Rivière