From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14236 invoked by alias); 13 Nov 2001 12:23:33 -0000 Mailing-List: contact cgen-help@sourceware.cygnus.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cgen-owner@sourceware.cygnus.com Received: (qmail 14108 invoked from network); 13 Nov 2001 12:23:23 -0000 Subject: Would supporting MIT SCALE Head & Tails be hard? To: cgen@sources.redhat.com Date: Tue, 02 Oct 2001 08:15:00 -0000 Reply-To: RDBrown@mira.net,RBrown64@csc.com.au X-Mailer: ELM [version 2.4ME+ PL66 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Message-Id: From: RDBrown@mira.net X-SW-Source: 2001-q4/txt/msg00000.txt.bz2 "Heads and Tails: A Variable-Length Instruction Format Supporting Parallel Fetch and Decode" (PDF paper) Heidi Pan and Krste Asanovic To appear, International Conference on Compilers, Architecture, and Synthesis for Embedded Systems (CASES 2001), Atlanta, GA, November 2001. >From the Abstract The new head-and-tails (HAT) format splits each instruction into a fixed-length head and a variable-length tail, and packs heads and tails in separate sections within a larger fixed-length instruction bundle. The paper evaluated 128-bit and 256-bit instruction bundles. As an idle question, how hard would it be for cgen to support this?