From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 17417 invoked by alias); 6 Dec 2001 23:10:05 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 17290 invoked from network); 6 Dec 2001 23:10:01 -0000 Received: from unknown (HELO vlsi1.ultra.nyu.edu) (128.122.140.213) by sources.redhat.com with SMTP; 6 Dec 2001 23:10:01 -0000 Received: by vlsi1.ultra.nyu.edu (4.1/1.34) id AA04387; Thu, 6 Dec 01 18:04:35 EST Date: Thu, 06 Dec 2001 15:12:00 -0000 From: kenner@vlsi1.ultra.nyu.edu (Richard Kenner) Message-Id: <10112062304.AA04387@vlsi1.ultra.nyu.edu> To: jsm28@cam.ac.uk Subject: Re: ACATS Cc: gcc@gcc.gnu.org X-SW-Source: 2001-12/txt/msg00324.txt.bz2 The point of using the vendor branch is to have *both* the original form (on the vendor branch) and the GCC version (on the mainline) in CVS - and when a new version is imported onto the vendor branch, CVS should be able to help with a lot of the work of merging onto the mainline. The CVS experts here can probably provide step-by-step instructions for doing this and indications of the pitfalls. I think there's a disconnect here. The issue is that the *format*, not the *content*, of the two branches is fundamentally different. The ACATS cvs contains one file per test while the GCC ACATS cvs would contain one file per *compilation unit* and there are often multiple such per test.