From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 10094 invoked by alias); 6 May 2011 14:33:54 -0000 Received: (qmail 10083 invoked by uid 22791); 6 May 2011 14:33:53 -0000 X-SWARE-Spam-Status: No, hits=-1.8 required=5.0 tests=AWL,BAYES_00,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (38.113.113.100) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 06 May 2011 14:33:39 +0000 Received: (qmail 23779 invoked from network); 6 May 2011 14:33:38 -0000 Received: from unknown (HELO digraph.polyomino.org.uk) (joseph@127.0.0.2) by mail.codesourcery.com with ESMTPA; 6 May 2011 14:33:38 -0000 Received: from jsm28 (helo=localhost) by digraph.polyomino.org.uk with local-esmtp (Exim 4.72) (envelope-from ) id 1QIM65-0002O9-CY; Fri, 06 May 2011 14:33:37 +0000 Date: Fri, 06 May 2011 14:33:00 -0000 From: "Joseph S. Myers" To: Yao Qi cc: gdb-patches@sourceware.org Subject: Re: [patch, testsuite] Tests to _Complex type In-Reply-To: <4DC401D0.1050500@codesourcery.com> Message-ID: References: <4DC401D0.1050500@codesourcery.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2011-05/txt/msg00188.txt.bz2 On Fri, 6 May 2011, Yao Qi wrote: > When I am fixing GDB's inf-call in one port, I find tests to _Complex > type is quite limited, especially on argument passing, there is none. > This patch is to add more tests to _Complex type on argument passing and > vararg support. It would be good to include a much wider range of tests. structs and unions containing complex values (and arrays of complex values, and complex values mixed with real ones, etc.), complex values in many different argument positions, and after integer arguments as well as floating-point ones, all three types _Complex float, _Complex double, _Complex long double properly covered in these ways. For structs and unions and arrays therein, the sort of case you want to cover includes the ARM (VFP ABI) and IA64 rules on homogeneous aggregates. (The reason I don't suggest covering complex integer types is that they are a GNU extension with no documented GNU DWARF extension to represent them. In principle target-specific types such as __float80 and __float128 (on x86 and IA64) and their complex versions *should* be covered, but at present GCC doesn't support using _Complex with those type names so accessing the complex versions requires using mode attributes.) -- Joseph S. Myers joseph@codesourcery.com