From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 96492 invoked by alias); 21 May 2015 08:12:49 -0000 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 Received: (qmail 96461 invoked by uid 89); 21 May 2015 08:12:49 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.4 required=5.0 tests=AWL,BAYES_00,KAM_LAZY_DOMAIN_SECURITY,SPF_HELO_PASS,T_RP_MATCHES_RCVD autolearn=no version=3.3.2 X-HELO: mx1.redhat.com Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-GCM-SHA384 encrypted) ESMTPS; Thu, 21 May 2015 08:12:48 +0000 Received: from int-mx14.intmail.prod.int.phx2.redhat.com (int-mx14.intmail.prod.int.phx2.redhat.com [10.5.11.27]) by mx1.redhat.com (Postfix) with ESMTPS id 2754F5D; Thu, 21 May 2015 08:12:47 +0000 (UTC) Received: from [127.0.0.1] (ovpn01.gateway.prod.ext.ams2.redhat.com [10.39.146.11]) by int-mx14.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id t4L8CjSk019403; Thu, 21 May 2015 04:12:46 -0400 Message-ID: <555D937D.4010800@redhat.com> Date: Thu, 21 May 2015 08:12:00 -0000 From: Pedro Alves User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0 MIME-Version: 1.0 To: Andrew Burgess , gdb-patches@sourceware.org Subject: Re: [PATCH 0/4] layout command changes References: In-Reply-To: Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-SW-Source: 2015-05/txt/msg00529.txt.bz2 On 05/21/2015 12:17 AM, Andrew Burgess wrote: > This patch set replaces an earlier patch I posted here: > https://sourceware.org/ml/gdb-patches/2015-04/msg00185.html > > In the previous patch I had to jump through some hoops in order to > support completion of the layout names like $FREGS. This was pretty > annoying as I had not realised these layouts existed until I started > writting the completer code... > > ...but it turns out that those layout names don't work anyway, and > have not done so for some time. I didn't figure out exactly when they > broke, but I believe they were broken in 6.8. > > Still, it doesn't matter, as we have the 'tui regs' command, which > does work, and does allow the register set displayed in tui to be > changed. This is for the best anyway (I think), personally, I felt > that managing both the layout, and the choice of register set all from > the layout command was too much overloading. > > The first patch in this series removes the $FREGS style register set > names from the layout command, and cleans up all of the code relating > to them. Looks like this was really meant to switch to the matching registers layout when the user did "display $fpregs", etc. instead of manually specifying that layout. We have: static void display_command (char *arg, int from_tty) { struct format_data fmt; struct expression *expr; struct display *newobj; int display_it = 1; const char *exp = arg; #if defined(TUI) /* NOTE: cagney/2003-02-13 The `tui_active' was previously `tui_version'. */ if (tui_active && exp != NULL && *exp == '$') display_it = (tui_set_layout_for_display_command (exp) == TUI_FAILURE); #endif Doesn't your series effectively make this bit in display_command dead? (while before it would switch on the registers layout). (We should probably rename tui_set_layout_for_display_command too.) I had never noticed these special register layouts before either. I'm not at all adverse to removing them. Not all expressions that start with $ are registers, and probably a better idea would be to have a separate "displays" window (so displays would go to that window instead of the command window when the TUI is active), so that the tui could neatly show watched variables/random expressions too. > > The second patch is a much simpler version of command completion > support for layout names. > > The third and forth patches fix small tui related issues that I > spotted during testing. Thanks, Pedro Alves