From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 2270 invoked by alias); 16 Feb 2004 20:59:43 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 2212 invoked from network); 16 Feb 2004 20:59:42 -0000 Received: from unknown (HELO nevyn.them.org) (66.93.172.17) by sources.redhat.com with SMTP; 16 Feb 2004 20:59:42 -0000 Received: from drow by nevyn.them.org with local (Exim 4.30 #1 (Debian)) id 1Aspq9-0004KW-Qb; Mon, 16 Feb 2004 15:59:41 -0500 Date: Mon, 16 Feb 2004 20:59:00 -0000 From: Daniel Jacobowitz To: Elena Zannoni Cc: gdb-patches@sources.redhat.com Subject: Re: [rfa/symbol readers] Clean up setting symbols' names Message-ID: <20040216205941.GA16546@nevyn.them.org> Mail-Followup-To: Elena Zannoni , gdb-patches@sources.redhat.com References: <20040216193453.GA1667@nevyn.them.org> <16433.11537.105919.97207@localhost.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <16433.11537.105919.97207@localhost.redhat.com> User-Agent: Mutt/1.5.1i X-SW-Source: 2004-02/txt/msg00432.txt.bz2 On Mon, Feb 16, 2004 at 03:50:25PM -0500, Elena Zannoni wrote: > Daniel Jacobowitz writes: > > This patch does a couple of related things. > > > > 1. It adds a macro, SYMBOL_SET_LINKAGE_NAME, which is used to set a > > symbol's name when the name should not be demangled. Used for things like > > typedefs whose name comes from debug info. I did not change anything > > related to memory allocation when I did this - though in hpread I found a > > few inconsistencies. > > > > 2. It makes hpread use SYMBOL_SET_NAMES. It contained some fallback code > > to use the debug info if demangling failed; I've disabled it, because from > > the description it sounds like the right thing to do would be to fix the > > demangler instead. I can't tell whether the code is currently used or not, > > so I would appreciate HP/UX testing for this patch. > > > > In the process I also kill a couple of DEPRECATED_SYMBOL_NAME uses, and the > > redundant function add_psymbol_with_dem_name_to_list. > > > > No testsuite changes with DWARF2 or stabs+ on i386-linux. OK? > > > > NO. There are way too many unrelated changes in this patch. > > The changes to hpread to use SYMBOL_SET_NAMES should be > independent. However you are also changing the logic of the code in a > few places, not just mechanically replacing things. You are changing > printf's, deleting functions, etc etc. You say there are no regression > on i386-linux, did you test on hpux? No. Fancy that, that's why I asked for someone to do so. I don't have any way to test on HP/UX; the last time I tried to set up a test environment on the testdrive systems it took me a week and a half to get a baseline. The last time I tried to use it they'd upgraded libraries and none of my binaries worked. > The objfiles and symfile changes are also logically separate. > The SYMBOL_SET_LINKAGE_NAME macro can also be a separate thing. > > I shouldn't spend time reminding you of this. I fail to see how the symfile change is independent of anything, since the function is still used before my patch. I'll break the patch in thirds with splitdiff and test the thirds. -- Daniel Jacobowitz MontaVista Software Debian GNU/Linux Developer