From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 20855 invoked by alias); 15 Dec 2004 21:48:47 -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 20757 invoked from network); 15 Dec 2004 21:48:35 -0000 Received: from unknown (HELO arwen.tausq.org) (64.81.244.109) by sourceware.org with SMTP; 15 Dec 2004 21:48:35 -0000 Received: by arwen.tausq.org (Postfix, from userid 1000) id C6E446BE7F; Wed, 15 Dec 2004 13:48:33 -0800 (PST) Date: Wed, 15 Dec 2004 22:46:00 -0000 From: Randolph Chung To: gdb-patches@sources.redhat.com Subject: Re: [rfc/rfa] Dummy calls for hppa-hpux Message-ID: <20041215214833.GF29171@tausq.org> Reply-To: Randolph Chung References: <20041213172952.GO29171@tausq.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20041213172952.GO29171@tausq.org> X-GPG: for GPG key, see http://www.tausq.org/gpg.txt User-Agent: Mutt/1.5.6+20040722i X-SW-Source: 2004-12/txt/msg00390.txt.bz2 > hppa64-hp-hpux11.11 > # of expected passes 35 > # of unexpected failures 64 > # of known failures 3 > > the large number of failures in hppa64-hp-hpux11.11 are due to a > cascading failure early on in the test (we have some problems passing > function pointers into functions); however, since basic infcalls work > now, i suspect a much larger portion of gdb testcases will benefit. i've looked into these failures a bit further; hppa64 requires indirect function calls to be made via function descriptors (on hppa32, it can be made via either a function descriptor or a "straight" function pointer). when doing a call like "p t_func_values (func_val1, add)" where "func_val1" is a function pointer variable, and "add" is a local function, gdb seems to be passing a function descriptor value for "func_val1", but a regular pointer for "add". When t_func_values tries to call the corresponding functions, it fails because it tries to dereference the function pointer as a function descriptor. I haven't been able to determine where and how gdb resolves the value of "add" to pass to call_functions_by_hand (). any hints would be appreciated. btw, for this hpux dummy calls patch, if i don't get any other comments i'll check it in tomorrow. :) randolph -- Randolph Chung Debian GNU/Linux Developer, hppa/ia64 ports http://www.tausq.org/