From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from simark.ca (simark.ca [158.69.221.121]) by sourceware.org (Postfix) with ESMTPS id A58DD385782A for ; Sat, 19 Sep 2020 23:52:34 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org A58DD385782A Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=simark.ca Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=simark@simark.ca Received: from [10.0.0.11] (173-246-6-90.qc.cable.ebox.net [173.246.6.90]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by simark.ca (Postfix) with ESMTPSA id 2FA4F1E599; Sat, 19 Sep 2020 19:52:34 -0400 (EDT) Subject: Re: [PATCH 1/6] Remove some manual memory management from compile interface To: Tom Tromey Cc: gdb-patches@sourceware.org References: <20200809135258.8207-1-tom@tromey.com> <20200809135258.8207-2-tom@tromey.com> <6824ff00-ef2a-c1dc-8717-75111d3fd8b0@simark.ca> <87pn6h5o1s.fsf@tromey.com> From: Simon Marchi Message-ID: <266736d8-415c-b18e-507f-c8e5fbe84231@simark.ca> Date: Sat, 19 Sep 2020 19:52:33 -0400 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.12.0 MIME-Version: 1.0 In-Reply-To: <87pn6h5o1s.fsf@tromey.com> Content-Type: text/plain; charset=utf-8 Content-Language: fr Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-4.8 required=5.0 tests=BAYES_00, KAM_DMARC_STATUS, NICE_REPLY_A, SPF_HELO_PASS, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org X-BeenThere: gdb-patches@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gdb-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 19 Sep 2020 23:52:35 -0000 On 2020-09-19 7:45 p.m., Tom Tromey wrote: > Simon> You could now remove this `if (num_baseclasses > 0)`, there's no > Simon> reason for it anymore I think. > > I looked at this more deeply, and based on the compiler library, I think > the rule is that a 0-length array should be passed as nullptr here: > > status > marshall (connection *conn, const gcc_vbase_array *a) > { > size_t len; > > if (a) > len = a->n_elements; > else > len = (size_t)-1; > > if (!marshall_array_start (conn, 'v', len)) > return FAIL; > > if (!a) > return OK; > [...] > > > So, I changed this code to use your suggestion and then also modified > the final call: > > result = instance->plugin ().start_class_type > (name.get (), resuld, num_baseclasses > 0 ? &bases : nullptr, > filename, line); > > This didn't result in any test result changes, but I didn't look to see > if this case is not tested or something like that. Oh, I didn't realize that the change I proposed would change how we call the function. What you ended up doing makes sense, if that's what's the compiler library expects. Simon