From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 84109 invoked by alias); 12 Feb 2020 22:48:05 -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 84074 invoked by uid 89); 12 Feb 2020 22:48:04 -0000 Authentication-Results: sourceware.org; auth=none X-Spam-SWARE-Status: No, score=-5.3 required=5.0 tests=AWL,BAYES_00,SPF_HELO_PASS,SPF_PASS autolearn=ham version=3.3.1 spammy= X-HELO: simark.ca Received: from simark.ca (HELO simark.ca) (158.69.221.121) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Wed, 12 Feb 2020 22:48:02 +0000 Received: from [172.16.0.95] (192-222-181-218.qc.cable.ebox.net [192.222.181.218]) (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 BF2E01E10B; Wed, 12 Feb 2020 17:48:00 -0500 (EST) Subject: Re: [PATCH 5/8] Add per-unit obstack To: Tom Tromey Cc: gdb-patches@sourceware.org References: <20200208152758.29385-1-tom@tromey.com> <20200208152758.29385-6-tom@tromey.com> <8e13cd77-cbe8-4549-5ff3-2cb5fbe32b47@simark.ca> <87y2t7qvhu.fsf@tromey.com> From: Simon Marchi Message-ID: <29b6625c-b493-9f5e-2184-177bb951c343@simark.ca> Date: Wed, 12 Feb 2020 22:48:00 -0000 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.4.1 MIME-Version: 1.0 In-Reply-To: <87y2t7qvhu.fsf@tromey.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-SW-Source: 2020-02/txt/msg00472.txt.bz2 On 2020-02-12 5:41 p.m., Tom Tromey wrote: >>>>>> "Simon" == Simon Marchi writes: > > Simon> Would it be simpler to eliminate the obstack by making dwarf2_cie_table > Simon> and dwarf2_fde_table hold objects instead of pointers? On top of that, > Simon> it should be good for performance as well, at least when doing lookups. > > Simon> The only issue is that dwarf2_cie objects are referenced by address by > Simon> the FDEs, so we must make sure they don't get moved by the container. > Simon> But I believe std::unordered_map guarantees that (disabling copy and > Simon> assign on dwarf2_cie would verify it). Or worst case, the it could > Simon> be an std::unordered_map of std::unique_ptr. > > I think I would rather defer this change, like the other one, since it's > not directly related to the purpose of the series. > > Tom > No problem, I suggested this in case it actually made things simpler, but I am fine with what you have. Simon