From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtpng3.m.smailru.net (smtpng3.m.smailru.net [94.100.177.149]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dev.tarantool.org (Postfix) with ESMTPS id ADD65445320 for ; Wed, 15 Jul 2020 15:25:51 +0300 (MSK) References: <1594221263-6228-1-git-send-email-alyapunov@tarantool.org> From: Aleksandr Lyapunov Message-ID: <6cf593e8-e5e5-d837-28a7-bc36f67c8203@tarantool.org> Date: Wed, 15 Jul 2020 15:25:50 +0300 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Content-Language: en-US Subject: Re: [Tarantool-patches] [PATCH v2 00/16] Transaction engine for memtx engine List-Id: Tarantool development patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Vladislav Shpilevoy , tarantool-patches@dev.tarantool.org I hope it will be enable by default.. Let's leave it for performance part. As for me I don't cate about memory but initialization could be significantly longer. I think I can refactor and improve that part. On 15.07.2020 02:47, Vladislav Shpilevoy wrote: > A lot of new expenses come from struct txn becoming huge. > It could be mitigated if you would introduce a new object > which would store struct txn, psn, all the new rlists. And > which would be created only when needed. However it may be > wrong if you will enable the new manager by default with > keeping on_yield error. > > I sent some comments, but I didn't review the patchset whole > yet, it is huge.