From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.2 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id BB795C35247 for ; Thu, 6 Feb 2020 20:58:54 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 8A36421741 for ; Thu, 6 Feb 2020 20:58:54 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20150623.gappssmtp.com header.i=@kernel-dk.20150623.gappssmtp.com header.b="B/G+VDjD" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727698AbgBFU6y (ORCPT ); Thu, 6 Feb 2020 15:58:54 -0500 Received: from mail-io1-f66.google.com ([209.85.166.66]:34622 "EHLO mail-io1-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727711AbgBFU6y (ORCPT ); Thu, 6 Feb 2020 15:58:54 -0500 Received: by mail-io1-f66.google.com with SMTP id z193so7828639iof.1 for ; Thu, 06 Feb 2020 12:58:53 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20150623.gappssmtp.com; s=20150623; h=subject:to:references:from:message-id:date:user-agent:mime-version :in-reply-to:content-language:content-transfer-encoding; bh=AdYB/qVsQCfjtK6YLT21IXUFdwDIjEDegRvK53B7PwA=; b=B/G+VDjDJpCN2ldsbpWbT42RDlAK1aKN/ANvJL9g9zPgjF8DL2GfhXzXLB8uc1UwiS eA2UEaT/9e5qEZXmrGe66LyBxqxfD5LafMnuCVJTeuQHanGIibi8Oy68+8tkrYhlBVV/ ADl5wWfjdOEqS6VLF5go16dYq32duVQVNrjHoF9oyXxPBAoMTgaNDCfbYpPadozI0lpe 33gOOkXws19I84A02ZRwcJ8eI/XUE8yBQg0z4KjUGqWCIItQKeOysELLKTQc+lJwpwNs NSqVv/uizyC/WXnUR+E+boLOwSM+GEE36sPTEXHN1FXyUZx97SJYTUTkIWopUKwFsKh3 CuWg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:subject:to:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=AdYB/qVsQCfjtK6YLT21IXUFdwDIjEDegRvK53B7PwA=; b=OpCSufMXvSo6BGPdk4MznlUaTymN16FceuwJ9p3TZAQCsOr47nTsD5qzLPvNv+gE9P bOEVDQaEkzGavPSBVLdgieTB2a/Tf0816ZHVr7qz6fH5hiA94R3VvX1l5ZJLMI0/5vMz vOOAXCsKGKoOgKRr2rFoIkztz5FNdhR+XBJnqvfU0hGsAo16/9/t6C0tQCe1gz5dFARP xADrIxZXoc7eFH6LyK/LMA/JINv74A2f7F/dVNMUGWmSPSIVSqc4vgEw4YIhNp1urOsl AP9vFBjNkUrCLH0BqM3yqiViEYsTI3xPCU7uE1hrX2F/hqrCU+M9AHnqmmp0pDFa8qeh K3rQ== X-Gm-Message-State: APjAAAUQG31hlpekNKy4G0o/627jux6yOF7Ln+Qxg9qyq8Xu6YWVeMhD cWBGj6qQhu5zZ5dAWiWmY3sJZxzBgDk= X-Google-Smtp-Source: APXvYqx6GqYXoq9y83++pyl2zcO3nfQ0K9n4lzEkbOCZZnD5TUxbNMS+BXy0qLcL8dl5/3QGqS8RzQ== X-Received: by 2002:a6b:5109:: with SMTP id f9mr126416iob.86.1581022733488; Thu, 06 Feb 2020 12:58:53 -0800 (PST) Received: from [192.168.1.159] ([65.144.74.34]) by smtp.gmail.com with ESMTPSA id e16sm210235iog.62.2020.02.06.12.58.52 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Thu, 06 Feb 2020 12:58:52 -0800 (PST) Subject: Re: [PATCH] io_uring: fix deferred req iovec leak To: Pavel Begunkov , io-uring@vger.kernel.org, linux-kernel@vger.kernel.org References: <6e7207b6-95c4-4287-5872-fb05abf60e88@kernel.dk> <4f7f61d3-b3f9-43db-ad32-ee502dc06c8b@gmail.com> <28cacc0c-68e4-a9d1-bb5e-03dbeff8a586@kernel.dk> <37dc06c1-e7ee-a185-43a7-98883709f5b0@gmail.com> From: Jens Axboe Message-ID: <2095dbc7-88fb-9d4f-78a5-8577dda09a92@kernel.dk> Date: Thu, 6 Feb 2020 13:58:51 -0700 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: <37dc06c1-e7ee-a185-43a7-98883709f5b0@gmail.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: io-uring-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org On 2/6/20 1:39 PM, Pavel Begunkov wrote: > On 06/02/2020 23:16, Jens Axboe wrote: >> On 2/6/20 1:00 PM, Pavel Begunkov wrote: >>> On 06/02/2020 22:56, Jens Axboe wrote: >>>> On 2/6/20 10:16 AM, Pavel Begunkov wrote: >>>>> On 06/02/2020 20:04, Pavel Begunkov wrote: >>>>>> On 06/02/2020 19:51, Pavel Begunkov wrote: >>>>>>> After defer, a request will be prepared, that includes allocating iovec >>>>>>> if needed, and then submitted through io_wq_submit_work() but not custom >>>>>>> handler (e.g. io_rw_async()/io_sendrecv_async()). However, it'll leak >>>>>>> iovec, as it's in io-wq and the code goes as follows: >>>>>>> >>>>>>> io_read() { >>>>>>> if (!io_wq_current_is_worker()) >>>>>>> kfree(iovec); >>>>>>> } >>>>>>> >>>>>>> Put all deallocation logic in io_{read,write,send,recv}(), which will >>>>>>> leave the memory, if going async with -EAGAIN. >>>>>>> >>>>>> Interestingly, this will fail badly if it returns -EAGAIN from io-wq context. >>>>>> Apparently, I need to do v2. >>>>>> >>>>> Or not... >>>>> Jens, can you please explain what's with the -EAGAIN handling in >>>>> io_wq_submit_work()? Checking the code, it seems neither of >>>>> read/write/recv/send can return -EAGAIN from async context (i.e. >>>>> force_nonblock=false). Are there other ops that can do it? >>>> >>>> Nobody should return -EAGAIN with force_nonblock=false, they should >>>> end the io_kiocb inline for that. >>>> >>> >>> If so for those 4, then the patch should work well. >> >> Maybe I'm dense, but I'm not seeing the leak? We have two cases here: >> > > There is an example: > > 1. submit a read, which need defer. > > 2. io_req_defer() allocates ->io and goes io_req_defer_prep() -> io_read_prep(). > Let #vecs > UIO_FASTIOV, so the prep() in the presence of ->io will allocate iovec. > Note: that work.func is left io_wq_submit_work > > 3. At some point @io_wq calls io_wq_submit_work() -> io_issue_sqe() -> io_read(), > > 4. actual reading succeeds, and it's coming to finalisation and the following > code in particular. > > if (!io_wq_current_is_worker()) > kfree(iovec); > > 5. Because we're in io_wq, the cleanup will not be performed, even though we're > returning with success. And that's a leak. > > Do you see anything wrong with it? That's my bad, I didn't read the subject fully, this is specific to a deferred request. Patch looks good to me, and it cleans it up too which is always a nice win! -- Jens Axboe