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=-14.6 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI, MENTIONS_GIT_HOSTING,NICE_REPLY_A,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=ham 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 06E27C433DB for ; Fri, 5 Feb 2021 22:56:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id B02DD64E02 for ; Fri, 5 Feb 2021 22:56:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231788AbhBEW41 (ORCPT ); Fri, 5 Feb 2021 17:56:27 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:60178 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231921AbhBEMul (ORCPT ); Fri, 5 Feb 2021 07:50:41 -0500 Received: from mail-wr1-x42c.google.com (mail-wr1-x42c.google.com [IPv6:2a00:1450:4864:20::42c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id E31B0C061786 for ; Fri, 5 Feb 2021 04:49:57 -0800 (PST) Received: by mail-wr1-x42c.google.com with SMTP id v15so7569355wrx.4 for ; Fri, 05 Feb 2021 04:49:57 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=to:cc:references:from:autocrypt:subject:message-id:date:user-agent :mime-version:in-reply-to:content-language:content-transfer-encoding; bh=3DJdX6bRVV97dMnba7dy5xmVdJfTEwvUkprfGzXoEhQ=; b=uMRWm0o6PhPUGcNzAbYsC1P6GxPicDLmxGPH8xmAHuayGi7WdAXwvqfmhqJdT6/QIg GqJD2AZ8VMKm+7HfiGK1f1pt6fb204AktgpQwlr7+SA1wG85vPPjFzDqlQ+Z1SY/dbFH OLj95F/sAw4QS516YstK2B9UOcmx7Cur2g6CMebpU91wvlI+4hCNBts/ikuZleC0MS/v Oax/UHf47g+6LTi7+IlVV6pfeo8cg6qEBwCyZh2C4gDQibp1Q/FHjiTMF/hwTUHvhXtA 163JOeGTM1gkSe3DAFH2zSxArsK52/7jkae55ybnIAwzVu22D21yDu4q+RT5odgif0tI +mfw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:to:cc:references:from:autocrypt:subject :message-id:date:user-agent:mime-version:in-reply-to :content-language:content-transfer-encoding; bh=3DJdX6bRVV97dMnba7dy5xmVdJfTEwvUkprfGzXoEhQ=; b=smzg4c7NTtknXWlXTZSSE6AVITjGk3cV32LYcIaiU2WOJFy9Mwdv5FVujreNkvKQis BGSzKix3j+n79AYz+/vuJoJ0Sb6lLNyBPrnI66YPFS6r4z6kkhxv4cU5w/gHOjBVIhdC YUPx31Q1BowB5BuBmw65iXVEKNzRM2SlnkbUvONEvohccbko3QHzTVuVA2mTPnEwzTNe 4cp1xa7brKqnV/+Qyot6xwzPLbizyeNaBhmUfW6k3NXLMLGg36M77/eCMi/SWr7HgcgA HIR6qfsdKHMtdo9g+1zsZxU5ZC93+OPHu8DRy4yZ0G9oelIRhC7tSA7srXojzueCkBxq WhrQ== X-Gm-Message-State: AOAM533caAa5Nez6jsOOgxj7P8HMJ5+6oIF1vYGpOg2JOsMiSZseQW3Q OgOdlG2wT1Vmr96brkSQOj61oMBFJqqVcQ== X-Google-Smtp-Source: ABdhPJzg0i6SjDP/BSYjKy3fTOPoF3Pk+I/95S1Eoj3f1MLVRMfkw05Gk6AvsuOd9ZY+oLjDuLtwiw== X-Received: by 2002:a05:6000:2aa:: with SMTP id l10mr4873649wry.368.1612529396392; Fri, 05 Feb 2021 04:49:56 -0800 (PST) Received: from [192.168.8.177] ([85.255.234.187]) by smtp.gmail.com with ESMTPSA id r16sm11881509wrt.68.2021.02.05.04.49.54 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Fri, 05 Feb 2021 04:49:55 -0800 (PST) To: Jens Axboe , Victor Stewart Cc: io-uring References: <49743b61-3777-f152-e1d5-128a53803bcd@gmail.com> <4b44f4e1-c039-a6b6-711f-22952ce1abfb@kernel.dk> <06ceae30-7221-80e9-13e3-148cdf5e3c9f@kernel.dk> <8d75bf78-7361-0649-e5a3-1288fea1197f@gmail.com> <725fa06a-da7e-9918-49b4-7489672ff0b4@kernel.dk> <5c3d084f-88e4-3e86-3560-95d90bb9ffcd@gmail.com> <39bc0ff3-db02-8fc7-da5c-b2f5f0fc715e@gmail.com> From: Pavel Begunkov Autocrypt: addr=asml.silence@gmail.com; prefer-encrypt=mutual; keydata= mQINBFmKBOQBEAC76ZFxLAKpDw0bKQ8CEiYJRGn8MHTUhURL02/7n1t0HkKQx2K1fCXClbps bdwSHrhOWdW61pmfMbDYbTj6ZvGRvhoLWfGkzujB2wjNcbNTXIoOzJEGISHaPf6E2IQx1ik9 6uqVkK1OMb7qRvKH0i7HYP4WJzYbEWVyLiAxUj611mC9tgd73oqZ2pLYzGTqF2j6a/obaqha +hXuWTvpDQXqcOZJXIW43atprH03G1tQs7VwR21Q1eq6Yvy2ESLdc38EqCszBfQRMmKy+cfp W3U9Mb1w0L680pXrONcnlDBCN7/sghGeMHjGKfNANjPc+0hzz3rApPxpoE7HC1uRiwC4et83 CKnncH1l7zgeBT9Oa3qEiBlaa1ZCBqrA4dY+z5fWJYjMpwI1SNp37RtF8fKXbKQg+JuUjAa9 Y6oXeyEvDHMyJYMcinl6xCqCBAXPHnHmawkMMgjr3BBRzODmMr+CPVvnYe7BFYfoajzqzq+h EyXSl3aBf0IDPTqSUrhbmjj5OEOYgRW5p+mdYtY1cXeK8copmd+fd/eTkghok5li58AojCba jRjp7zVOLOjDlpxxiKhuFmpV4yWNh5JJaTbwCRSd04sCcDNlJj+TehTr+o1QiORzc2t+N5iJ NbILft19Izdn8U39T5oWiynqa1qCLgbuFtnYx1HlUq/HvAm+kwARAQABtDFQYXZlbCBCZWd1 bmtvdiAoc2lsZW5jZSkgPGFzbWwuc2lsZW5jZUBnbWFpbC5jb20+iQJOBBMBCAA4FiEE+6Ju PTjTbx479o3OWt5b1Glr+6UFAlmKBOQCGwMFCwkIBwIGFQgJCgsCBBYCAwECHgECF4AACgkQ Wt5b1Glr+6WxZA//QueaKHzgdnOikJ7NA/Vq8FmhRlwgtP0+E+w93kL+ZGLzS/cUCIjn2f4Q Mcutj2Neg0CcYPX3b2nJiKr5Vn0rjJ/suiaOa1h1KzyNTOmxnsqE5fmxOf6C6x+NKE18I5Jy xzLQoktbdDVA7JfB1itt6iWSNoOTVcvFyvfe5ggy6FSCcP+m1RlR58XxVLH+qlAvxxOeEr/e aQfUzrs7gqdSd9zQGEZo0jtuBiB7k98t9y0oC9Jz0PJdvaj1NZUgtXG9pEtww3LdeXP/TkFl HBSxVflzeoFaj4UAuy8+uve7ya/ECNCc8kk0VYaEjoVrzJcYdKP583iRhOLlZA6HEmn/+Gh9 4orG67HNiJlbFiW3whxGizWsrtFNLsSP1YrEReYk9j1SoUHHzsu+ZtNfKuHIhK0sU07G1OPN 2rDLlzUWR9Jc22INAkhVHOogOcc5ajMGhgWcBJMLCoi219HlX69LIDu3Y34uIg9QPZIC2jwr 24W0kxmK6avJr7+n4o8m6sOJvhlumSp5TSNhRiKvAHB1I2JB8Q1yZCIPzx+w1ALxuoWiCdwV M/azguU42R17IuBzK0S3hPjXpEi2sK/k4pEPnHVUv9Cu09HCNnd6BRfFGjo8M9kZvw360gC1 reeMdqGjwQ68o9x0R7NBRrtUOh48TDLXCANAg97wjPoy37dQE7e5Ag0EWYoE5AEQAMWS+aBV IJtCjwtfCOV98NamFpDEjBMrCAfLm7wZlmXy5I6o7nzzCxEw06P2rhzp1hIqkaab1kHySU7g dkpjmQ7Jjlrf6KdMP87mC/Hx4+zgVCkTQCKkIxNE76Ff3O9uTvkWCspSh9J0qPYyCaVta2D1 Sq5HZ8WFcap71iVO1f2/FEHKJNz/YTSOS/W7dxJdXl2eoj3gYX2UZNfoaVv8OXKaWslZlgqN jSg9wsTv1K73AnQKt4fFhscN9YFxhtgD/SQuOldE5Ws4UlJoaFX/yCoJL3ky2kC0WFngzwRF Yo6u/KON/o28yyP+alYRMBrN0Dm60FuVSIFafSqXoJTIjSZ6olbEoT0u17Rag8BxnxryMrgR dkccq272MaSS0eOC9K2rtvxzddohRFPcy/8bkX+t2iukTDz75KSTKO+chce62Xxdg62dpkZX xK+HeDCZ7gRNZvAbDETr6XI63hPKi891GeZqvqQVYR8e+V2725w+H1iv3THiB1tx4L2bXZDI DtMKQ5D2RvCHNdPNcZeldEoJwKoA60yg6tuUquvsLvfCwtrmVI2rL2djYxRfGNmFMrUDN1Xq F3xozA91q3iZd9OYi9G+M/OA01husBdcIzj1hu0aL+MGg4Gqk6XwjoSxVd4YT41kTU7Kk+/I 5/Nf+i88ULt6HanBYcY/+Daeo/XFABEBAAGJAjYEGAEIACAWIQT7om49ONNvHjv2jc5a3lvU aWv7pQUCWYoE5AIbDAAKCRBa3lvUaWv7pfmcEACKTRQ28b1y5ztKuLdLr79+T+LwZKHjX++P 4wKjEOECCcB6KCv3hP+J2GCXDOPZvdg/ZYZafqP68Yy8AZqkfa4qPYHmIdpODtRzZSL48kM8 LRzV8Rl7J3ItvzdBRxf4T/Zseu5U6ELiQdCUkPGsJcPIJkgPjO2ROG/ZtYa9DvnShNWPlp+R uPwPccEQPWO/NP4fJl2zwC6byjljZhW5kxYswGMLBwb5cDUZAisIukyAa8Xshdan6C2RZcNs rB3L7vsg/R8UCehxOH0C+NypG2GqjVejNZsc7bgV49EOVltS+GmGyY+moIzxsuLmT93rqyII 5rSbbcTLe6KBYcs24XEoo49Zm9oDA3jYvNpeYD8rDcnNbuZh9kTgBwFN41JHOPv0W2FEEWqe JsCwQdcOQ56rtezdCJUYmRAt3BsfjN3Jn3N6rpodi4Dkdli8HylM5iq4ooeb5VkQ7UZxbCWt UVMKkOCdFhutRmYp0mbv2e87IK4erwNHQRkHUkzbsuym8RVpAZbLzLPIYK/J3RTErL6Z99N2 m3J6pjwSJY/zNwuFPs9zGEnRO4g0BUbwGdbuvDzaq6/3OJLKohr5eLXNU3JkT+3HezydWm3W OPhauth7W0db74Qd49HXK0xe/aPrK+Cp+kU1HRactyNtF8jZQbhMCC8vMGukZtWaAwpjWiiH bA== Subject: Re: bug with fastpoll accept and sqpoll + IOSQE_FIXED_FILE Message-ID: Date: Fri, 5 Feb 2021 12:46:11 +0000 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.3.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org On 04/02/2021 16:50, Jens Axboe wrote: > On 2/3/21 4:49 AM, Pavel Begunkov wrote: >> On 02/02/2021 20:56, Pavel Begunkov wrote: >>> On 02/02/2021 20:48, Jens Axboe wrote: >>>> On 2/2/21 1:34 PM, Pavel Begunkov wrote: >>>>> On 02/02/2021 17:41, Pavel Begunkov wrote: >>>>>> On 02/02/2021 17:24, Jens Axboe wrote: >>>>>>> On 2/2/21 10:10 AM, Victor Stewart wrote: >>>>>>>>> Can you send the updated test app? >>>>>>>> >>>>>>>> https://gist.github.com/victorstewart/98814b65ed702c33480487c05b40eb56 >>>>>>>> >>>>>>>> same link i just updated the same gist >>>>>>> >>>>>>> And how are you running it? >>>>>> >>>>>> with SQPOLL with FIXED FLAG -> FAILURE: failed with error = ??? >>>>>> -> io_uring_wait_cqe_timeout() strangely returns -1, (-EPERM??) >>>>> >>>>> Ok, _io_uring_get_cqe() is just screwed twice >>>>> >>>>> TL;DR >>>>> we enter into it with submit=0, do an iteration, which decrements it, >>>>> then a second iteration passes submit=-1, which is returned back by >>>>> the kernel as a result and propagated back from liburing... >>>> >>>> Yep, that's what I came up with too. We really just need a clear way >>>> of knowing when to break out, and when to keep going. Eg if we've >>>> done a loop and don't end up calling the system call, then there's >>>> no point in continuing. >>> >>> We can bodge something up (and forget about it), and do much cleaner >>> for IORING_FEAT_EXT_ARG, because we don't have LIBURING_UDATA_TIMEOUT >>> reqs for it and so can remove peek and so on. >> >> This version looks reasonably simple, and even passes tests and all >> issues found by Victor's test. Didn't test it yet, but should behave >> similarly in regard of internal timeouts (pre IORING_FEAT_EXT_ARG). >> >> static int _io_uring_get_cqe(struct io_uring *ring, struct io_uring_cqe **cqe_ptr, >> struct get_data *data) >> { >> struct io_uring_cqe *cqe = NULL; >> int ret = 0, err; >> >> do { >> unsigned flags = 0; >> unsigned nr_available; >> bool enter = false; >> >> err = __io_uring_peek_cqe(ring, &cqe, &nr_available); >> if (err) >> break; >> >> /* IOPOLL won't proceed when there're not reaped CQEs */ >> if (cqe && (ring->flags & IORING_SETUP_IOPOLL)) >> data->wait_nr = 0; >> >> if (data->wait_nr > nr_available || cq_ring_needs_flush(ring)) { >> flags = IORING_ENTER_GETEVENTS | data->get_flags; >> enter = true; >> } >> if (data->submit) { >> sq_ring_needs_enter(ring, &flags); >> enter = true; >> } >> if (!enter) >> break; >> >> ret = __sys_io_uring_enter2(ring->ring_fd, data->submit, >> data->wait_nr, flags, data->arg, >> data->sz); >> if (ret < 0) { >> err = -errno; >> break; >> } >> data->submit -= ret; >> } while (1); >> >> *cqe_ptr = cqe; >> return err; >> } > > So here's my take on this - any rewrite of _io_uring_get_cqe() is going > to end up adding special cases, that's unfortunately just the nature of > the game. And since we're going to be doing a new liburing release very > shortly, this isn't a great time to add a rewrite of it. It'll certainly > introduce more bugs than it solves, and hence regressions, no matter how > careful we are. > > Hence my suggestion is to just patch this in a trivial kind of fashion, > even if it doesn't really make the function any prettier. But it'll be > safer for a release, and then we can rework the function after. > > With that in mind, here's my suggestion. The premise is if we go through > the loop and don't do io_uring_enter(), then there's no point in > continuing. That's the trivial fix. Your idea but imho cleaner below. +1 comment inline > diff --git a/src/queue.c b/src/queue.c > index 94f791e..4161aa7 100644 > --- a/src/queue.c > +++ b/src/queue.c > @@ -89,12 +89,13 @@ static int _io_uring_get_cqe(struct io_uring *ring, struct io_uring_cqe **cqe_pt > { > struct io_uring_cqe *cqe = NULL; > const int to_wait = data->wait_nr; > - int ret = 0, err; > + int err; > > do { > bool cq_overflow_flush = false; > unsigned flags = 0; > unsigned nr_available; > + int ret = -2; > > err = __io_uring_peek_cqe(ring, &cqe, &nr_available); > if (err) > @@ -117,7 +118,9 @@ static int _io_uring_get_cqe(struct io_uring *ring, struct io_uring_cqe **cqe_pt > ret = __sys_io_uring_enter2(ring->ring_fd, data->submit, > data->wait_nr, flags, data->arg, > data->sz); > - if (ret < 0) { > + if (ret == -2) { > + break; peek/wait_cqe expect that cqe_ptr is filled on return=0. Looks we need to return an error or hack up those functions. > + } else if (ret < 0) { > err = -errno; > } else if (ret == (int)data->submit) { > data->submit = 0; > diff --git a/src/queue.c b/src/queue.c index 94f791e..7d6f31d 100644 --- a/src/queue.c +++ b/src/queue.c @@ -112,11 +112,15 @@ static int _io_uring_get_cqe(struct io_uring *ring, struct io_uring_cqe **cqe_pt flags = IORING_ENTER_GETEVENTS | data->get_flags; if (data->submit) sq_ring_needs_enter(ring, &flags); - if (data->wait_nr > nr_available || data->submit || - cq_overflow_flush) - ret = __sys_io_uring_enter2(ring->ring_fd, data->submit, - data->wait_nr, flags, data->arg, - data->sz); + + if (data->wait_nr <= nr_available && !data->submit && + !cq_overflow_flush) { + err = ?; + break; + } + ret = __sys_io_uring_enter2(ring->ring_fd, data->submit, + data->wait_nr, flags, data->arg, + data->sz); if (ret < 0) { err = -errno; } else if (ret == (int)data->submit) { -- Pavel Begunkov