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=-18.7 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED, USER_AGENT_GIT autolearn=unavailable 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 A7038C83014 for ; Tue, 1 Dec 2020 09:20:28 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 5F71A2065C for ; Tue, 1 Dec 2020 09:20:28 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="aHEUdwUz" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2390742AbgLAJTc (ORCPT ); Tue, 1 Dec 2020 04:19:32 -0500 Received: from mail.kernel.org ([198.145.29.99]:47718 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2389644AbgLAJKS (ORCPT ); Tue, 1 Dec 2020 04:10:18 -0500 Received: from localhost (83-86-74-64.cable.dynamic.v4.ziggo.nl [83.86.74.64]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPSA id 0E0F9221FF; Tue, 1 Dec 2020 09:09:30 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1606813771; bh=IBKo153u6tCyF/G1i86IGoLG+Sjwvg5SFXRm9IGDGj0=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=aHEUdwUzWNRuYkhInCWDHOYhXji1sdh0McWHnUkULkTAkHMa76ZjQh2xO8qIWgUat Q1geRIH/+pDS9JoRSsjKH1TJJiUVD+SGMrEWAlEc2W6x1iESrjvT0sAAS2ClqEUemq DQCUqJw8657bLKWTURjLD+Yl19fnXbsVvi0bvHYc= From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Jens Axboe , Sasha Levin Subject: [PATCH 5.9 054/152] io_uring: handle -EOPNOTSUPP on path resolution Date: Tue, 1 Dec 2020 09:52:49 +0100 Message-Id: <20201201084719.013832675@linuxfoundation.org> X-Mailer: git-send-email 2.29.2 In-Reply-To: <20201201084711.707195422@linuxfoundation.org> References: <20201201084711.707195422@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: stable@vger.kernel.org From: Jens Axboe [ Upstream commit 944d1444d53f5a213457e5096db370cfd06923d4 ] Any attempt to do path resolution on /proc/self from an async worker will yield -EOPNOTSUPP. We can safely do that resolution from the task itself, and without blocking, so retry it from there. Ideally io_uring would know this upfront and not have to go through the worker thread to find out, but that doesn't currently seem feasible. Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin --- fs/io_uring.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 185c206e5683f..5d9f8e40b93d3 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -436,6 +436,7 @@ struct io_sr_msg { struct io_open { struct file *file; int dfd; + bool ignore_nonblock; struct filename *filename; struct open_how how; unsigned long nofile; @@ -3591,6 +3592,7 @@ static int __io_openat_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe return ret; } req->open.nofile = rlimit(RLIMIT_NOFILE); + req->open.ignore_nonblock = false; req->flags |= REQ_F_NEED_CLEANUP; return 0; } @@ -3638,7 +3640,7 @@ static int io_openat2(struct io_kiocb *req, bool force_nonblock) struct file *file; int ret; - if (force_nonblock) + if (force_nonblock && !req->open.ignore_nonblock) return -EAGAIN; ret = build_open_flags(&req->open.how, &op); @@ -3653,6 +3655,21 @@ static int io_openat2(struct io_kiocb *req, bool force_nonblock) if (IS_ERR(file)) { put_unused_fd(ret); ret = PTR_ERR(file); + /* + * A work-around to ensure that /proc/self works that way + * that it should - if we get -EOPNOTSUPP back, then assume + * that proc_self_get_link() failed us because we're in async + * context. We should be safe to retry this from the task + * itself with force_nonblock == false set, as it should not + * block on lookup. Would be nice to know this upfront and + * avoid the async dance, but doesn't seem feasible. + */ + if (ret == -EOPNOTSUPP && io_wq_current_is_worker()) { + req->open.ignore_nonblock = true; + refcount_inc(&req->refs); + io_req_task_queue(req); + return 0; + } } else { fsnotify_open(file); fd_install(ret, file); -- 2.27.0