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=-6.0 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SIGNED_OFF_BY, 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 17653C10F11 for ; Wed, 24 Apr 2019 17:43:36 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id D100A2064A for ; Wed, 24 Apr 2019 17:43:35 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1556127815; bh=Np8TE3AqgGnnNFegyow15CxHEu4a2ScslFafqJmnklY=; h=From:To:Cc:Subject:Date:In-Reply-To:References:List-ID:From; b=Xmf0v+4jZteHfXj6EYas0dgjbDNrQwE/E6B9sdPbSTguJKBbObWE1r2pUDAtVEQ5s aEiKZO1Tzv2hm//yDRs1ZDQ+RR5FJRNYRyKXls8hv7wirgJdHLeTvl/+g1/TsfSLnx w0iZLYpRaN6BLbFqdE8LvJ/4j9AHNhCH8czGQhGI= Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2404102AbfDXRne (ORCPT ); Wed, 24 Apr 2019 13:43:34 -0400 Received: from mail.kernel.org ([198.145.29.99]:34090 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2391849AbfDXRfP (ORCPT ); Wed, 24 Apr 2019 13:35:15 -0400 Received: from localhost (62-193-50-229.as16211.net [62.193.50.229]) (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 98B0C21903; Wed, 24 Apr 2019 17:35:13 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1556127314; bh=Np8TE3AqgGnnNFegyow15CxHEu4a2ScslFafqJmnklY=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=f3xbtCtzLEjDCgqFTVUW4f1KF474EP4d69dyhfQVXb/nTlEXV3+kqXOhQ+TNLr7gP 8LLPuRUttyKQmzAgWk1Xf4RN3TrHA7BtPITzB0Nys7syrQaA8v/bf5JaHVwbg9zLv/ j+b1+SJBqdEw6S33hGAcD8t2/t60Cs8cO7QK+NrY= From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Aurelien Aptel , Steve French , Pavel Shilovsky Subject: [PATCH 5.0 044/115] CIFS: keep FileInfo handle live during oplock break Date: Wed, 24 Apr 2019 19:09:40 +0200 Message-Id: <20190424170927.609641495@linuxfoundation.org> X-Mailer: git-send-email 2.21.0 In-Reply-To: <20190424170924.797924502@linuxfoundation.org> References: <20190424170924.797924502@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Aurelien Aptel commit b98749cac4a695f084a5ff076f4510b23e353ecd upstream. In the oplock break handler, writing pending changes from pages puts the FileInfo handle. If the refcount reaches zero it closes the handle and waits for any oplock break handler to return, thus causing a deadlock. To prevent this situation: * We add a wait flag to cifsFileInfo_put() to decide whether we should wait for running/pending oplock break handlers * We keep an additionnal reference of the SMB FileInfo handle so that for the rest of the handler putting the handle won't close it. - The ref is bumped everytime we queue the handler via the cifs_queue_oplock_break() helper. - The ref is decremented at the end of the handler This bug was triggered by xfstest 464. Also important fix to address the various reports of oops in smb2_push_mandatory_locks Signed-off-by: Aurelien Aptel Signed-off-by: Steve French Reviewed-by: Pavel Shilovsky CC: Stable Signed-off-by: Greg Kroah-Hartman --- fs/cifs/cifsglob.h | 2 ++ fs/cifs/file.c | 30 +++++++++++++++++++++++++----- fs/cifs/misc.c | 25 +++++++++++++++++++++++-- fs/cifs/smb2misc.c | 6 +++--- 4 files changed, 53 insertions(+), 10 deletions(-) --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -1303,6 +1303,7 @@ cifsFileInfo_get_locked(struct cifsFileI } struct cifsFileInfo *cifsFileInfo_get(struct cifsFileInfo *cifs_file); +void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_hdlr); void cifsFileInfo_put(struct cifsFileInfo *cifs_file); #define CIFS_CACHE_READ_FLG 1 @@ -1824,6 +1825,7 @@ GLOBAL_EXTERN spinlock_t gidsidlock; #endif /* CONFIG_CIFS_ACL */ void cifs_oplock_break(struct work_struct *work); +void cifs_queue_oplock_break(struct cifsFileInfo *cfile); extern const struct slow_work_ops cifs_oplock_break_ops; extern struct workqueue_struct *cifsiod_wq; --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -360,13 +360,31 @@ cifsFileInfo_get(struct cifsFileInfo *ci return cifs_file; } -/* - * Release a reference on the file private data. This may involve closing - * the filehandle out on the server. Must be called without holding - * tcon->open_file_lock and cifs_file->file_info_lock. +/** + * cifsFileInfo_put - release a reference of file priv data + * + * Always potentially wait for oplock handler. See _cifsFileInfo_put(). */ void cifsFileInfo_put(struct cifsFileInfo *cifs_file) { + _cifsFileInfo_put(cifs_file, true); +} + +/** + * _cifsFileInfo_put - release a reference of file priv data + * + * This may involve closing the filehandle @cifs_file out on the + * server. Must be called without holding tcon->open_file_lock and + * cifs_file->file_info_lock. + * + * If @wait_for_oplock_handler is true and we are releasing the last + * reference, wait for any running oplock break handler of the file + * and cancel any pending one. If calling this function from the + * oplock break handler, you need to pass false. + * + */ +void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_handler) +{ struct inode *inode = d_inode(cifs_file->dentry); struct cifs_tcon *tcon = tlink_tcon(cifs_file->tlink); struct TCP_Server_Info *server = tcon->ses->server; @@ -414,7 +432,8 @@ void cifsFileInfo_put(struct cifsFileInf spin_unlock(&tcon->open_file_lock); - oplock_break_cancelled = cancel_work_sync(&cifs_file->oplock_break); + oplock_break_cancelled = wait_oplock_handler ? + cancel_work_sync(&cifs_file->oplock_break) : false; if (!tcon->need_reconnect && !cifs_file->invalidHandle) { struct TCP_Server_Info *server = tcon->ses->server; @@ -4480,6 +4499,7 @@ void cifs_oplock_break(struct work_struc cinode); cifs_dbg(FYI, "Oplock release rc = %d\n", rc); } + _cifsFileInfo_put(cfile, false /* do not wait for ourself */); cifs_done_oplock_break(cinode); } --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -501,8 +501,7 @@ is_valid_oplock_break(char *buffer, stru CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &pCifsInode->flags); - queue_work(cifsoplockd_wq, - &netfile->oplock_break); + cifs_queue_oplock_break(netfile); netfile->oplock_break_cancelled = false; spin_unlock(&tcon->open_file_lock); @@ -607,6 +606,28 @@ void cifs_put_writer(struct cifsInodeInf spin_unlock(&cinode->writers_lock); } +/** + * cifs_queue_oplock_break - queue the oplock break handler for cfile + * + * This function is called from the demultiplex thread when it + * receives an oplock break for @cfile. + * + * Assumes the tcon->open_file_lock is held. + * Assumes cfile->file_info_lock is NOT held. + */ +void cifs_queue_oplock_break(struct cifsFileInfo *cfile) +{ + /* + * Bump the handle refcount now while we hold the + * open_file_lock to enforce the validity of it for the oplock + * break handler. The matching put is done at the end of the + * handler. + */ + cifsFileInfo_get(cfile); + + queue_work(cifsoplockd_wq, &cfile->oplock_break); +} + void cifs_done_oplock_break(struct cifsInodeInfo *cinode) { clear_bit(CIFS_INODE_PENDING_OPLOCK_BREAK, &cinode->flags); --- a/fs/cifs/smb2misc.c +++ b/fs/cifs/smb2misc.c @@ -555,7 +555,7 @@ smb2_tcon_has_lease(struct cifs_tcon *tc clear_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &cinode->flags); - queue_work(cifsoplockd_wq, &cfile->oplock_break); + cifs_queue_oplock_break(cfile); kfree(lw); return true; } @@ -719,8 +719,8 @@ smb2_is_valid_oplock_break(char *buffer, CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &cinode->flags); spin_unlock(&cfile->file_info_lock); - queue_work(cifsoplockd_wq, - &cfile->oplock_break); + + cifs_queue_oplock_break(cfile); spin_unlock(&tcon->open_file_lock); spin_unlock(&cifs_tcp_ses_lock);