Received: by 2002:a05:6358:3188:b0:123:57c1:9b43 with SMTP id q8csp37099841rwd; Tue, 11 Jul 2023 09:24:07 -0700 (PDT) X-Google-Smtp-Source: APBJJlHF6vlrdOtgAq7q68mLpsh4uqwlmeu5pGmo5GoG2KrU4wpeD+2hfouITQH6PmtfTtJl1M5C X-Received: by 2002:a05:6a00:a1f:b0:668:806d:c2d6 with SMTP id p31-20020a056a000a1f00b00668806dc2d6mr16907127pfh.12.1689092646957; Tue, 11 Jul 2023 09:24:06 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1689092646; cv=none; d=google.com; s=arc-20160816; b=oLJiCKFJeoGed5EjGDGVo8Joj1Q/NesIYUVCCeLntuJv0t4KOKoomDzOCqTqSoKr5v VGSjnOP57a5SbqJ/kzteZO4wl3VrchX4c9ehYG8mo8WXT8SERCtzbekQ7b5SBXsTl2EF SxnVU932cipa0+gyC6QzEoigYfrMXqT4O14xRTmfLMrXHbBwGBUQXUrE5ATthDepsb3Z OJXBXZWFnf+BcA/75ozvppX0NZGxAHpnhMzDT6d/KeFurJaHMuhiVobzXn3aqOe5KSsy /J2cC6wdjGoKR0aZCy9mjUpVNoqYryxPfaSFZPV9HkY1+Ex3cpixtcF00KmRImg5EE26 2NhQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:in-reply-to:content-disposition:mime-version :references:message-id:subject:cc:to:from:date:dkim-signature; bh=Xabd5+tKdxzACjm3r640WGDoM3dIXGfIhrKtcv2p6XU=; fh=gA6YOWNFp98UrO2u+0ZWcWDLjqDah2DYZcaKYoPGUag=; b=GB7xlAu686e/2BpeIIwHyB9iyN7sKwVdW7ncPEHiV/bUh57sp1ic8pDgrrTelKMOyY V2QBtRJk8GTIpWMilV1OrkBpCZiKIIcbvtqKJlV2o1QvJLSHEFN15pLQ11AgUej8B2yq len5kTAg0YPr+SyUbxQ4p8hSQPJfWPN7lXveEK3YMtl5fpJHf/HItunlgzn9ehLayhIE TDTCFD4A9E2dD8kEViSziZ0XLFmeXyFGXh9HB+P2nam1uhttCsY6mz3ykGdmKFf2d1b/ MlQmF7lodR1m9VsiIKy9Q48vVh1+tHqnMTZaep+LuTqNdV2HHW7xnOWHJj650Au1yeG3 661Q== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=rc8vn0uB; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Return-Path: Received: from out1.vger.email (out1.vger.email. [2620:137:e000::1:20]) by mx.google.com with ESMTP id r22-20020a63ce56000000b005503f266742si1609368pgi.647.2023.07.11.09.23.53; Tue, 11 Jul 2023 09:24:06 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20; Authentication-Results: mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=rc8vn0uB; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230419AbjGKQQt (ORCPT + 99 others); Tue, 11 Jul 2023 12:16:49 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51160 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229917AbjGKQQp (ORCPT ); Tue, 11 Jul 2023 12:16:45 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [IPv6:2604:1380:4641:c500::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 1ABE6DD for ; Tue, 11 Jul 2023 09:16:43 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id 9415C6152D for ; Tue, 11 Jul 2023 16:16:42 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id D884CC433C8; Tue, 11 Jul 2023 16:16:41 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1689092202; bh=rRZeIKQQb+qwxmnDhPPGlaH96szkShA7ICpYM/PVFi8=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=rc8vn0uB9tPlcuUwT5F8z9iXu2yjLff6cxv55rsFfQqgAPO1+3f+1w9VBYv3olzho YDwfEt9d3ky19mIVamGyNP+E8o3EPre00Ep35OHLtcBHxsxB/7069m2r/7ZesBekMy lHtUNe89yatns9oOXMd2DZZrIqxr6lg32G5kWq3Wqfhny2BJvYjLz0gAupiKbOZ/+i o52yfuUg5u9hIV6On71abHqtD8aYa1XEuf8dsEsz+9/8W2TmdJBFnFF4GlBcKgFe4J F5ssC8XIX9KFHeGRmDpH4Apm2WDDllBSTunPfRDPCiHNkFyOmo+p7VQwwSf9UqIY3s m1WyXJe7Q3lrg== Date: Tue, 11 Jul 2023 09:16:40 -0700 From: Jaegeuk Kim To: Chao Yu Cc: linux-f2fs-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org, syzbot+e5600587fa9cbf8e3826@syzkaller.appspotmail.com Subject: Re: [PATCH] f2fs: fix potential deadlock by reordering w/ i_sem Message-ID: References: <20230710061021.2303432-1-chao@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-Spam-Status: No, score=-4.4 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,RCVD_IN_DNSWL_MED, SPF_HELO_NONE,SPF_PASS,T_SCC_BODY_TEXT_LINE,URIBL_BLOCKED autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 07/11, Chao Yu wrote: > On 2023/7/11 1:18, Jaegeuk Kim wrote: > > On 07/10, Chao Yu wrote: > > > syzbot reports deadlock bug as below: > > > > > > -> #1 (&fi->i_sem){+.+.}-{3:3}: > > > down_write+0x3a/0x50 kernel/locking/rwsem.c:1573 > > > f2fs_down_write fs/f2fs/f2fs.h:2133 [inline] > > > f2fs_add_inline_entry+0x3a8/0x760 fs/f2fs/inline.c:644 > > > f2fs_add_dentry+0xba/0x1e0 fs/f2fs/dir.c:784 > > > f2fs_do_add_link+0x21e/0x340 fs/f2fs/dir.c:827 > > > f2fs_add_link fs/f2fs/f2fs.h:3554 [inline] > > > f2fs_create+0x32c/0x530 fs/f2fs/namei.c:377 > > > lookup_open fs/namei.c:3492 [inline] > > > open_last_lookups fs/namei.c:3560 [inline] > > > path_openat+0x13e7/0x3180 fs/namei.c:3790 > > > do_filp_open+0x234/0x490 fs/namei.c:3820 > > > do_sys_openat2+0x13e/0x1d0 fs/open.c:1407 > > > do_sys_open fs/open.c:1422 [inline] > > > __do_sys_open fs/open.c:1430 [inline] > > > __se_sys_open fs/open.c:1426 [inline] > > > __x64_sys_open+0x225/0x270 fs/open.c:1426 > > > do_syscall_x64 arch/x86/entry/common.c:50 [inline] > > > do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 > > > entry_SYSCALL_64_after_hwframe+0x63/0xcd > > > > > > -> #0 (&fi->i_xattr_sem){.+.+}-{3:3}: > > > check_prev_add kernel/locking/lockdep.c:3142 [inline] > > > check_prevs_add kernel/locking/lockdep.c:3261 [inline] > > > validate_chain kernel/locking/lockdep.c:3876 [inline] > > > __lock_acquire+0x39ff/0x7f70 kernel/locking/lockdep.c:5144 > > > lock_acquire+0x1e3/0x520 kernel/locking/lockdep.c:5761 > > > down_read+0x47/0x2f0 kernel/locking/rwsem.c:1520 > > > f2fs_down_read fs/f2fs/f2fs.h:2108 [inline] > > > f2fs_getxattr+0xb8/0x1460 fs/f2fs/xattr.c:532 > > > __f2fs_get_acl+0x52/0x8e0 fs/f2fs/acl.c:179 > > > f2fs_acl_create fs/f2fs/acl.c:377 [inline] > > > f2fs_init_acl+0xd7/0x9a0 fs/f2fs/acl.c:420 > > > f2fs_init_inode_metadata+0x824/0x1190 fs/f2fs/dir.c:558 > > > f2fs_do_tmpfile+0x34/0x170 fs/f2fs/dir.c:839 > > > __f2fs_tmpfile+0x1f9/0x380 fs/f2fs/namei.c:884 > > > f2fs_ioc_start_atomic_write+0x4a3/0x9e0 fs/f2fs/file.c:2099 > > > __f2fs_ioctl+0x1b5c/0xb770 > > > vfs_ioctl fs/ioctl.c:51 [inline] > > > __do_sys_ioctl fs/ioctl.c:870 [inline] > > > __se_sys_ioctl+0xf8/0x170 fs/ioctl.c:856 > > > do_syscall_x64 arch/x86/entry/common.c:50 [inline] > > > do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 > > > entry_SYSCALL_64_after_hwframe+0x63/0xcd > > > > > > The root cause is as below reversed lock order: > > > - f2fs_create > > > - f2fs_add_dentry > > > - f2fs_down_write(&F2FS_I(dir)->i_xattr_sem) > > > - f2fs_ioc_start_atomic_write > > > - __f2fs_tmpfile > > > - f2fs_do_tmpfile > > > - f2fs_down_write(&F2FS_I(inode)->i_sem) > > > - f2fs_init_inode_metadata > > > - f2fs_init_acl > > > - __f2fs_get_acl > > > - f2fs_getxattr > > > - f2fs_down_read(&F2FS_I(dir)->i_xattr_sem) > > > - f2fs_add_inline_entry > > > - f2fs_down_write(&F2FS_I(inode)->i_sem) > > > > How is it possible to get two inode are identical? > > Oh, so, it looks like a false positive? > > BTW, except commit message, what do you think of the change itself? > We can using i_sem to instead i_xattr_sem, so that 1) it can shrink > inode's size, 2) it's more clear that use common i_sem lock to protect > inode's {,x}attr access. If then, xattr can be blocked by unnecessary i_sem lock? > > Thanks, > > > > > > > > > We can break the dependency of deadlock by below change: > > > - use i_sem to keep order of {get,set}xattr instead of i_xattr_sem > > > - keep below lock order in inode operation to avoid deadlock: > > > dir->i_sem -> inode->i_sem > > > dir->i_sem -> dpage_lock > > > > > > Fixes: 5eda1ad1aaff ("f2fs: fix deadlock in i_xattr_sem and inode page lock") > > > Reported-by: syzbot+e5600587fa9cbf8e3826@syzkaller.appspotmail.com > > > Closes: https://lore.kernel.org/linux-f2fs-devel/00000000000096797d06001a359d@google.com > > > Signed-off-by: Chao Yu > > > --- > > > fs/f2fs/acl.c | 36 +++++++++++++++++++----------------- > > > fs/f2fs/acl.h | 4 ++-- > > > fs/f2fs/dir.c | 22 ++++++++++++++++------ > > > fs/f2fs/f2fs.h | 1 - > > > fs/f2fs/super.c | 5 ++--- > > > fs/f2fs/verity.c | 5 +++-- > > > fs/f2fs/xattr.c | 37 +++++++++++++++++++++---------------- > > > fs/f2fs/xattr.h | 19 +++++++++++++------ > > > 8 files changed, 76 insertions(+), 53 deletions(-) > > > > > > diff --git a/fs/f2fs/acl.c b/fs/f2fs/acl.c > > > index ec2aeccb69a3..af58b38e953c 100644 > > > --- a/fs/f2fs/acl.c > > > +++ b/fs/f2fs/acl.c > > > @@ -166,7 +166,7 @@ static void *f2fs_acl_to_disk(struct f2fs_sb_info *sbi, > > > } > > > static struct posix_acl *__f2fs_get_acl(struct inode *inode, int type, > > > - struct page *dpage) > > > + struct page *dpage, bool locked) > > > { > > > int name_index = F2FS_XATTR_INDEX_POSIX_ACL_DEFAULT; > > > void *value = NULL; > > > @@ -176,13 +176,13 @@ static struct posix_acl *__f2fs_get_acl(struct inode *inode, int type, > > > if (type == ACL_TYPE_ACCESS) > > > name_index = F2FS_XATTR_INDEX_POSIX_ACL_ACCESS; > > > - retval = f2fs_getxattr(inode, name_index, "", NULL, 0, dpage); > > > + retval = f2fs_getxattr(inode, name_index, "", NULL, 0, dpage, locked); > > > if (retval > 0) { > > > value = f2fs_kmalloc(F2FS_I_SB(inode), retval, GFP_F2FS_ZERO); > > > if (!value) > > > return ERR_PTR(-ENOMEM); > > > retval = f2fs_getxattr(inode, name_index, "", value, > > > - retval, dpage); > > > + retval, dpage, locked); > > > } > > > if (retval > 0) > > > @@ -201,7 +201,7 @@ struct posix_acl *f2fs_get_acl(struct inode *inode, int type, bool rcu) > > > if (rcu) > > > return ERR_PTR(-ECHILD); > > > - return __f2fs_get_acl(inode, type, NULL); > > > + return __f2fs_get_acl(inode, type, NULL, false); > > > } > > > static int f2fs_acl_update_mode(struct mnt_idmap *idmap, > > > @@ -226,9 +226,9 @@ static int f2fs_acl_update_mode(struct mnt_idmap *idmap, > > > return 0; > > > } > > > -static int __f2fs_set_acl(struct mnt_idmap *idmap, > > > - struct inode *inode, int type, > > > - struct posix_acl *acl, struct page *ipage) > > > +static int __f2fs_set_acl(struct mnt_idmap *idmap, struct inode *inode, > > > + int type, struct posix_acl *acl, > > > + struct page *ipage, bool locked) > > > { > > > int name_index; > > > void *value = NULL; > > > @@ -266,7 +266,8 @@ static int __f2fs_set_acl(struct mnt_idmap *idmap, > > > } > > > } > > > - error = f2fs_setxattr(inode, name_index, "", value, size, ipage, 0); > > > + error = f2fs_setxattr(inode, name_index, "", value, size, > > > + ipage, 0, locked); > > > kfree(value); > > > if (!error) > > > @@ -284,7 +285,7 @@ int f2fs_set_acl(struct mnt_idmap *idmap, struct dentry *dentry, > > > if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) > > > return -EIO; > > > - return __f2fs_set_acl(idmap, inode, type, acl, NULL); > > > + return __f2fs_set_acl(idmap, inode, type, acl, NULL, false); > > > } > > > /* > > > @@ -362,7 +363,7 @@ static int f2fs_acl_create_masq(struct posix_acl *acl, umode_t *mode_p) > > > static int f2fs_acl_create(struct inode *dir, umode_t *mode, > > > struct posix_acl **default_acl, struct posix_acl **acl, > > > - struct page *dpage) > > > + struct page *dpage, bool locked) > > > { > > > struct posix_acl *p; > > > struct posix_acl *clone; > > > @@ -374,7 +375,7 @@ static int f2fs_acl_create(struct inode *dir, umode_t *mode, > > > if (S_ISLNK(*mode) || !IS_POSIXACL(dir)) > > > return 0; > > > - p = __f2fs_get_acl(dir, ACL_TYPE_DEFAULT, dpage); > > > + p = __f2fs_get_acl(dir, ACL_TYPE_DEFAULT, dpage, locked); > > > if (!p || p == ERR_PTR(-EOPNOTSUPP)) { > > > *mode &= ~current_umask(); > > > return 0; > > > @@ -412,28 +413,29 @@ static int f2fs_acl_create(struct inode *dir, umode_t *mode, > > > } > > > int f2fs_init_acl(struct inode *inode, struct inode *dir, struct page *ipage, > > > - struct page *dpage) > > > + struct page *dpage, bool locked) > > > { > > > struct posix_acl *default_acl = NULL, *acl = NULL; > > > int error; > > > - error = f2fs_acl_create(dir, &inode->i_mode, &default_acl, &acl, dpage); > > > + error = f2fs_acl_create(dir, &inode->i_mode, &default_acl, &acl, > > > + dpage, locked); > > > if (error) > > > return error; > > > f2fs_mark_inode_dirty_sync(inode, true); > > > if (default_acl) { > > > - error = __f2fs_set_acl(NULL, inode, ACL_TYPE_DEFAULT, default_acl, > > > - ipage); > > > + error = __f2fs_set_acl(NULL, inode, ACL_TYPE_DEFAULT, > > > + default_acl, ipage, locked); > > > posix_acl_release(default_acl); > > > } else { > > > inode->i_default_acl = NULL; > > > } > > > if (acl) { > > > if (!error) > > > - error = __f2fs_set_acl(NULL, inode, ACL_TYPE_ACCESS, acl, > > > - ipage); > > > + error = __f2fs_set_acl(NULL, inode, ACL_TYPE_ACCESS, > > > + acl, ipage, locked); > > > posix_acl_release(acl); > > > } else { > > > inode->i_acl = NULL; > > > diff --git a/fs/f2fs/acl.h b/fs/f2fs/acl.h > > > index 94ebfbfbdc6f..9c14b6f549c6 100644 > > > --- a/fs/f2fs/acl.h > > > +++ b/fs/f2fs/acl.h > > > @@ -37,13 +37,13 @@ extern struct posix_acl *f2fs_get_acl(struct inode *, int, bool); > > > extern int f2fs_set_acl(struct mnt_idmap *, struct dentry *, > > > struct posix_acl *, int); > > > extern int f2fs_init_acl(struct inode *, struct inode *, struct page *, > > > - struct page *); > > > + struct page *, bool); > > > #else > > > #define f2fs_get_acl NULL > > > #define f2fs_set_acl NULL > > > static inline int f2fs_init_acl(struct inode *inode, struct inode *dir, > > > - struct page *ipage, struct page *dpage) > > > + struct page *ipage, struct page *dpage, bool locked) > > > { > > > return 0; > > > } > > > diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c > > > index d635c58cf5a3..4b5c62e18d67 100644 > > > --- a/fs/f2fs/dir.c > > > +++ b/fs/f2fs/dir.c > > > @@ -540,6 +540,8 @@ struct page *f2fs_init_inode_metadata(struct inode *inode, struct inode *dir, > > > int err; > > > if (is_inode_flag_set(inode, FI_NEW_INODE)) { > > > + struct f2fs_xattr_arg xarg; > > > + > > > page = f2fs_new_inode_page(inode); > > > if (IS_ERR(page)) > > > return page; > > > @@ -555,12 +557,16 @@ struct page *f2fs_init_inode_metadata(struct inode *inode, struct inode *dir, > > > put_page(page); > > > } > > > - err = f2fs_init_acl(inode, dir, page, dpage); > > > + err = f2fs_init_acl(inode, dir, page, dpage, true); > > > if (err) > > > goto put_error; > > > + xarg.page = page; > > > + xarg.locked = true; > > > + > > > err = f2fs_init_security(inode, dir, > > > - fname ? fname->usr_fname : NULL, page); > > > + fname ? fname->usr_fname : NULL, > > > + &xarg); > > > if (err) > > > goto put_error; > > > @@ -775,18 +781,20 @@ int f2fs_add_dentry(struct inode *dir, const struct f2fs_filename *fname, > > > { > > > int err = -EAGAIN; > > > + f2fs_down_read(&F2FS_I(dir)->i_sem); > > > + > > > if (f2fs_has_inline_dentry(dir)) { > > > /* > > > - * Should get i_xattr_sem to keep the lock order: > > > - * i_xattr_sem -> inode_page lock used by f2fs_setxattr. > > > + * Should get i_sem to keep the lock order: > > > + * i_sem -> inode_page lock used by f2fs_setxattr. > > > */ > > > - f2fs_down_read(&F2FS_I(dir)->i_xattr_sem); > > > err = f2fs_add_inline_entry(dir, fname, inode, ino, mode); > > > - f2fs_up_read(&F2FS_I(dir)->i_xattr_sem); > > > } > > > if (err == -EAGAIN) > > > err = f2fs_add_regular_entry(dir, fname, inode, ino, mode); > > > + f2fs_up_read(&F2FS_I(dir)->i_sem); > > > + > > > f2fs_update_time(F2FS_I_SB(dir), REQ_TIME); > > > return err; > > > } > > > @@ -835,6 +843,7 @@ int f2fs_do_tmpfile(struct inode *inode, struct inode *dir) > > > struct page *page; > > > int err = 0; > > > + f2fs_down_write(&F2FS_I(dir)->i_sem); > > > f2fs_down_write(&F2FS_I(inode)->i_sem); > > > page = f2fs_init_inode_metadata(inode, dir, NULL, NULL); > > > if (IS_ERR(page)) { > > > @@ -847,6 +856,7 @@ int f2fs_do_tmpfile(struct inode *inode, struct inode *dir) > > > f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); > > > fail: > > > f2fs_up_write(&F2FS_I(inode)->i_sem); > > > + f2fs_up_write(&F2FS_I(dir)->i_sem); > > > return err; > > > } > > > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h > > > index c7cb2177b252..60ec032be48d 100644 > > > --- a/fs/f2fs/f2fs.h > > > +++ b/fs/f2fs/f2fs.h > > > @@ -838,7 +838,6 @@ struct f2fs_inode_info { > > > /* avoid racing between foreground op and gc */ > > > struct f2fs_rwsem i_gc_rwsem[2]; > > > - struct f2fs_rwsem i_xattr_sem; /* avoid racing between reading and changing EAs */ > > > int i_extra_isize; /* size of extra space located in i_addr */ > > > kprojid_t i_projid; /* id for project quota */ > > > diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c > > > index ca31163da00a..c72fda24cffd 100644 > > > --- a/fs/f2fs/super.c > > > +++ b/fs/f2fs/super.c > > > @@ -1418,7 +1418,6 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb) > > > INIT_LIST_HEAD(&fi->gdirty_list); > > > init_f2fs_rwsem(&fi->i_gc_rwsem[READ]); > > > init_f2fs_rwsem(&fi->i_gc_rwsem[WRITE]); > > > - init_f2fs_rwsem(&fi->i_xattr_sem); > > > /* Will be used by directory only */ > > > fi->i_dir_level = F2FS_SB(sb)->dir_level; > > > @@ -3163,7 +3162,7 @@ static int f2fs_get_context(struct inode *inode, void *ctx, size_t len) > > > { > > > return f2fs_getxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION, > > > F2FS_XATTR_NAME_ENCRYPTION_CONTEXT, > > > - ctx, len, NULL); > > > + ctx, len, NULL, false); > > > } > > > static int f2fs_set_context(struct inode *inode, const void *ctx, size_t len, > > > @@ -3183,7 +3182,7 @@ static int f2fs_set_context(struct inode *inode, const void *ctx, size_t len, > > > return f2fs_setxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION, > > > F2FS_XATTR_NAME_ENCRYPTION_CONTEXT, > > > - ctx, len, fs_data, XATTR_CREATE); > > > + ctx, len, fs_data, XATTR_CREATE, true); > > > } > > > static const union fscrypt_policy *f2fs_get_dummy_policy(struct super_block *sb) > > > diff --git a/fs/f2fs/verity.c b/fs/f2fs/verity.c > > > index 4fc95f353a7a..e181528b1f26 100644 > > > --- a/fs/f2fs/verity.c > > > +++ b/fs/f2fs/verity.c > > > @@ -181,7 +181,7 @@ static int f2fs_end_enable_verity(struct file *filp, const void *desc, > > > /* Set the verity xattr. */ > > > err = f2fs_setxattr(inode, F2FS_XATTR_INDEX_VERITY, > > > F2FS_XATTR_NAME_VERITY, &dloc, sizeof(dloc), > > > - NULL, XATTR_CREATE); > > > + NULL, XATTR_CREATE, false); > > > if (err) > > > goto cleanup; > > > @@ -226,7 +226,8 @@ static int f2fs_get_verity_descriptor(struct inode *inode, void *buf, > > > /* Get the descriptor location */ > > > res = f2fs_getxattr(inode, F2FS_XATTR_INDEX_VERITY, > > > - F2FS_XATTR_NAME_VERITY, &dloc, sizeof(dloc), NULL); > > > + F2FS_XATTR_NAME_VERITY, &dloc, sizeof(dloc), > > > + NULL, false); > > > if (res < 0 && res != -ERANGE) > > > return res; > > > if (res != sizeof(dloc) || dloc.version != cpu_to_le32(F2FS_VERIFY_VER)) { > > > diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c > > > index 476b186b90a6..a6a611f8a771 100644 > > > --- a/fs/f2fs/xattr.c > > > +++ b/fs/f2fs/xattr.c > > > @@ -61,7 +61,7 @@ static int f2fs_xattr_generic_get(const struct xattr_handler *handler, > > > return -EINVAL; > > > } > > > return f2fs_getxattr(inode, handler->flags, name, > > > - buffer, size, NULL); > > > + buffer, size, NULL, false); > > > } > > > static int f2fs_xattr_generic_set(const struct xattr_handler *handler, > > > @@ -84,7 +84,7 @@ static int f2fs_xattr_generic_set(const struct xattr_handler *handler, > > > return -EINVAL; > > > } > > > return f2fs_setxattr(inode, handler->flags, name, > > > - value, size, NULL, flags); > > > + value, size, NULL, flags, false); > > > } > > > static bool f2fs_xattr_user_list(struct dentry *dentry) > > > @@ -136,15 +136,16 @@ static int f2fs_xattr_advise_set(const struct xattr_handler *handler, > > > #ifdef CONFIG_F2FS_FS_SECURITY > > > static int f2fs_initxattrs(struct inode *inode, const struct xattr *xattr_array, > > > - void *page) > > > + void *fs_data) > > > { > > > + struct f2fs_xattr_arg *xarg = (struct f2fs_xattr_arg *)fs_data; > > > const struct xattr *xattr; > > > int err = 0; > > > for (xattr = xattr_array; xattr->name != NULL; xattr++) { > > > err = f2fs_setxattr(inode, F2FS_XATTR_INDEX_SECURITY, > > > xattr->name, xattr->value, > > > - xattr->value_len, (struct page *)page, 0); > > > + xattr->value_len, xarg->page, 0, xarg->locked); > > > if (err < 0) > > > break; > > > } > > > @@ -152,10 +153,11 @@ static int f2fs_initxattrs(struct inode *inode, const struct xattr *xattr_array, > > > } > > > int f2fs_init_security(struct inode *inode, struct inode *dir, > > > - const struct qstr *qstr, struct page *ipage) > > > + const struct qstr *qstr, > > > + struct f2fs_xattr_arg *xarg) > > > { > > > return security_inode_init_security(inode, dir, qstr, > > > - &f2fs_initxattrs, ipage); > > > + &f2fs_initxattrs, xarg); > > > } > > > #endif > > > @@ -512,7 +514,8 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize, > > > } > > > int f2fs_getxattr(struct inode *inode, int index, const char *name, > > > - void *buffer, size_t buffer_size, struct page *ipage) > > > + void *buffer, size_t buffer_size, > > > + struct page *ipage, bool locked) > > > { > > > struct f2fs_xattr_entry *entry = NULL; > > > int error; > > > @@ -528,12 +531,12 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name, > > > if (len > F2FS_NAME_LEN) > > > return -ERANGE; > > > - if (!ipage) > > > - f2fs_down_read(&F2FS_I(inode)->i_xattr_sem); > > > + if (!locked) > > > + f2fs_down_read(&F2FS_I(inode)->i_sem); > > > error = lookup_all_xattrs(inode, ipage, index, len, name, > > > &entry, &base_addr, &base_size, &is_inline); > > > - if (!ipage) > > > - f2fs_up_read(&F2FS_I(inode)->i_xattr_sem); > > > + if (!locked) > > > + f2fs_up_read(&F2FS_I(inode)->i_sem); > > > if (error) > > > return error; > > > @@ -567,9 +570,9 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size) > > > int error; > > > size_t rest = buffer_size; > > > - f2fs_down_read(&F2FS_I(inode)->i_xattr_sem); > > > + f2fs_down_read(&F2FS_I(inode)->i_sem); > > > error = read_all_xattrs(inode, NULL, &base_addr); > > > - f2fs_up_read(&F2FS_I(inode)->i_xattr_sem); > > > + f2fs_up_read(&F2FS_I(inode)->i_sem); > > > if (error) > > > return error; > > > @@ -775,7 +778,7 @@ static int __f2fs_setxattr(struct inode *inode, int index, > > > int f2fs_setxattr(struct inode *inode, int index, const char *name, > > > const void *value, size_t size, > > > - struct page *ipage, int flags) > > > + struct page *ipage, int flags, bool locked) > > > { > > > struct f2fs_sb_info *sbi = F2FS_I_SB(inode); > > > int err; > > > @@ -796,9 +799,11 @@ int f2fs_setxattr(struct inode *inode, int index, const char *name, > > > f2fs_balance_fs(sbi, true); > > > f2fs_lock_op(sbi); > > > - f2fs_down_write(&F2FS_I(inode)->i_xattr_sem); > > > + if (!locked) > > > + f2fs_down_write(&F2FS_I(inode)->i_sem); > > > err = __f2fs_setxattr(inode, index, name, value, size, ipage, flags); > > > - f2fs_up_write(&F2FS_I(inode)->i_xattr_sem); > > > + if (!locked) > > > + f2fs_up_write(&F2FS_I(inode)->i_sem); > > > f2fs_unlock_op(sbi); > > > f2fs_update_time(sbi, REQ_TIME); > > > diff --git a/fs/f2fs/xattr.h b/fs/f2fs/xattr.h > > > index b1811c392e6f..170e4a49af31 100644 > > > --- a/fs/f2fs/xattr.h > > > +++ b/fs/f2fs/xattr.h > > > @@ -52,6 +52,11 @@ struct f2fs_xattr_entry { > > > char e_name[]; /* attribute name */ > > > }; > > > +struct f2fs_xattr_arg { > > > + struct page *page; /* inode page */ > > > + bool locked; /* indicate i_sem is locked */ > > > +}; > > > + > > > #define XATTR_HDR(ptr) ((struct f2fs_xattr_header *)(ptr)) > > > #define XATTR_ENTRY(ptr) ((struct f2fs_xattr_entry *)(ptr)) > > > #define XATTR_FIRST_ENTRY(ptr) (XATTR_ENTRY(XATTR_HDR(ptr) + 1)) > > > @@ -128,9 +133,9 @@ extern const struct xattr_handler f2fs_xattr_security_handler; > > > extern const struct xattr_handler *f2fs_xattr_handlers[]; > > > extern int f2fs_setxattr(struct inode *, int, const char *, > > > - const void *, size_t, struct page *, int); > > > + const void *, size_t, struct page *, int, bool); > > > extern int f2fs_getxattr(struct inode *, int, const char *, void *, > > > - size_t, struct page *); > > > + size_t, struct page *, bool locked); > > > extern ssize_t f2fs_listxattr(struct dentry *, char *, size_t); > > > extern int f2fs_init_xattr_caches(struct f2fs_sb_info *); > > > extern void f2fs_destroy_xattr_caches(struct f2fs_sb_info *); > > > @@ -140,13 +145,13 @@ extern void f2fs_destroy_xattr_caches(struct f2fs_sb_info *); > > > #define f2fs_listxattr NULL > > > static inline int f2fs_setxattr(struct inode *inode, int index, > > > const char *name, const void *value, size_t size, > > > - struct page *page, int flags) > > > + struct page *page, int flags, bool locked) > > > { > > > return -EOPNOTSUPP; > > > } > > > static inline int f2fs_getxattr(struct inode *inode, int index, > > > const char *name, void *buffer, > > > - size_t buffer_size, struct page *dpage) > > > + size_t buffer_size, struct page *dpage, bool locked) > > > { > > > return -EOPNOTSUPP; > > > } > > > @@ -156,10 +161,12 @@ static inline void f2fs_destroy_xattr_caches(struct f2fs_sb_info *sbi) { } > > > #ifdef CONFIG_F2FS_FS_SECURITY > > > extern int f2fs_init_security(struct inode *, struct inode *, > > > - const struct qstr *, struct page *); > > > + const struct qstr *, > > > + struct f2fs_xattr_arg *xarg); > > > #else > > > static inline int f2fs_init_security(struct inode *inode, struct inode *dir, > > > - const struct qstr *qstr, struct page *ipage) > > > + const struct qstr *qstr, > > > + struct f2fs_xattr_arg *xarg) > > > { > > > return 0; > > > } > > > -- > > > 2.40.1