From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753478AbdKFOkR (ORCPT ); Mon, 6 Nov 2017 09:40:17 -0500 Received: from mx1.redhat.com ([209.132.183.28]:52124 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753625AbdKFOkL (ORCPT ); Mon, 6 Nov 2017 09:40:11 -0500 DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 09B8BC058ECA Authentication-Results: ext-mx08.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx08.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=marcandre.lureau@redhat.com From: =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: aarcange@redhat.com, hughd@google.com, nyc@holomorphy.com, mike.kravetz@oracle.com, =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= Subject: [PATCH v2 4/9] hugetlbfs: implement memfd sealing Date: Mon, 6 Nov 2017 15:39:39 +0100 Message-Id: <20171106143944.13821-5-marcandre.lureau@redhat.com> In-Reply-To: <20171106143944.13821-1-marcandre.lureau@redhat.com> References: <20171106143944.13821-1-marcandre.lureau@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Mon, 06 Nov 2017 14:40:11 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Implements memfd sealing, similar to shmem: - WRITE: deny fallocate(PUNCH_HOLE). mmap() write is denied in memfd_add_seals(). write() doesn't exist for hugetlbfs. - SHRINK: added similar check as shmem_setattr() - GROW: added similar check as shmem_setattr() & shmem_fallocate() Except write() operation that doesn't exist with hugetlbfs, that should make sealing as close as it can be to shmem support. Signed-off-by: Marc-André Lureau Reviewed-by: Mike Kravetz --- fs/hugetlbfs/inode.c | 29 +++++++++++++++++++++++++++-- include/linux/hugetlb.h | 1 + 2 files changed, 28 insertions(+), 2 deletions(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index f57aab929e41..01f5aa6ea57a 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -510,8 +510,16 @@ static long hugetlbfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) if (hole_end > hole_start) { struct address_space *mapping = inode->i_mapping; + struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); inode_lock(inode); + + /* protected by i_mutex */ + if (info->seals & F_SEAL_WRITE) { + inode_unlock(inode); + return -EPERM; + } + i_mmap_lock_write(mapping); if (!RB_EMPTY_ROOT(&mapping->i_mmap.rb_root)) hugetlb_vmdelete_list(&mapping->i_mmap, @@ -529,6 +537,7 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset, loff_t len) { struct inode *inode = file_inode(file); + struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); struct address_space *mapping = inode->i_mapping; struct hstate *h = hstate_inode(inode); struct vm_area_struct pseudo_vma; @@ -560,6 +569,11 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset, if (error) goto out; + if ((info->seals & F_SEAL_GROW) && offset + len > inode->i_size) { + error = -EPERM; + goto out; + } + /* * Initialize a pseudo vma as this is required by the huge page * allocation routines. If NUMA is configured, use page index @@ -650,6 +664,7 @@ static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) struct hstate *h = hstate_inode(inode); int error; unsigned int ia_valid = attr->ia_valid; + struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); BUG_ON(!inode); @@ -658,10 +673,17 @@ static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) return error; if (ia_valid & ATTR_SIZE) { + loff_t oldsize = inode->i_size; + loff_t newsize = attr->ia_size; + error = -EINVAL; - if (attr->ia_size & ~huge_page_mask(h)) + if (newsize & ~huge_page_mask(h)) return -EINVAL; - error = hugetlb_vmtruncate(inode, attr->ia_size); + /* protected by i_mutex */ + if ((newsize < oldsize && (info->seals & F_SEAL_SHRINK)) || + (newsize > oldsize && (info->seals & F_SEAL_GROW))) + return -EPERM; + error = hugetlb_vmtruncate(inode, newsize); if (error) return error; } @@ -713,6 +735,8 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, inode = new_inode(sb); if (inode) { + struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); + inode->i_ino = get_next_ino(); inode_init_owner(inode, dir, mode); lockdep_set_class(&inode->i_mapping->i_mmap_rwsem, @@ -720,6 +744,7 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, inode->i_mapping->a_ops = &hugetlbfs_aops; inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode); inode->i_mapping->private_data = resv_map; + info->seals = F_SEAL_SEAL; switch (mode & S_IFMT) { default: init_special_inode(inode, mode, dev); diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 590a77433a14..2a21c59a9952 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -282,6 +282,7 @@ static inline struct hugetlbfs_sb_info *HUGETLBFS_SB(struct super_block *sb) struct hugetlbfs_inode_info { struct shared_policy policy; struct inode vfs_inode; + unsigned int seals; }; static inline struct hugetlbfs_inode_info *HUGETLBFS_I(struct inode *inode) -- 2.15.0.rc0.40.gaefcc5f6f