| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Thu, 11 Jun 2020 20:04:06 +0900
From: Daeho Jeong <daeho43@...il.com>
To: Chao Yu <yuchao0@...wei.com>
Cc: linux-kernel@...r.kernel.org,
linux-f2fs-devel@...ts.sourceforge.net, kernel-team@...roid.com,
Daeho Jeong <daehojeong@...gle.com>
Subject: Re: [f2fs-dev] [PATCH v2] f2fs: add F2FS_IOC_SEC_TRIM_FILE ioctl
2020년 6월 11일 (목) 오후 5:56, Chao Yu <yuchao0@...wei.com>님이 작성:
>
> On 2020/6/11 11:16, Daeho Jeong wrote:
> > From: Daeho Jeong <daehojeong@...gle.com>
> >
> > Added a new ioctl to send discard commands or/and zero out
> > to whole data area of a regular file for security reason.
> >
> > Signed-off-by: Daeho Jeong <daehojeong@...gle.com>
> > ---
> > fs/f2fs/f2fs.h | 8 +++
> > fs/f2fs/file.c | 143 +++++++++++++++++++++++++++++++++++++++++++++++++
> > 2 files changed, 151 insertions(+)
> >
> > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> > index c812fb8e2d9c..ca139fac5a73 100644
> > --- a/fs/f2fs/f2fs.h
> > +++ b/fs/f2fs/f2fs.h
> > @@ -434,6 +434,7 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal,
> > _IOR(F2FS_IOCTL_MAGIC, 18, __u64)
> > #define F2FS_IOC_RESERVE_COMPRESS_BLOCKS \
> > _IOR(F2FS_IOCTL_MAGIC, 19, __u64)
> > +#define F2FS_IOC_SEC_TRIM_FILE _IOW(F2FS_IOCTL_MAGIC, 20, __u32)
> >
> > #define F2FS_IOC_GET_VOLUME_NAME FS_IOC_GETFSLABEL
> > #define F2FS_IOC_SET_VOLUME_NAME FS_IOC_SETFSLABEL
> > @@ -453,6 +454,13 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal,
> > #define F2FS_GOING_DOWN_METAFLUSH 0x3 /* going down with meta flush */
> > #define F2FS_GOING_DOWN_NEED_FSCK 0x4 /* going down to trigger fsck */
> >
> > +/*
> > + * Flags used by F2FS_IOC_SEC_TRIM_FILE
> > + */
> > +#define F2FS_TRIM_FILE_DISCARD 0x1 /* send discard command */
> > +#define F2FS_TRIM_FILE_ZEROOUT 0x2 /* zero out */
> > +#define F2FS_TRIM_FILE_MASK 0x3
> > +
> > #if defined(__KERNEL__) && defined(CONFIG_COMPAT)
> > /*
> > * ioctl commands in 32 bit emulation
> > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> > index dfa1ac2d751a..ba9b7ec5d6bf 100644
> > --- a/fs/f2fs/file.c
> > +++ b/fs/f2fs/file.c
> > @@ -3749,6 +3749,146 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg)
> > return ret;
> > }
> >
> > +static int f2fs_secure_erase(struct block_device *bdev, block_t block,
> > + block_t len, u32 flags)
> > +{
> > + struct request_queue *q = bdev_get_queue(bdev);
> > + sector_t sector = SECTOR_FROM_BLOCK(block);
> > + sector_t nr_sects = SECTOR_FROM_BLOCK(len);
> > + int ret = 0;
> > +
> > + if (!q)
> > + return -ENXIO;
> > +
> > + if (flags & F2FS_TRIM_FILE_DISCARD)
> > + ret = blkdev_issue_discard(bdev, sector, nr_sects, GFP_NOFS,
> > + blk_queue_secure_erase(q) ?
> > + BLKDEV_DISCARD_SECURE : 0);
> > +
> > + if (!ret && (flags & F2FS_TRIM_FILE_ZEROOUT))
> > + ret = blkdev_issue_zeroout(bdev, sector, nr_sects, GFP_NOFS, 0);
> > +
> > + return ret;
> > +}
> > +
> > +static int f2fs_sec_trim_file(struct file *filp, unsigned long arg)
> > +{
> > + struct inode *inode = file_inode(filp);
> > + struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
> > + struct address_space *mapping = inode->i_mapping;
> > + struct block_device *prev_bdev = NULL;
> > + pgoff_t index, pg_start = 0, pg_end;
> > + block_t prev_block = 0, len = 0;
> > + u32 flags;
> > + int ret = 0;
> > +
> > + if (!(filp->f_mode & FMODE_WRITE))
> > + return -EBADF;
> > +
> > + if (get_user(flags, (u32 __user *)arg))
> > + return -EFAULT;
> > + if (flags == 0 || (flags & ~F2FS_TRIM_FILE_MASK))
> > + return -EINVAL;
> > +
> > + if ((flags & F2FS_TRIM_FILE_DISCARD) && !f2fs_hw_support_discard(sbi))
> > + return -EOPNOTSUPP;
> > +
> > + file_start_write(filp);
>
> Now, I'm a little confused about when we need to call __mnt_want_write_file(),
> you know, vfs_write() still will call this function when updating time.
> - __generic_file_write_iter
> - file_update_time
> - __mnt_want_write_file
>
> And previously, f2fs ioctl uses mnt_{want,drop}_write_file() whenever there is
> any updates on fs/file, if Eric is correct, we need to clean up most of ioctl
> interface as well.
I also saw most filesytem codes use just mnt_{want,drop}_write_file()
and actually it doesn't affect code working. It's a matter of doing a
redundant job or not.
AFAIUI, if the file is not open for writing (FMODE_WRITE), we have to
call mnt_want_write_file() to increase mnt_writers.
In this case, we already checked it has FMODE_WRITE flag.
>
> > + inode_lock(inode);
> > +
> > + if (!S_ISREG(inode->i_mode) || f2fs_is_atomic_file(inode) ||
> > + f2fs_compressed_file(inode)) {
> > + ret = -EINVAL;
> > + goto err;
> > + }
> > +
> > + if (!inode->i_size)
> > + goto err;
> > + pg_end = DIV_ROUND_UP(inode->i_size, PAGE_SIZE);
> > +
> > + ret = f2fs_convert_inline_inode(inode);
> > + if (ret)
> > + goto err;
> > +
> > + down_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]);
> > + down_write(&F2FS_I(inode)->i_mmap_sem);
> > +
> > + ret = filemap_write_and_wait(mapping);
> > + if (ret)
> > + goto out;
> > +
> > + truncate_inode_pages(mapping, 0);
> > +
> > + for (index = pg_start; index < pg_end;) {
> > + struct dnode_of_data dn;
> > + unsigned int end_offset;
> > +
> > + set_new_dnode(&dn, inode, NULL, NULL, 0);
> > + ret = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE);
> > + if (ret)
>
> if (ret == -ENOENT) {
> index = f2fs_get_next_page_offset(&dn, index);
> continue;
> }
Got it.
>
> > + goto out;
> > +
> > + end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
> > + if (pg_end < end_offset + index)
> > + end_offset = pg_end - index;
> > +
> > + for (; dn.ofs_in_node < end_offset;
> > + dn.ofs_in_node++, index++) {
> > + struct block_device *cur_bdev;
> > + block_t blkaddr = f2fs_data_blkaddr(&dn);
> > +
> > + if (__is_valid_data_blkaddr(blkaddr)) {
> > + if (!f2fs_is_valid_blkaddr(F2FS_I_SB(inode),
> > + blkaddr, DATA_GENERIC_ENHANCE)) {
> > + ret = -EFSCORRUPTED;
> > + goto out;
>
>
> if we goto out label here, we will miss to call f2fs_put_dnode()?
>
Oops, I missed this, when I changed the code flow. Thanks!
> > + }
> > + } else
>
> How about this?
>
> if (!__is_valid_data_blkaddr())
> continue;
>
> if (!f2fs_is_valid_blkaddr(F2FS_I_SB(inode), blkaddr, DATA_GENERIC_ENHANCE)) {
> ret = -EFSCORRUPTED;
> goto out;
> }
>
Looks better.
> > + continue;
> > +
> > + cur_bdev = f2fs_target_device(sbi, blkaddr, NULL);
> > + if (f2fs_is_multi_device(sbi)) {
> > + int i = f2fs_target_device_index(sbi, blkaddr);
> > +
> > + blkaddr -= FDEV(i).start_blk;
> > + }
> > +
> > + if (len) {
> > + if (prev_bdev == cur_bdev &&
> > + blkaddr == prev_block + len) {
> > + len++;
> > + } else {
> > + ret = f2fs_secure_erase(prev_bdev,
> > + prev_block, len, flags);
> > + if (ret)
> > + goto out;
>
> Ditto,
Got it.
Powered by blists - more mailing lists