2019-10-09 03:21:32

by Randall Huang

[permalink] [raw]
Subject: [PATCH] f2fs: fix to avoid memory leakage in f2fs_listxattr

In f2fs_listxattr, there is no boundary check before
memcpy e_name to buffer.
If the e_name_len is corrupted,
unexpected memory contents may be returned to the buffer.

Signed-off-by: Randall Huang <[email protected]>
---
fs/f2fs/xattr.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index b32c45621679..acc3663970cd 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -538,8 +538,9 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
{
struct inode *inode = d_inode(dentry);
+ nid_t xnid = F2FS_I(inode)->i_xattr_nid;
struct f2fs_xattr_entry *entry;
- void *base_addr;
+ void *base_addr, *last_base_addr;
int error = 0;
size_t rest = buffer_size;

@@ -549,6 +550,8 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
if (error)
return error;

+ last_base_addr = (void *)base_addr + XATTR_SIZE(xnid, inode);
+
list_for_each_xattr(entry, base_addr) {
const struct xattr_handler *handler =
f2fs_xattr_handler(entry->e_name_index);
@@ -559,6 +562,15 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
if (!handler || (handler->list && !handler->list(dentry)))
continue;

+ if ((void *)(entry) + sizeof(__u32) > last_base_addr ||
+ (void *)XATTR_NEXT_ENTRY(entry) > last_base_addr) {
+ f2fs_err(F2FS_I_SB(inode), "inode (%lu) has corrupted xattr",
+ inode->i_ino);
+ set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK);
+ error = -EFSCORRUPTED;
+ goto cleanup;
+ }
+
prefix = xattr_prefix(handler);
prefix_len = strlen(prefix);
size = prefix_len + entry->e_name_len + 1;
--
2.23.0.581.g78d2f28ef7-goog


2019-10-15 07:31:36

by Chao Yu

[permalink] [raw]
Subject: Re: [PATCH] f2fs: fix to avoid memory leakage in f2fs_listxattr

Hi Randall,

On 2019/10/9 11:20, Randall Huang wrote:
> In f2fs_listxattr, there is no boundary check before
> memcpy e_name to buffer.
> If the e_name_len is corrupted,
> unexpected memory contents may be returned to the buffer.
>
> Signed-off-by: Randall Huang <[email protected]>
> ---
> fs/f2fs/xattr.c | 14 +++++++++++++-
> 1 file changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
> index b32c45621679..acc3663970cd 100644
> --- a/fs/f2fs/xattr.c
> +++ b/fs/f2fs/xattr.c
> @@ -538,8 +538,9 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
> ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
> {
> struct inode *inode = d_inode(dentry);
> + nid_t xnid = F2FS_I(inode)->i_xattr_nid;
> struct f2fs_xattr_entry *entry;
> - void *base_addr;
> + void *base_addr, *last_base_addr;
> int error = 0;
> size_t rest = buffer_size;
>
> @@ -549,6 +550,8 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
> if (error)
> return error;
>
> + last_base_addr = (void *)base_addr + XATTR_SIZE(xnid, inode);
> +
> list_for_each_xattr(entry, base_addr) {
> const struct xattr_handler *handler =
> f2fs_xattr_handler(entry->e_name_index);
> @@ -559,6 +562,15 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
> if (!handler || (handler->list && !handler->list(dentry)))
> continue;
>
> + if ((void *)(entry) + sizeof(__u32) > last_base_addr ||
> + (void *)XATTR_NEXT_ENTRY(entry) > last_base_addr) {
> + f2fs_err(F2FS_I_SB(inode), "inode (%lu) has corrupted xattr",
> + inode->i_ino);
> + set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK);
> + error = -EFSCORRUPTED;
> + goto cleanup;
> + }

Could you relocate sanity check to the place before we check handler? As I'm
thinking we should always check validation of current entry before using its
field (entry->index).

Thanks,

> +
> prefix = xattr_prefix(handler);
> prefix_len = strlen(prefix);
> size = prefix_len + entry->e_name_len + 1;
>

2019-10-18 23:36:16

by Randall Huang

[permalink] [raw]
Subject: [PATCH v2] f2fs: fix to avoid memory leakage in f2fs_listxattr

In f2fs_listxattr, there is no boundary check before
memcpy e_name to buffer.
If the e_name_len is corrupted,
unexpected memory contents may be returned to the buffer.

Signed-off-by: Randall Huang <[email protected]>
---
v2:
relocate sanity check
---
fs/f2fs/xattr.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index 181900af2576..296b3189448a 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -539,8 +539,9 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
{
struct inode *inode = d_inode(dentry);
+ nid_t xnid = F2FS_I(inode)->i_xattr_nid;
struct f2fs_xattr_entry *entry;
- void *base_addr;
+ void *base_addr, *last_base_addr;
int error = 0;
size_t rest = buffer_size;

@@ -550,6 +551,8 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
if (error)
return error;

+ last_base_addr = (void *)base_addr + XATTR_SIZE(xnid, inode);
+
list_for_each_xattr(entry, base_addr) {
const struct xattr_handler *handler =
f2fs_xattr_handler(entry->e_name_index);
@@ -557,6 +560,15 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
size_t prefix_len;
size_t size;

+ if ((void *)(entry) + sizeof(__u32) > last_base_addr ||
+ (void *)XATTR_NEXT_ENTRY(entry) > last_base_addr) {
+ f2fs_err(F2FS_I_SB(inode), "inode (%lu) has corrupted xattr",
+ inode->i_ino);
+ set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK);
+ error = -EFSCORRUPTED;
+ goto cleanup;
+ }
+
if (!handler || (handler->list && !handler->list(dentry)))
continue;

--
2.23.0.866.gb869b98d4c-goog

2019-10-22 12:08:18

by Chao Yu

[permalink] [raw]
Subject: Re: [PATCH v2] f2fs: fix to avoid memory leakage in f2fs_listxattr

On 2019/10/18 14:56, Randall Huang wrote:
> In f2fs_listxattr, there is no boundary check before
> memcpy e_name to buffer.
> If the e_name_len is corrupted,
> unexpected memory contents may be returned to the buffer.
>
> Signed-off-by: Randall Huang <[email protected]>

Reviewed-by: Chao Yu <[email protected]>

Thanks,