[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1504622676-2992-2-git-send-email-joe.lawrence@redhat.com>
Date: Tue, 5 Sep 2017 10:44:34 -0400
From: Joe Lawrence <joe.lawrence@...hat.com>
To: linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org
Cc: Alexander Viro <viro@...iv.linux.org.uk>,
"Luis R. Rodriguez" <mcgrof@...nel.org>,
Kees Cook <keescook@...omium.org>,
Mikulas Patocka <mpatocka@...hat.com>,
Michael Kerrisk <mtk.manpages@...il.com>
Subject: [PATCH RFC 1/3] pipe: avoid round_pipe_size() nr_pages overflow on 32-bit
The round_pipe_size() function contains a right-bit-shift expression
which may overflow, which would cause undefined results in a subsequent
roundup_pow_of_two() call.
static inline unsigned int round_pipe_size(unsigned int size)
{
unsigned long nr_pages;
nr_pages = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
return roundup_pow_of_two(nr_pages) << PAGE_SHIFT;
}
PAGE_SIZE is defined as (1UL << PAGE_SHIFT), so:
- 4 bytes wide on 32-bit (0 to 0xffffffff)
- 8 bytes wide on 64-bit (0 to 0xffffffffffffffff)
That means that 32-bit round_pipe_size(), nr_pages may overflow to 0:
size=0x00000000 nr_pages=0x0
size=0x00000001 nr_pages=0x1
size=0xfffff000 nr_pages=0xfffff
size=0xfffff001 nr_pages=0x0 << !
size=0xffffffff nr_pages=0x0 << !
This is bad because roundup_pow_of_two(n) is undefined when n == 0!
64-bit is not a problem* as the unsigned int size is 4 bytes wide
(similar to 32-bit) and the larger, 8 byte wide unsigned long, is
sufficient to handle the largest value of the bit shift expression:
size=0xffffffff nr_pages=100000
(*On 64-bit, round_pipe_size(0xffffffff) will later overflow when the
result of roundup_pow_of_two(0x100000) is shifted left by PAGE_SHIFT.
This behavior is at least defined by the language, so callers can safely
sanity-check a 0 return value from round_pipe_size.)
Modify round_pipe_size() to return 0 if n == 0, reset the original
pipe_max_size value, and update callers to handle accordingly.
Reported-by: Mikulas Patocka <mpatocka@...hat.com>
Signed-off-by: Joe Lawrence <joe.lawrence@...hat.com>
---
fs/pipe.c | 18 ++++++++++++++++--
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/fs/pipe.c b/fs/pipe.c
index 97e5be897753..fa28910b3c59 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -1017,13 +1017,16 @@ static int fifo_open(struct inode *inode, struct file *filp)
/*
* Currently we rely on the pipe array holding a power-of-2 number
- * of pages.
+ * of pages. Returns 0 on error.
*/
static inline unsigned int round_pipe_size(unsigned int size)
{
unsigned long nr_pages;
nr_pages = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ if (nr_pages == 0)
+ return 0;
+
return roundup_pow_of_two(nr_pages) << PAGE_SHIFT;
}
@@ -1039,6 +1042,8 @@ static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long arg)
long ret = 0;
size = round_pipe_size(arg);
+ if (size == 0)
+ return -EINVAL;
nr_pages = size >> PAGE_SHIFT;
if (!nr_pages)
@@ -1122,13 +1127,22 @@ static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long arg)
int pipe_proc_fn(struct ctl_table *table, int write, void __user *buf,
size_t *lenp, loff_t *ppos)
{
+ unsigned int orig_pipe_max_size;
+ unsigned int rounded_pipe_max_size;
int ret;
+ orig_pipe_max_size = pipe_max_size;
ret = proc_dointvec_minmax(table, write, buf, lenp, ppos);
if (ret < 0 || !write)
return ret;
- pipe_max_size = round_pipe_size(pipe_max_size);
+ rounded_pipe_max_size = round_pipe_size(pipe_max_size);
+ if (rounded_pipe_max_size == 0) {
+ pipe_max_size = orig_pipe_max_size;
+ return -EINVAL;
+ }
+
+ pipe_max_size = rounded_pipe_max_size;
return ret;
}
--
1.8.3.1
Powered by blists - more mailing lists