sysfs: fix off-by-one error in fill_read_buffer
PATCH 1/1 sysfs: fix off-by-one error in fill_read_buffer
Hi,everyone.
I found that there is a off-by-one problem in the following code.
Version: 2.6.24-rc2
File: fs/sysfs/file.c:118-122
Function: fill_read_buffer
--------------------------------------------------------------------
count = ops->show(kobj, attr_sd->s_attr.attr, buffer->page); sysfs_put_active_two(attr_sd); BUG_ON(count > (ssize_t)PAGE_SIZE);
--------------------------------------------------------------------
Because according to the specification of the sysfs and the implement of the show methods, the show methods return the number of bytes which would be generated for the given input, excluding the trailing null.So if the return value of the show methods equals PAGE_SIZE - 1, the buffer is full in fact. And if the return value equals PAGE_SIZE, the resulting string was already truncated,or buffer overflow occurred.
This patch fixes an off-by-one error in fill_read_buffer.
--- fs/sysfs/file.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
fs/sysfs/file.c +1 -1
@@ -119,7 +119,7 @@ static int fill_read_buffer(struct dentry * dentry, struct sysfs_buffer * buffersysfs_put_active_two (attr_sd); -BUG_ON (count > (ssize_t )PAGE_SIZE ); +BUG_ON (count >= (ssize_t )PAGE_SIZE );if (count >=0 ) { buffer->needs_read_fill =0 ; buffer->count = count;
Signature · 1 line
1.5.3
Quoted 21 lines · Miao Xie
Hi,everyone.
I found that there is a off-by-one problem in the following code.Version: 2.6.24-rc2 File: fs/sysfs/file.c:118-122 Function: fill_read_buffer -------------------------------------------------------------------- count = ops->show(kobj, attr_sd->s_attr.attr, buffer->page); sysfs_put_active_two(attr_sd); BUG_ON(count > (ssize_t)PAGE_SIZE);--------------------------------------------------------------------
Because according to the specification of the sysfs and the implement of the
show methods, the show methods return the number of bytes which would be generated for the given input, excluding the trailing null.So if the return value of the show methods equalsPAGE_SIZE- 1, the buffer is full in fact. And if the return value equalsPAGE_SIZE, the resulting string was already truncated,or buffer overflow occurred.
Quoted 3 lines
This patch fixes an off-by-one error in
fill_read_buffer.Signed-off-by Miao Xie <[email protected]>
It isn't strictly a bug. If the ->show() op fills full PAGE_SIZE and returns PAGE_SIZE, the user will get full PAGE_SIZE bytes correctly, so it will work. However, considering normal use cases, return value of PAGE_SIZE very likely indicates an error condition, so considering it a BUG condition is a good idea.
Miao, can you please note that the code works fine with PAGE_SIZE return but it's likely to indicate truncated result or overflow in normal use cases as a comment on top of the BUG_ON()?
Thanks.
Signature · 1 line
tejun
Quoted 11 lines · on 2007-11-12 22:08 Tejun Heo wrote:
It isn't strictly a bug. If the ->show() op fills full
PAGE_SIZEand returnsPAGE_SIZE, the user will get fullPAGE_SIZEbytes correctly, so it will work. However, considering normal use cases, return value ofPAGE_SIZEvery likely indicates an error condition, so considering it a BUG condition is a good idea.Miao, can you please note that the code works fine with
PAGE_SIZEreturn but it's likely to indicate truncated result or overflow in normal use cases as a comment on top of theBUG_ON()?Thanks.
OK, I did it.
--- fs/sysfs/file.c | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-)
fs/sysfs/file.c +3 -1
@@ -119,7 +119,9 @@ static int fill_read_buffer(struct dentry * dentry, struct sysfs_buffer * buffersysfs_put_active_two (attr_sd); -BUG_ON (count > (ssize_t )PAGE_SIZE ); +/* the code works fine with PAGE_SIZE return but it's likely to +indicate truncated result or overflow in normal use cases. */ +BUG_ON (count >= (ssize_t )PAGE_SIZE );if (count >=0 ) { buffer->needs_read_fill =0 ; buffer->count = count;
Signature · 1 line
1.5.3
Tejun Heo
Acked-byQuoted 11 lines · Miao Xie
Quoted 10 lines · on 2007-11-12 22:08 Tejun Heo wrote:
It isn't strictly a bug. If the ->show() op fills full
PAGE_SIZEand returnsPAGE_SIZE, the user will get fullPAGE_SIZEbytes correctly, so it will work. However, considering normal use cases, return value ofPAGE_SIZEvery likely indicates an error condition, so considering it a BUG condition is a good idea.Miao, can you please note that the code works fine with
PAGE_SIZEreturn but it's likely to indicate truncated result or overflow in normal use cases as a comment on top of theBUG_ON()?
Quoted 5 lines
Quoted 1 line
Thanks.
OK, I did it.
Signed-off-by Miao Xie <[email protected]>
Greg, can you please push this one? Thanks.
Signature · 1 line
tejun