HID: i2c-hid: allocate hid buffers for real worst case

Commit Message

Dmitry Torokhov Sept. 8, 2017, 5:55 p.m. UTC

From: Adrian Salido <[e-mail]>

The buffer allocation is not currently accounting for an extra byte for
the report id. This can cause an out of bounds access in function
i2c_hid_set_or_send_report() with reportID > 15.

Signed-off-by: Guenter Roeck <[e-mail]>
Signed-off-by: Dmitry Torokhov <[e-mail]>
---
 drivers/hid/i2c-hid/i2c-hid.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

Jiri Kosina Sept. 13, 2017, 2:02 p.m. UTC | #1

On Fri, 8 Sep 2017, Dmitry Torokhov wrote:

> From: Adrian Salido <[e-mail]>
> 
> The buffer allocation is not currently accounting for an extra byte for
> the report id. This can cause an out of bounds access in function
> i2c_hid_set_or_send_report() with reportID > 15.
> 
> Signed-off-by: Guenter Roeck <[e-mail]>
> Signed-off-by: Dmitry Torokhov <[e-mail]>

Missing signoff from the patch author?

Also, I think this should have Cc: stable, right?

Thanks,

Dmitry Torokhov Sept. 13, 2017, 2:49 p.m. UTC | #2

On Wed, Sep 13, 2017 at 07:02:05AM -0700, Jiri Kosina wrote:
> On Fri, 8 Sep 2017, Dmitry Torokhov wrote:
> 
> > From: Adrian Salido <[e-mail]>
> > 
> > The buffer allocation is not currently accounting for an extra byte for
> > the report id. This can cause an out of bounds access in function
> > i2c_hid_set_or_send_report() with reportID > 15.
> > 
> > Signed-off-by: Guenter Roeck <[e-mail]>
> > Signed-off-by: Dmitry Torokhov <[e-mail]>
> 
> Missing signoff from the patch author?

Oops, I must have cut it off on accident while removing ChromeOS
specific tags, the original commit is here:

https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/475212

> 
> Also, I think this should have Cc: stable, right?

I usually let maintainers decide, but yes.

Thanks.

Jiri Kosina Sept. 13, 2017, 2:52 p.m. UTC | #3

On Wed, 13 Sep 2017, Dmitry Torokhov wrote:

> > > From: Adrian Salido <[e-mail]>
> > > 
> > > The buffer allocation is not currently accounting for an extra byte for
> > > the report id. This can cause an out of bounds access in function
> > > i2c_hid_set_or_send_report() with reportID > 15.
> > > 
> > > Signed-off-by: Guenter Roeck <[e-mail]>
> > > Signed-off-by: Dmitry Torokhov <[e-mail]>
> > 
> > Missing signoff from the patch author?
> 
> Oops, I must have cut it off on accident while removing ChromeOS
> specific tags, the original commit is here:
> 
> https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/475212

Ok, thanks, will use that one. How about 

	Reviewed-by: Benson Leung <[e-mail]>

which is missing in the mail you've sent, but is there in the above 
reference commit?

> > Also, I think this should have Cc: stable, right?
> 
> I usually let maintainers decide, but yes.

I'll be adding it. Thanks,

Benson Leung Sept. 13, 2017, 3 p.m. UTC | #4

Hi Jiri,

On Wed, Sep 13, 2017 at 07:52:20AM -0700, Jiri Kosina wrote:
> > Oops, I must have cut it off on accident while removing ChromeOS
> > specific tags, the original commit is here:
> > 
> > https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/475212
> 
> Ok, thanks, will use that one. How about 
> 
> 	Reviewed-by: Benson Leung <[e-mail]>
> 
> which is missing in the mail you've sent, but is there in the above 
> reference commit?

Submission looks good to me. Go ahead and add.
Reviewed-by: Benson Leung <[e-mail]>

Thanks,
Benson

Jiri Kosina Sept. 13, 2017, 4:18 p.m. UTC | #5

On Fri, 8 Sep 2017, Dmitry Torokhov wrote:

> From: Adrian Salido <[e-mail]>
> 
> The buffer allocation is not currently accounting for an extra byte for
> the report id. This can cause an out of bounds access in function
> i2c_hid_set_or_send_report() with reportID > 15.
> 
> Signed-off-by: Guenter Roeck <[e-mail]>
> Signed-off-by: Dmitry Torokhov <[e-mail]>

I've added the missing tags and applied to for-4.14/upstream-fixes

Patch

diff --git a/drivers/hid/i2c-hid/i2c-hid.c b/drivers/hid/i2c-hid/i2c-hid.c
index 77396145d2d0..9145c2129a96 100644
--- a/drivers/hid/i2c-hid/i2c-hid.c
+++ b/drivers/hid/i2c-hid/i2c-hid.c
@@ -543,7 +543,8 @@  static int i2c_hid_alloc_buffers(struct i2c_hid *ihid, size_t report_size)
 {
 	/* the worst case is computed from the set_report command with a
 	 * reportID > 15 and the maximum report length */
-	int args_len = sizeof(__u8) + /* optional ReportID byte */
+	int args_len = sizeof(__u8) + /* ReportID */
+		       sizeof(__u8) + /* optional ReportID byte */
 		       sizeof(__u16) + /* data register */
 		       sizeof(__u16) + /* size of the report */
 		       report_size; /* report */