]> git.baikalelectronics.ru Git - kernel.git/commit
usb: usbtmc: Fix bug in pipe direction for control transfers
authorAlan Stern <stern@rowland.harvard.edu>
Thu, 3 Mar 2022 21:00:17 +0000 (16:00 -0500)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 15 Mar 2022 17:45:31 +0000 (18:45 +0100)
commitdd7054c333f63ec984c806d64018a1a5e31af4c6
tree79b2a8d1e782d2841fdd8ac88390add13b98651a
parent6145b99ec7e43d3f18258507961ea6dde246630a
usb: usbtmc: Fix bug in pipe direction for control transfers

The syzbot fuzzer reported a minor bug in the usbtmc driver:

usb 5-1: BOGUS control dir, pipe 80001e80 doesn't match bRequestType 0
WARNING: CPU: 0 PID: 3813 at drivers/usb/core/urb.c:412
usb_submit_urb+0x13a5/0x1970 drivers/usb/core/urb.c:410
Modules linked in:
CPU: 0 PID: 3813 Comm: syz-executor122 Not tainted
5.17.0-rc5-syzkaller-00306-g2293be58d6a1 #0
...
Call Trace:
 <TASK>
 usb_start_wait_urb+0x113/0x530 drivers/usb/core/message.c:58
 usb_internal_control_msg drivers/usb/core/message.c:102 [inline]
 usb_control_msg+0x2a5/0x4b0 drivers/usb/core/message.c:153
 usbtmc_ioctl_request drivers/usb/class/usbtmc.c:1947 [inline]

The problem is that usbtmc_ioctl_request() uses usb_rcvctrlpipe() for
all of its transfers, whether they are in or out.  It's easy to fix.

CC: <stable@vger.kernel.org>
Reported-and-tested-by: syzbot+a48e3d1a875240cab5de@syzkaller.appspotmail.com
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Link: https://lore.kernel.org/r/YiEsYTPEE6lOCOA5@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/usb/class/usbtmc.c