[Linux-kernel-mentees] [PATCH v2] net: usb: rtl8150: don't incorrectly assign random MAC addresses
joe at perches.com
Sun Oct 11 20:14:40 UTC 2020
On Sun, 2020-10-11 at 22:31 +0300, Petko Manolov wrote:
> On 20-10-11 11:33:00, Joe Perches wrote:
> > On Sun, 2020-10-11 at 10:59 -0700, Jakub Kicinski wrote:
> > > On Sun, 11 Oct 2020 23:00:30 +0530 Anant Thazhemadam wrote:
> > > > In set_ethernet_addr(), if get_registers() succeeds, the ethernet address
> > > > that was read must be copied over. Otherwise, a random ethernet address
> > > > must be assigned.
> > > >
> > > > get_registers() returns 0 if successful, and negative error number
> > > > otherwise. However, in set_ethernet_addr(), this return value is
> > > > incorrectly checked.
> > > >
> > > > Since this return value will never be equal to sizeof(node_id), a
> > > > random MAC address will always be generated and assigned to the
> > > > device; even in cases when get_registers() is successful.
> > > >
> > > > Correctly modifying the condition that checks if get_registers() was
> > > > successful or not fixes this problem, and copies the ethernet address
> > > > appropriately.
> > There are many unchecked uses of set_registers and get_registers
> > in this file.
> > If failures are really expected, then it might be better to fix
> > them up too.
> Checking the return value of each get/set_registers() is going to be a PITA and
> not very helpful. Doing so when setting the MAC address _does_ make sense as in
> that case it is not a hard error.
> In almost all other occasions if usb_control_msg_send/recv() return an error i'd
> rather dump an error message (from within get/set_registers()) and let the user
> decide whether to get rid of this adapter or start debugging it.
Your code, your choices...
Consider using _once or _ratelimited output too.
More information about the Linux-kernel-mentees