On Fri, Jul 23, 2021 at 12:43 AM Jiri Slaby [off-list ref] wrote:

When a tty driver pointer is used as a return value of struct
console's device() hook, don't store a semi-state into global variable
which holds the tty driver. It could mean console::device() would return
a bogus value. This is important esp. after the next patch where we
switch from alloc_tty_driver to tty_alloc_driver. tty_alloc_driver
returns ERR_PTR in case of error and that might have unexpected results
as the code doesn't expect this.

Signed-off-by: Jiri Slaby <redacted>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Helge Deller <deller@gmx.de>
Cc: Chris Zankel <chris@zankel.net>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Laurentiu Tudor <laurentiu.tudor@nxp.com>
Cc: Felipe Balbi <balbi@kernel.org>
---
 arch/m68k/emu/nfcon.c                  | 27 +++++++++++---------
 arch/parisc/kernel/pdc_cons.c          | 28 +++++++++++----------
 arch/xtensa/platforms/iss/console.c    | 33 +++++++++++++-----------
 drivers/tty/amiserial.c                | 35 ++++++++++++++------------
 drivers/tty/ehv_bytechan.c             | 28 ++++++++++++---------
 drivers/tty/hvc/hvsi.c                 | 35 ++++++++++++++------------
 drivers/usb/gadget/function/u_serial.c | 32 ++++++++++++-----------
 7 files changed, 119 insertions(+), 99 deletions(-)

Reviewed-by: Max Filippov <jcmvbkbc@gmail.com>

-- 
Thanks.
-- Max