linux-toradex.git/drivers/bluetooth/hci_ldisc.c, branch v6.3-rc3

Bluetooth: hci_{ldisc,serdev}: check percpu_init_rwsem() failure

2022-09-19T17:33:39+00:00

syzbot is reporting NULL pointer dereference at hci_uart_tty_close() [1],
for rcu_sync_enter() is called without rcu_sync_init() due to
hci_uart_tty_open() ignoring percpu_init_rwsem() failure.

While we are at it, fix that hci_uart_register_device() ignores
percpu_init_rwsem() failure and hci_uart_unregister_device() does not
call percpu_free_rwsem().

Link: https://syzkaller.appspot.com/bug?extid=576dfca25381fb6fbc5f [1]
Reported-by: syzbot 
Signed-off-by: Tetsuo Handa 
Fixes: 67d2f8781b9f00d1 ("Bluetooth: hci_ldisc: Allow sleeping while proto locks are held.")
Fixes: d73e172816652772 ("Bluetooth: hci_serdev: Init hci_uart proto_lock to avoid oops")
Signed-off-by: Luiz Augusto von Dentz

tty: remove file from tty_ldisc_ops::ioctl and compat_ioctl

2021-11-25T17:36:27+00:00

After the previous patches, noone needs 'file' parameter in neither
ioctl hook from tty_ldisc_ops. So remove 'file' from both of them.

Cc: Marcel Holtmann 
Cc: Johan Hedberg 
Cc: Luiz Augusto von Dentz 
Cc: Wolfgang Grandegger 
Cc: Marc Kleine-Budde 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Andreas Koensgen 
Cc: Paul Mackerras 
Acked-by: Krzysztof Kozlowski  [NFC]
Acked-by: Dmitry Torokhov 
Acked-by: Marc Kleine-Budde 
Signed-off-by: Jiri Slaby 
Link: https://lore.kernel.org/r/20211122094529.24171-1-jslaby@suse.cz
Signed-off-by: Greg Kroah-Hartman

Merge tag 'tty-5.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty

2021-11-04T16:09:37+00:00

Pull tty / serial driver updates from Greg KH:
 "Here is the big set of tty and serial driver updates for 5.16-rc1.

  Nothing major in here at all, just lots of tiny serial and tty driver
  updates for various reported things, and some good cleanups. These
  include:

   - more good tty api cleanups from Jiri

   - stm32 serial driver updates

   - softlockup fix for non-preempt systems under high serial load

   - rpmsg serial driver update

   - 8250 drivers updates and fixes

   - n_gsm line discipline fixes and updates as people are finally
     starting to use it.

  All of these have been in linux-next for a while now with no reported
  issues"

* tag 'tty-5.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty: (86 commits)
  tty: Fix extra "not" in TTY_DRIVER_REAL_RAW description
  serial: cpm_uart: Protect udbg definitions by CONFIG_SERIAL_CPM_CONSOLE
  tty: rpmsg: Define tty name via constant string literal
  tty: rpmsg: Add pr_fmt() to prefix messages
  tty: rpmsg: Use dev_err_probe() in ->probe()
  tty: rpmsg: Unify variable used to keep an error code
  tty: rpmsg: Assign returned id to a local variable
  serial: stm32: push DMA RX data before suspending
  serial: stm32: terminate / restart DMA transfer at suspend / resume
  serial: stm32: rework RX dma initialization and release
  serial: 8250_pci: Remove empty stub pci_quatech_exit()
  serial: 8250_pci: Replace custom pci_match_id() implementation
  serial: xilinx_uartps: Fix race condition causing stuck TX
  serial: sunzilog: Mark sunzilog_putchar() __maybe_unused
  Revert "tty: hvc: pass DMA capable memory to put_chars()"
  Revert "virtio-console: remove unnecessary kmemdup()"
  serial: 8250_pci: Replace dev_*() by pci_*() macros
  serial: 8250_pci: Get rid of redundant 'else' keyword
  serial: 8250_pci: Refactor the loop in pci_ite887x_init()
  tty: add rpmsg driver
  ...

tty: remove file from n_tty_ioctl_helper

2021-09-22T14:59:13+00:00

After the previous patch, there are no users of 'file' in
n_tty_ioctl_helper. So remove it also from there.

Cc: Marcel Holtmann 
Cc: Johan Hedberg 
Cc: Luiz Augusto von Dentz 
Cc: Paul Mackerras 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Krzysztof Kozlowski 
Acked-by: Krzysztof Kozlowski 
Signed-off-by: Jiri Slaby 
Link: https://lore.kernel.org/r/20210914091134.17426-6-jslaby@suse.cz
Signed-off-by: Greg Kroah-Hartman

Bluetooth: hci_ldisc: require CAP_NET_ADMIN to attach N_HCI ldisc

2021-09-22T14:14:21+00:00

Any unprivileged user can attach N_HCI ldisc and send packets coming from a
virtual controller by using PTYs.

Require initial namespace CAP_NET_ADMIN to do that.

Signed-off-by: Thadeu Lima de Souza Cascardo 
Signed-off-by: Marcel Holtmann

tty: no checking of tty_unregister_ldisc

2021-05-13T14:57:17+00:00

tty_unregister_ldisc now returns 0 = success. No need to check the
return value. In fact, the users only warned if an error occured and
didn't do anything useful anyway -- the ldisc module was unloaded in any
case.

Signed-off-by: Jiri Slaby 
Cc: William Hubbs 
Cc: Chris Brannon 
Cc: Kirk Reiser 
Cc: Samuel Thibault 
Cc: Marcel Holtmann 
Cc: Johan Hedberg 
Cc: Luiz Augusto von Dentz 
Cc: Arnd Bergmann 
Cc: Wolfgang Grandegger 
Cc: Marc Kleine-Budde 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Andreas Koensgen 
Cc: Paul Mackerras 
Cc: Rodolfo Giometti 
Cc: Peter Ujfalusi 
Cc: Liam Girdwood 
Cc: Mark Brown 
Cc: Jaroslav Kysela 
Cc: Takashi Iwai 
Link: https://lore.kernel.org/r/20210505091928.22010-19-jslaby@suse.cz
Signed-off-by: Greg Kroah-Hartman

tty: make tty_ldisc_ops a param in tty_unregister_ldisc

2021-05-13T14:57:16+00:00

Make tty_unregister_ldisc symmetric to tty_register_ldisc by accepting
struct tty_ldisc_ops as a parameter instead of ldisc number. This avoids
checking of the ldisc number bounds in tty_unregister_ldisc.

Signed-off-by: Jiri Slaby 
Cc: William Hubbs 
Cc: Chris Brannon 
Cc: Kirk Reiser 
Cc: Samuel Thibault 
Cc: Marcel Holtmann 
Cc: Johan Hedberg 
Cc: Luiz Augusto von Dentz 
Cc: Dmitry Torokhov 
Cc: Arnd Bergmann 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Wolfgang Grandegger 
Cc: Marc Kleine-Budde 
Cc: Andreas Koensgen 
Cc: Paul Mackerras 
Cc: Rodolfo Giometti 
Cc: Peter Ujfalusi 
Cc: Liam Girdwood 
Cc: Mark Brown 
Cc: Jaroslav Kysela 
Cc: Takashi Iwai 
Link: https://lore.kernel.org/r/20210505091928.22010-17-jslaby@suse.cz
Signed-off-by: Greg Kroah-Hartman

tty: set tty_ldisc_ops::num statically

2021-05-13T14:57:16+00:00

There is no reason to pass the ldisc number to tty_register_ldisc
separately. Just set it in the already defined tty_ldisc_ops in all the
ldiscs.

This simplifies tty_register_ldisc a bit too (no need to set the num
member there).

Signed-off-by: Jiri Slaby 
Cc: William Hubbs 
Cc: Chris Brannon 
Cc: Kirk Reiser 
Cc: Samuel Thibault 
Cc: Marcel Holtmann 
Cc: Johan Hedberg 
Cc: Luiz Augusto von Dentz 
Cc: Dmitry Torokhov 
Cc: Arnd Bergmann 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Wolfgang Grandegger 
Cc: Marc Kleine-Budde 
Cc: Andreas Koensgen 
Cc: Paul Mackerras 
Cc: Rodolfo Giometti 
Cc: Peter Ujfalusi 
Cc: Liam Girdwood 
Cc: Mark Brown 
Cc: Jaroslav Kysela 
Cc: Takashi Iwai 
Link: https://lore.kernel.org/r/20210505091928.22010-15-jslaby@suse.cz
Signed-off-by: Greg Kroah-Hartman

tty: make fp of tty_ldisc_ops::receive_buf{,2} const

2021-05-13T14:57:16+00:00

Char pointer (cp) passed to tty_ldisc_ops::receive_buf{,2} is const.
There is no reason for flag pointer (fp) not to be too. So switch it in
the definition and all uses.

Signed-off-by: Jiri Slaby 
Cc: William Hubbs 
Cc: Chris Brannon 
Cc: Kirk Reiser 
Cc: Samuel Thibault 
Cc: Marcel Holtmann 
Cc: Johan Hedberg 
Cc: Luiz Augusto von Dentz 
Cc: Dmitry Torokhov 
Cc: Arnd Bergmann 
Cc: "David S. Miller" 
Cc: Jakub Kicinski 
Cc: Wolfgang Grandegger 
Cc: Marc Kleine-Budde 
Cc: Andreas Koensgen 
Cc: Paul Mackerras 
Cc: Liam Girdwood 
Cc: Mark Brown 
Cc: Jaroslav Kysela 
Cc: Takashi Iwai 
Cc: Peter Ujfalusi 
Link: https://lore.kernel.org/r/20210505091928.22010-12-jslaby@suse.cz
Signed-off-by: Greg Kroah-Hartman

tty: remove TTY_LDISC_MAGIC

2021-03-10T08:34:06+00:00

First, it is never checked. Second, use of it as a debugging aid is
at least questionable. With the current tools, I don't think anyone used
this kind of thing for debugging purposes for years.

On the top of that, e.g. serdev does not set this field of tty_ldisc_ops
at all.

So get rid of this legacy.

Signed-off-by: Jiri Slaby 
Link: https://lore.kernel.org/r/20210302062214.29627-8-jslaby@suse.cz
Signed-off-by: Greg Kroah-Hartman