pull-request: bluetooth 2023-10-11

pull-request: bluetooth 2023-10-11

[Luiz Augusto von Dentz]

The following changes since commit a950a5921db450c74212327f69950ff03419483a:

  net/smc: Fix pos miscalculation in statistics (2023-10-11 10:36:35 +0100)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth.git tags/for-net-2023-10-11

for you to fetch changes up to fc11ae648f03e97d00853d2ea045c909e64d0f36:

  Bluetooth: hci_sock: Correctly bounds check and pad HCI_MON_NEW_INDEX name (2023-10-11 11:19:11 -0700)

----------------------------------------------------------------
bluetooth pull request for net:

 - Fix race when opening vhci device
 - Avoid memcmp() out of bounds warning
 - Correctly bounds check and pad HCI_MON_NEW_INDEX name
 - Fix using memcmp when comparing keys
 - Ignore error return for hci_devcd_register() in btrtl
 - Always check if connection is alive before deleting
 - Fix a refcnt underflow problem for hci_conn

----------------------------------------------------------------
Arkadiusz Bokowy (1):
      Bluetooth: vhci: Fix race when opening vhci device

Arnd Bergmann (1):
      Bluetooth: avoid memcmp() out of bounds warning

Edward AD (1):
      Bluetooth: hci_sock: fix slab oob read in create_monitor_event

Iulia Tanasescu (1):
      Bluetooth: ISO: Fix invalid context error

Kees Cook (1):
      Bluetooth: hci_sock: Correctly bounds check and pad HCI_MON_NEW_INDEX name

Lee, Chun-Yi (2):
      Bluetooth: hci_event: Ignore NULL link key
      Bluetooth: Reject connection with the device which has same BD_ADDR

Luiz Augusto von Dentz (2):
      Bluetooth: hci_event: Fix using memcmp when comparing keys
      Bluetooth: hci_event: Fix coding style

Max Chou (1):
      Bluetooth: btrtl: Ignore error return for hci_devcd_register()

Pauli Virtanen (1):
      Bluetooth: hci_sync: always check if connection is alive before deleting

Ziyang Xuan (1):
      Bluetooth: Fix a refcnt underflow problem for hci_conn

 drivers/bluetooth/btrtl.c       | 10 +++------
 drivers/bluetooth/hci_vhci.c    |  3 +++
 include/net/bluetooth/hci_mon.h |  2 +-
 net/bluetooth/hci_conn.c        |  9 ++++++++
 net/bluetooth/hci_event.c       | 48 +++++++++++++++++++++++++++++++++--------
 net/bluetooth/hci_sock.c        |  3 ++-
 net/bluetooth/hci_sync.c        | 26 +++++++++++-----------
 7 files changed, 69 insertions(+), 32 deletions(-)

Re: pull-request: bluetooth 2023-10-11

[Jakub Kicinski]

On Wed, 11 Oct 2023 12:15:24 -0700 Luiz Augusto von Dentz wrote:
>  - Fix race when opening vhci device
>  - Avoid memcmp() out of bounds warning
>  - Correctly bounds check and pad HCI_MON_NEW_INDEX name
>  - Fix using memcmp when comparing keys
>  - Ignore error return for hci_devcd_register() in btrtl
>  - Always check if connection is alive before deleting
>  - Fix a refcnt underflow problem for hci_conn

Commit: fc11ae648f03 ("Bluetooth: hci_sock: Correctly bounds check and pad HCI_MON_NEW_INDEX name")
	Fixes tag: Fixes: 78480de55a96 ("Bluetooth: hci_sock: fix slab oob read in create_monitor_event")
	Has these problem(s):
		- Target SHA1 does not exist

Commit: 6f0ff718ed67 ("Bluetooth: avoid memcmp() out of bounds warning")
	Fixes tag: Fixes: d70e44fef8621 ("Bluetooth: Reject connection with the device which has same BD_ADDR")
	Has these problem(s):
		- Target SHA1 does not exist

Commit: b03f32b195df ("Bluetooth: hci_event: Fix coding style")
	Fixes tag: Fixes: d70e44fef862 ("Bluetooth: Reject connection with the device which has same BD_ADDR")
	Has these problem(s):
		- Target SHA1 does not exist

Commit: a9500f272b3b ("Bluetooth: hci_event: Fix using memcmp when comparing keys")
	Fixes tag: Fixes: fe7a9da4fa54 ("Bluetooth: hci_event: Ignore NULL link key")
	Has these problem(s):
		- Target SHA1 does not exist

:(

Re: pull-request: bluetooth 2023-10-11

[Luiz Augusto von Dentz]

Hi Jakub,

On Fri, Oct 13, 2023 at 5:54 PM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Wed, 11 Oct 2023 12:15:24 -0700 Luiz Augusto von Dentz wrote:
> >  - Fix race when opening vhci device
> >  - Avoid memcmp() out of bounds warning
> >  - Correctly bounds check and pad HCI_MON_NEW_INDEX name
> >  - Fix using memcmp when comparing keys
> >  - Ignore error return for hci_devcd_register() in btrtl
> >  - Always check if connection is alive before deleting
> >  - Fix a refcnt underflow problem for hci_conn
>
> Commit: fc11ae648f03 ("Bluetooth: hci_sock: Correctly bounds check and pad HCI_MON_NEW_INDEX name")
>         Fixes tag: Fixes: 78480de55a96 ("Bluetooth: hci_sock: fix slab oob read in create_monitor_event")
>         Has these problem(s):
>                 - Target SHA1 does not exist
>
> Commit: 6f0ff718ed67 ("Bluetooth: avoid memcmp() out of bounds warning")
>         Fixes tag: Fixes: d70e44fef8621 ("Bluetooth: Reject connection with the device which has same BD_ADDR")
>         Has these problem(s):
>                 - Target SHA1 does not exist
>
> Commit: b03f32b195df ("Bluetooth: hci_event: Fix coding style")
>         Fixes tag: Fixes: d70e44fef862 ("Bluetooth: Reject connection with the device which has same BD_ADDR")
>         Has these problem(s):
>                 - Target SHA1 does not exist
>
> Commit: a9500f272b3b ("Bluetooth: hci_event: Fix using memcmp when comparing keys")
>         Fixes tag: Fixes: fe7a9da4fa54 ("Bluetooth: hci_event: Ignore NULL link key")
>         Has these problem(s):
>                 - Target SHA1 does not exist
>
> :(

Not sure what I'm doing wrong, because verify_fixes.sh doesn't
actually warn these for me, or perhaps it needs to be used in a clean
tree/new clone since it may match commit ids of different
branchs/remotes? Anyway, I'm preparing a new pull-request.

-- 
Luiz Augusto von Dentz

Re: pull-request: bluetooth 2023-10-11

[Paolo Abeni]

On Fri, 2023-10-13 at 20:12 -0700, Luiz Augusto von Dentz wrote:
> Hi Jakub,
> 
> On Fri, Oct 13, 2023 at 5:54 PM Jakub Kicinski <kuba@kernel.org> wrote:
> > 
> > On Wed, 11 Oct 2023 12:15:24 -0700 Luiz Augusto von Dentz wrote:
> > >  - Fix race when opening vhci device
> > >  - Avoid memcmp() out of bounds warning
> > >  - Correctly bounds check and pad HCI_MON_NEW_INDEX name
> > >  - Fix using memcmp when comparing keys
> > >  - Ignore error return for hci_devcd_register() in btrtl
> > >  - Always check if connection is alive before deleting
> > >  - Fix a refcnt underflow problem for hci_conn
> > 
> > Commit: fc11ae648f03 ("Bluetooth: hci_sock: Correctly bounds check and pad HCI_MON_NEW_INDEX name")
> >         Fixes tag: Fixes: 78480de55a96 ("Bluetooth: hci_sock: fix slab oob read in create_monitor_event")
> >         Has these problem(s):
> >                 - Target SHA1 does not exist
> > 
> > Commit: 6f0ff718ed67 ("Bluetooth: avoid memcmp() out of bounds warning")
> >         Fixes tag: Fixes: d70e44fef8621 ("Bluetooth: Reject connection with the device which has same BD_ADDR")
> >         Has these problem(s):
> >                 - Target SHA1 does not exist
> > 
> > Commit: b03f32b195df ("Bluetooth: hci_event: Fix coding style")
> >         Fixes tag: Fixes: d70e44fef862 ("Bluetooth: Reject connection with the device which has same BD_ADDR")
> >         Has these problem(s):
> >                 - Target SHA1 does not exist
> > 
> > Commit: a9500f272b3b ("Bluetooth: hci_event: Fix using memcmp when comparing keys")
> >         Fixes tag: Fixes: fe7a9da4fa54 ("Bluetooth: hci_event: Ignore NULL link key")
> >         Has these problem(s):
> >                 - Target SHA1 does not exist
> > 
> > :(
> 
> Not sure what I'm doing wrong, because verify_fixes.sh doesn't
> actually warn these for me, or perhaps it needs to be used in a clean
> tree/new clone since it may match commit ids of different
> branchs/remotes? 

Indeed verify_fixes will match any commit in your tree. I personally
stumbled upon similar issues while doing backports - when references to
different trees converged in my repo.

One things that did help me is using a specific, clean, clone with a
single remote for all the netdev patches.

Cheers,

Paolo

> Anyway, I'm preparing a new pull-request.