pull request: bluetooth-next 2023-06-27

pull request: bluetooth-next 2023-06-27

[Luiz Augusto von Dentz]

The following changes since commit ae230642190a51b85656d6da2df744d534d59544:

  Merge branch 'af_unix-followup-fixes-for-so_passpidfd' (2023-06-27 10:50:25 -0700)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git tags/for-net-next-2023-06-27

for you to fetch changes up to e63d8ed98082395ca509163f386f5b04f53872b3:

  Bluetooth: msft: Extended monitor tracking by address filter (2023-06-27 11:52:58 -0700)

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

 - Add Reialtek devcoredump support
 - Add support for device 6655:8771
 - Add extended monitor tracking by address filter
 - Add support for connecting multiple BISes
 - Add support to reset via ACPI DSM for Intel controllers
 - Add support for MT7922 used in Asus Ally
 - Add support Mediatek MT7925
 - Fixes for use-after-free in L2CAP

----------------------------------------------------------------
Claudia Draghicescu (1):
      Bluetooth: Check for ISO support in controller

Dan Gora (2):
      Bluetooth: btrtl: Add missing MODULE_FIRMWARE declarations
      Bluetooth: btusb: Add device 6655:8771 to device tables

Hilda Wu (2):
      Bluetooth: btrtl: Add Realtek devcoredump support
      Bluetooth: msft: Extended monitor tracking by address filter

Iulia Tanasescu (2):
      Bluetooth: ISO: Add support for connecting multiple BISes
      Bluetooth: ISO: Support multiple BIGs

Ivan Orlov (1):
      Bluetooth: hci_sysfs: make bt_class a static const structure

Jiapeng Chong (1):
      Bluetooth: hci_conn: Use kmemdup() to replace kzalloc + memcpy

Johan Hovold (3):
      Bluetooth: fix invalid-bdaddr quirk for non-persistent setup
      Bluetooth: fix use-bdaddr-property quirk
      Bluetooth: hci_bcm: do not mark valid bd_addr as invalid

Kiran K (1):
      Bluetooth: btintel: Add support to reset bluetooth via ACPI DSM

Luiz Augusto von Dentz (6):
      Bluetooth: Consolidate code around sk_alloc into a helper function
      Bluetooth: Init sk_peer_* on bt_sock_alloc
      Bluetooth: hci_sock: Forward credentials to monitor
      Bluetooth: MGMT: Fix marking SCAN_RSP as not connectable
      Bluetooth: ISO: Rework sync_interval to be sync_factor
      Bluetooth: hci_event: Fix parsing of CIS Established Event

Matthew Anderson (1):
      Bluetooth: btusb: Add MT7922 bluetooth ID for the Asus Ally

Max Chou (1):
      Bluetooth: btrtl: Correct the length of the HCI command for drop fw

Min-Hua Chen (1):
      Bluetooth: btqca: use le32_to_cpu for ver.soc_id

Pauli Virtanen (6):
      Bluetooth: ISO: use hci_sync for setting CIG parameters
      Bluetooth: ISO: do not emit new LE Create CIS if previous is pending
      Bluetooth: hci_event: fix Set CIG Parameters error status handling
      Bluetooth: use RCU for hci_conn_params and iterate safely in hci_sync
      Bluetooth: hci_event: call disconnect callback before deleting conn
      Bluetooth: ISO: fix iso_conn related locking and validity issues

Peter Tsao (1):
      Bluetooth: btusb: Add support Mediatek MT7925

Sai Teja Aluvala (2):
      Bluetooth: hci_qca: Add qcom devcoredump sysfs support
      Bluetooth: hci_qca: Add qcom devcoredump support

Sungwoo Kim (1):
      Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb

Zhengping Jiang (1):
      Bluetooth: L2CAP: Fix use-after-free

 drivers/bluetooth/btintel.c       | 121 +++++++++++
 drivers/bluetooth/btintel.h       |   2 +
 drivers/bluetooth/btmtk.c         |   1 +
 drivers/bluetooth/btmtk.h         |   5 +
 drivers/bluetooth/btqca.c         |   2 +-
 drivers/bluetooth/btrtl.c         | 181 +++++++++++++----
 drivers/bluetooth/btrtl.h         |  13 ++
 drivers/bluetooth/btusb.c         | 173 ++++++++++++++--
 drivers/bluetooth/hci_bcm.c       |   3 +-
 drivers/bluetooth/hci_qca.c       | 150 +++++++++-----
 include/net/bluetooth/bluetooth.h |   5 +-
 include/net/bluetooth/hci.h       |  11 +
 include/net/bluetooth/hci_core.h  |  45 ++++-
 include/net/bluetooth/hci_sync.h  |   2 +-
 include/net/bluetooth/mgmt.h      |   3 +
 net/bluetooth/af_bluetooth.c      |  45 +++++
 net/bluetooth/bnep/sock.c         |  10 +-
 net/bluetooth/hci_conn.c          | 315 +++++++++++++++++------------
 net/bluetooth/hci_core.c          |  38 +++-
 net/bluetooth/hci_event.c         | 190 ++++++++++++------
 net/bluetooth/hci_sock.c          |  77 ++++++-
 net/bluetooth/hci_sync.c          | 265 +++++++++++++++++-------
 net/bluetooth/hci_sysfs.c         |  14 +-
 net/bluetooth/hidp/sock.c         |  10 +-
 net/bluetooth/iso.c               |  97 +++++----
 net/bluetooth/l2cap_core.c        |   5 +
 net/bluetooth/l2cap_sock.c        |  31 +--
 net/bluetooth/mgmt.c              |  32 +--
 net/bluetooth/msft.c              | 412 ++++++++++++++++++++++++++++++++++++--
 net/bluetooth/rfcomm/sock.c       |  13 +-
 net/bluetooth/sco.c               |  10 +-
 31 files changed, 1753 insertions(+), 528 deletions(-)

Re: pull request: bluetooth-next 2023-06-27

[Jakub Kicinski]

On Tue, 27 Jun 2023 12:10:04 -0700 Luiz Augusto von Dentz wrote:
> bluetooth-next pull request for net-next:
> 
>  - Add Reialtek devcoredump support
>  - Add support for device 6655:8771
>  - Add extended monitor tracking by address filter
>  - Add support for connecting multiple BISes
>  - Add support to reset via ACPI DSM for Intel controllers
>  - Add support for MT7922 used in Asus Ally
>  - Add support Mediatek MT7925
>  - Fixes for use-after-free in L2CAP

As you probably realized these came in a little late for our main pull
request for this merge window. Can we cut this down a little bit?
Stick to the fixes and changes which you have the most confidence in
and try to keep the new lines under 1k LoC?

I had a look thru and these changes look like stuff we can definitely
pull:

 a8d0b0440b7f ("Bluetooth: btrtl: Add missing MODULE_FIRMWARE declarations")
 349cae7e8d84 ("Bluetooth: btusb: Add device 6655:8771 to device tables")
 afdbe6303877 ("Bluetooth: btqca: use le32_to_cpu for ver.soc_id")
 d1b10da77355 ("Bluetooth: L2CAP: Fix use-after-free")
 c1121a116d5f ("Bluetooth: fix invalid-bdaddr quirk for non-persistent setup")
 2f8b38e5eba4 ("Bluetooth: fix use-bdaddr-property quirk")
 317af9ba6fff ("Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb")
 a6cfe4261f5e ("Bluetooth: hci_bcm: do not mark valid bd_addr as invalid")
 20b3370a6bfb ("Bluetooth: ISO: use hci_sync for setting CIG parameters")
 29a3b409a3f2 ("Bluetooth: hci_event: fix Set CIG Parameters error status handling")
 48d15256595b ("Bluetooth: MGMT: Fix marking SCAN_RSP as not connectable")
 f145eeb779c3 ("Bluetooth: ISO: Rework sync_interval to be sync_factor")
 0d39e82e1a7b ("Bluetooth: hci_sysfs: make bt_class a static const structure")
 8649851b1945 ("Bluetooth: hci_event: Fix parsing of CIS Established Event")
 5b611951e075 ("Bluetooth: btusb: Add MT7922 bluetooth ID for the Asus Ally")
 00b51ce9f603 ("Bluetooth: hci_conn: Use kmemdup() to replace kzalloc + memcpy")

You can throw in a few more things you think are important and are
unlikely to cause regressions.

Re: pull request: bluetooth-next 2023-06-27

[Luiz Augusto von Dentz]

Hi Jakub,

On Wed, Jun 28, 2023 at 7:38 PM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Tue, 27 Jun 2023 12:10:04 -0700 Luiz Augusto von Dentz wrote:
> > bluetooth-next pull request for net-next:
> >
> >  - Add Reialtek devcoredump support
> >  - Add support for device 6655:8771
> >  - Add extended monitor tracking by address filter
> >  - Add support for connecting multiple BISes
> >  - Add support to reset via ACPI DSM for Intel controllers
> >  - Add support for MT7922 used in Asus Ally
> >  - Add support Mediatek MT7925
> >  - Fixes for use-after-free in L2CAP
>
> As you probably realized these came in a little late for our main pull
> request for this merge window. Can we cut this down a little bit?
> Stick to the fixes and changes which you have the most confidence in
> and try to keep the new lines under 1k LoC?

Yeah, sorry about that I had a business trip which sort of messed up
the pull request.

> I had a look thru and these changes look like stuff we can definitely
> pull:
>
>  a8d0b0440b7f ("Bluetooth: btrtl: Add missing MODULE_FIRMWARE declarations")
>  349cae7e8d84 ("Bluetooth: btusb: Add device 6655:8771 to device tables")
>  afdbe6303877 ("Bluetooth: btqca: use le32_to_cpu for ver.soc_id")
>  d1b10da77355 ("Bluetooth: L2CAP: Fix use-after-free")
>  c1121a116d5f ("Bluetooth: fix invalid-bdaddr quirk for non-persistent setup")
>  2f8b38e5eba4 ("Bluetooth: fix use-bdaddr-property quirk")
>  317af9ba6fff ("Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb")
>  a6cfe4261f5e ("Bluetooth: hci_bcm: do not mark valid bd_addr as invalid")
>  20b3370a6bfb ("Bluetooth: ISO: use hci_sync for setting CIG parameters")
>  29a3b409a3f2 ("Bluetooth: hci_event: fix Set CIG Parameters error status handling")
>  48d15256595b ("Bluetooth: MGMT: Fix marking SCAN_RSP as not connectable")
>  f145eeb779c3 ("Bluetooth: ISO: Rework sync_interval to be sync_factor")
>  0d39e82e1a7b ("Bluetooth: hci_sysfs: make bt_class a static const structure")
>  8649851b1945 ("Bluetooth: hci_event: Fix parsing of CIS Established Event")
>  5b611951e075 ("Bluetooth: btusb: Add MT7922 bluetooth ID for the Asus Ally")
>  00b51ce9f603 ("Bluetooth: hci_conn: Use kmemdup() to replace kzalloc + memcpy")
>
> You can throw in a few more things you think are important and are
> unlikely to cause regressions.

Yeah, those seem to be the most important ones, do you want me to redo
the pull-request or perhaps you can just cherry-pick them?

-- 
Luiz Augusto von Dentz

Re: pull request: bluetooth-next 2023-06-27

[Jakub Kicinski]

On Wed, 28 Jun 2023 22:01:05 -0700 Luiz Augusto von Dentz wrote:
> >  a8d0b0440b7f ("Bluetooth: btrtl: Add missing MODULE_FIRMWARE declarations")
> >  349cae7e8d84 ("Bluetooth: btusb: Add device 6655:8771 to device tables")
> >  afdbe6303877 ("Bluetooth: btqca: use le32_to_cpu for ver.soc_id")
> >  d1b10da77355 ("Bluetooth: L2CAP: Fix use-after-free")
> >  c1121a116d5f ("Bluetooth: fix invalid-bdaddr quirk for non-persistent setup")
> >  2f8b38e5eba4 ("Bluetooth: fix use-bdaddr-property quirk")
> >  317af9ba6fff ("Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb")
> >  a6cfe4261f5e ("Bluetooth: hci_bcm: do not mark valid bd_addr as invalid")
> >  20b3370a6bfb ("Bluetooth: ISO: use hci_sync for setting CIG parameters")
> >  29a3b409a3f2 ("Bluetooth: hci_event: fix Set CIG Parameters error status handling")
> >  48d15256595b ("Bluetooth: MGMT: Fix marking SCAN_RSP as not connectable")
> >  f145eeb779c3 ("Bluetooth: ISO: Rework sync_interval to be sync_factor")
> >  0d39e82e1a7b ("Bluetooth: hci_sysfs: make bt_class a static const structure")
> >  8649851b1945 ("Bluetooth: hci_event: Fix parsing of CIS Established Event")
> >  5b611951e075 ("Bluetooth: btusb: Add MT7922 bluetooth ID for the Asus Ally")
> >  00b51ce9f603 ("Bluetooth: hci_conn: Use kmemdup() to replace kzalloc + memcpy")
> >
> > You can throw in a few more things you think are important and are
> > unlikely to cause regressions.  
> 
> Yeah, those seem to be the most important ones, do you want me to redo
> the pull-request or perhaps you can just cherry-pick them?

Nothing to add to that list?
Let me see if I can cherry-pick them cleanly.

Re: pull request: bluetooth-next 2023-06-27

[Jakub Kicinski]

On Thu, 29 Jun 2023 08:22:41 -0700 Jakub Kicinski wrote:
> On Wed, 28 Jun 2023 22:01:05 -0700 Luiz Augusto von Dentz wrote:
> > >  a8d0b0440b7f ("Bluetooth: btrtl: Add missing MODULE_FIRMWARE declarations")
> > >  349cae7e8d84 ("Bluetooth: btusb: Add device 6655:8771 to device tables")
> > >  afdbe6303877 ("Bluetooth: btqca: use le32_to_cpu for ver.soc_id")
> > >  d1b10da77355 ("Bluetooth: L2CAP: Fix use-after-free")
> > >  c1121a116d5f ("Bluetooth: fix invalid-bdaddr quirk for non-persistent setup")
> > >  2f8b38e5eba4 ("Bluetooth: fix use-bdaddr-property quirk")
> > >  317af9ba6fff ("Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb")
> > >  a6cfe4261f5e ("Bluetooth: hci_bcm: do not mark valid bd_addr as invalid")
> > >  20b3370a6bfb ("Bluetooth: ISO: use hci_sync for setting CIG parameters")
> > >  29a3b409a3f2 ("Bluetooth: hci_event: fix Set CIG Parameters error status handling")
> > >  48d15256595b ("Bluetooth: MGMT: Fix marking SCAN_RSP as not connectable")
> > >  f145eeb779c3 ("Bluetooth: ISO: Rework sync_interval to be sync_factor")
> > >  0d39e82e1a7b ("Bluetooth: hci_sysfs: make bt_class a static const structure")
> > >  8649851b1945 ("Bluetooth: hci_event: Fix parsing of CIS Established Event")
> > >  5b611951e075 ("Bluetooth: btusb: Add MT7922 bluetooth ID for the Asus Ally")
> > >  00b51ce9f603 ("Bluetooth: hci_conn: Use kmemdup() to replace kzalloc + memcpy")
> > >
> > > You can throw in a few more things you think are important and are
> > > unlikely to cause regressions.    
> > 
> > Yeah, those seem to be the most important ones, do you want me to redo
> > the pull-request or perhaps you can just cherry-pick them?  
> 
> Nothing to add to that list?
> Let me see if I can cherry-pick them cleanly.

I pushed these to net now, hopefully I didn't mess it up :)

Re: pull request: bluetooth-next 2023-06-27

[Luiz Augusto von Dentz]

Hi Jakub,

On Thu, Jun 29, 2023 at 10:59 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Thu, 29 Jun 2023 08:22:41 -0700 Jakub Kicinski wrote:
> > On Wed, 28 Jun 2023 22:01:05 -0700 Luiz Augusto von Dentz wrote:
> > > >  a8d0b0440b7f ("Bluetooth: btrtl: Add missing MODULE_FIRMWARE declarations")
> > > >  349cae7e8d84 ("Bluetooth: btusb: Add device 6655:8771 to device tables")
> > > >  afdbe6303877 ("Bluetooth: btqca: use le32_to_cpu for ver.soc_id")
> > > >  d1b10da77355 ("Bluetooth: L2CAP: Fix use-after-free")
> > > >  c1121a116d5f ("Bluetooth: fix invalid-bdaddr quirk for non-persistent setup")
> > > >  2f8b38e5eba4 ("Bluetooth: fix use-bdaddr-property quirk")
> > > >  317af9ba6fff ("Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_ready_cb")
> > > >  a6cfe4261f5e ("Bluetooth: hci_bcm: do not mark valid bd_addr as invalid")
> > > >  20b3370a6bfb ("Bluetooth: ISO: use hci_sync for setting CIG parameters")
> > > >  29a3b409a3f2 ("Bluetooth: hci_event: fix Set CIG Parameters error status handling")
> > > >  48d15256595b ("Bluetooth: MGMT: Fix marking SCAN_RSP as not connectable")
> > > >  f145eeb779c3 ("Bluetooth: ISO: Rework sync_interval to be sync_factor")
> > > >  0d39e82e1a7b ("Bluetooth: hci_sysfs: make bt_class a static const structure")
> > > >  8649851b1945 ("Bluetooth: hci_event: Fix parsing of CIS Established Event")
> > > >  5b611951e075 ("Bluetooth: btusb: Add MT7922 bluetooth ID for the Asus Ally")
> > > >  00b51ce9f603 ("Bluetooth: hci_conn: Use kmemdup() to replace kzalloc + memcpy")
> > > >
> > > > You can throw in a few more things you think are important and are
> > > > unlikely to cause regressions.
> > >
> > > Yeah, those seem to be the most important ones, do you want me to redo
> > > the pull-request or perhaps you can just cherry-pick them?
> >
> > Nothing to add to that list?
> > Let me see if I can cherry-pick them cleanly.
>
> I pushed these to net now, hopefully I didn't mess it up :)

Great, thanks. I guess I will change the frequency we do pull request
to net-next going forward, perhaps something doing it
bi-weekly/monthly would be better to avoid risking missing the merge
window if that happens to conflict with some event, etc.

-- 
Luiz Augusto von Dentz

Re: pull request: bluetooth-next 2023-06-27

[Jakub Kicinski]

On Thu, 29 Jun 2023 11:34:34 -0700 Luiz Augusto von Dentz wrote:
> > > Nothing to add to that list?
> > > Let me see if I can cherry-pick them cleanly.  
> >
> > I pushed these to net now, hopefully I didn't mess it up :)  
> 
> Great, thanks. I guess I will change the frequency we do pull request
> to net-next going forward, perhaps something doing it
> bi-weekly/monthly would be better to avoid risking missing the merge
> window if that happens to conflict with some event, etc.

Maybe every 3 weeks would be optimal? Basically the week after -rc3,
the week after -rc6 and then a small catch up right before the merge
window if needed? Obviously easier said than done as life tends to 
not align with fixed schedules..

Re: pull request: bluetooth-next 2023-06-27

[Luiz Augusto von Dentz]

Hi Jakub,

On Thu, Jun 29, 2023 at 11:43 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Thu, 29 Jun 2023 11:34:34 -0700 Luiz Augusto von Dentz wrote:
> > > > Nothing to add to that list?
> > > > Let me see if I can cherry-pick them cleanly.
> > >
> > > I pushed these to net now, hopefully I didn't mess it up :)
> >
> > Great, thanks. I guess I will change the frequency we do pull request
> > to net-next going forward, perhaps something doing it
> > bi-weekly/monthly would be better to avoid risking missing the merge
> > window if that happens to conflict with some event, etc.
>
> Maybe every 3 weeks would be optimal? Basically the week after -rc3,
> the week after -rc6 and then a small catch up right before the merge
> window if needed? Obviously easier said than done as life tends to
> not align with fixed schedules..

Sounds great, will put a calendar reminding me to send pull-requests
every 3 weeks.

-- 
Luiz Augusto von Dentz