Skip to contentSkip to conversation

Revert "Bluetooth: core: Fix missing power_on work cancel on HCI close"

PATCH In review Max Krummenacher 8 participants 16 messages Jun 14, 2022 – Jul 5, 2022

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Revert "Bluetooth: core: Fix missing power_on work cancel on HCI close"

From: Max Krummenacher <[email protected]>

This reverts commit ff7f2926114d3a50f5ffe461a9bce8d761748da5.

The commit ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close") introduced between v5.18 and v5.19-rc1 makes going to suspend freeze. v5.19-rc2 is equally affected.

This has been seen on a Colibri iMX6ULL WB which has a Marvell 8997 based WiFi / Bluetooth module connected over SDIO.

With 'v5.18' or 'v5.19-rc1 with said commit reverted' a suspend/resume cycle looks as follows and the device is functional after the resume:

root@imx6ull:~# rfkill
ID TYPE      DEVICE    SOFT      HARD
 0 bluetooth hci0   blocked unblocked
 1 wlan      phy0   blocked unblocked
root@imx6ull:~# echo enabled > /sys/class/tty/ttymxc0/power/wakeup
root@imx6ull:~# date;echo mem > /sys/power/state;date
Tue Jun 14 14:43:03 UTC 2022
[ 6393.464497] PM: suspend entry (deep)
[ 6393.529398] Filesystems sync: 0.064 seconds
[ 6393.594006] Freezing user space processes ... (elapsed 0.015 seconds) done.
[ 6393.610266] OOM killer disabled.
[ 6393.610285] Freezing remaining freezable tasks ... (elapsed 0.013 seconds) done.
[ 6393.623727] printk: Suspending console(s) (use no_console_suspend to debug)

~~ suspended until console initiates the resume

[ 6394.023552] fec 20b4000.ethernet eth0: Link is Down
[ 6394.049902] PM: suspend devices took 0.300 seconds
[ 6394.091654] Disabling non-boot CPUs ...
[ 6394.565896] PM: resume devices took 0.440 seconds
[ 6394.681350] OOM killer enabled.
[ 6394.681369] Restarting tasks ... done.
[ 6394.741157] random: crng reseeded on system resumption
[ 6394.813135] PM: suspend exit
Tue Jun 14 14:43:11 UTC 2022
[ 6396.403873] fec 20b4000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
[ 6396.404347] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready

root@imx6ull:~#

With 'v5.19-rc1' suspend freezes in the suspend phase, i.e. power consumption is not lowered and no wakeup source initiates a wakup.

root@imx6ull:~# rfkill
ID TYPE      DEVICE    SOFT      HARD
 0 bluetooth hci0   blocked unblocked
 1 wlan      phy0   blocked unblocked
root@imx6ull:~# echo enabled > /sys/class/tty/ttymxc0/power/wakeup
root@imx6ull:~# date;echo mem > /sys/power/state;date
Tue Jun 14 12:40:38 UTC 2022
[  122.476333] PM: suspend entry (deep)
[  122.556012] Filesystems sync: 0.079 seconds

~~ no further kernel output

If one first unbinds the bluetooth device driver, suspend / resume works as expected also with 'v5.19-rc1':

root@imx6ull:~# echo mmc1:0001:2 > /sys/bus/sdio/drivers/btmrvl_sdio/unbind
root@imx6ull:~# rfkill
ID TYPE DEVICE    SOFT      HARD
 1 wlan phy0   blocked unblocked
root@imx6ull:~# echo enabled > /sys/class/tty/ttymxc0/power/wakeup
root@imx6ull:~# date;echo mem > /sys/power/state;date
Tue Jun 14 14:59:26 UTC 2022
[  123.530310] PM: suspend entry (deep)
[  123.595432] Filesystems sync: 0.064 seconds
[  123.672478] Freezing user space processes ... (elapsed 0.028 seconds) done.
[  123.701848] OOM killer disabled.
[  123.701869] Freezing remaining freezable tasks ... (elapsed 0.007 seconds) done.
[  123.709993] printk: Suspending console(s) (use no_console_suspend to debug)
[  124.097772] fec 20b4000.ethernet eth0: Link is Down
[  124.124795] PM: suspend devices took 0.280 seconds
[  124.165893] Disabling non-boot CPUs ...
[  124.632959] PM: resume devices took 0.430 seconds
[  124.750164] OOM killer enabled.
[  124.750187] Restarting tasks ... done.
[  124.827899] random: crng reseeded on system resumption
[  124.923183] PM: suspend exit
Tue Jun 14 14:59:31 UTC 2022
[  127.520321] fec 20b4000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
[  127.520514] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
root@imx6ull:~#

---

 net/bluetooth/hci_core.c | 2 ++
 net/bluetooth/hci_sync.c | 1 -
 2 files changed, 2 insertions(+), 1 deletion(-)
net/bluetooth/hci_core.c +2 -0
index 59a5c1341c26..19df3905c5f8 100644
@@ -2675,6 +2675,8 @@ void hci_unregister_dev(struct hci_dev *hdev)
 	list_del(&hdev->list);
 	write_unlock(&hci_dev_list_lock);
 
+	cancel_work_sync(&hdev->power_on);
+
 	hci_cmd_sync_clear(hdev);
 
 	if (!test_bit(HCI_QUIRK_NO_SUSPEND_NOTIFIER, &hdev->quirks))
net/bluetooth/hci_sync.c +0 -1
index 286d6767f017..1739e8cb3291 100644
@@ -4088,7 +4088,6 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 
 	bt_dev_dbg(hdev, "");
 
-	cancel_work_sync(&hdev->power_on);
 	cancel_delayed_work(&hdev->power_off);
 	cancel_delayed_work(&hdev->ncmd_timer);
 
Signature · 1 line
2.20.1
#1 lore raw

This is an automated email and please do not reply to this email.

Dear Submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
While preparing the CI tests, the patches you submitted couldn't be applied to the current HEAD of the repository.

----- Output -----
error: patch failed: net/bluetooth/hci_core.c:2675
error: net/bluetooth/hci_core.c: patch does not apply
hint: Use 'git am --show-current-patch' to see the failed patch

Please resolve the issue and submit the patches again.

---
Regards,
Linux Bluetooth

#2 lore raw
In reply to Max Krummenacher

Marcel, Vasyl,
any comment on this?

Francesco

Quoted 123 lines · Max Krummenacher

From: Max Krummenacher <[email protected]>

This reverts commit ff7f2926114d3a50f5ffe461a9bce8d761748da5.

The commit ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close") introduced between v5.18 and v5.19-rc1 makes going to suspend freeze. v5.19-rc2 is equally affected.

This has been seen on a Colibri iMX6ULL WB which has a Marvell 8997 based WiFi / Bluetooth module connected over SDIO.

With 'v5.18' or 'v5.19-rc1 with said commit reverted' a suspend/resume cycle looks as follows and the device is functional after the resume:

root@imx6ull:~# rfkill
ID TYPE      DEVICE    SOFT      HARD
 0 bluetooth hci0   blocked unblocked
 1 wlan      phy0   blocked unblocked
root@imx6ull:~# echo enabled > /sys/class/tty/ttymxc0/power/wakeup
root@imx6ull:~# date;echo mem > /sys/power/state;date
Tue Jun 14 14:43:03 UTC 2022
[ 6393.464497] PM: suspend entry (deep)
[ 6393.529398] Filesystems sync: 0.064 seconds
[ 6393.594006] Freezing user space processes ... (elapsed 0.015 seconds) done.
[ 6393.610266] OOM killer disabled.
[ 6393.610285] Freezing remaining freezable tasks ... (elapsed 0.013 seconds) done.
[ 6393.623727] printk: Suspending console(s) (use no_console_suspend to debug)

~~ suspended until console initiates the resume

[ 6394.023552] fec 20b4000.ethernet eth0: Link is Down
[ 6394.049902] PM: suspend devices took 0.300 seconds
[ 6394.091654] Disabling non-boot CPUs ...
[ 6394.565896] PM: resume devices took 0.440 seconds
[ 6394.681350] OOM killer enabled.
[ 6394.681369] Restarting tasks ... done.
[ 6394.741157] random: crng reseeded on system resumption
[ 6394.813135] PM: suspend exit
Tue Jun 14 14:43:11 UTC 2022
[ 6396.403873] fec 20b4000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
[ 6396.404347] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready

root@imx6ull:~#

With 'v5.19-rc1' suspend freezes in the suspend phase, i.e. power consumption is not lowered and no wakeup source initiates a wakup.

root@imx6ull:~# rfkill
ID TYPE      DEVICE    SOFT      HARD
 0 bluetooth hci0   blocked unblocked
 1 wlan      phy0   blocked unblocked
root@imx6ull:~# echo enabled > /sys/class/tty/ttymxc0/power/wakeup
root@imx6ull:~# date;echo mem > /sys/power/state;date
Tue Jun 14 12:40:38 UTC 2022
[  122.476333] PM: suspend entry (deep)
[  122.556012] Filesystems sync: 0.079 seconds

~~ no further kernel output

If one first unbinds the bluetooth device driver, suspend / resume works as expected also with 'v5.19-rc1':

root@imx6ull:~# echo mmc1:0001:2 > /sys/bus/sdio/drivers/btmrvl_sdio/unbind
root@imx6ull:~# rfkill
ID TYPE DEVICE    SOFT      HARD
 1 wlan phy0   blocked unblocked
root@imx6ull:~# echo enabled > /sys/class/tty/ttymxc0/power/wakeup
root@imx6ull:~# date;echo mem > /sys/power/state;date
Tue Jun 14 14:59:26 UTC 2022
[  123.530310] PM: suspend entry (deep)
[  123.595432] Filesystems sync: 0.064 seconds
[  123.672478] Freezing user space processes ... (elapsed 0.028 seconds) done.
[  123.701848] OOM killer disabled.
[  123.701869] Freezing remaining freezable tasks ... (elapsed 0.007 seconds) done.
[  123.709993] printk: Suspending console(s) (use no_console_suspend to debug)
[  124.097772] fec 20b4000.ethernet eth0: Link is Down
[  124.124795] PM: suspend devices took 0.280 seconds
[  124.165893] Disabling non-boot CPUs ...
[  124.632959] PM: resume devices took 0.430 seconds
[  124.750164] OOM killer enabled.
[  124.750187] Restarting tasks ... done.
[  124.827899] random: crng reseeded on system resumption
[  124.923183] PM: suspend exit
Tue Jun 14 14:59:31 UTC 2022
[  127.520321] fec 20b4000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
[  127.520514] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
root@imx6ull:~#

---

 net/bluetooth/hci_core.c | 2 ++
 net/bluetooth/hci_sync.c | 1 -
 2 files changed, 2 insertions(+), 1 deletion(-)
net/bluetooth/hci_core.c +2 -0
index 59a5c1341c26..19df3905c5f8 100644
@@ -2675,6 +2675,8 @@ void hci_unregister_dev(struct hci_dev *hdev)
 	list_del(&hdev->list);
 	write_unlock(&hci_dev_list_lock);
 
+	cancel_work_sync(&hdev->power_on);
+
 	hci_cmd_sync_clear(hdev);
 
 	if (!test_bit(HCI_QUIRK_NO_SUSPEND_NOTIFIER, &hdev->quirks))
net/bluetooth/hci_sync.c +0 -1
index 286d6767f017..1739e8cb3291 100644
@@ -4088,7 +4088,6 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 
 	bt_dev_dbg(hdev, "");
 
-	cancel_work_sync(&hdev->power_on);
 	cancel_delayed_work(&hdev->power_off);
 	cancel_delayed_work(&hdev->ncmd_timer);
 
Signature · 1 line
2.20.1
maintainer
#3 lore raw
Quoted 2 lines · On Tue, 21 Jun 2022 13:55:14 +0200 Francesco Dolcini wrote:

Marcel, Vasyl,
any comment on this?

+1 is there an ETA on the fix getting into net?

Quoted 14 lines · Max Krummenacher
Quoted 13 lines

From: Max Krummenacher <[email protected]>

This reverts commit ff7f2926114d3a50f5ffe461a9bce8d761748da5.

The commit ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close") introduced between v5.18 and v5.19-rc1 makes going to suspend freeze. v5.19-rc2 is equally affected.

This has been seen on a Colibri iMX6ULL WB which has a Marvell 8997 based WiFi / Bluetooth module connected over SDIO.

With 'v5.18' or 'v5.19-rc1 with said commit reverted' a suspend/resume cycle looks as follows and the device is functional after the resume:

#4 lore raw
In reply to Max Krummenacher

W dniu 14.06.2022 o 20:17, Max Krummenacher pisze:

Quoted 11 lines

From: Max Krummenacher <[email protected]>

This reverts commit ff7f2926114d3a50f5ffe461a9bce8d761748da5.

The commit ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close") introduced between v5.18 and v5.19-rc1 makes going to suspend freeze. v5.19-rc2 is equally affected.

This has been seen on a Colibri iMX6ULL WB which has a Marvell 8997 based WiFi / Bluetooth module connected over SDIO.
[...]

Hello,

commit ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

causes problems also on my laptop (HP 17-by0001nw with a Realtek Bluetooth adapter).

I have Bluetooth disabled by default on startup (via systemd-rfkill.service ) and vanilla kernel 5.19.0-rc5 fails to suspend (the screen turns black, but I am then able to touch a trackpad and log in). Reverting that commit on top of 5.19.0-rc5 fixes the issue.

On bare 5.19.0-rc5, after startup, the kworker/u9:0+hci0 process hangs indefinitely with this stacktrace (obtained through "cat /proc/163/stack" )

        [<0>] __flush_work+0x143/0x220
        [<0>] __cancel_work_timer+0x122/0x1a0
        [<0>] cancel_work_sync+0x10/0x20
        [<0>] hci_dev_close_sync+0x2a/0x550 [bluetooth]
        [<0>] hci_dev_do_close+0x2a/0x60 [bluetooth]
        [<0>] hci_power_on+0x91/0x200 [bluetooth]
        [<0>] process_one_work+0x21c/0x3c0
        [<0>] worker_thread+0x4a/0x3a0
        [<0>] kthread+0xcf/0xf0
        [<0>] ret_from_fork+0x22/0x30

It appears that the hci_power_on() function calls hci_dev_do_close() in this block:

    if (hci_dev_test_flag(hdev, HCI_RFKILLED) ||
         /* [...] */ ) {
        hci_dev_clear_flag(hdev, HCI_AUTO_OFF);
        hci_dev_do_close(hdev);
    } else /* [...] */

which then calls hci_dev_close_sync(). With the problematic commit, that function calls

       cancel_work_sync(&hdev->power_on)

which tries to cancel the execution of the hci_power_on() function's itself, which leads to a deadlock.

When trying to suspend, the "/lib/systemd/systemd-sleep suspend" process hangs on

        [<0>] hci_suspend_dev+0x87/0xf0 [bluetooth]
        [<0>] hci_suspend_notifier+0x38/0x80 [bluetooth]
        [<0>] notifier_call_chain_robust+0x5e/0xc0
        [<0>] blocking_notifier_call_chain_robust+0x42/0x60
        [<0>] pm_notifier_call_chain_robust+0x1d/0x40
        [<0>] pm_suspend+0x116/0x5a0
        [<0>] state_store+0x82/0xe0
        [<0>] kobj_attr_store+0x12/0x20
        [<0>] sysfs_kf_write+0x3e/0x50
        [<0>] kernfs_fop_write_iter+0x138/0x1c0
        [<0>] new_sync_write+0x104/0x180
        [<0>] vfs_write+0x1d7/0x260
        [<0>] ksys_write+0x67/0xe0
        [<0>] __x64_sys_write+0x1a/0x20
        [<0>] do_syscall_64+0x3b/0x90
        [<0>] entry_SYSCALL_64_after_hwframe+0x46/0xb0

My device is:

Bus 001 Device 005: ID 0bda:b00b Realtek Semiconductor Corp. Bluetooth Radio

So probably the
commit ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close") should be reverted or corrected.

Greetings,

Mateusz Jończyk

Quoted 32 lines

---

 net/bluetooth/hci_core.c | 2 ++
 net/bluetooth/hci_sync.c | 1 -
 2 files changed, 2 insertions(+), 1 deletion(-)
net/bluetooth/hci_core.c +2 -0
index 59a5c1341c26..19df3905c5f8 100644
@@ -2675,6 +2675,8 @@ void hci_unregister_dev(struct hci_dev *hdev)
 	list_del(&hdev->list);
 	write_unlock(&hci_dev_list_lock);
 
+	cancel_work_sync(&hdev->power_on);
+
 	hci_cmd_sync_clear(hdev);
 
 	if (!test_bit(HCI_QUIRK_NO_SUSPEND_NOTIFIER, &hdev->quirks))
net/bluetooth/hci_sync.c +0 -1
index 286d6767f017..1739e8cb3291 100644
@@ -4088,7 +4088,6 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 
 	bt_dev_dbg(hdev, "");
 
-	cancel_work_sync(&hdev->power_on);
 	cancel_delayed_work(&hdev->power_off);
 	cancel_delayed_work(&hdev->ncmd_timer);
 
#5 lore raw
Quoted 9 lines · Mateusz Jończyk

W dniu 14.06.2022 o 20:17, Max Krummenacher pisze:

Quoted 7 lines

From: Max Krummenacher <[email protected]>

This reverts commit ff7f2926114d3a50f5ffe461a9bce8d761748da5.

The commit ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close") introduced between v5.18 and v5.19-rc1 makes going to suspend freeze. v5.19-rc2 is equally affected.

The following follow up commit fixes the issue for me without the revert.
https://lore.kernel.org/all/[email protected]/

Thanks
Max

#6 lore raw

PATCH 1/1 Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

In reply to Max Krummenacher

cancel_work_sync(&hdev->power_on) was moved to hci_dev_close_sync in commit [1] to ensure that power_on work is canceled after HCI interface down.

But, in certain cases power_on work function may call hci_dev_close_sync
itself: hci_power_on -> hci_dev_do_close -> hci_dev_close_sync -> cancel_work_sync(&hdev->power_on), causing deadlock. In particular, this happens when device is rfkilled on boot. To avoid deadlock, move power_on work canceling out of hci_dev_do_close/hci_dev_close_sync.

Deadlock introduced by commit [1] was reported in [2,3] as broken suspend. Suspend did not work because hdev->req_lock held as result of power_on work deadlock. In fact, other BT features were not working. It was not observed when testing [1] since it was verified without rfkill in place.

NOTE: It is not needed to cancel power_on work from other places where hci_dev_do_close/hci_dev_close_sync is called in case:
* Requests were serialized due to hdev->req_workqueue. The power_on work is first in that workqueue.
* hci_rfkill_set_block which won't close device anyway until HCI_SETUP is on.
* hci_sock_release which runs after hci_sock_bind which ensures HCI_SETUP was cleared.

As result, behaviour is the same as in pre-dd06ed7 commit, except power_on work cancel added to hci_dev_close.

[1]: commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close") [2]: https://lore.kernel.org/lkml/[email protected]/
[2]: https://lore.kernel.org/lkml/[email protected]/

Fixes commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close")
---
 net/bluetooth/hci_core.c | 3 +++
 net/bluetooth/hci_sync.c | 1 -
 2 files changed, 3 insertions(+), 1 deletion(-)
net/bluetooth/hci_core.c +3 -0
index 59a5c1341c26..a0f99baafd35 100644
@@ -571,6 +571,7 @@ int hci_dev_close(__u16 dev)
 		goto done;
 	}
 
+	cancel_work_sync(&hdev->power_on);
 	if (hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF))
 		cancel_delayed_work(&hdev->power_off);
 
@@ -2675,6 +2676,8 @@ void hci_unregister_dev(struct hci_dev *hdev)
 	list_del(&hdev->list);
 	write_unlock(&hci_dev_list_lock);
 
+	cancel_work_sync(&hdev->power_on);
+
 	hci_cmd_sync_clear(hdev);
 
 	if (!test_bit(HCI_QUIRK_NO_SUSPEND_NOTIFIER, &hdev->quirks))
net/bluetooth/hci_sync.c +0 -1
index 286d6767f017..1739e8cb3291 100644
@@ -4088,7 +4088,6 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 
 	bt_dev_dbg(hdev, "");
 
-	cancel_work_sync(&hdev->power_on);
 	cancel_delayed_work(&hdev->power_off);
 	cancel_delayed_work(&hdev->ncmd_timer);
 
Signature · 1 line
2.30.2
author Tested-by on v1 1/1
#7 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

Quoted 76 lines · Vasyl Vavrychuk

cancel_work_sync(&hdev->power_on) was moved to hci_dev_close_sync in commit [1] to ensure that power_on work is canceled after HCI interface down.

But, in certain cases power_on work function may call hci_dev_close_sync
itself: hci_power_on -> hci_dev_do_close -> hci_dev_close_sync -> cancel_work_sync(&hdev->power_on), causing deadlock. In particular, this happens when device is rfkilled on boot. To avoid deadlock, move power_on work canceling out of hci_dev_do_close/hci_dev_close_sync.

Deadlock introduced by commit [1] was reported in [2,3] as broken suspend. Suspend did not work because hdev->req_lock held as result of power_on work deadlock. In fact, other BT features were not working. It was not observed when testing [1] since it was verified without rfkill in place.

NOTE: It is not needed to cancel power_on work from other places where hci_dev_do_close/hci_dev_close_sync is called in case:
* Requests were serialized due to hdev->req_workqueue. The power_on work is first in that workqueue.
* hci_rfkill_set_block which won't close device anyway until HCI_SETUP is on.
* hci_sock_release which runs after hci_sock_bind which ensures HCI_SETUP was cleared.

As result, behaviour is the same as in pre-dd06ed7 commit, except power_on work cancel added to hci_dev_close.

[1]: commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close") [2]: https://lore.kernel.org/lkml/[email protected]/
[2]: https://lore.kernel.org/lkml/[email protected]/

Fixes commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close")
---
 net/bluetooth/hci_core.c | 3 +++
 net/bluetooth/hci_sync.c | 1 -
 2 files changed, 3 insertions(+), 1 deletion(-)
net/bluetooth/hci_core.c +3 -0
index 59a5c1341c26..a0f99baafd35 100644
@@ -571,6 +571,7 @@ int hci_dev_close(__u16 dev)
                goto done;
        }
 
+       cancel_work_sync(&hdev->power_on);
        if (hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF))
                cancel_delayed_work(&hdev->power_off);
 
@@ -2675,6 +2676,8 @@ void hci_unregister_dev(struct hci_dev *hdev)
        list_del(&hdev->list);
        write_unlock(&hci_dev_list_lock);
 
+       cancel_work_sync(&hdev->power_on);
+
        hci_cmd_sync_clear(hdev);
 
        if (!test_bit(HCI_QUIRK_NO_SUSPEND_NOTIFIER, &hdev->quirks))
net/bluetooth/hci_sync.c +0 -2
index 286d6767f017..1739e8cb3291 100644
@@ -4088,7 +4088,6 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 
        bt_dev_dbg(hdev, "");
 
-       cancel_work_sync(&hdev->power_on);
        cancel_delayed_work(&hdev->power_off);
        cancel_delayed_work(&hdev->ncmd_timer);
 
--
Quoted 2 lines

2.30.2

This fixes the issue I described in [1]. I.e. The kernel no longer freezes while going to suspend.

Thanks!
Max

#8 lore raw

RE: Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

In reply to Vasyl Vavrychuk

This is an automated email and please do not reply to this email.

Dear Submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
While preparing the CI tests, the patches you submitted couldn't be applied to the current HEAD of the repository.

----- Output -----
error: patch failed: net/bluetooth/hci_core.c:2675
error: net/bluetooth/hci_core.c: patch does not apply
hint: Use 'git am --show-current-patch' to see the failed patch

Please resolve the issue and submit the patches again.

---
Regards,
Linux Bluetooth

#9 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

In reply to Vasyl Vavrychuk

Hello Vasyl,

Quoted 1 line · Vasyl Vavrychuk
Fixes commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

This fixes tag is broken, dd06ed7ad057 does not exist on torvalds/master, and the commit word should be removed.

Should be:

Fixes ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

Francesco

#10 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

Hi,

Quoted 4 lines · Francesco Dolcini

Hello Vasyl,

On Tue, Jul 05, 2022 at 03:59:31PM +0300, Vasyl Vavrychuk wrote:

Quoted 8 lines
Quoted 1 line
Fixes commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

This fixes tag is broken, dd06ed7ad057 does not exist on torvalds/master, and the commit word should be removed.

Should be:

Fixes ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

Ive rebased the patch on top of bluetooth-next and fixed the hash, lets see if passes CI I might just go ahead and push it.

Signature · 1 line
Luiz Augusto von Dentz
maintainer
#11 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

Quoted 11 lines · On Tue, 5 Jul 2022 10:26:08 -0700 Luiz Augusto von Dentz wrote:
Quoted 9 lines · Francesco Dolcini

Hello Vasyl,

Quoted 1 line · Vasyl Vavrychuk
Fixes commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

This fixes tag is broken, dd06ed7ad057 does not exist on torvalds/master, and the commit word should be removed.

Quoted 6 lines
Quoted 3 lines

Should be:

Fixes ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

Ive rebased the patch on top of bluetooth-next and fixed the hash, lets see if passes CI I might just go ahead and push it.

Thanks for pushing it along, the final version can got thru bluetooth -> -> net and into 5.19, right?

#12 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

Hi Jakub,

Quoted 15 lines · Jakub Kicinski
Quoted 13 lines · On Tue, 5 Jul 2022 10:26:08 -0700 Luiz Augusto von Dentz wrote:
Quoted 11 lines · Francesco Dolcini

Hello Vasyl,

Quoted 1 line · Vasyl Vavrychuk
Fixes commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

This fixes tag is broken, dd06ed7ad057 does not exist on torvalds/master, and the commit word should be removed.

Should be:

Quoted 7 lines
Quoted 4 lines
Quoted 1 line
Fixes ff7f2926114d ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

Ive rebased the patch on top of bluetooth-next and fixed the hash, lets see if passes CI I might just go ahead and push it.

Thanks for pushing it along, the final version can got thru bluetooth -> -> net and into 5.19, right?

Yep, I will send the pull request in a moment.

Signature · 1 line
Luiz Augusto von Dentz
maintainer
#13 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

Quoted 7 lines · On Tue, 5 Jul 2022 12:00:43 -0700 Luiz Augusto von Dentz wrote:
Quoted 5 lines
Quoted 2 lines

Ive rebased the patch on top of bluetooth-next and fixed the hash, lets see if passes CI I might just go ahead and push it.

Thanks for pushing it along, the final version can got thru bluetooth -> -> net and into 5.19, right?

Yep, I will send the pull request in a moment.

Perfect, thank you!!

#14 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

In reply to Vasyl Vavrychuk

W dniu 5.07.2022 o 14:59, Vasyl Vavrychuk pisze:

Quoted 28 lines

cancel_work_sync(&hdev->power_on) was moved to hci_dev_close_sync in commit [1] to ensure that power_on work is canceled after HCI interface down.

But, in certain cases power_on work function may call hci_dev_close_sync
itself: hci_power_on -> hci_dev_do_close -> hci_dev_close_sync -> cancel_work_sync(&hdev->power_on), causing deadlock. In particular, this happens when device is rfkilled on boot. To avoid deadlock, move power_on work canceling out of hci_dev_do_close/hci_dev_close_sync.

Deadlock introduced by commit [1] was reported in [2,3] as broken suspend. Suspend did not work because hdev->req_lock held as result of power_on work deadlock. In fact, other BT features were not working. It was not observed when testing [1] since it was verified without rfkill in place.

NOTE: It is not needed to cancel power_on work from other places where hci_dev_do_close/hci_dev_close_sync is called in case:
* Requests were serialized due to hdev->req_workqueue. The power_on work is first in that workqueue.
* hci_rfkill_set_block which won't close device anyway until HCI_SETUP is on.
* hci_sock_release which runs after hci_sock_bind which ensures HCI_SETUP was cleared.

As result, behaviour is the same as in pre-dd06ed7 commit, except power_on work cancel added to hci_dev_close.

Quoted 8 lines

[1]: commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close") [2]: https://lore.kernel.org/lkml/[email protected]/
[2]: https://lore.kernel.org/lkml/[email protected]/

Fixes commit dd06ed7ad057 ("Bluetooth: core: Fix missing power_on work cancel on HCI close")

Works well: suspend (with bluetooth on and also off), hibernation, sending files, rfkill.

Thank you.

Greetings,

Mateusz Jończyk

#15 lore raw

Re: [PATCH] Bluetooth: core: Fix deadlock due to `cancel_work_sync(&hdev->power_on)` from hci_power_on_sync.

In reply to Vasyl Vavrychuk
This patch was applied to netdev/net.git (master) by Luiz Augusto von Dentz <[email protected]>:

Hello:

This patch was applied to netdev/net.git (master) by Luiz Augusto von Dentz <[email protected]>:

Quoted 11 lines · On Tue, 5 Jul 2022 15:59:31 +0300 you wrote:

cancel_work_sync(&hdev->power_on) was moved to hci_dev_close_sync in commit [1] to ensure that power_on work is canceled after HCI interface down.

But, in certain cases power_on work function may call hci_dev_close_sync
itself: hci_power_on -> hci_dev_do_close -> hci_dev_close_sync -> cancel_work_sync(&hdev->power_on), causing deadlock. In particular, this happens when device is rfkilled on boot. To avoid deadlock, move power_on work canceling out of hci_dev_do_close/hci_dev_close_sync.

[...]

Here is the summary with links:
- Bluetooth: core: Fix deadlock due to cancel_work_sync(&hdev->power_on) from hci_power_on_sync.

    https://git.kernel.org/netdev/net/c/e36bea6e78ab

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html