Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
---
Changes from v1:
- added Co-developed-by tag
arch/arm/boot/dts/sun8i-h2-plus-bananapi-m2-zero.dts | 3 ++-
arch/arm/boot/dts/sun8i-h3-nanopi.dtsi | 1 +
arch/arm/boot/dts/sun8i-h3-orangepi-2.dts | 3 ++-
arch/arm/boot/dts/sun8i-h3-orangepi-pc.dts | 3 ++-
arch/arm/boot/dts/sunxi-bananapi-m2-plus.dtsi | 3 ++-
arch/arm/boot/dts/sunxi-libretech-all-h3-cc.dtsi | 1 +
6 files changed, 10 insertions(+), 4 deletions(-)
On Tue, Nov 30, 2021 at 12:55 AM Jernej Skrabec
[off-list ref] wrote:
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
It might have been that after shutdown there was really no way to
"power on" the board with these GPIO power keys, so we didn't use
KEY_POWER for them?
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
On Tue, Nov 30, 2021 at 1:29 AM Jernej Škrabec [off-list ref] wrote:
Dne ponedeljek, 29. november 2021 ob 18:16:31 CET je Chen-Yu Tsai napisal(a):
quoted
On Tue, Nov 30, 2021 at 12:55 AM Jernej Skrabec
[off-list ref] wrote:
quoted
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
It might have been that after shutdown there was really no way to
"power on" the board with these GPIO power keys, so we didn't use
KEY_POWER for them?
KEY_POWER is actually processed by userspace, so it would still make sense to
power down board if power key is pressed. Correct me if I'm wrong, but in
combination with wfi it could work before SCP FW was available?
I guess it could? But where would it do wfi, and what would execute
after wfi?
IIRC the kernel's halt sequence doesn't really clean up interrupts,
so wfi isn't going to get very far.
quoted
quoted
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Dne ponedeljek, 29. november 2021 ob 18:16:31 CET je Chen-Yu Tsai napisal(a):
On Tue, Nov 30, 2021 at 12:55 AM Jernej Skrabec
[off-list ref] wrote:
quoted
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
It might have been that after shutdown there was really no way to
"power on" the board with these GPIO power keys, so we didn't use
KEY_POWER for them?
KEY_POWER is actually processed by userspace, so it would still make sense to
power down board if power key is pressed. Correct me if I'm wrong, but in
combination with wfi it could work before SCP FW was available?
quoted
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
From: Maxime Ripard <hidden> Date: 2021-12-01 08:21:56
On Mon, 29 Nov 2021 17:55:10 +0100, Jernej Skrabec wrote:
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
Adjust them, so they can function as intended by designer.
[...]
Hi,
On Mon, Nov 29, 2021 at 05:55:10PM +0100, Jernej Skrabec wrote:
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
This patch results in the following traceback when rebooting an
orangepi-pc qemu emulation.
[ 30.899594]
[ 30.899685] ============================================
[ 30.899757] WARNING: possible recursive locking detected
[ 30.899938] 5.17.0-rc3-00394-gc849047c2473 #1 Not tainted
[ 30.900055] --------------------------------------------
[ 30.900124] init/307 is trying to acquire lock:
[ 30.900246] c2dfe27c (&irq_desc_lock_class){-.-.}-{2:2}, at: __irq_get_desc_lock+0x58/0xa0
[ 30.900900]
[ 30.900900] but task is already holding lock:
[ 30.900974] c3c0ac7c (&irq_desc_lock_class){-.-.}-{2:2}, at: __irq_get_desc_lock+0x58/0xa0
[ 30.901101]
[ 30.901101] other info that might help us debug this:
[ 30.901188] Possible unsafe locking scenario:
[ 30.901188]
[ 30.901262] CPU0
[ 30.901301] ----
[ 30.901339] lock(&irq_desc_lock_class);
[ 30.901411] lock(&irq_desc_lock_class);
[ 30.901480]
[ 30.901480] *** DEADLOCK ***
[ 30.901480]
[ 30.901554] May be due to missing lock nesting notation
[ 30.901554]
[ 30.901657] 4 locks held by init/307:
[ 30.901724] #0: c1f29f18 (system_transition_mutex){+.+.}-{3:3}, at: __do_sys_reboot+0x90/0x23c
[ 30.901889] #1: c20f7760 (&dev->mutex){....}-{3:3}, at: device_shutdown+0xf4/0x224
[ 30.902016] #2: c2e804d8 (&dev->mutex){....}-{3:3}, at: device_shutdown+0x104/0x224
[ 30.902138] #3: c3c0ac7c (&irq_desc_lock_class){-.-.}-{2:2}, at: __irq_get_desc_lock+0x58/0xa0
[ 30.902281]
[ 30.902281] stack backtrace:
[ 30.902462] CPU: 0 PID: 307 Comm: init Not tainted 5.17.0-rc3-00394-gc849047c2473 #1
[ 30.902572] Hardware name: Allwinner sun8i Family
[ 30.902781] unwind_backtrace from show_stack+0x10/0x14
[ 30.902895] show_stack from dump_stack_lvl+0x68/0x90
[ 30.902970] dump_stack_lvl from __lock_acquire+0x1680/0x31a0
[ 30.903047] __lock_acquire from lock_acquire+0x148/0x3dc
[ 30.903118] lock_acquire from _raw_spin_lock_irqsave+0x50/0x6c
[ 30.903197] _raw_spin_lock_irqsave from __irq_get_desc_lock+0x58/0xa0
[ 30.903282] __irq_get_desc_lock from irq_set_irq_wake+0x2c/0x19c
[ 30.903366] irq_set_irq_wake from irq_set_irq_wake+0x13c/0x19c
[ 30.903442] irq_set_irq_wake from gpio_keys_suspend+0x80/0x1a4
[ 30.903523] gpio_keys_suspend from gpio_keys_shutdown+0x10/0x2c
[ 30.903603] gpio_keys_shutdown from device_shutdown+0x180/0x224
[ 30.903685] device_shutdown from __do_sys_reboot+0x134/0x23c
[ 30.903764] __do_sys_reboot from ret_fast_syscall+0x0/0x1c
[ 30.903894] Exception stack(0xc584ffa8 to 0xc584fff0)
[ 30.904013] ffa0: 01234567 000c623f fee1dead 28121969 01234567 00000000
[ 30.904117] ffc0: 01234567 000c623f 00000001 00000058 000d85c0 00000000 00000000 00000000
[ 30.904213] ffe0: 000d8298 be84ddf4 000918bc b6eb0edc
[ 30.905189] reboot: Restarting system
The warning is no longer seen after reverting this patch.
The problem exists but is not seen in v5.17-rc4 because a bug in commit
8df89a7cbc63 ("pinctrl-sunxi: don't call pinctrl_gpio_direction()")
hides it. That problem is fixed with commit 3c5412cdec9f ("pinctrl-sunxi:
sunxi_pinctrl_gpio_direction_in/output: use correct offset") in linux-next,
and the traceback is seen there.
Guenter
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Hi!
Dne torek, 15. februar 2022 ob 01:27:32 CET je Guenter Roeck napisal(a):
Hi,
On Mon, Nov 29, 2021 at 05:55:10PM +0100, Jernej Skrabec wrote:
quoted
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
This patch results in the following traceback when rebooting an
orangepi-pc qemu emulation.
[ 30.899594]
[ 30.899685] ============================================
[ 30.899757] WARNING: possible recursive locking detected
[ 30.899938] 5.17.0-rc3-00394-gc849047c2473 #1 Not tainted
[ 30.900055] --------------------------------------------
[ 30.900124] init/307 is trying to acquire lock:
[ 30.900246] c2dfe27c (&irq_desc_lock_class){-.-.}-{2:2}, at:
__irq_get_desc_lock+0x58/0xa0
[ 30.900900]
[ 30.900900] but task is already holding lock:
[ 30.900974] c3c0ac7c (&irq_desc_lock_class){-.-.}-{2:2}, at:
__irq_get_desc_lock+0x58/0xa0
[ 30.901101]
[ 30.901101] other info that might help us debug this:
[ 30.901188] Possible unsafe locking scenario:
[ 30.901188]
[ 30.901262] CPU0
[ 30.901301] ----
[ 30.901339] lock(&irq_desc_lock_class);
[ 30.901411] lock(&irq_desc_lock_class);
[ 30.901480]
[ 30.901480] *** DEADLOCK ***
[ 30.901480]
[ 30.901554] May be due to missing lock nesting notation
[ 30.901554]
[ 30.901657] 4 locks held by init/307:
[ 30.901724] #0: c1f29f18 (system_transition_mutex){+.+.}-{3:3}, at:
[ 30.904213] ffe0: 000d8298 be84ddf4 000918bc b6eb0edc
[ 30.905189] reboot: Restarting system
The warning is no longer seen after reverting this patch.
The problem exists but is not seen in v5.17-rc4 because a bug in commit
8df89a7cbc63 ("pinctrl-sunxi: don't call pinctrl_gpio_direction()")
hides it. That problem is fixed with commit 3c5412cdec9f ("pinctrl-sunxi:
sunxi_pinctrl_gpio_direction_in/output: use correct offset") in linux-next,
and the traceback is seen there.
Hm... These DT changes were tested with many users on older kernels for some
time now and new properties conform to bindings. Should we revert pinctrl
changes?
Best regards,
Jernej
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Hi!
Dne torek, 15. februar 2022 ob 01:27:32 CET je Guenter Roeck napisal(a):
quoted
Hi,
On Mon, Nov 29, 2021 at 05:55:10PM +0100, Jernej Skrabec wrote:
quoted
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
This patch results in the following traceback when rebooting an
orangepi-pc qemu emulation.
[ 30.899594]
[ 30.899685] ============================================
[ 30.899757] WARNING: possible recursive locking detected
[ 30.899938] 5.17.0-rc3-00394-gc849047c2473 #1 Not tainted
[ 30.900055] --------------------------------------------
[ 30.900124] init/307 is trying to acquire lock:
[ 30.900246] c2dfe27c (&irq_desc_lock_class){-.-.}-{2:2}, at:
__irq_get_desc_lock+0x58/0xa0
quoted
[ 30.900900]
[ 30.900900] but task is already holding lock:
[ 30.900974] c3c0ac7c (&irq_desc_lock_class){-.-.}-{2:2}, at:
__irq_get_desc_lock+0x58/0xa0
quoted
[ 30.901101]
[ 30.901101] other info that might help us debug this:
[ 30.901188] Possible unsafe locking scenario:
[ 30.901188]
[ 30.901262] CPU0
[ 30.901301] ----
[ 30.901339] lock(&irq_desc_lock_class);
[ 30.901411] lock(&irq_desc_lock_class);
[ 30.901480]
[ 30.901480] *** DEADLOCK ***
[ 30.901480]
[ 30.901554] May be due to missing lock nesting notation
[ 30.901554]
[ 30.901657] 4 locks held by init/307:
[ 30.901724] #0: c1f29f18 (system_transition_mutex){+.+.}-{3:3}, at:
[ 30.904213] ffe0: 000d8298 be84ddf4 000918bc b6eb0edc
[ 30.905189] reboot: Restarting system
The warning is no longer seen after reverting this patch.
The problem exists but is not seen in v5.17-rc4 because a bug in commit
8df89a7cbc63 ("pinctrl-sunxi: don't call pinctrl_gpio_direction()")
hides it. That problem is fixed with commit 3c5412cdec9f ("pinctrl-sunxi:
sunxi_pinctrl_gpio_direction_in/output: use correct offset") in linux-next,
and the traceback is seen there.
Hm... These DT changes were tested with many users on older kernels for some
time now and new properties conform to bindings. Should we revert pinctrl
changes?
I don't think those changes were tested with orangepi-pc on real hardware.
Maybe I didn't explain it clearly enough: Commit 8df89a7cbc63 does _not_
introduce the problem. It hides the problem. Reverting commit 8df89a7cbc63
won't help but result in exactly the same backtrace (I tried).
Some more details: This commit introduces "wakeup-source;" to various
orangepi-pc nodes. This triggers in a call to sunxi_pinctrl_irq_set_wake(),
which did not happen before and which may result in the traceback.
Guenter
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Samuel Holland <samuel@sholland.org> Date: 2022-02-16 04:11:44
On 2/15/22 4:20 PM, Guenter Roeck wrote:
On 2/15/22 12:34, Jernej Škrabec wrote:
quoted
Hi!
Dne torek, 15. februar 2022 ob 01:27:32 CET je Guenter Roeck napisal(a):
quoted
Hi,
On Mon, Nov 29, 2021 at 05:55:10PM +0100, Jernej Skrabec wrote:
quoted
Several H3 and one H2+ board have power key nodes, which are slightly
off. Some are missing wakeup-source property and some have BTN_0 code
assigned instead of KEY_POWER.
Adjust them, so they can function as intended by designer.
Co-developed-by: Michael Klein <redacted>
Signed-off-by: Michael Klein <redacted>
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
This patch results in the following traceback when rebooting an
orangepi-pc qemu emulation.
[ 30.899594]
[ 30.899685] ============================================
[ 30.899757] WARNING: possible recursive locking detected
[ 30.899938] 5.17.0-rc3-00394-gc849047c2473 #1 Not tainted
[ 30.900055] --------------------------------------------
[ 30.900124] init/307 is trying to acquire lock:
[ 30.900246] c2dfe27c (&irq_desc_lock_class){-.-.}-{2:2}, at:
__irq_get_desc_lock+0x58/0xa0
quoted
[ 30.900900]
[ 30.900900] but task is already holding lock:
[ 30.900974] c3c0ac7c (&irq_desc_lock_class){-.-.}-{2:2}, at:
__irq_get_desc_lock+0x58/0xa0
quoted
[ 30.901101]
[ 30.901101] other info that might help us debug this:
[ 30.901188] Possible unsafe locking scenario:
[ 30.901188]
[ 30.901262] CPU0
[ 30.901301] ----
[ 30.901339] lock(&irq_desc_lock_class);
[ 30.901411] lock(&irq_desc_lock_class);
[ 30.901480]
[ 30.901480] *** DEADLOCK ***
[ 30.901480]
[ 30.901554] May be due to missing lock nesting notation
[ 30.901554]
[ 30.901657] 4 locks held by init/307:
[ 30.901724] #0: c1f29f18 (system_transition_mutex){+.+.}-{3:3}, at:
[ 30.904213] ffe0: 000d8298 be84ddf4 000918bc b6eb0edc
[ 30.905189] reboot: Restarting system
The warning is no longer seen after reverting this patch.
The problem exists but is not seen in v5.17-rc4 because a bug in commit
8df89a7cbc63 ("pinctrl-sunxi: don't call pinctrl_gpio_direction()")
hides it. That problem is fixed with commit 3c5412cdec9f ("pinctrl-sunxi:
sunxi_pinctrl_gpio_direction_in/output: use correct offset") in linux-next,
and the traceback is seen there.
Hm... These DT changes were tested with many users on older kernels for some
time now and new properties conform to bindings. Should we revert pinctrl
changes?
I don't think those changes were tested with orangepi-pc on real hardware.
Maybe I didn't explain it clearly enough: Commit 8df89a7cbc63 does _not_
introduce the problem. It hides the problem. Reverting commit 8df89a7cbc63
won't help but result in exactly the same backtrace (I tried).
Some more details: This commit introduces "wakeup-source;" to various
orangepi-pc nodes. This triggers in a call to sunxi_pinctrl_irq_set_wake(),
which did not happen before and which may result in the traceback.
I don't think there is any real issue here. The two irq_desc's being locked are
always different, so there is no deadlock. This recursive irq_set_irq_wake seems
to be a reasonably common pattern in GPIO drivers, and several of them contain
code to silence lockdep. I've sent a patch adding a copy of that to the sunxi
driver:
https://lore.kernel.org/lkml/20220216040037.22730-1-samuel@sholland.org/
Please see if it works for you.
Regards,
Samuel
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
On Tue, Feb 15, 2022 at 10:11:36PM -0600, Samuel Holland wrote:
[ ... ]
I don't think there is any real issue here. The two irq_desc's being locked are
always different, so there is no deadlock. This recursive irq_set_irq_wake seems
to be a reasonably common pattern in GPIO drivers, and several of them contain
code to silence lockdep. I've sent a patch adding a copy of that to the sunxi
driver:
https://lore.kernel.org/lkml/20220216040037.22730-1-samuel@sholland.org/
Please see if it works for you.
Yes, it does. The warning is gone with your patch applied on top of linux-next.
Thanks,
Guenter
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel