Skip to contentSkip to conversation

pinctrl: intel: Stop setting IRQF_NO_THREAD ?

RFC Hans de Goede 3 participants 6 messages Jun 21, 2025 – May 5 GPIO

lore
Share & tools
lore
#0 lore raw

Hi,

While debugging the following lockdep report:

=============================
[ BUG: Invalid wait context ]
...
swapper/10/0 is trying to lock:
ffff88819c271888 (&tp->xfer_wait){....}-{3:3},
at: __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
...

 Call Trace:
 <IRQ>
 ...
 __raw_spin_lock_irqsave (./include/linux/spinlock_api_smp.h:111)
 __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
 vsc_tp_isr (drivers/misc/mei/vsc-tp.c:110) mei_vsc_hw
 __handle_irq_event_percpu (kernel/irq/handle.c:158)
 handle_irq_event (kernel/irq/handle.c:195 kernel/irq/handle.c:210)
 handle_edge_irq (kernel/irq/chip.c:833)
 ...
 </IRQ>

I realized after a while that the root-cause here is the IRQF_NO_THREAD usage in pinctrl-intel.c. This was introduced in 1a7d1cb81eb2 ("pinctrl:
intel: Prevent force threading of the interrupt handler") to avoid problems caused by using request_irq() for what should be a chained irq handler (which itself is a workaround because of a shared IRQ on some platforms).

Generally speaking using IRQF_NO_THREAD is undesirable for 2 reasons:

1. It introduces extra latency on PREEMPT-RT kernels
2. Setting IRQF_NO_THREAD requires all interrupt handlers for GPIO interrupts to use raw-spinlocks only since normal spinlocks can sleep in PREEMPT-RT kernels and with IRQF_NO_THREAD the interrupt handlers will run in an atomic context

2. is what is causing the lockdep report above, by simply using a wake_up(&wq_head) call in an interrupt handler, since wait-queues use normal spinlocks not raw spinlocks.

I've tried just removing the IRQF_NO_THREAD flag and that fixes the lockdep report. I've also tried reproducing the problem for which the flag was added in commit 1a7d1cb81eb2 by using a kernel with CONFIG_IRQ_FORCED_THREADING and "threadirqs" on the kernel commandline. And the problem not reproduce. I'm not sure this is 100% proof that the flag is no longer necessary though ...

So 2 questions:

1. Should we maybe just drop the flag ?
2. Or should we have 2 different code-paths for GPIO controllers with/without shared IRQs and use a chained-irq approach for the not shared case, to at least reduce the usage of the flag ?

Regards,

Hans

p.s. For the 2 different code paths approach I believe we can use intel_pinctrl_probe_by_uid() to identify platforms which use a shared IRQ. AFAICT on all platforms which use intel_pinctrl_probe_by_hid() there will only be a single GPIO controller instance.

#1 lore raw

Hi,

Quoted 41 lines · Hans de Goede

Hi,

While debugging the following lockdep report:

=============================
[ BUG: Invalid wait context ]
...
swapper/10/0 is trying to lock:
ffff88819c271888 (&tp->xfer_wait){....}-{3:3},
at: __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
...

 Call Trace:
 <IRQ>
 ...
 __raw_spin_lock_irqsave (./include/linux/spinlock_api_smp.h:111)
 __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
 vsc_tp_isr (drivers/misc/mei/vsc-tp.c:110) mei_vsc_hw
 __handle_irq_event_percpu (kernel/irq/handle.c:158)
 handle_irq_event (kernel/irq/handle.c:195 kernel/irq/handle.c:210)
 handle_edge_irq (kernel/irq/chip.c:833)
 ...
 </IRQ>

I realized after a while that the root-cause here is the IRQF_NO_THREAD usage in pinctrl-intel.c. This was introduced in 1a7d1cb81eb2 ("pinctrl:
intel: Prevent force threading of the interrupt handler") to avoid problems caused by using request_irq() for what should be a chained irq handler (which itself is a workaround because of a shared IRQ on some platforms).

Generally speaking using IRQF_NO_THREAD is undesirable for 2 reasons:

1. It introduces extra latency on PREEMPT-RT kernels
2. Setting IRQF_NO_THREAD requires all interrupt handlers for GPIO interrupts to use raw-spinlocks only since normal spinlocks can sleep in PREEMPT-RT kernels and with IRQF_NO_THREAD the interrupt handlers will run in an atomic context

2. is what is causing the lockdep report above, by simply using a wake_up(&wq_head) call in an interrupt handler, since wait-queues use normal spinlocks not raw spinlocks.

Quoted 6 lines

I've tried just removing the IRQF_NO_THREAD flag and that fixes the lockdep report. I've also tried reproducing the problem for which the flag was added in commit 1a7d1cb81eb2 by using a kernel with CONFIG_IRQ_FORCED_THREADING and "threadirqs" on the kernel commandline. And the problem not reproduce. I'm not sure this is 100% proof that the flag is no longer necessary though ...

Can you try also with CONFIG_PREEMPT_RT and see if that triggers the issue? If not then:

Quoted 6 lines

So 2 questions:

1. Should we maybe just drop the flag ?
2. Or should we have 2 different code-paths for GPIO controllers with/without shared IRQs and use a chained-irq approach for the not shared case, to at least reduce the usage of the flag ?

I would just drop the flag then.

#2 lore raw
Quoted 51 lines · Mika Westerberg
Quoted 39 lines · Hans de Goede

While debugging the following lockdep report:

=============================
[ BUG: Invalid wait context ]
...
swapper/10/0 is trying to lock:
ffff88819c271888 (&tp->xfer_wait){....}-{3:3},
at: __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
...

 Call Trace:
 <IRQ>
 ...
 __raw_spin_lock_irqsave (./include/linux/spinlock_api_smp.h:111)
 __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
 vsc_tp_isr (drivers/misc/mei/vsc-tp.c:110) mei_vsc_hw
 __handle_irq_event_percpu (kernel/irq/handle.c:158)
 handle_irq_event (kernel/irq/handle.c:195 kernel/irq/handle.c:210)
 handle_edge_irq (kernel/irq/chip.c:833)
 ...
 </IRQ>

I realized after a while that the root-cause here is the IRQF_NO_THREAD usage in pinctrl-intel.c. This was introduced in 1a7d1cb81eb2 ("pinctrl:
intel: Prevent force threading of the interrupt handler") to avoid problems caused by using request_irq() for what should be a chained irq handler (which itself is a workaround because of a shared IRQ on some platforms).

Generally speaking using IRQF_NO_THREAD is undesirable for 2 reasons:

1. It introduces extra latency on PREEMPT-RT kernels
2. Setting IRQF_NO_THREAD requires all interrupt handlers for GPIO interrupts to use raw-spinlocks only since normal spinlocks can sleep in PREEMPT-RT kernels and with IRQF_NO_THREAD the interrupt handlers will run in an atomic context

2. is what is causing the lockdep report above, by simply using a wake_up(&wq_head) call in an interrupt handler, since wait-queues use normal spinlocks not raw spinlocks.

Quoted 6 lines

I've tried just removing the IRQF_NO_THREAD flag and that fixes the lockdep report. I've also tried reproducing the problem for which the flag was added in commit 1a7d1cb81eb2 by using a kernel with CONFIG_IRQ_FORCED_THREADING and "threadirqs" on the kernel commandline. And the problem not reproduce. I'm not sure this is 100% proof that the flag is no longer necessary though ...

Can you try also with CONFIG_PREEMPT_RT and see if that triggers the issue? If not then:

Quoted 8 lines
Quoted 6 lines

So 2 questions:

1. Should we maybe just drop the flag ?
2. Or should we have 2 different code-paths for GPIO controllers with/without shared IRQs and use a chained-irq approach for the not shared case, to at least reduce the usage of the flag ?

I would just drop the flag then.

Hans, any conclusion on this?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#3 lore raw

Hi Andy,

Quoted 54 lines · Andy Shevchenko
Quoted 53 lines · Mika Westerberg
Quoted 39 lines · Hans de Goede

While debugging the following lockdep report:

=============================
[ BUG: Invalid wait context ]
...
swapper/10/0 is trying to lock:
ffff88819c271888 (&tp->xfer_wait){....}-{3:3},
at: __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
...

 Call Trace:
 <IRQ>
 ...
 __raw_spin_lock_irqsave (./include/linux/spinlock_api_smp.h:111)
 __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
 vsc_tp_isr (drivers/misc/mei/vsc-tp.c:110) mei_vsc_hw
 __handle_irq_event_percpu (kernel/irq/handle.c:158)
 handle_irq_event (kernel/irq/handle.c:195 kernel/irq/handle.c:210)
 handle_edge_irq (kernel/irq/chip.c:833)
 ...
 </IRQ>

I realized after a while that the root-cause here is the IRQF_NO_THREAD usage in pinctrl-intel.c. This was introduced in 1a7d1cb81eb2 ("pinctrl:
intel: Prevent force threading of the interrupt handler") to avoid problems caused by using request_irq() for what should be a chained irq handler (which itself is a workaround because of a shared IRQ on some platforms).

Generally speaking using IRQF_NO_THREAD is undesirable for 2 reasons:

1. It introduces extra latency on PREEMPT-RT kernels
2. Setting IRQF_NO_THREAD requires all interrupt handlers for GPIO interrupts to use raw-spinlocks only since normal spinlocks can sleep in PREEMPT-RT kernels and with IRQF_NO_THREAD the interrupt handlers will run in an atomic context

2. is what is causing the lockdep report above, by simply using a wake_up(&wq_head) call in an interrupt handler, since wait-queues use normal spinlocks not raw spinlocks.

Quoted 6 lines

I've tried just removing the IRQF_NO_THREAD flag and that fixes the lockdep report. I've also tried reproducing the problem for which the flag was added in commit 1a7d1cb81eb2 by using a kernel with CONFIG_IRQ_FORCED_THREADING and "threadirqs" on the kernel commandline. And the problem not reproduce. I'm not sure this is 100% proof that the flag is no longer necessary though ...

Can you try also with CONFIG_PREEMPT_RT and see if that triggers the issue? If not then:

Quoted 2 lines

So 2 questions:

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines

1. Should we maybe just drop the flag ?
2. Or should we have 2 different code-paths for GPIO controllers with/without shared IRQs and use a chained-irq approach for the not shared case, to at least reduce the usage of the flag ?

I would just drop the flag then.

Hans, any conclusion on this?

I worked around this issue in the affected driver. I have not looked further into actually dropping IRQF_NO_THREAD from the Intel pinctrl/ GPIO drivers.

I do think that dropping IRQF_NO_THREAD from the Intel pinctrl/GPIO drivers is probably a good idea, but this will need someone to drive this forward including dealing with any regressions this may lead to.

Regards,

Hans

#4 lore raw
Quoted 60 lines · Hans de Goede
Quoted 59 lines · Andy Shevchenko
Quoted 58 lines · Mika Westerberg
Quoted 39 lines · Hans de Goede

While debugging the following lockdep report:

=============================
[ BUG: Invalid wait context ]
...
swapper/10/0 is trying to lock:
ffff88819c271888 (&tp->xfer_wait){....}-{3:3},
at: __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
...

 Call Trace:
 <IRQ>
 ...
 __raw_spin_lock_irqsave (./include/linux/spinlock_api_smp.h:111)
 __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
 vsc_tp_isr (drivers/misc/mei/vsc-tp.c:110) mei_vsc_hw
 __handle_irq_event_percpu (kernel/irq/handle.c:158)
 handle_irq_event (kernel/irq/handle.c:195 kernel/irq/handle.c:210)
 handle_edge_irq (kernel/irq/chip.c:833)
 ...
 </IRQ>

I realized after a while that the root-cause here is the IRQF_NO_THREAD usage in pinctrl-intel.c. This was introduced in 1a7d1cb81eb2 ("pinctrl:
intel: Prevent force threading of the interrupt handler") to avoid problems caused by using request_irq() for what should be a chained irq handler (which itself is a workaround because of a shared IRQ on some platforms).

Generally speaking using IRQF_NO_THREAD is undesirable for 2 reasons:

1. It introduces extra latency on PREEMPT-RT kernels
2. Setting IRQF_NO_THREAD requires all interrupt handlers for GPIO interrupts to use raw-spinlocks only since normal spinlocks can sleep in PREEMPT-RT kernels and with IRQF_NO_THREAD the interrupt handlers will run in an atomic context

2. is what is causing the lockdep report above, by simply using a wake_up(&wq_head) call in an interrupt handler, since wait-queues use normal spinlocks not raw spinlocks.

Quoted 6 lines

I've tried just removing the IRQF_NO_THREAD flag and that fixes the lockdep report. I've also tried reproducing the problem for which the flag was added in commit 1a7d1cb81eb2 by using a kernel with CONFIG_IRQ_FORCED_THREADING and "threadirqs" on the kernel commandline. And the problem not reproduce. I'm not sure this is 100% proof that the flag is no longer necessary though ...

Can you try also with CONFIG_PREEMPT_RT and see if that triggers the issue? If not then:

Quoted 6 lines

So 2 questions:

1. Should we maybe just drop the flag ?
2. Or should we have 2 different code-paths for GPIO controllers with/without shared IRQs and use a chained-irq approach for the not shared case, to at least reduce the usage of the flag ?

Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I would just drop the flag then.

Hans, any conclusion on this?

I worked around this issue in the affected driver.

Is it upstream? Can you share the commit ID or patch in ML (if it's ready) for that? (Just for the record.)

Quoted 7 lines

I have not looked
further into actually dropping IRQF_NO_THREAD from the Intel pinctrl/ GPIO drivers.

I do think that dropping IRQF_NO_THREAD from the Intel pinctrl/GPIO drivers is probably a good idea, but this will need someone to drive this forward including dealing with any regressions this may lead to.

Thanks for clarifying the state of affairs!

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#5 lore raw

Hi Andy,

Quoted 61 lines · Andy Shevchenko
Quoted 60 lines · Hans de Goede
Quoted 59 lines · Andy Shevchenko
Quoted 58 lines · Mika Westerberg
Quoted 39 lines · Hans de Goede

While debugging the following lockdep report:

=============================
[ BUG: Invalid wait context ]
...
swapper/10/0 is trying to lock:
ffff88819c271888 (&tp->xfer_wait){....}-{3:3},
at: __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
...

 Call Trace:
 <IRQ>
 ...
 __raw_spin_lock_irqsave (./include/linux/spinlock_api_smp.h:111)
 __wake_up (kernel/sched/wait.c:106 kernel/sched/wait.c:127)
 vsc_tp_isr (drivers/misc/mei/vsc-tp.c:110) mei_vsc_hw
 __handle_irq_event_percpu (kernel/irq/handle.c:158)
 handle_irq_event (kernel/irq/handle.c:195 kernel/irq/handle.c:210)
 handle_edge_irq (kernel/irq/chip.c:833)
 ...
 </IRQ>

I realized after a while that the root-cause here is the IRQF_NO_THREAD usage in pinctrl-intel.c. This was introduced in 1a7d1cb81eb2 ("pinctrl:
intel: Prevent force threading of the interrupt handler") to avoid problems caused by using request_irq() for what should be a chained irq handler (which itself is a workaround because of a shared IRQ on some platforms).

Generally speaking using IRQF_NO_THREAD is undesirable for 2 reasons:

1. It introduces extra latency on PREEMPT-RT kernels
2. Setting IRQF_NO_THREAD requires all interrupt handlers for GPIO interrupts to use raw-spinlocks only since normal spinlocks can sleep in PREEMPT-RT kernels and with IRQF_NO_THREAD the interrupt handlers will run in an atomic context

2. is what is causing the lockdep report above, by simply using a wake_up(&wq_head) call in an interrupt handler, since wait-queues use normal spinlocks not raw spinlocks.

Quoted 6 lines

I've tried just removing the IRQF_NO_THREAD flag and that fixes the lockdep report. I've also tried reproducing the problem for which the flag was added in commit 1a7d1cb81eb2 by using a kernel with CONFIG_IRQ_FORCED_THREADING and "threadirqs" on the kernel commandline. And the problem not reproduce. I'm not sure this is 100% proof that the flag is no longer necessary though ...

Can you try also with CONFIG_PREEMPT_RT and see if that triggers the issue? If not then:

Quoted 6 lines

So 2 questions:

1. Should we maybe just drop the flag ?
2. Or should we have 2 different code-paths for GPIO controllers with/without shared IRQs and use a chained-irq approach for the not shared case, to at least reduce the usage of the flag ?

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I would just drop the flag then.

Hans, any conclusion on this?

I worked around this issue in the affected driver.

Is it upstream? Can you share the commit ID or patch in ML (if it's ready) for that? (Just for the record.)

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=cee3dba7b7416c02ff3

Regards,

Hans