Re: [PATCH v2] Do not mark ACPI devices as irq safe

[Andi Shyti]

Hi Breno,

You don't need to resend the patch. Because the changes are only
in the commit log, I can take care of them.

First of all, we need to fix the title to be:

"i2c: tegra: Do not mark ACPI devices as irq safe"

On Tue, Aug 13, 2024 at 09:12:53AM GMT, Breno Leitao wrote:
> On ACPI machines, the tegra i2c module encounters an issue due to a
> mutex being called inside a spinlock. This leads to the following bug:
> 
> 	BUG: sleeping function called from invalid context at kernel/locking/mutex.c:585
> 	in_atomic(): 0, irqs_disabled(): 1, non_block: 0, pid: 1282, name: kssif0010
> 	preempt_count: 0, expected: 0
> 	RCU nest depth: 0, expected: 0
> 	irq event stamp: 0
> 
> 	Call trace:
> 	__might_sleep
> 	__mutex_lock_common
> 	mutex_lock_nested
> 	acpi_subsys_runtime_resume
> 	rpm_resume
> 	tegra_i2c_xfer

We can keep the trace as:

	BUG: sleeping function called from invalid context at kernel/locking/mutex.c:585
	...

	Call trace:
	__might_sleep
	__mutex_lock_common
	mutex_lock_nested
	acpi_subsys_runtime_resume
	rpm_resume
	tegra_i2c_xfer

> The problem arises because during __pm_runtime_resume(), the spinlock
> &dev->power.lock is acquired before rpm_resume() is called. Later,
> rpm_resume() invokes acpi_subsys_runtime_resume(), which relies on
> mutexes, triggering the error.
> 
> To address this issue, devices on ACPI are now marked as not IRQ-safe,
> considering the dependency of acpi_subsys_runtime_resume() on mutexes.
> 
> Co-developed-by: Michael van der Westhuizen <rmikey@xxxxxxxx>
> Signed-off-by: Michael van der Westhuizen <rmikey@xxxxxxxx>
> Signed-off-by: Breno Leitao <leitao@xxxxxxxxxx>
> Reviewed-by: Dmitry Osipenko <digetx@xxxxxxxxx>
> Reviewed-by: Andy Shevchenko <andy@xxxxxxxxxx>

I haven't seen Andy explicitly tagging this patch. Andy, can we
keep it? Or have I missed it.

Besides, you also need:

Fixes: ede2299f7101 ("i2c: tegra: Support atomic transfers")
Cc: <stable@xxxxxxxxxxxxxxx> # v5.6+

Can you please check whether this is right?

This patch won't apply, though, as far as 5.6 so you should
expect to provide some support for the stable backport.

Andi