accel/ivpu: Remove lockdep_assert_irqs_disabled()

[PATCH] accel/ivpu: Remove lockdep_assert_irqs_disabled()

Posted by Maarten Lankhorst 2 months, 3 weeks ago

This breaks on PREEMPT_RT, and should be unneeded since
lockdep can track irq disabled status itself.

Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
---
 drivers/accel/ivpu/ivpu_ipc.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c
index 39f83225c1815..5f00809d448af 100644
--- a/drivers/accel/ivpu/ivpu_ipc.c
+++ b/drivers/accel/ivpu/ivpu_ipc.c
@@ -141,7 +141,6 @@ ivpu_ipc_rx_msg_add(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons,
 	struct ivpu_ipc_rx_msg *rx_msg;
 
 	lockdep_assert_held(&ipc->cons_lock);
-	lockdep_assert_irqs_disabled();
 
 	rx_msg = kzalloc(sizeof(*rx_msg), GFP_ATOMIC);
 	if (!rx_msg) {
-- 
2.45.2

Re: [PATCH] accel/ivpu: Remove lockdep_assert_irqs_disabled()

Posted by Jacek Lawrynowicz 2 months, 2 weeks ago

Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>

On 7/15/2025 6:59 PM, Maarten Lankhorst wrote:
> This breaks on PREEMPT_RT, and should be unneeded since
> lockdep can track irq disabled status itself.
> 
> Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
> ---
>  drivers/accel/ivpu/ivpu_ipc.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c
> index 39f83225c1815..5f00809d448af 100644
> --- a/drivers/accel/ivpu/ivpu_ipc.c
> +++ b/drivers/accel/ivpu/ivpu_ipc.c
> @@ -141,7 +141,6 @@ ivpu_ipc_rx_msg_add(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons,
>  	struct ivpu_ipc_rx_msg *rx_msg;
>  
>  	lockdep_assert_held(&ipc->cons_lock);
> -	lockdep_assert_irqs_disabled();
>  
>  	rx_msg = kzalloc(sizeof(*rx_msg), GFP_ATOMIC);
>  	if (!rx_msg) {

Re: [PATCH] accel/ivpu: Remove lockdep_assert_irqs_disabled()

Posted by Maarten Lankhorst 2 months, 2 weeks ago

Hey,

Den 2025-07-22 kl. 09:53, skrev Jacek Lawrynowicz:
> Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>

Thanks, pushed to misc-next.

Best regards,
~Maarten
 
> On 7/15/2025 6:59 PM, Maarten Lankhorst wrote:
>> This breaks on PREEMPT_RT, and should be unneeded since
>> lockdep can track irq disabled status itself.
>>
>> Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
>> ---
>>  drivers/accel/ivpu/ivpu_ipc.c | 1 -
>>  1 file changed, 1 deletion(-)
>>
>> diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c
>> index 39f83225c1815..5f00809d448af 100644
>> --- a/drivers/accel/ivpu/ivpu_ipc.c
>> +++ b/drivers/accel/ivpu/ivpu_ipc.c
>> @@ -141,7 +141,6 @@ ivpu_ipc_rx_msg_add(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons,
>>  	struct ivpu_ipc_rx_msg *rx_msg;
>>  
>>  	lockdep_assert_held(&ipc->cons_lock);
>> -	lockdep_assert_irqs_disabled();
>>  
>>  	rx_msg = kzalloc(sizeof(*rx_msg), GFP_ATOMIC);
>>  	if (!rx_msg) {
>