Re: [PATCH 02/20] dma-buf: rework stub fence initialisation v2

["Christian König" <christian.koenig@amd.com>]

On 11/4/25 16:01, Tvrtko Ursulin wrote:
> On 31/10/2025 13:16, Christian König wrote:
>>   /**
>>    * dma_fence_get_stub - return a signaled fence
>>    *
>> - * Return a stub fence which is already signaled. The fence's
>> - * timestamp corresponds to the first time after boot this
>> - * function is called.
>> + * Return a stub fence which is already signaled. The fence's timestamp
>> + * corresponds to the initialisation time of the linux kernel.
>>    */
>>   struct dma_fence *dma_fence_get_stub(void)
>>   {
>> -    spin_lock(&dma_fence_stub_lock);
>> -    if (!dma_fence_stub.ops) {
>> -        dma_fence_init(&dma_fence_stub,
>> -                   &dma_fence_stub_ops,
>> -                   &dma_fence_stub_lock,
>> -                   0, 0);
>> -
>> -        set_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT,
>> -            &dma_fence_stub.flags);
>> -
>> -        dma_fence_signal_locked(&dma_fence_stub);
>> -    }
>> -    spin_unlock(&dma_fence_stub_lock);
>> -
>>       return dma_fence_get(&dma_fence_stub);
> 
> Actually could you check if this could be demoted to static inline? It strikes me pointless to export a symbol for such a trivial wrapper.

I thought about that previously as well, but the answer is "No it can't".

We would need to expose the dma_fence_stub symbol then and the reason we have the function in the first place is to avoid that.

Regards,
Christian.

> 
> Regards,
> 
> Tvrtko
> 
>>   }
>>   EXPORT_SYMBOL(dma_fence_get_stub);
>