Re: [PATCH v5 7/9] vpci: split vpci_process_pending()

On 08.07.2026 23:02, Stewart Hildebrand wrote:
> Create a new function to separate the mapping logic from the pdev
> validation and lock-acquiring logic. Slightly consolidate exit paths for
> hardware_domain and !hardware_domain.
> 
> No functional change.

Is this then just for the sake of splitting, perhaps indeed helping clarity some?
Or is this split also going to be leveraged later?

> --- a/xen/drivers/vpci/header.c
> +++ b/xen/drivers/vpci/header.c
> @@ -102,23 +102,12 @@ static int cf_check map_range(
>      return rc;
>  }
>  
> -bool vpci_process_pending(struct vcpu *v)
> +static int process_map_task(const struct pci_dev *pdev, bool map)

In the end it'll of course be Roger's call, yet I wonder: What's the significance
of "task" in the new function's name?

Jan