Re: [PATCH 2/5] VIRT: Support runtime irq_bypass consumer

[Alex Williamson <alex.williamson@redhat.com>]

On Thu, 2015-12-03 at 10:22 -0800, Yunhong Jiang wrote:
> Extend the irq_bypass manager to support runtime consumers. A runtime
> irq_bypass consumer can handle interrupt when an interrupt triggered. A
> runtime consumer has it's handle_irq() function set and passing a
> irq_context for the irq handling.
> 
> A producer keep a link for the runtime consumers, so that it can invoke
> each consumer's handle_irq() when irq invoked.
> 
> Currently the irq_bypass manager has several code path assuming there is
> only one consumer/producer pair for each token. For example, when
> register the producer, it exits the loop after finding one match
> consumer.  This is updated to support both static consumer (like for
> Posted Interrupt consumer) and runtime consumer.
> 
> Signed-off-by: Yunhong Jiang <yunhong.jiang@linux.intel.com>
> ---
>  include/linux/irqbypass.h |  8 +++++
>  virt/lib/irqbypass.c      | 82 +++++++++++++++++++++++++++++++++++------------
>  2 files changed, 69 insertions(+), 21 deletions(-)
> 
> diff --git a/include/linux/irqbypass.h b/include/linux/irqbypass.h
> index 1551b5b2f4c2..d5bec0c7be3a 100644
> --- a/include/linux/irqbypass.h
> +++ b/include/linux/irqbypass.h
> @@ -12,6 +12,7 @@
>  #define IRQBYPASS_H
>  
>  #include 
> +#include 
>  
>  struct irq_bypass_consumer;
>  
> @@ -47,6 +48,9 @@ struct irq_bypass_consumer;
>   */
>  struct irq_bypass_producer {
>  	struct list_head node;
> +	/* Update side is synchronized by the lock on irqbypass.c */
> +	struct srcu_struct srcu;
> +	struct list_head consumers;
>  	void *token;
>  	int irq;
>  	int (*add_consumer)(struct irq_bypass_producer *,

Documentation?

> @@ -61,6 +65,7 @@ struct irq_bypass_producer {
>   * struct irq_bypass_consumer - IRQ bypass consumer definition
>   * @node: IRQ bypass manager private list management
>   * @token: opaque token to match between producer and consumer
> + * @sibling: consumers with same token list management
>   * @add_producer: Connect the IRQ consumer to an IRQ producer
>   * @del_producer: Disconnect the IRQ consumer from an IRQ producer
>   * @stop: Perform any quiesce operations necessary prior to add/del (optional)

What about @handle_irq and @irq_context?

> @@ -73,6 +78,7 @@ struct irq_bypass_producer {
>   */
>  struct irq_bypass_consumer {
>  	struct list_head node;
> +	struct list_head sibling;
>  	void *token;
>  	int (*add_producer)(struct irq_bypass_consumer *,
>  			    struct irq_bypass_producer *);
> @@ -80,6 +86,8 @@ struct irq_bypass_consumer {
>  			     struct irq_bypass_producer *);
>  	void (*stop)(struct irq_bypass_consumer *);
>  	void (*start)(struct irq_bypass_consumer *);
> +	int (*handle_irq)(void *arg);

If we called this with a pointer to the consumer, like the other
functions, the consumer could embed arg (irq_context) into their own
structure, or in this case, do a container_of and avoid storing the
irqfd pointer entirely.

> +	void *irq_context;
>  };
>  
>  int irq_bypass_register_producer(struct irq_bypass_producer *);