Re: [PATCH] Abstracted Priority Inheritance for RT

[Daniel Walker]

On Wed, 1 Jun 2005, Ingo Molnar wrote:

> i'd rather not slow things down by callbacks and other abstraction 
> before seeing how things want to integrate in fact. Do we really need 
> the callbacks?

I think it would be hard to do without a way to signal when a waiter
changes priorties. Since other structures could handle it differently.

Another problem is that there needs to be a clear way to know which
structure owns the rt_mutex_waiter . Something in there needs to be 
unique. It can't be assumed anymore that everything is an rt_mutex. 

The lock owner could be put into the rt_mutex_waiter structure. Which
would make the structure bigger, but it's usually stack space. This would
also create some duplicate data since every waiter would need to hold the
owners task_struct pointer. 


Daniel

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/