RE: [RFC/PATCH] FUSYN Realtime & Robust mutexes for Linux try 2.1

[Perez-Gonzalez, Inaky]

> From: Bernhard Kuhn [mailto:bkuhn@xxxxxxxxxxxxxx]

> > This code proposes an implementation of kernel based mutexes,
> 
> Pretty interessting stuff! I will inspect if i could combine
> it with the "real-time interrupts" i recently described
> (http://www.linuxdevices.com/articles/AT6105045931.html).

I saw it, nice trick!

> Currently i'm protecting critical areas with "prioritized
> spinlocks" that don't provide a priority inversion aviodance
> scheme. Having "real" mutexes with priority inheritence
> should be pretty helpfull to make the kernel hard real time
> aware.

Scratch, scratch. Well, the mutexes (fulocks) are implemented
using spinlocks; I mean, the spinlock protects the fulock
structure that describes the mutex (wait list, owner, etc),
so I am afraid we'd be chasing our tail.

What's special about the prioritized spinlocks? I don't remember
having read about that in the article.

Iñaky Pérez-González -- Not speaking for Intel -- all opinions are my own (and my fault)
-
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/