> Partly the paths are nothing like as serialized as you might think.
> Sct has also done some patches (which DaveM then extended a lot) to
> drop kernel locks during cached data copies too.
Something I see often, is that having two processes running on an SMP
box, accessing different devices, on different spindles -- when one
process is writing data out, the other pretty much blocks in a D
state.
This is frustrating to say the least and I haven't the faintest idea
on how I might fix this.....
-cw
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/