On Fri, 9 Jun 2000, Tigran Aivazian wrote:
> >
> > Now that we have access to MSRs from userspace, is there any reason
> > this could not be done better there ?
> >
>
> yes, there is - the lack of ability to bind a userspace process to a CPU
> is the main reason why I implemented it differently from the other
> "classical" implementations.
heh, I just noticed that Linux msr driver is actually more clever than
what I called "classical" msr driver (I shouldn't say that - my friend and
colleague actually wrote the one for UnixWare7!), i.e. in Linux you don't
need to bind to a CPU since /dev/cpu/msr? is actually an array of one
minor per CPU.
So, yes, I agree - you can rewrite this whole thing entirely in userspace.
Regards,
Tigran
-
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/
This archive was generated by hypermail 2b29 : Thu Jun 15 2000 - 21:00:18 EST