[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Xen-devel] Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable
- To: Zachary Amsden <zach@xxxxxxxxxx>
- From: Andi Kleen <ak@xxxxxx>
- Date: 20 Mar 2007 21:02:42 +0100
- Date: Tue, 20 Mar 2007 21:02:42 +0100
- Cc: Jeremy Fitzhardinge <jeremy@xxxxxxxx>, xen-devel@xxxxxxxxxxxxxxxxxxx, akpm@xxxxxxxxxxxxxxxxxxxx, netdev@xxxxxxxxxxxxxxx, Rusty Russell <rusty@xxxxxxxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, chrisw@xxxxxxxxxxxx, virtualization@xxxxxxxxxxxxxx, "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx>, anthony@xxxxxxxxxxxxx, mingo@xxxxxxx, Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>, David Miller <davem@xxxxxxxxxxxxx>
- Delivery-date: Tue, 20 Mar 2007 13:06:16 -0700
- List-id: Xen developer discussion <xen-devel.lists.xensource.com>
> Worth it on 32-bit. On AMD64, probably not. On Intel 64-bit, maybe,
> but less important than in P4 days.
Well most of Intel 64bit is P4 -- and Intel is still shipping
millions more of them each quarter.
> This could change character completely if used at the tail of a function
> where you now have
>
> sti; 1: ret
>
> Which generates an interrupt holdoff on the ret, an unusual thing to do.
Unusual yes, but I don't see how it should cause problems. Or do you
have anything specific in mind?
Worse is probably that on K8 this case might cause a pipeline stall
if there isn't a prefix in front of the ret.
-Andi
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
