Re: [Xen-devel] [Patch][RESEND] Add hardware CR8 acceleration for TPRaccessing

[Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>]

Why set the TPR threshold to non-zero when current TPR is less than highest
IRR? The TPR is not gating injection of the interrupt, so why would we care
if its value becomes even smaller -- it's already small enough?

Couldn't we have
    if (highest_irr == -1) || (highest_irr > tpr)
        TPR_thresh=0
    else
        TPR_thresh = highest_irr>>4

 -- Keir

On 30/10/06 07:02, "Cui, Dexuan" <dexuan.cui@xxxxxxxxx> wrote:

> Any advice about the patch cr8-acceleration-3.patch?
> Hi Keir, could you give some comments? Thanks!
>  -- Dexuan
> 
> -----Original Message-----
> From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
> [mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Cui, Dexuan
> Sent: 2006年10月25日 11:12
> To: Keir.Fraser@xxxxxxxxxxxx
> Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
> Subject: [Xen-devel] [Patch][RESEND] Add hardware CR8 acceleration for
> TPRaccessing
> 
> x64 Windows uses CR8 to access TPR very frequently. This patch enables
> TPR shadow and allows mov-from/to-CR8 to access it directly; tests
> indicates it can boost greatly the performance of x64 Windows 2003.
> 
> Thanks
>  -- Dexuan
> 
> Signed-off-by: Eddie Dong <eddie.dong@xxxxxxxxx>
> Signed-off-by: Yunhong Jiang <yunhong.jiang@xxxxxxxxx>
> Signed-off-by: Dexuan Cui <dexuan.cui@xxxxxxxxx>
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-devel



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel