[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH] enable hpet after s3 resume


  • To: "Lu, Guanqun" <guanqun.lu@xxxxxxxxx>, xen-devel组 <xen-devel@xxxxxxxxxxxxxxxxxxx>
  • From: Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
  • Date: Tue, 18 Nov 2008 09:57:27 +0000
  • Cc:
  • Delivery-date: Tue, 18 Nov 2008 01:58:43 -0800
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>
  • Thread-index: AclJXv30XOVi3v8HTtK9waDYksrQrwABRIDO
  • Thread-topic: [Xen-devel] [PATCH] enable hpet after s3 resume

On 18/11/08 09:21, "Lu, Guanqun" <guanqun.lu@xxxxxxxxx> wrote:

> Some timer source requires re-initialization after S3 resume,
> due to reset when backing to S0. Some cleanup to allow such
> change generic.
> 
> Signed-off-by Guanqun Lu <guanqun.lu@xxxxxxxxx>
> Signed-off-by Kevin Tian <kevin.tian@xxxxxxxxx>

The hpet.c changes are ugly. What's so special about hpet_broadcast_init()
that it needs to distinguish between boot and S3 resume? In the latter case
it executes with APs still offline and IRQs disabled, so what's wrong with
rewriting the fixmap entry, re-initialising the spinlock and hpet_event
info?

Rather than passing a param to plt_src.init() instead define a
plt_src.resume() for HPET only, doing only hpet_setup(). Again, I thinkw e
ought to be able to avoid a resume/not-resume boolean pretty easily.

Please split yor cleanup of time.c (the listing of platform timer structures
and iterating over them into a separate patch [so send two patches, 1/2 and
2/2]).

 -- Keir



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


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.