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

Re: [Xen-devel] [PATCH] x86/S3: Restore broken vcpu affinity on resume



On 27/03/13 12:04, Ben Guthro wrote:

On 03/27/2013 08:01 AM, George Dunlap wrote:
On Tue, Mar 26, 2013 at 5:20 PM, Ben Guthro <benjamin.guthro@xxxxxxxxxx> wrote:
When in SYS_STATE_suspend, and going through the cpu_disable_scheduler
path, save a copy of the current cpu affinity, and mark a flag to
restore it later.

Later, in the resume process, when enabling nonboot cpus restore these
affinities.

This is the second submission of this patch.
Primary differences from the first patch is to fix formatting problems.
However, when doing so, I tested with another patch in the
cpu_disable_scheduler() path that is also appropriate here.

Signed-off-by: Ben Guthro <benjamin.guthro@xxxxxxxxxx>
Overall looks fine to me; just a few comments below.

diff --git a/xen/common/cpupool.c b/xen/common/cpupool.c
index 10b10f8..7a04f5e 100644
--- a/xen/common/cpupool.c
+++ b/xen/common/cpupool.c
@@ -19,13 +19,10 @@
   #include <xen/sched-if.h>
   #include <xen/cpu.h>

-#define for_each_cpupool(ptr)    \
-    for ((ptr) = &cpupool_list; *(ptr) != NULL; (ptr) = &((*(ptr))->next))
-
You're taking this out because it's not used, I presume?

Since you'll probably be sending another patch anyway (see below), I
think it would be better if you pull this out into a specific
"clean-up" patch.
No. This was moved to an h file to allow use elsewhere.
I'm in the process of looking into Jan's suggestion of eliminating the
need for it by moving some code into thaw_domains()

Oh, right -- sorry, I missed it way down at the bottom of sched-if.h.  N/m.

 -George


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel


 


Rackspace

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