RE: [Xen-ia64-devel] [PATCH] Reload KRs when domain switch

["Tian, Kevin" <kevin.tian@xxxxxxxxx>]

Same confusion to me. Just know I found the reason. Last time when we claimed 
domU working, we're using xenlinux-ia64.hg and I just confirmed that tip still 
read/write KRs by shared page. Actually this is the first time we tried to boot 
domU on sparse directory where guest is modified to access physical KRs 
directly. So you see this issue jumping out to bring us big troubles since we 
have assumption that it should work before. ;-)

Thanks,
Kevin

>-----Original Message-----
>From: Magenheimer, Dan (HP Labs Fort Collins) [mailto:dan.magenheimer@xxxxxx]
>Sent: 2005年11月1日 10:59
>To: Tian, Kevin; xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>Subject: RE: [Xen-ia64-devel] [PATCH] Reload KRs when domain switch
>
>Another good catch!
>
>At first I was confused by this:  The guest has owned the KR's
>for a long time (Eddie submitted a patch a couple of months
>ago I think).  However, until recently, xenlinux wasn't actually
>using the KRs!
>
>Dan
>
>> -----Original Message-----
>> From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
>> [mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf
>> Of Tian, Kevin
>> Sent: Monday, October 31, 2005 6:55 PM
>> To: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
>> Subject: [Xen-ia64-devel] [PATCH] Reload KRs when domain switch
>>
>> Hi, Dan,
>>
>> This is next block which made dom0 in nested dtlb miss forever. The
>> reason is that KR is not reloaded after domain switch, and
>> thus invalid
>> KRs are used. Several TLB miss related handlers in guest depend on KR7
>> to walk current page table. Then invalid KR value made everything
>> messed.
>>
>> Signed-off-by Kevin Tian <Kevin.tian@xxxxxxxxx>
>>
>> Thanks,
>> Kevin
>>

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel