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

RE: [Xen-devel] [PATCH] Fixing iopl and ioperm


  • To: "Keir Fraser" <Keir.Fraser@xxxxxxxxxxxx>
  • From: "Nakajima, Jun" <jun.nakajima@xxxxxxxxx>
  • Date: Mon, 13 Jun 2005 16:06:34 -0700
  • Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
  • Delivery-date: Mon, 13 Jun 2005 23:06:33 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>
  • Thread-index: AcVwaXK8N6xvUM3OQdePRmdyyiW1ZgAAVa6w
  • Thread-topic: [Xen-devel] [PATCH] Fixing iopl and ioperm

Keir Fraser wrote:
> On 11 Jun 2005, at 06:45, Nakajima, Jun wrote:
> 
>> With this patch, x86 and x86-64 share ioport.c, fixing LTP iopl and
>> ioperm testcase failures (on both). We found an iopl testcase
>> failing even on x86 xenlinux. 
>> 
>> Now x86-64 xenlinux should have the same results regarding the LTP
>> testcases (as far as we tested).
> 
> Why does this patch modify the i386 ioport.c so much? I would expect
> that you ought to be able to use the xen/i386 ioport.c with no
> modification at all. The different function prototypes for sys_iopl
> between i386 and x86_64 shouldn't matter -- x86_64 ought to work with
> teh existing xen/i386 function prototype and function implementation.
> 
That part is basically coming from the implementation of set_bitmap; x86
and x86-64 uses different implemenations. We'll try that in a sperate
patch; I think the x86-64 one is simpler.

Well I heard the current x86 (_not_ 64-bit) xenlinux failed one of LTP
iopl/ioperm testcases (and it passes on the native Linux). So I expect
some changes are required there. Maybe that patch should go first. I'll
double-check it.

Jun
---
Intel Open Source Technology Center

_______________________________________________
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®.