From mboxrd@z Thu Jan  1 00:00:00 1970
From: Mukesh Rathor <mukesh.rathor@oracle.com>
Subject: Re: [BUGFIX][PATCH 4/4] XEN_DOMCTL_gdbsx_guestmemio:
 always do the copyback.
Date: Tue, 7 Jan 2014 15:01:48 -0800
Message-ID: <20140107150148.4cbf1a73@mantra.us.oracle.com>
References: <1388857936-664-1-git-send-email-dslutz@verizon.com>
	<1388857936-664-5-git-send-email-dslutz@verizon.com>
	<20140106175349.6cbd190b@mantra.us.oracle.com>
	<1389088824.31766.105.camel@kazak.uk.xensource.com>
	<1389088937.31766.107.camel@kazak.uk.xensource.com>
	<52CC2A2F.7010700@terremark.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xen.org>
In-Reply-To: <52CC2A2F.7010700@terremark.com>
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: Don Slutz <dslutz@verizon.com>
Cc: Keir Fraser <keir@xen.org>, Ian Campbell <Ian.Campbell@citrix.com>, Stefano Stabellini <stefano.stabellini@eu.citrix.com>, George Dunlap <george.dunlap@eu.citrix.com>, Ian Jackson <ian.jackson@eu.citrix.com>, xen-devel@lists.xen.org, Jan Beulich <jbeulich@suse.com>
List-Id: xen-devel@lists.xenproject.org

On Tue, 07 Jan 2014 11:24:15 -0500
Don Slutz <dslutz@verizon.com> wrote:

> On 01/07/14 05:02, Ian Campbell wrote:
> > On Tue, 2014-01-07 at 10:00 +0000, Ian Campbell wrote:
> >> On Mon, 2014-01-06 at 17:53 -0800, Mukesh Rathor wrote:
> >>> On Sat,  4 Jan 2014 12:52:16 -0500
> >>> Don Slutz <dslutz@verizon.com> wrote:
> >>>
> >>>> The gdbsx code expects that domctl->u.gdbsx_guest_memio.remain is
> >>>> returned.
> >>>>

..... 

> I had assumed that this patch (which is not need to "fix" the bugs I
> found), was to be dropped in v2.  However I will agree that currently
> there is no way to know about partial success.  The untested change:
> 
> diff --git a/xen/arch/x86/domctl.c b/xen/arch/x86/domctl.c
> index ef6c140..0add07e 100644
> --- a/xen/arch/x86/domctl.c
> +++ b/xen/arch/x86/domctl.c
> @@ -43,7 +43,7 @@ static int gdbsx_guest_mem_io(
>       iop->remain = dbg_rw_mem(
>           (dbgva_t)iop->gva, (dbgbyte_t *)l_uva, iop->len, domid,
>           iop->gwr, iop->pgd3val);
> -    return (iop->remain ? -EFAULT : 0);
> +    return 0;
>   }
> 
>   long arch_do_domctl(
> 
> 
> Would appear to allow partial success to be reported and also meet
> with remain not to be looked at with an error.

No, the partial success is relevant in other cases, like EAGAIN,
but not EFAULT. If we make it pre-emptible in future to return
EAGAIN, we'd need to make sure remain was honored. Again, think of it
this way, if the first copyin failed, remain would have not been 
initialized.

So, because now the only cause of unfinished copy from dbg_rw_mem is 
EFAULT, we should leave above xen code alone, and just change gdbsx.

thanks
mukesh