WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

Re: [Xen-devel] [IA64] Weekly benchmark results [2009ww16]

To: KUWAMURA Shin'ya <kuwa@xxxxxxxxxxxxxx>
Subject: Re: [Xen-devel] [IA64] Weekly benchmark results [2009ww16]
From: Isaku Yamahata <yamahata@xxxxxxxxxxxxx>
Date: Fri, 24 Apr 2009 21:09:10 +0900
Cc: xen-devel@xxxxxxxxxxxxxxxxxxx, xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
Delivery-date: Fri, 24 Apr 2009 05:09:34 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <20090424.172743.78359305.kuwa@xxxxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <20090424.172743.78359305.kuwa@xxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mutt/1.5.6i
Can you please try this patch?

linux/blktap: fix vma_close() for partial munmap.

the previous vma_close() didn't take partial 
munmap into acount. This patch fixes it.

Signed-off-by: Isaku Yamahata <yamahata@xxxxxxxxxxxxx>

diff --git a/drivers/xen/blktap/blktap.c b/drivers/xen/blktap/blktap.c
--- a/drivers/xen/blktap/blktap.c
+++ b/drivers/xen/blktap/blktap.c
@@ -295,6 +295,7 @@ static inline int OFFSET_TO_SEG(int offs
  */
 struct tap_vma_priv {
        tap_blkif_t *info;
+       struct kref kref;
        struct page *map[];
 };
 
@@ -392,19 +393,31 @@ static pte_t blktap_clear_pte(struct vm_
        return copy;
 }
 
+static void blktap_vma_open(struct vm_area_struct *vma)
+{
+       struct tap_vma_priv *priv = vma->vm_private_data;
+       kref_get(&priv->kref);
+}
+
+
+static void blktap_vma_release(struct kref *kref)
+{
+       struct tap_vma_priv *priv = container_of(kref,
+                                                struct tap_vma_priv, kref);
+       priv->info->vma = NULL;
+       kfree(priv);
+}
+
 static void blktap_vma_close(struct vm_area_struct *vma)
 {
        struct tap_vma_priv *priv = vma->vm_private_data;
-
-       if (priv) {
-               priv->info->vma = NULL;
-               kfree(priv);
-       }
+       kref_put(&priv->kref, blktap_vma_release);
 }
 
 struct vm_operations_struct blktap_vm_ops = {
        nopage:   blktap_nopage,
        zap_pte:  blktap_clear_pte,
+       open:     blktap_vma_open,
        close:    blktap_vma_close,
 };
 
@@ -707,6 +720,7 @@ static int blktap_mmap(struct file *filp
                goto fail;
        }
        priv->info = info;
+       kref_init(&priv->kref);
 
        vma->vm_private_data = priv;
        vma->vm_flags |= VM_FOREIGN;


On Fri, Apr 24, 2009 at 05:27:43PM +0900, KUWAMURA Shin'ya wrote:
> Hi,
> 
> I report a benchmark result of this week on IPF using
> ia64/xen-unstable and ia64/linux-2.6.18-xen.
> 
> - The following cset broke tap:aio.
>   863: blktap: don't access deallocated data
>   http://xenbits.xensource.com/linux-2.6.18-xen.hg?rev/464a925d73f1
>   Guests hung up while booting, so we used file instead of tap:aio.
> 
> - The issue that HVM domains cannot read the contents of HDD is fixed.
> 
> TEST ENVIRONMENT
>     Machine          : Tiger4
>     Kernel           : 2.6.18.8-xen
>     Changeset        : 19562:5599cc1e0a84 (ia64/xen-unstable)
>                      863:464a925d73f1   (ia64/linux-2.6.18-xen)
>                      136:eb97cb2936d6   (efi-vfirmware)
>                      08251d3            (qemu-xen-unstable)
>     Dom0 OS          : RHEL4 U2 (2P)
>     DomU OS          : RHEL4 U2 (8P, using tap:aio)
>     DomVTi OS        : RHEL4 U2 (8P, with PV-on-HVM drivers)
>     CONFIG_QEMU      : ioemu-remote
> 
> TEST RESULTS
>   DomU: # file
>     unixbench4.1.0    : Pass
>     bonnie++-1.03     : Pass
>     ltp-full-20070930 : Pass
>     iozone3_191       : Pass
>     lmbench-3.0-a5    : Pass
>   DomVTi: # file
>     unixbench4.1.0    : Pass
>     bonnie++-1.03     : Pass
>     ltp-full-20070930 : Pass
>     iozone3_191       : Pass
>     lmbench-3.0-a5    : Pass
> 
> Best regards,
> KUWAMURA Shin'ya and FUJITSU members
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-devel
> 

-- 
yamahata

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