patches for DPDK stable branches
 help / color / mirror / Atom feed
* [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling
@ 2018-08-02 17:21 Maxime Coquelin
  2018-08-03  2:30 ` Tiwei Bie
  2018-08-03 11:27 ` [dpdk-stable] " Jens Freimann
  0 siblings, 2 replies; 6+ messages in thread
From: Maxime Coquelin @ 2018-08-02 17:21 UTC (permalink / raw)
  To: tiwei.bie, zhihong.wang, jfreimann, dev; +Cc: stable, Maxime Coquelin

IOTLB entries contain the host virtual address of the guest
pages. When receiving a new VHOST_USER_SET_MEM_TABLE request,
the previous regions get unmapped, so the IOTLB entries, if any,
will be invalid. It does cause the vhost-user process to
segfault.

This patch introduces a new function to flush the IOTLB cache,
and call it as soon as the backend handles a VHOST_USER_SET_MEM
request.

Fixes: 69c90e98f483 ("vhost: enable IOMMU support")
Cc: stable@dpdk.org

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
Changes since v1:
- Fix indentation (Stephen)
- Fix double iotlb-lock lock

 lib/librte_vhost/iotlb.c      | 10 ++++++++--
 lib/librte_vhost/iotlb.h      |  2 +-
 lib/librte_vhost/vhost_user.c |  5 +++++
 3 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/lib/librte_vhost/iotlb.c b/lib/librte_vhost/iotlb.c
index c11ebcaac..c6354fef7 100644
--- a/lib/librte_vhost/iotlb.c
+++ b/lib/librte_vhost/iotlb.c
@@ -303,6 +303,13 @@ vhost_user_iotlb_cache_find(struct vhost_virtqueue *vq, uint64_t iova,
 	return vva;
 }
 
+void
+vhost_user_iotlb_flush_all(struct vhost_virtqueue *vq)
+{
+	vhost_user_iotlb_cache_remove_all(vq);
+	vhost_user_iotlb_pending_remove_all(vq);
+}
+
 int
 vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
 {
@@ -315,8 +322,7 @@ vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
 		 * The cache has already been initialized,
 		 * just drop all cached and pending entries.
 		 */
-		vhost_user_iotlb_cache_remove_all(vq);
-		vhost_user_iotlb_pending_remove_all(vq);
+		vhost_user_iotlb_flush_all(vq);
 	}
 
 #ifdef RTE_LIBRTE_VHOST_NUMA
diff --git a/lib/librte_vhost/iotlb.h b/lib/librte_vhost/iotlb.h
index e7083e37b..60b9e4c57 100644
--- a/lib/librte_vhost/iotlb.h
+++ b/lib/librte_vhost/iotlb.h
@@ -73,7 +73,7 @@ void vhost_user_iotlb_pending_insert(struct vhost_virtqueue *vq, uint64_t iova,
 						uint8_t perm);
 void vhost_user_iotlb_pending_remove(struct vhost_virtqueue *vq, uint64_t iova,
 						uint64_t size, uint8_t perm);
-
+void vhost_user_iotlb_flush_all(struct vhost_virtqueue *vq);
 int vhost_user_iotlb_init(struct virtio_net *dev, int vq_index);
 
 #endif /* _VHOST_IOTLB_H_ */
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index dc53ff712..a2d4c9ffc 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -813,6 +813,11 @@ vhost_user_set_mem_table(struct virtio_net **pdev, struct VhostUserMsg *pmsg)
 		dev->mem = NULL;
 	}
 
+	/* Flush IOTLB cache as previous HVAs are now invalid */
+	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
+		for (i = 0; i < dev->nr_vring; i++)
+			vhost_user_iotlb_flush_all(dev->virtqueue[i]);
+
 	dev->nr_guest_pages = 0;
 	if (!dev->guest_pages) {
 		dev->max_guest_pages = 8;
-- 
2.14.4

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling
  2018-08-02 17:21 [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling Maxime Coquelin
@ 2018-08-03  2:30 ` Tiwei Bie
  2018-08-03  7:54   ` Maxime Coquelin
  2018-08-03 11:27 ` [dpdk-stable] " Jens Freimann
  1 sibling, 1 reply; 6+ messages in thread
From: Tiwei Bie @ 2018-08-03  2:30 UTC (permalink / raw)
  To: Maxime Coquelin; +Cc: zhihong.wang, jfreimann, dev, stable

On Thu, Aug 02, 2018 at 07:21:22PM +0200, Maxime Coquelin wrote:
> IOTLB entries contain the host virtual address of the guest
> pages. When receiving a new VHOST_USER_SET_MEM_TABLE request,
> the previous regions get unmapped, so the IOTLB entries, if any,
> will be invalid. It does cause the vhost-user process to
> segfault.
> 
> This patch introduces a new function to flush the IOTLB cache,
> and call it as soon as the backend handles a VHOST_USER_SET_MEM
> request.
> 
> Fixes: 69c90e98f483 ("vhost: enable IOMMU support")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
> ---
> Changes since v1:
> - Fix indentation (Stephen)
> - Fix double iotlb-lock lock
> 
>  lib/librte_vhost/iotlb.c      | 10 ++++++++--
>  lib/librte_vhost/iotlb.h      |  2 +-
>  lib/librte_vhost/vhost_user.c |  5 +++++
>  3 files changed, 14 insertions(+), 3 deletions(-)
> 
> diff --git a/lib/librte_vhost/iotlb.c b/lib/librte_vhost/iotlb.c
> index c11ebcaac..c6354fef7 100644
> --- a/lib/librte_vhost/iotlb.c
> +++ b/lib/librte_vhost/iotlb.c
> @@ -303,6 +303,13 @@ vhost_user_iotlb_cache_find(struct vhost_virtqueue *vq, uint64_t iova,
>  	return vva;
>  }
>  
> +void
> +vhost_user_iotlb_flush_all(struct vhost_virtqueue *vq)
> +{
> +	vhost_user_iotlb_cache_remove_all(vq);
> +	vhost_user_iotlb_pending_remove_all(vq);
> +}
> +
>  int
>  vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
>  {
> @@ -315,8 +322,7 @@ vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
>  		 * The cache has already been initialized,
>  		 * just drop all cached and pending entries.
>  		 */
> -		vhost_user_iotlb_cache_remove_all(vq);
> -		vhost_user_iotlb_pending_remove_all(vq);
> +		vhost_user_iotlb_flush_all(vq);
>  	}
>  
>  #ifdef RTE_LIBRTE_VHOST_NUMA
> diff --git a/lib/librte_vhost/iotlb.h b/lib/librte_vhost/iotlb.h
> index e7083e37b..60b9e4c57 100644
> --- a/lib/librte_vhost/iotlb.h
> +++ b/lib/librte_vhost/iotlb.h
> @@ -73,7 +73,7 @@ void vhost_user_iotlb_pending_insert(struct vhost_virtqueue *vq, uint64_t iova,
>  						uint8_t perm);
>  void vhost_user_iotlb_pending_remove(struct vhost_virtqueue *vq, uint64_t iova,
>  						uint64_t size, uint8_t perm);
> -
> +void vhost_user_iotlb_flush_all(struct vhost_virtqueue *vq);
>  int vhost_user_iotlb_init(struct virtio_net *dev, int vq_index);
>  
>  #endif /* _VHOST_IOTLB_H_ */
> diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
> index dc53ff712..a2d4c9ffc 100644
> --- a/lib/librte_vhost/vhost_user.c
> +++ b/lib/librte_vhost/vhost_user.c
> @@ -813,6 +813,11 @@ vhost_user_set_mem_table(struct virtio_net **pdev, struct VhostUserMsg *pmsg)
>  		dev->mem = NULL;
>  	}
>  
> +	/* Flush IOTLB cache as previous HVAs are now invalid */
> +	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
> +		for (i = 0; i < dev->nr_vring; i++)
> +			vhost_user_iotlb_flush_all(dev->virtqueue[i]);

Why is the pending list also flushed?

Thanks,
Tiwei

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling
  2018-08-03  2:30 ` Tiwei Bie
@ 2018-08-03  7:54   ` Maxime Coquelin
  2018-08-03  8:37     ` Tiwei Bie
  0 siblings, 1 reply; 6+ messages in thread
From: Maxime Coquelin @ 2018-08-03  7:54 UTC (permalink / raw)
  To: Tiwei Bie; +Cc: zhihong.wang, jfreimann, dev, stable



On 08/03/2018 04:30 AM, Tiwei Bie wrote:
> On Thu, Aug 02, 2018 at 07:21:22PM +0200, Maxime Coquelin wrote:
>> IOTLB entries contain the host virtual address of the guest
>> pages. When receiving a new VHOST_USER_SET_MEM_TABLE request,
>> the previous regions get unmapped, so the IOTLB entries, if any,
>> will be invalid. It does cause the vhost-user process to
>> segfault.
>>
>> This patch introduces a new function to flush the IOTLB cache,
>> and call it as soon as the backend handles a VHOST_USER_SET_MEM
>> request.
>>
>> Fixes: 69c90e98f483 ("vhost: enable IOMMU support")
>> Cc: stable@dpdk.org
>>
>> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>> ---
>> Changes since v1:
>> - Fix indentation (Stephen)
>> - Fix double iotlb-lock lock
>>
>>   lib/librte_vhost/iotlb.c      | 10 ++++++++--
>>   lib/librte_vhost/iotlb.h      |  2 +-
>>   lib/librte_vhost/vhost_user.c |  5 +++++
>>   3 files changed, 14 insertions(+), 3 deletions(-)
>>
>> diff --git a/lib/librte_vhost/iotlb.c b/lib/librte_vhost/iotlb.c
>> index c11ebcaac..c6354fef7 100644
>> --- a/lib/librte_vhost/iotlb.c
>> +++ b/lib/librte_vhost/iotlb.c
>> @@ -303,6 +303,13 @@ vhost_user_iotlb_cache_find(struct vhost_virtqueue *vq, uint64_t iova,
>>   	return vva;
>>   }
>>   
>> +void
>> +vhost_user_iotlb_flush_all(struct vhost_virtqueue *vq)
>> +{
>> +	vhost_user_iotlb_cache_remove_all(vq);
>> +	vhost_user_iotlb_pending_remove_all(vq);
>> +}
>> +
>>   int
>>   vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
>>   {
>> @@ -315,8 +322,7 @@ vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
>>   		 * The cache has already been initialized,
>>   		 * just drop all cached and pending entries.
>>   		 */
>> -		vhost_user_iotlb_cache_remove_all(vq);
>> -		vhost_user_iotlb_pending_remove_all(vq);
>> +		vhost_user_iotlb_flush_all(vq);
>>   	}
>>   
>>   #ifdef RTE_LIBRTE_VHOST_NUMA
>> diff --git a/lib/librte_vhost/iotlb.h b/lib/librte_vhost/iotlb.h
>> index e7083e37b..60b9e4c57 100644
>> --- a/lib/librte_vhost/iotlb.h
>> +++ b/lib/librte_vhost/iotlb.h
>> @@ -73,7 +73,7 @@ void vhost_user_iotlb_pending_insert(struct vhost_virtqueue *vq, uint64_t iova,
>>   						uint8_t perm);
>>   void vhost_user_iotlb_pending_remove(struct vhost_virtqueue *vq, uint64_t iova,
>>   						uint64_t size, uint8_t perm);
>> -
>> +void vhost_user_iotlb_flush_all(struct vhost_virtqueue *vq);
>>   int vhost_user_iotlb_init(struct virtio_net *dev, int vq_index);
>>   
>>   #endif /* _VHOST_IOTLB_H_ */
>> diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
>> index dc53ff712..a2d4c9ffc 100644
>> --- a/lib/librte_vhost/vhost_user.c
>> +++ b/lib/librte_vhost/vhost_user.c
>> @@ -813,6 +813,11 @@ vhost_user_set_mem_table(struct virtio_net **pdev, struct VhostUserMsg *pmsg)
>>   		dev->mem = NULL;
>>   	}
>>   
>> +	/* Flush IOTLB cache as previous HVAs are now invalid */
>> +	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
>> +		for (i = 0; i < dev->nr_vring; i++)
>> +			vhost_user_iotlb_flush_all(dev->virtqueue[i]);
> 
> Why is the pending list also flushed?

As it might be asynchronous, I think it is better to flush the pending
list too.

For example, the backend request a translation just before the guest
remove the driver, the IOVA requested might not be valid anymore and
so no reply will be sent by QEMU. So the request would remain in the
pending list forever.

I don't doing that is mandatory, but it does nor hurt IMHO.

Maxime
> Thanks,
> Tiwei
> 

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling
  2018-08-03  7:54   ` Maxime Coquelin
@ 2018-08-03  8:37     ` Tiwei Bie
  2018-08-04 23:49       ` [dpdk-stable] [dpdk-dev] " Thomas Monjalon
  0 siblings, 1 reply; 6+ messages in thread
From: Tiwei Bie @ 2018-08-03  8:37 UTC (permalink / raw)
  To: Maxime Coquelin; +Cc: zhihong.wang, jfreimann, dev, stable

On Fri, Aug 03, 2018 at 09:54:21AM +0200, Maxime Coquelin wrote:
> On 08/03/2018 04:30 AM, Tiwei Bie wrote:
> > On Thu, Aug 02, 2018 at 07:21:22PM +0200, Maxime Coquelin wrote:
[...]
> >>   
> >>   #endif /* _VHOST_IOTLB_H_ */
> >> diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
> >> index dc53ff712..a2d4c9ffc 100644
> >> --- a/lib/librte_vhost/vhost_user.c
> >> +++ b/lib/librte_vhost/vhost_user.c
> >> @@ -813,6 +813,11 @@ vhost_user_set_mem_table(struct virtio_net **pdev, struct VhostUserMsg *pmsg)
> >>   		dev->mem = NULL;
> >>   	}
> >>   
> >> +	/* Flush IOTLB cache as previous HVAs are now invalid */
> >> +	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
> >> +		for (i = 0; i < dev->nr_vring; i++)
> >> +			vhost_user_iotlb_flush_all(dev->virtqueue[i]);
> > 
> > Why is the pending list also flushed?
> 
> As it might be asynchronous, I think it is better to flush the pending
> list too.
> 
> For example, the backend request a translation just before the guest
> remove the driver, the IOVA requested might not be valid anymore and
> so no reply will be sent by QEMU. So the request would remain in the
> pending list forever.
> 
> I don't doing that is mandatory, but it does nor hurt IMHO.

Yeah, it doesn't hurt. I was just curious about
why you want to do that. :)

Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling
  2018-08-02 17:21 [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling Maxime Coquelin
  2018-08-03  2:30 ` Tiwei Bie
@ 2018-08-03 11:27 ` Jens Freimann
  1 sibling, 0 replies; 6+ messages in thread
From: Jens Freimann @ 2018-08-03 11:27 UTC (permalink / raw)
  To: Maxime Coquelin; +Cc: tiwei.bie, zhihong.wang, dev, stable

On Thu, Aug 02, 2018 at 07:21:22PM +0200, Maxime Coquelin wrote:
>IOTLB entries contain the host virtual address of the guest
>pages. When receiving a new VHOST_USER_SET_MEM_TABLE request,
>the previous regions get unmapped, so the IOTLB entries, if any,
>will be invalid. It does cause the vhost-user process to
>segfault.
>
>This patch introduces a new function to flush the IOTLB cache,
>and call it as soon as the backend handles a VHOST_USER_SET_MEM
>request.
>
>Fixes: 69c90e98f483 ("vhost: enable IOMMU support")
>Cc: stable@dpdk.org
>
>Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>---
>Changes since v1:
>- Fix indentation (Stephen)
>- Fix double iotlb-lock lock
>
> lib/librte_vhost/iotlb.c      | 10 ++++++++--
> lib/librte_vhost/iotlb.h      |  2 +-
> lib/librte_vhost/vhost_user.c |  5 +++++
> 3 files changed, 14 insertions(+), 3 deletions(-)
>

Reviewed-by: Jens Freimann <jfreimann@redhat.com>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-stable] [dpdk-dev] [PATCH v2] vhost: flush IOTLB cache on new mem table handling
  2018-08-03  8:37     ` Tiwei Bie
@ 2018-08-04 23:49       ` Thomas Monjalon
  0 siblings, 0 replies; 6+ messages in thread
From: Thomas Monjalon @ 2018-08-04 23:49 UTC (permalink / raw)
  To: Maxime Coquelin; +Cc: dev, Tiwei Bie, zhihong.wang, jfreimann, stable

03/08/2018 10:37, Tiwei Bie:
> On Fri, Aug 03, 2018 at 09:54:21AM +0200, Maxime Coquelin wrote:
> > On 08/03/2018 04:30 AM, Tiwei Bie wrote:
> > > On Thu, Aug 02, 2018 at 07:21:22PM +0200, Maxime Coquelin wrote:
> [...]
> > >>   
> > >>   #endif /* _VHOST_IOTLB_H_ */
> > >> diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
> > >> index dc53ff712..a2d4c9ffc 100644
> > >> --- a/lib/librte_vhost/vhost_user.c
> > >> +++ b/lib/librte_vhost/vhost_user.c
> > >> @@ -813,6 +813,11 @@ vhost_user_set_mem_table(struct virtio_net **pdev, struct VhostUserMsg *pmsg)
> > >>   		dev->mem = NULL;
> > >>   	}
> > >>   
> > >> +	/* Flush IOTLB cache as previous HVAs are now invalid */
> > >> +	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
> > >> +		for (i = 0; i < dev->nr_vring; i++)
> > >> +			vhost_user_iotlb_flush_all(dev->virtqueue[i]);
> > > 
> > > Why is the pending list also flushed?
> > 
> > As it might be asynchronous, I think it is better to flush the pending
> > list too.
> > 
> > For example, the backend request a translation just before the guest
> > remove the driver, the IOVA requested might not be valid anymore and
> > so no reply will be sent by QEMU. So the request would remain in the
> > pending list forever.
> > 
> > I don't doing that is mandatory, but it does nor hurt IMHO.
> 
> Yeah, it doesn't hurt. I was just curious about
> why you want to do that. :)
> 
> Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>

Applied, thanks

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2018-08-04 23:49 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-08-02 17:21 [dpdk-stable] [PATCH v2] vhost: flush IOTLB cache on new mem table handling Maxime Coquelin
2018-08-03  2:30 ` Tiwei Bie
2018-08-03  7:54   ` Maxime Coquelin
2018-08-03  8:37     ` Tiwei Bie
2018-08-04 23:49       ` [dpdk-stable] [dpdk-dev] " Thomas Monjalon
2018-08-03 11:27 ` [dpdk-stable] " Jens Freimann

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).