| Message ID | 1520590048-151284-1-git-send-email-junjie.j.chen@intel.com (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Headers | show |
| Context | Check | Description |
|---|---|---|
| ci/checkpatch | success | coding style OK |
| ci/Intel-compilation | success | Compilation OK |
On 3/9/2018 6:07 PM, Junjie Chen wrote: > In vhost-switch example, when binding nic to vfio-pci with iommu enabled, > dequeue zero copy cannot work in VM2NIC mode due to no iommu dma mapping > is setup for guest memory currently. > > Signed-off-by: Junjie Chen <junjie.j.chen@intel.com> > --- > Changes in V3: > - update limitation to iommu > Changes in V2: > - add doc in vhost lib > > doc/guides/prog_guide/vhost_lib.rst | 5 +++++ > doc/guides/sample_app_ug/vhost.rst | 5 ++++- > 2 files changed, 9 insertions(+), 1 deletion(-) > > diff --git a/doc/guides/prog_guide/vhost_lib.rst b/doc/guides/prog_guide/vhost_lib.rst > index 18227b6..06d214f 100644 > --- a/doc/guides/prog_guide/vhost_lib.rst > +++ b/doc/guides/prog_guide/vhost_lib.rst > @@ -83,6 +83,11 @@ The following is an overview of some key Vhost API functions: > of those segments, thus the fewer the segments, the quicker we will get > the mapping. NOTE: we may speed it by using tree searching in future. > > + * zero copy does not work when using driver with iommu mode currently, this Considering FreeBSD driver nic_uio does not support iommu, we can make "driver" more explicitly. For example, "driver with iommu mode" -> "vfio-pci with iommu mode" > + is because we don't setup iommu dma mapping for guest memory. > For example, when you bind device to vfio-pci driver, you need to set driver to work in noiommu mode. Reword the above sentence a little bit: "If vfio-pci is a must in your case, insert vfio-pci kernel module in noiommu mode." > + > - ``RTE_VHOST_USER_IOMMU_SUPPORT`` > > IOMMU support will be enabled when this flag is set. It is disabled by > diff --git a/doc/guides/sample_app_ug/vhost.rst b/doc/guides/sample_app_ug/vhost.rst > index a4bdc6a..f0bb169 100644 > --- a/doc/guides/sample_app_ug/vhost.rst > +++ b/doc/guides/sample_app_ug/vhost.rst > @@ -147,7 +147,10 @@ retries on an RX burst, it takes effect only when rx retry is enabled. The > default value is 15. > > **--dequeue-zero-copy** > -Dequeue zero copy will be enabled when this option is given. > +Dequeue zero copy will be enabled when this option is given. it is worth to > +note that if NIC is binded to driver with iommu enabled, dequeue zero copy > +cannot work at VM2NIC mode (vm2vm=0) due to currently we don't setup iommu > +dma mapping for guest memory. > > **--vlan-strip 0|1** > VLAN strip option is removed, because different NICs have different behaviors
Hi Jianfeng. > -----Original Message----- > From: Tan, Jianfeng > Sent: Monday, March 12, 2018 11:15 AM > To: Chen, Junjie J <junjie.j.chen@intel.com>; yliu@fridaylinux.org; > maxime.coquein@redhat.com > Cc: dev@dpdk.org > Subject: Re: [PATCH v3] doc: add driver limitation for vhost dequeue zero copy > > > > On 3/9/2018 6:07 PM, Junjie Chen wrote: > > In vhost-switch example, when binding nic to vfio-pci with iommu > > enabled, dequeue zero copy cannot work in VM2NIC mode due to no iommu > > dma mapping is setup for guest memory currently. > > > > Signed-off-by: Junjie Chen <junjie.j.chen@intel.com> > > --- > > Changes in V3: > > - update limitation to iommu > > Changes in V2: > > - add doc in vhost lib > > > > doc/guides/prog_guide/vhost_lib.rst | 5 +++++ > > doc/guides/sample_app_ug/vhost.rst | 5 ++++- > > 2 files changed, 9 insertions(+), 1 deletion(-) > > > > diff --git a/doc/guides/prog_guide/vhost_lib.rst > > b/doc/guides/prog_guide/vhost_lib.rst > > index 18227b6..06d214f 100644 > > --- a/doc/guides/prog_guide/vhost_lib.rst > > +++ b/doc/guides/prog_guide/vhost_lib.rst > > @@ -83,6 +83,11 @@ The following is an overview of some key Vhost API > functions: > > of those segments, thus the fewer the segments, the quicker we will > get > > the mapping. NOTE: we may speed it by using tree searching in > future. > > > > + * zero copy does not work when using driver with iommu mode > > + currently, this > > Considering FreeBSD driver nic_uio does not support iommu, we can make > "driver" more explicitly. For example, > > "driver with iommu mode" -> "vfio-pci with iommu mode" Will update. > > > + is because we don't setup iommu dma mapping for guest memory. > > > For example, when you bind device to vfio-pci driver, you need to set driver to > work in noiommu mode. > > Reword the above sentence a little bit: > "If vfio-pci is a must in your case, insert vfio-pci kernel module in noiommu > mode." Will update, thanks. > > + > > - ``RTE_VHOST_USER_IOMMU_SUPPORT`` > > > > IOMMU support will be enabled when this flag is set. It is > > disabled by diff --git a/doc/guides/sample_app_ug/vhost.rst > > b/doc/guides/sample_app_ug/vhost.rst > > index a4bdc6a..f0bb169 100644 > > --- a/doc/guides/sample_app_ug/vhost.rst > > +++ b/doc/guides/sample_app_ug/vhost.rst > > @@ -147,7 +147,10 @@ retries on an RX burst, it takes effect only when rx > retry is enabled. The > > default value is 15. > > > > **--dequeue-zero-copy** > > -Dequeue zero copy will be enabled when this option is given. > > +Dequeue zero copy will be enabled when this option is given. it is > > +worth to note that if NIC is binded to driver with iommu enabled, > > +dequeue zero copy cannot work at VM2NIC mode (vm2vm=0) due to > > +currently we don't setup iommu dma mapping for guest memory. > > > > **--vlan-strip 0|1** > > VLAN strip option is removed, because different NICs have different > > behaviors
diff --git a/doc/guides/prog_guide/vhost_lib.rst b/doc/guides/prog_guide/vhost_lib.rst index 18227b6..06d214f 100644 --- a/doc/guides/prog_guide/vhost_lib.rst +++ b/doc/guides/prog_guide/vhost_lib.rst @@ -83,6 +83,11 @@ The following is an overview of some key Vhost API functions: of those segments, thus the fewer the segments, the quicker we will get the mapping. NOTE: we may speed it by using tree searching in future. + * zero copy does not work when using driver with iommu mode currently, this + is because we don't setup iommu dma mapping for guest memory. For example, + when you bind device to vfio-pci driver, you need to set driver to work + in noiommu mode. + - ``RTE_VHOST_USER_IOMMU_SUPPORT`` IOMMU support will be enabled when this flag is set. It is disabled by diff --git a/doc/guides/sample_app_ug/vhost.rst b/doc/guides/sample_app_ug/vhost.rst index a4bdc6a..f0bb169 100644 --- a/doc/guides/sample_app_ug/vhost.rst +++ b/doc/guides/sample_app_ug/vhost.rst @@ -147,7 +147,10 @@ retries on an RX burst, it takes effect only when rx retry is enabled. The default value is 15. **--dequeue-zero-copy** -Dequeue zero copy will be enabled when this option is given. +Dequeue zero copy will be enabled when this option is given. it is worth to +note that if NIC is binded to driver with iommu enabled, dequeue zero copy +cannot work at VM2NIC mode (vm2vm=0) due to currently we don't setup iommu +dma mapping for guest memory. **--vlan-strip 0|1** VLAN strip option is removed, because different NICs have different behaviors
In vhost-switch example, when binding nic to vfio-pci with iommu enabled, dequeue zero copy cannot work in VM2NIC mode due to no iommu dma mapping is setup for guest memory currently. Signed-off-by: Junjie Chen <junjie.j.chen@intel.com> --- Changes in V3: - update limitation to iommu Changes in V2: - add doc in vhost lib doc/guides/prog_guide/vhost_lib.rst | 5 +++++ doc/guides/sample_app_ug/vhost.rst | 5 ++++- 2 files changed, 9 insertions(+), 1 deletion(-)