DPDK patches and discussions
 help / color / mirror / Atom feed
* [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver
@ 2015-01-29  1:41 Jingjing Wu
  2015-01-29  1:41 ` [dpdk-dev] [PATCH 1/2] i40e: fix the bug when configuring vsi Jingjing Wu
                   ` (5 more replies)
  0 siblings, 6 replies; 12+ messages in thread
From: Jingjing Wu @ 2015-01-29  1:41 UTC (permalink / raw)
  To: dev

Enable SRIOV switch in i40e driver. With this patch set, SRIOV switch
can be done on Fortville NICs.

Jingjing Wu (2):
  i40e: fix the bug when configuring vsi
  i40e: enable internal switch of pf

 lib/librte_pmd_i40e/i40e_ethdev.c | 38 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

-- 
1.9.3

^ permalink raw reply	[flat|nested] 12+ messages in thread

* [dpdk-dev] [PATCH 1/2] i40e: fix the bug when configuring vsi
  2015-01-29  1:41 [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Jingjing Wu
@ 2015-01-29  1:41 ` Jingjing Wu
  2015-01-29  1:41 ` [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf Jingjing Wu
                   ` (4 subsequent siblings)
  5 siblings, 0 replies; 12+ messages in thread
From: Jingjing Wu @ 2015-01-29  1:41 UTC (permalink / raw)
  To: dev

In i40e_vsi_config_tc_queue_mapping, should add a flag to indicate
another valid setting by OR operation, but not set this flag to
valid_sections, otherwise it will overwrite the flags set before.

Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
---
 lib/librte_pmd_i40e/i40e_ethdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c b/lib/librte_pmd_i40e/i40e_ethdev.c
index b47a3d2..fe758c2 100644
--- a/lib/librte_pmd_i40e/i40e_ethdev.c
+++ b/lib/librte_pmd_i40e/i40e_ethdev.c
@@ -2632,7 +2632,7 @@ i40e_vsi_config_tc_queue_mapping(struct i40e_vsi *vsi,
 			rte_cpu_to_le_16(I40E_AQ_VSI_QUE_MAP_CONTIG);
 		info->queue_mapping[0] = rte_cpu_to_le_16(vsi->base_queue);
 	}
-	info->valid_sections =
+	info->valid_sections |=
 		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_QUEUE_MAP_VALID);
 
 	return I40E_SUCCESS;
-- 
1.9.3

^ permalink raw reply	[flat|nested] 12+ messages in thread

* [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
  2015-01-29  1:41 [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Jingjing Wu
  2015-01-29  1:41 ` [dpdk-dev] [PATCH 1/2] i40e: fix the bug when configuring vsi Jingjing Wu
@ 2015-01-29  1:41 ` Jingjing Wu
  2015-01-29  1:56   ` Qiu, Michael
  2015-02-04  7:41 ` [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Liu, Jijiang
                   ` (3 subsequent siblings)
  5 siblings, 1 reply; 12+ messages in thread
From: Jingjing Wu @ 2015-01-29  1:41 UTC (permalink / raw)
  To: dev

This patch enables PF's internal switch by setting ALLOWLOOPBACK
flag when VEB is created. With this patch, traffic from PF can be
switched on the VEB.

Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
---
 lib/librte_pmd_i40e/i40e_ethdev.c | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c b/lib/librte_pmd_i40e/i40e_ethdev.c
index fe758c2..94fd36c 100644
--- a/lib/librte_pmd_i40e/i40e_ethdev.c
+++ b/lib/librte_pmd_i40e/i40e_ethdev.c
@@ -2854,6 +2854,40 @@ i40e_vsi_dump_bw_config(struct i40e_vsi *vsi)
 	return 0;
 }
 
+/*
+ * i40e_enable_pf_lb
+ * @pf: pointer to the pf structure
+ *
+ * allow loopback on pf
+ */
+static inline void
+i40e_enable_pf_lb(struct i40e_pf *pf)
+{
+	struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+	struct i40e_vsi_context ctxt;
+	int ret;
+
+	memset(&ctxt, 0, sizeof(ctxt));
+	ctxt.seid = pf->main_vsi_seid;
+	ctxt.pf_num = hw->pf_id;
+	ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
+	if (ret) {
+		PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d, aq_err %d",
+			ret, hw->aq.asq_last_status);
+		return;
+	}
+	ctxt.flags = I40E_AQ_VSI_TYPE_PF;
+	ctxt.info.valid_sections =
+		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
+	ctxt.info.switch_id |=
+		rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
+
+	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
+	if (ret)
+		PMD_DRV_LOG(ERR, "update vsi switch failed, aq_err=%d\n",
+			hw->aq.asq_last_status);
+}
+
 /* Setup a VSI */
 struct i40e_vsi *
 i40e_vsi_setup(struct i40e_pf *pf,
@@ -2889,6 +2923,8 @@ i40e_vsi_setup(struct i40e_pf *pf,
 			PMD_DRV_LOG(ERR, "VEB setup failed");
 			return NULL;
 		}
+		/* set ALLOWLOOPBACk on pf, when veb is created */
+		i40e_enable_pf_lb(pf);
 	}
 
 	vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);
-- 
1.9.3

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
  2015-01-29  1:41 ` [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf Jingjing Wu
@ 2015-01-29  1:56   ` Qiu, Michael
  2015-01-29  4:57     ` Wu, Jingjing
  0 siblings, 1 reply; 12+ messages in thread
From: Qiu, Michael @ 2015-01-29  1:56 UTC (permalink / raw)
  To: Wu, Jingjing, dev

On 1/29/2015 9:42 AM, Jingjing Wu wrote:
> This patch enables PF's internal switch by setting ALLOWLOOPBACK
> flag when VEB is created. With this patch, traffic from PF can be
> switched on the VEB.
>
> Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
> ---
>  lib/librte_pmd_i40e/i40e_ethdev.c | 36 ++++++++++++++++++++++++++++++++++++
>  1 file changed, 36 insertions(+)
>
> diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c b/lib/librte_pmd_i40e/i40e_ethdev.c
> index fe758c2..94fd36c 100644
> --- a/lib/librte_pmd_i40e/i40e_ethdev.c
> +++ b/lib/librte_pmd_i40e/i40e_ethdev.c
> @@ -2854,6 +2854,40 @@ i40e_vsi_dump_bw_config(struct i40e_vsi *vsi)
>  	return 0;
>  }
>  
> +/*
> + * i40e_enable_pf_lb
> + * @pf: pointer to the pf structure
> + *
> + * allow loopback on pf
> + */
> +static inline void
> +i40e_enable_pf_lb(struct i40e_pf *pf)
> +{
> +	struct i40e_hw *hw = I40E_PF_TO_HW(pf);
> +	struct i40e_vsi_context ctxt;
> +	int ret;
> +
> +	memset(&ctxt, 0, sizeof(ctxt));
> +	ctxt.seid = pf->main_vsi_seid;
> +	ctxt.pf_num = hw->pf_id;
> +	ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
> +	if (ret) {
> +		PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d, aq_err %d",
> +			ret, hw->aq.asq_last_status);
> +		return;
> +	}
> +	ctxt.flags = I40E_AQ_VSI_TYPE_PF;
> +	ctxt.info.valid_sections =
> +		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);

Here does it need to be "|=" ? As ctxt.infowill be filled in
i40e_aq_get_vsi_params(), I don't know if it has other issue for
override this filled by "=".

Thanks,
Michael
> +	ctxt.info.switch_id |=
> +		rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
> +
> +	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
> +	if (ret)
> +		PMD_DRV_LOG(ERR, "update vsi switch failed, aq_err=%d\n",
> +			hw->aq.asq_last_status);
> +}
> +
>  /* Setup a VSI */
>  struct i40e_vsi *
>  i40e_vsi_setup(struct i40e_pf *pf,
> @@ -2889,6 +2923,8 @@ i40e_vsi_setup(struct i40e_pf *pf,
>  			PMD_DRV_LOG(ERR, "VEB setup failed");
>  			return NULL;
>  		}
> +		/* set ALLOWLOOPBACk on pf, when veb is created */
> +		i40e_enable_pf_lb(pf);
>  	}
>  
>  	vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
  2015-01-29  1:56   ` Qiu, Michael
@ 2015-01-29  4:57     ` Wu, Jingjing
  2015-01-29  6:06       ` Qiu, Michael
  0 siblings, 1 reply; 12+ messages in thread
From: Wu, Jingjing @ 2015-01-29  4:57 UTC (permalink / raw)
  To: Qiu, Michael, dev

Hi, Michael

> -----Original Message-----
> From: Qiu, Michael
> Sent: Thursday, January 29, 2015 9:56 AM
> To: Wu, Jingjing; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
> 
> On 1/29/2015 9:42 AM, Jingjing Wu wrote:
> > This patch enables PF's internal switch by setting ALLOWLOOPBACK flag
> > when VEB is created. With this patch, traffic from PF can be switched
> > on the VEB.
> >
> > Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
> > ---
> >  lib/librte_pmd_i40e/i40e_ethdev.c | 36
> > ++++++++++++++++++++++++++++++++++++
> >  1 file changed, 36 insertions(+)
> >
> > diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
> > b/lib/librte_pmd_i40e/i40e_ethdev.c
> > index fe758c2..94fd36c 100644
> > --- a/lib/librte_pmd_i40e/i40e_ethdev.c
> > +++ b/lib/librte_pmd_i40e/i40e_ethdev.c
> > @@ -2854,6 +2854,40 @@ i40e_vsi_dump_bw_config(struct i40e_vsi *vsi)
> >  	return 0;
> >  }
> >
> > +/*
> > + * i40e_enable_pf_lb
> > + * @pf: pointer to the pf structure
> > + *
> > + * allow loopback on pf
> > + */
> > +static inline void
> > +i40e_enable_pf_lb(struct i40e_pf *pf) {
> > +	struct i40e_hw *hw = I40E_PF_TO_HW(pf);
> > +	struct i40e_vsi_context ctxt;
> > +	int ret;
> > +
> > +	memset(&ctxt, 0, sizeof(ctxt));
> > +	ctxt.seid = pf->main_vsi_seid;
> > +	ctxt.pf_num = hw->pf_id;
> > +	ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
> > +	if (ret) {
> > +		PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d,
> aq_err %d",
> > +			ret, hw->aq.asq_last_status);
> > +		return;
> > +	}
> > +	ctxt.flags = I40E_AQ_VSI_TYPE_PF;
> > +	ctxt.info.valid_sections =
> > +		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
> 
> Here does it need to be "|=" ? As ctxt.infowill be filled in
> i40e_aq_get_vsi_params(), I don't know if it has other issue for override this
> filled by "=".
> 
> Thanks,
> Michael

You can look at the following lines. What we called is i40e_aq_update_vsi_params.
So we need only set the flag we want to update.

Thanks
Jingjing

> > +	ctxt.info.switch_id |=
> > +		rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
> > +
> > +	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
> > +	if (ret)
> > +		PMD_DRV_LOG(ERR, "update vsi switch failed,
> aq_err=%d\n",
> > +			hw->aq.asq_last_status);
> > +}
> > +
> >  /* Setup a VSI */
> >  struct i40e_vsi *
> >  i40e_vsi_setup(struct i40e_pf *pf,
> > @@ -2889,6 +2923,8 @@ i40e_vsi_setup(struct i40e_pf *pf,
> >  			PMD_DRV_LOG(ERR, "VEB setup failed");
> >  			return NULL;
> >  		}
> > +		/* set ALLOWLOOPBACk on pf, when veb is created */
> > +		i40e_enable_pf_lb(pf);
> >  	}
> >
> >  	vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
  2015-01-29  4:57     ` Wu, Jingjing
@ 2015-01-29  6:06       ` Qiu, Michael
  2015-01-29  6:26         ` Wu, Jingjing
  0 siblings, 1 reply; 12+ messages in thread
From: Qiu, Michael @ 2015-01-29  6:06 UTC (permalink / raw)
  To: Wu, Jingjing, dev

On 1/29/2015 12:57 PM, Wu, Jingjing wrote:
> Hi, Michael
>
>> -----Original Message-----
>> From: Qiu, Michael
>> Sent: Thursday, January 29, 2015 9:56 AM
>> To: Wu, Jingjing; dev@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
>>
>> On 1/29/2015 9:42 AM, Jingjing Wu wrote:
>>> This patch enables PF's internal switch by setting ALLOWLOOPBACK flag
>>> when VEB is created. With this patch, traffic from PF can be switched
>>> on the VEB.
>>>
>>> Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
>>> ---
>>>  lib/librte_pmd_i40e/i40e_ethdev.c | 36
>>> ++++++++++++++++++++++++++++++++++++
>>>  1 file changed, 36 insertions(+)
>>>
>>> diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
>>> b/lib/librte_pmd_i40e/i40e_ethdev.c
>>> index fe758c2..94fd36c 100644
>>> --- a/lib/librte_pmd_i40e/i40e_ethdev.c
>>> +++ b/lib/librte_pmd_i40e/i40e_ethdev.c
>>> @@ -2854,6 +2854,40 @@ i40e_vsi_dump_bw_config(struct i40e_vsi *vsi)
>>>  	return 0;
>>>  }
>>>
>>> +/*
>>> + * i40e_enable_pf_lb
>>> + * @pf: pointer to the pf structure
>>> + *
>>> + * allow loopback on pf
>>> + */
>>> +static inline void
>>> +i40e_enable_pf_lb(struct i40e_pf *pf) {
>>> +	struct i40e_hw *hw = I40E_PF_TO_HW(pf);
>>> +	struct i40e_vsi_context ctxt;
>>> +	int ret;
>>> +
>>> +	memset(&ctxt, 0, sizeof(ctxt));
>>> +	ctxt.seid = pf->main_vsi_seid;
>>> +	ctxt.pf_num = hw->pf_id;
>>> +	ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
>>> +	if (ret) {
>>> +		PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d,
>> aq_err %d",
>>> +			ret, hw->aq.asq_last_status);
>>> +		return;
>>> +	}
>>> +	ctxt.flags = I40E_AQ_VSI_TYPE_PF;
>>> +	ctxt.info.valid_sections =
>>> +		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
>> Here does it need to be "|=" ? As ctxt.infowill be filled in
>> i40e_aq_get_vsi_params(), I don't know if it has other issue for override this
>> filled by "=".
>>
>> Thanks,
>> Michael
> You can look at the following lines. What we called is i40e_aq_update_vsi_params.
> So we need only set the flag we want to update.

Sorry, I make a mistake, what I mean is:

1. ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
    here will fill the the field  ctxt.info of struct i40e_vsi_context
ctxt right?
    So ctxt.info is get from other place.

2. Then:

+	ctxt.info.valid_sections =
+		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);

Has been override by assignment a value, so I just confuse whether it has some issue.

If I'm wrong, please ignore.


Thanks,
Michael

> Thanks
> Jingjing
>
>>> +	ctxt.info.switch_id |=
>>> +		rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
>>> +
>>> +	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
>>> +	if (ret)
>>> +		PMD_DRV_LOG(ERR, "update vsi switch failed,
>> aq_err=%d\n",
>>> +			hw->aq.asq_last_status);
>>> +}
>>> +
>>>  /* Setup a VSI */
>>>  struct i40e_vsi *
>>>  i40e_vsi_setup(struct i40e_pf *pf,
>>> @@ -2889,6 +2923,8 @@ i40e_vsi_setup(struct i40e_pf *pf,
>>>  			PMD_DRV_LOG(ERR, "VEB setup failed");
>>>  			return NULL;
>>>  		}
>>> +		/* set ALLOWLOOPBACk on pf, when veb is created */
>>> +		i40e_enable_pf_lb(pf);
>>>  	}
>>>
>>>  	vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);
>


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
  2015-01-29  6:06       ` Qiu, Michael
@ 2015-01-29  6:26         ` Wu, Jingjing
  2015-01-29  6:52           ` Qiu, Michael
  0 siblings, 1 reply; 12+ messages in thread
From: Wu, Jingjing @ 2015-01-29  6:26 UTC (permalink / raw)
  To: Qiu, Michael, dev

Hi, Michael

> -----Original Message-----
> From: Qiu, Michael
> Sent: Thursday, January 29, 2015 2:06 PM
> To: Wu, Jingjing; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
> 
> On 1/29/2015 12:57 PM, Wu, Jingjing wrote:
> > Hi, Michael
> >
> >> -----Original Message-----
> >> From: Qiu, Michael
> >> Sent: Thursday, January 29, 2015 9:56 AM
> >> To: Wu, Jingjing; dev@dpdk.org
> >> Subject: Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of
> >> pf
> >>
> >> On 1/29/2015 9:42 AM, Jingjing Wu wrote:
> >>> This patch enables PF's internal switch by setting ALLOWLOOPBACK
> >>> flag when VEB is created. With this patch, traffic from PF can be
> >>> switched on the VEB.
> >>>
> >>> Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
> >>> ---
> >>>  lib/librte_pmd_i40e/i40e_ethdev.c | 36
> >>> ++++++++++++++++++++++++++++++++++++
> >>>  1 file changed, 36 insertions(+)
> >>>
> >>> diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
> >>> b/lib/librte_pmd_i40e/i40e_ethdev.c
> >>> index fe758c2..94fd36c 100644
> >>> --- a/lib/librte_pmd_i40e/i40e_ethdev.c
> >>> +++ b/lib/librte_pmd_i40e/i40e_ethdev.c
> >>> @@ -2854,6 +2854,40 @@ i40e_vsi_dump_bw_config(struct i40e_vsi
> *vsi)
> >>>  	return 0;
> >>>  }
> >>>
> >>> +/*
> >>> + * i40e_enable_pf_lb
> >>> + * @pf: pointer to the pf structure
> >>> + *
> >>> + * allow loopback on pf
> >>> + */
> >>> +static inline void
> >>> +i40e_enable_pf_lb(struct i40e_pf *pf) {
> >>> +	struct i40e_hw *hw = I40E_PF_TO_HW(pf);
> >>> +	struct i40e_vsi_context ctxt;
> >>> +	int ret;
> >>> +
> >>> +	memset(&ctxt, 0, sizeof(ctxt));
> >>> +	ctxt.seid = pf->main_vsi_seid;
> >>> +	ctxt.pf_num = hw->pf_id;
> >>> +	ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
> >>> +	if (ret) {
> >>> +		PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d,
> >> aq_err %d",
> >>> +			ret, hw->aq.asq_last_status);
> >>> +		return;
> >>> +	}
> >>> +	ctxt.flags = I40E_AQ_VSI_TYPE_PF;
> >>> +	ctxt.info.valid_sections =
> >>> +		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
> >> Here does it need to be "|=" ? As ctxt.infowill be filled in
> >> i40e_aq_get_vsi_params(), I don't know if it has other issue for
> >> override this filled by "=".
> >>
> >> Thanks,
> >> Michael
> > You can look at the following lines. What we called is
> i40e_aq_update_vsi_params.
> > So we need only set the flag we want to update.
> 
> Sorry, I make a mistake, what I mean is:
> 
> 1. ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
>     here will fill the the field  ctxt.info of struct i40e_vsi_context ctxt right?
>     So ctxt.info is get from other place.
> 
> 2. Then:
> 
> +	ctxt.info.valid_sections =
> +		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
> 
> Has been override by assignment a value, so I just confuse whether it has
> some issue.
> 
> If I'm wrong, please ignore. 
> 
> 
> Thanks,
> Michael
> 
I get your idea now. Some elements in ctxt is meaningless and not set when getting, and others are meaningful when
updating. The valid_sections is only meaningful when setting. If one flag in valid_section is set, it means the
hw need to process corresponding section.

> > Thanks
> > Jingjing
> >
> >>> +	ctxt.info.switch_id |=
> >>> +		rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
> >>> +
> >>> +	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
> >>> +	if (ret)
> >>> +		PMD_DRV_LOG(ERR, "update vsi switch failed,
> >> aq_err=%d\n",
> >>> +			hw->aq.asq_last_status);
> >>> +}
> >>> +
> >>>  /* Setup a VSI */
> >>>  struct i40e_vsi *
> >>>  i40e_vsi_setup(struct i40e_pf *pf,
> >>> @@ -2889,6 +2923,8 @@ i40e_vsi_setup(struct i40e_pf *pf,
> >>>  			PMD_DRV_LOG(ERR, "VEB setup failed");
> >>>  			return NULL;
> >>>  		}
> >>> +		/* set ALLOWLOOPBACk on pf, when veb is created */
> >>> +		i40e_enable_pf_lb(pf);
> >>>  	}
> >>>
> >>>  	vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);
> >

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
  2015-01-29  6:26         ` Wu, Jingjing
@ 2015-01-29  6:52           ` Qiu, Michael
  0 siblings, 0 replies; 12+ messages in thread
From: Qiu, Michael @ 2015-01-29  6:52 UTC (permalink / raw)
  To: Wu, Jingjing, dev

On 1/29/2015 2:27 PM, Wu, Jingjing wrote:
> Hi, Michael
>
>> -----Original Message-----
>> From: Qiu, Michael
>> Sent: Thursday, January 29, 2015 2:06 PM
>> To: Wu, Jingjing; dev@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf
>>
>> On 1/29/2015 12:57 PM, Wu, Jingjing wrote:
>>> Hi, Michael
>>>
>>>> -----Original Message-----
>>>> From: Qiu, Michael
>>>> Sent: Thursday, January 29, 2015 9:56 AM
>>>> To: Wu, Jingjing; dev@dpdk.org
>>>> Subject: Re: [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of
>>>> pf
>>>>
>>>> On 1/29/2015 9:42 AM, Jingjing Wu wrote:
>>>>> This patch enables PF's internal switch by setting ALLOWLOOPBACK
>>>>> flag when VEB is created. With this patch, traffic from PF can be
>>>>> switched on the VEB.
>>>>>
>>>>> Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
>>>>> ---
>>>>>  lib/librte_pmd_i40e/i40e_ethdev.c | 36
>>>>> ++++++++++++++++++++++++++++++++++++
>>>>>  1 file changed, 36 insertions(+)
>>>>>
>>>>> diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
>>>>> b/lib/librte_pmd_i40e/i40e_ethdev.c
>>>>> index fe758c2..94fd36c 100644
>>>>> --- a/lib/librte_pmd_i40e/i40e_ethdev.c
>>>>> +++ b/lib/librte_pmd_i40e/i40e_ethdev.c
>>>>> @@ -2854,6 +2854,40 @@ i40e_vsi_dump_bw_config(struct i40e_vsi
>> *vsi)
>>>>>  	return 0;
>>>>>  }
>>>>>
>>>>> +/*
>>>>> + * i40e_enable_pf_lb
>>>>> + * @pf: pointer to the pf structure
>>>>> + *
>>>>> + * allow loopback on pf
>>>>> + */
>>>>> +static inline void
>>>>> +i40e_enable_pf_lb(struct i40e_pf *pf) {
>>>>> +	struct i40e_hw *hw = I40E_PF_TO_HW(pf);
>>>>> +	struct i40e_vsi_context ctxt;
>>>>> +	int ret;
>>>>> +
>>>>> +	memset(&ctxt, 0, sizeof(ctxt));
>>>>> +	ctxt.seid = pf->main_vsi_seid;
>>>>> +	ctxt.pf_num = hw->pf_id;
>>>>> +	ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
>>>>> +	if (ret) {
>>>>> +		PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d,
>>>> aq_err %d",
>>>>> +			ret, hw->aq.asq_last_status);
>>>>> +		return;
>>>>> +	}
>>>>> +	ctxt.flags = I40E_AQ_VSI_TYPE_PF;
>>>>> +	ctxt.info.valid_sections =
>>>>> +		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
>>>> Here does it need to be "|=" ? As ctxt.infowill be filled in
>>>> i40e_aq_get_vsi_params(), I don't know if it has other issue for
>>>> override this filled by "=".
>>>>
>>>> Thanks,
>>>> Michael
>>> You can look at the following lines. What we called is
>> i40e_aq_update_vsi_params.
>>> So we need only set the flag we want to update.
>> Sorry, I make a mistake, what I mean is:
>>
>> 1. ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
>>     here will fill the the field  ctxt.info of struct i40e_vsi_context ctxt right?
>>     So ctxt.info is get from other place.
>>
>> 2. Then:
>>
>> +	ctxt.info.valid_sections =
>> +		rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
>>
>> Has been override by assignment a value, so I just confuse whether it has
>> some issue.
>>
>> If I'm wrong, please ignore. 
>>
>>
>> Thanks,
>> Michael
>>
> I get your idea now. Some elements in ctxt is meaningless and not set when getting, and others are meaningful when
> updating. The valid_sections is only meaningful when setting. If one flag in valid_section is set, it means the
> hw need to process corresponding section.

OK, as it meaningless, I agree with you.

Thanks,
Michael
>>> Thanks
>>> Jingjing
>>>
>>>>> +	ctxt.info.switch_id |=
>>>>> +		rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
>>>>> +
>>>>> +	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
>>>>> +	if (ret)
>>>>> +		PMD_DRV_LOG(ERR, "update vsi switch failed,
>>>> aq_err=%d\n",
>>>>> +			hw->aq.asq_last_status);
>>>>> +}
>>>>> +
>>>>>  /* Setup a VSI */
>>>>>  struct i40e_vsi *
>>>>>  i40e_vsi_setup(struct i40e_pf *pf,
>>>>> @@ -2889,6 +2923,8 @@ i40e_vsi_setup(struct i40e_pf *pf,
>>>>>  			PMD_DRV_LOG(ERR, "VEB setup failed");
>>>>>  			return NULL;
>>>>>  		}
>>>>> +		/* set ALLOWLOOPBACk on pf, when veb is created */
>>>>> +		i40e_enable_pf_lb(pf);
>>>>>  	}
>>>>>
>>>>>  	vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);
>


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver
  2015-01-29  1:41 [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Jingjing Wu
  2015-01-29  1:41 ` [dpdk-dev] [PATCH 1/2] i40e: fix the bug when configuring vsi Jingjing Wu
  2015-01-29  1:41 ` [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf Jingjing Wu
@ 2015-02-04  7:41 ` Liu, Jijiang
  2015-02-09  8:55 ` Zhang, Helin
                   ` (2 subsequent siblings)
  5 siblings, 0 replies; 12+ messages in thread
From: Liu, Jijiang @ 2015-02-04  7:41 UTC (permalink / raw)
  To: Wu, Jingjing, dev



> -----Original Message-----
> From: Wu, Jingjing
> Sent: Thursday, January 29, 2015 9:42 AM
> To: dev@dpdk.org
> Cc: Wu, Jingjing; Zhang, Helin; Chen, Jing D; Cao, Min
> Subject: [PATCH 0/2] enable SRIOV switch in i40e driver
> 
> Enable SRIOV switch in i40e driver. With this patch set, SRIOV switch can be
> done on Fortville NICs.
> 
> Jingjing Wu (2):
>   i40e: fix the bug when configuring vsi
>   i40e: enable internal switch of pf
> 
>  lib/librte_pmd_i40e/i40e_ethdev.c | 38
> +++++++++++++++++++++++++++++++++++++-
>  1 file changed, 37 insertions(+), 1 deletion(-)
> 
> --
> 1.9.3

Acked-by:  Jijiang Liu <Jijiang.liu@intel.com>

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver
  2015-01-29  1:41 [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Jingjing Wu
                   ` (2 preceding siblings ...)
  2015-02-04  7:41 ` [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Liu, Jijiang
@ 2015-02-09  8:55 ` Zhang, Helin
  2015-02-15  6:26 ` Cao, Min
  2015-02-15  6:30 ` Cao, Min
  5 siblings, 0 replies; 12+ messages in thread
From: Zhang, Helin @ 2015-02-09  8:55 UTC (permalink / raw)
  To: Wu, Jingjing; +Cc: dev



> -----Original Message-----
> From: Wu, Jingjing
> Sent: Thursday, January 29, 2015 9:42 AM
> To: dev@dpdk.org
> Cc: Wu, Jingjing; Zhang, Helin; Chen, Jing D; Cao, Min
> Subject: [PATCH 0/2] enable SRIOV switch in i40e driver
> 
> Enable SRIOV switch in i40e driver. With this patch set, SRIOV switch can be
> done on Fortville NICs.
> 
> Jingjing Wu (2):
>   i40e: fix the bug when configuring vsi
>   i40e: enable internal switch of pf
> 
>  lib/librte_pmd_i40e/i40e_ethdev.c | 38
> +++++++++++++++++++++++++++++++++++++-
>  1 file changed, 37 insertions(+), 1 deletion(-)
> 
> --
> 1.9.3

Applied.

Thanks to Jingjing and Jijiang!

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver
  2015-01-29  1:41 [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Jingjing Wu
                   ` (3 preceding siblings ...)
  2015-02-09  8:55 ` Zhang, Helin
@ 2015-02-15  6:26 ` Cao, Min
  2015-02-15  6:30 ` Cao, Min
  5 siblings, 0 replies; 12+ messages in thread
From: Cao, Min @ 2015-02-15  6:26 UTC (permalink / raw)
  To: Wu, Jingjing, dev

Test by: min.cao <min.cao@intel.com>
Patch name: 		[PATCH 0/2] enable SRIOV switch in i40e driver
Test Flag: 			Tested-by
Tester name: 		min.cao@intel.com
Result summary:		total 1 cases, 1 passed, 0 failed

Test Case 1:		
Name:				packet forwarding of SRIOV switch in i40e driver
Environment:		OS: Fedora20 3.11.10-301.fc20.x86_64
				gcc (GCC) 4.8.2
				CPU: Intel(R) Xeon(R) CPU E5-2680 0 @ 2.70GHz
				NIC: Fortville eagle 
Test result:		PASSED
Detail:                 packet forwarding of SRIOV switch in i40e driver are successful between 2 vms.

-----Original Message-----
From: Wu, Jingjing 
Sent: Thursday, January 29, 2015 9:42 AM
To: dev@dpdk.org
Cc: Wu, Jingjing; Zhang, Helin; Chen, Jing D; Cao, Min
Subject: [PATCH 0/2] enable SRIOV switch in i40e driver

Enable SRIOV switch in i40e driver. With this patch set, SRIOV switch
can be done on Fortville NICs.

Jingjing Wu (2):
  i40e: fix the bug when configuring vsi
  i40e: enable internal switch of pf

 lib/librte_pmd_i40e/i40e_ethdev.c | 38 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

-- 
1.9.3

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver
  2015-01-29  1:41 [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Jingjing Wu
                   ` (4 preceding siblings ...)
  2015-02-15  6:26 ` Cao, Min
@ 2015-02-15  6:30 ` Cao, Min
  5 siblings, 0 replies; 12+ messages in thread
From: Cao, Min @ 2015-02-15  6:30 UTC (permalink / raw)
  To: Cao, Min, Wu, Jingjing, dev

Tested-by: min.cao <min.cao@intel.com>
Patch name: 		[PATCH 0/2] enable SRIOV switch in i40e driver
Test Flag: 			Tested-by
Tester name: 		min.cao@intel.com
Result summary:		total 1 cases, 1 passed, 0 failed

Test Case 1:		
Name:				packet forwarding of SRIOV switch in i40e driver
Environment:		OS: Fedora20 3.11.10-301.fc20.x86_64
				gcc (GCC) 4.8.2
				CPU: Intel(R) Xeon(R) CPU E5-2680 0 @ 2.70GHz
				NIC: Fortville eagle 
Test result:		PASSED
Detail:                 packet forwarding of SRIOV switch in i40e driver are successful between 2 vms.

-----Original Message-----
From: Wu, Jingjing 
Sent: Thursday, January 29, 2015 9:42 AM
To: dev@dpdk.org
Cc: Wu, Jingjing; Zhang, Helin; Chen, Jing D; Cao, Min
Subject: [PATCH 0/2] enable SRIOV switch in i40e driver

Enable SRIOV switch in i40e driver. With this patch set, SRIOV switch
can be done on Fortville NICs.

Jingjing Wu (2):
  i40e: fix the bug when configuring vsi
  i40e: enable internal switch of pf

 lib/librte_pmd_i40e/i40e_ethdev.c | 38 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

-- 
1.9.3

^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, other threads:[~2015-02-15  6:30 UTC | newest]

Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-01-29  1:41 [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Jingjing Wu
2015-01-29  1:41 ` [dpdk-dev] [PATCH 1/2] i40e: fix the bug when configuring vsi Jingjing Wu
2015-01-29  1:41 ` [dpdk-dev] [PATCH 2/2] i40e: enable internal switch of pf Jingjing Wu
2015-01-29  1:56   ` Qiu, Michael
2015-01-29  4:57     ` Wu, Jingjing
2015-01-29  6:06       ` Qiu, Michael
2015-01-29  6:26         ` Wu, Jingjing
2015-01-29  6:52           ` Qiu, Michael
2015-02-04  7:41 ` [dpdk-dev] [PATCH 0/2] enable SRIOV switch in i40e driver Liu, Jijiang
2015-02-09  8:55 ` Zhang, Helin
2015-02-15  6:26 ` Cao, Min
2015-02-15  6:30 ` Cao, Min

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).