* [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support
@ 2019-11-18 0:43 Qi Zhang
2019-11-18 5:57 ` Yang, Qiming
2019-11-19 5:52 ` Ye Xiaolong
0 siblings, 2 replies; 5+ messages in thread
From: Qi Zhang @ 2019-11-18 0:43 UTC (permalink / raw)
To: qiming.yang, yahui.cao, beilei.xing; +Cc: dev, Qi Zhang
Since not all data paths support flow mark, the driver need
a hint from application to select the correct data path if
flow mark is required. The patch introduce a devarg
"flow-mark-support" as a workaround solution, since a standard
way is still ongoing.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
v3:
- add "experimental notification" in ice.rst
v2:
- fix typo
doc/guides/nics/ice.rst | 12 ++++++++++++
drivers/net/ice/ice_ethdev.c | 11 ++++++++++-
drivers/net/ice/ice_ethdev.h | 1 +
drivers/net/ice/ice_rxtx_vec_common.h | 5 +++++
4 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/doc/guides/nics/ice.rst b/doc/guides/nics/ice.rst
index 1a426438d..d1b80c2fa 100644
--- a/doc/guides/nics/ice.rst
+++ b/doc/guides/nics/ice.rst
@@ -80,6 +80,18 @@ Runtime Config Options
-w 80:00.0,pipeline-mode-support=1
+- ``Flow Mark Support`` (default ``0``)
+
+ This is a hint to the driver to select the data path that support flow mark extraction
+ by default.
+ NOTE: This is an experimental devarg, it will be removed when any of below conditions
+ is ready.
+ 1) all data path support flow mark (currently vPMD does not)
+ 2) a new offload like RTE_DEV_RX_OFFLOAD_FLOW_MARK be introduced as a standard way to hint.
+ Example::
+
+ -w 80:00.0,flow-mark-support=1
+
- ``Protocol extraction for per queue``
Configure the RX queues to do protocol extraction into mbuf for protocol
diff --git a/drivers/net/ice/ice_ethdev.c b/drivers/net/ice/ice_ethdev.c
index abf00d404..9f2cb2f40 100644
--- a/drivers/net/ice/ice_ethdev.c
+++ b/drivers/net/ice/ice_ethdev.c
@@ -23,11 +23,13 @@
/* devargs */
#define ICE_SAFE_MODE_SUPPORT_ARG "safe-mode-support"
#define ICE_PIPELINE_MODE_SUPPORT_ARG "pipeline-mode-support"
+#define ICE_FLOW_MARK_SUPPORT_ARG "flow-mark-support"
#define ICE_PROTO_XTR_ARG "proto_xtr"
static const char * const ice_valid_args[] = {
ICE_SAFE_MODE_SUPPORT_ARG,
ICE_PIPELINE_MODE_SUPPORT_ARG,
+ ICE_FLOW_MARK_SUPPORT_ARG,
ICE_PROTO_XTR_ARG,
NULL
};
@@ -1987,6 +1989,12 @@ static int ice_parse_devargs(struct rte_eth_dev *dev)
ret = rte_kvargs_process(kvlist, ICE_PIPELINE_MODE_SUPPORT_ARG,
&parse_bool, &ad->devargs.pipe_mode_support);
+ if (ret)
+ goto bail;
+
+ ret = rte_kvargs_process(kvlist, ICE_FLOW_MARK_SUPPORT_ARG,
+ &parse_bool, &ad->devargs.flow_mark_support);
+ printf("flow_mark = %d\n", ad->devargs.flow_mark_support);
bail:
rte_kvargs_free(kvlist);
@@ -4571,7 +4579,8 @@ RTE_PMD_REGISTER_KMOD_DEP(net_ice, "* igb_uio | uio_pci_generic | vfio-pci");
RTE_PMD_REGISTER_PARAM_STRING(net_ice,
ICE_PROTO_XTR_ARG "=[queue:]<vlan|ipv4|ipv6|ipv6_flow|tcp>"
ICE_SAFE_MODE_SUPPORT_ARG "=<0|1>"
- ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>");
+ ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>"
+ ICE_FLOW_MARK_SUPPORT_ARG "=<0|1>");
RTE_INIT(ice_init_log)
{
diff --git a/drivers/net/ice/ice_ethdev.h b/drivers/net/ice/ice_ethdev.h
index 4a0d37b32..4d35339a7 100644
--- a/drivers/net/ice/ice_ethdev.h
+++ b/drivers/net/ice/ice_ethdev.h
@@ -391,6 +391,7 @@ struct ice_devargs {
int safe_mode_support;
uint8_t proto_xtr_dflt;
int pipe_mode_support;
+ int flow_mark_support;
uint8_t proto_xtr[ICE_MAX_QUEUE_NUM];
};
diff --git a/drivers/net/ice/ice_rxtx_vec_common.h b/drivers/net/ice/ice_rxtx_vec_common.h
index 080ca4175..086428898 100644
--- a/drivers/net/ice/ice_rxtx_vec_common.h
+++ b/drivers/net/ice/ice_rxtx_vec_common.h
@@ -268,6 +268,11 @@ ice_rx_vec_dev_check_default(struct rte_eth_dev *dev)
{
int i;
struct ice_rx_queue *rxq;
+ struct ice_adapter *ad =
+ ICE_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
+
+ if (ad->devargs.flow_mark_support)
+ return -1;
for (i = 0; i < dev->data->nb_rx_queues; i++) {
rxq = dev->data->rx_queues[i];
--
2.13.6
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support
2019-11-18 0:43 [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support Qi Zhang
@ 2019-11-18 5:57 ` Yang, Qiming
2019-11-19 5:52 ` Ye Xiaolong
1 sibling, 0 replies; 5+ messages in thread
From: Yang, Qiming @ 2019-11-18 5:57 UTC (permalink / raw)
To: Zhang, Qi Z, Cao, Yahui, Xing, Beilei; +Cc: dev
Hi,
> -----Original Message-----
> From: Zhang, Qi Z
> Sent: Monday, November 18, 2019 8:43 AM
> To: Yang, Qiming <qiming.yang@intel.com>; Cao, Yahui
> <yahui.cao@intel.com>; Xing, Beilei <beilei.xing@intel.com>
> Cc: dev@dpdk.org; Zhang, Qi Z <qi.z.zhang@intel.com>
> Subject: [PATCH v3] net/ice: add flow mark hint support
>
> Since not all data paths support flow mark, the driver need a hint from
> application to select the correct data path if flow mark is required. The patch
> introduce a devarg "flow-mark-support" as a workaround solution, since a
> standard way is still ongoing.
>
> Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
> ---
>
> v3:
> - add "experimental notification" in ice.rst
>
> v2:
> - fix typo
>
> doc/guides/nics/ice.rst | 12 ++++++++++++
> drivers/net/ice/ice_ethdev.c | 11 ++++++++++-
> drivers/net/ice/ice_ethdev.h | 1 +
> drivers/net/ice/ice_rxtx_vec_common.h | 5 +++++
> 4 files changed, 28 insertions(+), 1 deletion(-)
>
> diff --git a/doc/guides/nics/ice.rst b/doc/guides/nics/ice.rst index
> 1a426438d..d1b80c2fa 100644
> --- a/doc/guides/nics/ice.rst
> +++ b/doc/guides/nics/ice.rst
> @@ -80,6 +80,18 @@ Runtime Config Options
>
> -w 80:00.0,pipeline-mode-support=1
>
> +- ``Flow Mark Support`` (default ``0``)
> +
> + This is a hint to the driver to select the data path that support
> + flow mark extraction by default.
> + NOTE: This is an experimental devarg, it will be removed when any of
> + below conditions is ready.
> + 1) all data path support flow mark (currently vPMD does not)
> + 2) a new offload like RTE_DEV_RX_OFFLOAD_FLOW_MARK be introduced
> as a standard way to hint.
> + Example::
> +
> + -w 80:00.0,flow-mark-support=1
> +
> - ``Protocol extraction for per queue``
>
> Configure the RX queues to do protocol extraction into mbuf for protocol
> diff --git a/drivers/net/ice/ice_ethdev.c b/drivers/net/ice/ice_ethdev.c index
> abf00d404..9f2cb2f40 100644
> --- a/drivers/net/ice/ice_ethdev.c
> +++ b/drivers/net/ice/ice_ethdev.c
> @@ -23,11 +23,13 @@
> /* devargs */
> #define ICE_SAFE_MODE_SUPPORT_ARG "safe-mode-support"
> #define ICE_PIPELINE_MODE_SUPPORT_ARG "pipeline-mode-support"
> +#define ICE_FLOW_MARK_SUPPORT_ARG "flow-mark-support"
> #define ICE_PROTO_XTR_ARG "proto_xtr"
>
> static const char * const ice_valid_args[] = {
> ICE_SAFE_MODE_SUPPORT_ARG,
> ICE_PIPELINE_MODE_SUPPORT_ARG,
> + ICE_FLOW_MARK_SUPPORT_ARG,
> ICE_PROTO_XTR_ARG,
> NULL
> };
> @@ -1987,6 +1989,12 @@ static int ice_parse_devargs(struct rte_eth_dev
> *dev)
>
> ret = rte_kvargs_process(kvlist, ICE_PIPELINE_MODE_SUPPORT_ARG,
> &parse_bool, &ad-
> >devargs.pipe_mode_support);
> + if (ret)
> + goto bail;
> +
> + ret = rte_kvargs_process(kvlist, ICE_FLOW_MARK_SUPPORT_ARG,
> + &parse_bool, &ad-
> >devargs.flow_mark_support);
> + printf("flow_mark = %d\n", ad->devargs.flow_mark_support);
>
> bail:
> rte_kvargs_free(kvlist);
> @@ -4571,7 +4579,8 @@ RTE_PMD_REGISTER_KMOD_DEP(net_ice, "*
> igb_uio | uio_pci_generic | vfio-pci");
> RTE_PMD_REGISTER_PARAM_STRING(net_ice,
> ICE_PROTO_XTR_ARG
> "=[queue:]<vlan|ipv4|ipv6|ipv6_flow|tcp>"
> ICE_SAFE_MODE_SUPPORT_ARG "=<0|1>"
> - ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>");
> + ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>"
> + ICE_FLOW_MARK_SUPPORT_ARG "=<0|1>");
>
> RTE_INIT(ice_init_log)
> {
> diff --git a/drivers/net/ice/ice_ethdev.h b/drivers/net/ice/ice_ethdev.h index
> 4a0d37b32..4d35339a7 100644
> --- a/drivers/net/ice/ice_ethdev.h
> +++ b/drivers/net/ice/ice_ethdev.h
> @@ -391,6 +391,7 @@ struct ice_devargs {
> int safe_mode_support;
> uint8_t proto_xtr_dflt;
> int pipe_mode_support;
> + int flow_mark_support;
> uint8_t proto_xtr[ICE_MAX_QUEUE_NUM];
> };
>
> diff --git a/drivers/net/ice/ice_rxtx_vec_common.h
> b/drivers/net/ice/ice_rxtx_vec_common.h
> index 080ca4175..086428898 100644
> --- a/drivers/net/ice/ice_rxtx_vec_common.h
> +++ b/drivers/net/ice/ice_rxtx_vec_common.h
> @@ -268,6 +268,11 @@ ice_rx_vec_dev_check_default(struct rte_eth_dev
> *dev) {
> int i;
> struct ice_rx_queue *rxq;
> + struct ice_adapter *ad =
> + ICE_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
> +
> + if (ad->devargs.flow_mark_support)
> + return -1;
>
> for (i = 0; i < dev->data->nb_rx_queues; i++) {
> rxq = dev->data->rx_queues[i];
> --
> 2.13.6
Acked-by: Qiming Yang <qiming.yang@intel.com>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support
2019-11-18 0:43 [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support Qi Zhang
2019-11-18 5:57 ` Yang, Qiming
@ 2019-11-19 5:52 ` Ye Xiaolong
2019-11-19 6:03 ` Zhang, Qi Z
1 sibling, 1 reply; 5+ messages in thread
From: Ye Xiaolong @ 2019-11-19 5:52 UTC (permalink / raw)
To: Qi Zhang; +Cc: qiming.yang, yahui.cao, beilei.xing, dev
On 11/18, Qi Zhang wrote:
>Since not all data paths support flow mark, the driver need
s/need/needs
>a hint from application to select the correct data path if
>flow mark is required. The patch introduce a devarg
s/introduce/introduces
>"flow-mark-support" as a workaround solution, since a standard
>way is still ongoing.
>
>Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
>---
>
>v3:
>- add "experimental notification" in ice.rst
>
>v2:
>- fix typo
>
> doc/guides/nics/ice.rst | 12 ++++++++++++
> drivers/net/ice/ice_ethdev.c | 11 ++++++++++-
> drivers/net/ice/ice_ethdev.h | 1 +
> drivers/net/ice/ice_rxtx_vec_common.h | 5 +++++
> 4 files changed, 28 insertions(+), 1 deletion(-)
>
>diff --git a/doc/guides/nics/ice.rst b/doc/guides/nics/ice.rst
>index 1a426438d..d1b80c2fa 100644
>--- a/doc/guides/nics/ice.rst
>+++ b/doc/guides/nics/ice.rst
>@@ -80,6 +80,18 @@ Runtime Config Options
>
> -w 80:00.0,pipeline-mode-support=1
>
>+- ``Flow Mark Support`` (default ``0``)
>+
>+ This is a hint to the driver to select the data path that support flow mark extraction
s/support/supports
>+ by default.
>+ NOTE: This is an experimental devarg, it will be removed when any of below conditions
>+ is ready.
>+ 1) all data path support flow mark (currently vPMD does not)
s/path/paths
>+ 2) a new offload like RTE_DEV_RX_OFFLOAD_FLOW_MARK be introduced as a standard way to hint.
>+ Example::
>+
>+ -w 80:00.0,flow-mark-support=1
>+
> - ``Protocol extraction for per queue``
>
> Configure the RX queues to do protocol extraction into mbuf for protocol
>diff --git a/drivers/net/ice/ice_ethdev.c b/drivers/net/ice/ice_ethdev.c
>index abf00d404..9f2cb2f40 100644
>--- a/drivers/net/ice/ice_ethdev.c
>+++ b/drivers/net/ice/ice_ethdev.c
>@@ -23,11 +23,13 @@
> /* devargs */
> #define ICE_SAFE_MODE_SUPPORT_ARG "safe-mode-support"
> #define ICE_PIPELINE_MODE_SUPPORT_ARG "pipeline-mode-support"
>+#define ICE_FLOW_MARK_SUPPORT_ARG "flow-mark-support"
> #define ICE_PROTO_XTR_ARG "proto_xtr"
>
> static const char * const ice_valid_args[] = {
> ICE_SAFE_MODE_SUPPORT_ARG,
> ICE_PIPELINE_MODE_SUPPORT_ARG,
>+ ICE_FLOW_MARK_SUPPORT_ARG,
> ICE_PROTO_XTR_ARG,
> NULL
> };
>@@ -1987,6 +1989,12 @@ static int ice_parse_devargs(struct rte_eth_dev *dev)
>
> ret = rte_kvargs_process(kvlist, ICE_PIPELINE_MODE_SUPPORT_ARG,
> &parse_bool, &ad->devargs.pipe_mode_support);
>+ if (ret)
>+ goto bail;
>+
>+ ret = rte_kvargs_process(kvlist, ICE_FLOW_MARK_SUPPORT_ARG,
>+ &parse_bool, &ad->devargs.flow_mark_support);
Do we need error handling here?
>+ printf("flow_mark = %d\n", ad->devargs.flow_mark_support);
Use PMD_INIT_LOG instead, and in order to provide more informative message, what
about we directly say "flow_mark is 0, vector path won't be selected." or something
similar according to the value of flow_makr_support.
>
> bail:
> rte_kvargs_free(kvlist);
>@@ -4571,7 +4579,8 @@ RTE_PMD_REGISTER_KMOD_DEP(net_ice, "* igb_uio | uio_pci_generic | vfio-pci");
> RTE_PMD_REGISTER_PARAM_STRING(net_ice,
> ICE_PROTO_XTR_ARG "=[queue:]<vlan|ipv4|ipv6|ipv6_flow|tcp>"
> ICE_SAFE_MODE_SUPPORT_ARG "=<0|1>"
>- ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>");
>+ ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>"
>+ ICE_FLOW_MARK_SUPPORT_ARG "=<0|1>");
>
> RTE_INIT(ice_init_log)
> {
>diff --git a/drivers/net/ice/ice_ethdev.h b/drivers/net/ice/ice_ethdev.h
>index 4a0d37b32..4d35339a7 100644
>--- a/drivers/net/ice/ice_ethdev.h
>+++ b/drivers/net/ice/ice_ethdev.h
>@@ -391,6 +391,7 @@ struct ice_devargs {
> int safe_mode_support;
> uint8_t proto_xtr_dflt;
> int pipe_mode_support;
>+ int flow_mark_support;
> uint8_t proto_xtr[ICE_MAX_QUEUE_NUM];
> };
>
>diff --git a/drivers/net/ice/ice_rxtx_vec_common.h b/drivers/net/ice/ice_rxtx_vec_common.h
>index 080ca4175..086428898 100644
>--- a/drivers/net/ice/ice_rxtx_vec_common.h
>+++ b/drivers/net/ice/ice_rxtx_vec_common.h
>@@ -268,6 +268,11 @@ ice_rx_vec_dev_check_default(struct rte_eth_dev *dev)
> {
> int i;
> struct ice_rx_queue *rxq;
>+ struct ice_adapter *ad =
>+ ICE_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
>+
>+ if (ad->devargs.flow_mark_support)
Or we can add hint info here.
Thanks,
Xiaolong
>+ return -1;
>
> for (i = 0; i < dev->data->nb_rx_queues; i++) {
> rxq = dev->data->rx_queues[i];
>--
>2.13.6
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support
2019-11-19 5:52 ` Ye Xiaolong
@ 2019-11-19 6:03 ` Zhang, Qi Z
2019-11-19 6:08 ` Ye Xiaolong
0 siblings, 1 reply; 5+ messages in thread
From: Zhang, Qi Z @ 2019-11-19 6:03 UTC (permalink / raw)
To: Ye, Xiaolong; +Cc: Yang, Qiming, Cao, Yahui, Xing, Beilei, dev
> -----Original Message-----
> From: Ye, Xiaolong <xiaolong.ye@intel.com>
> Sent: Tuesday, November 19, 2019 1:53 PM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>
> Cc: Yang, Qiming <qiming.yang@intel.com>; Cao, Yahui
> <yahui.cao@intel.com>; Xing, Beilei <beilei.xing@intel.com>; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support
>
> On 11/18, Qi Zhang wrote:
> >Since not all data paths support flow mark, the driver need
>
> s/need/needs
>
> >a hint from application to select the correct data path if flow mark is
> >required. The patch introduce a devarg
>
> s/introduce/introduces
>
> >"flow-mark-support" as a workaround solution, since a standard way is
> >still ongoing.
> >
> >Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
> >---
> >
> >v3:
> >- add "experimental notification" in ice.rst
> >
> >v2:
> >- fix typo
> >
> > doc/guides/nics/ice.rst | 12 ++++++++++++
> > drivers/net/ice/ice_ethdev.c | 11 ++++++++++-
> > drivers/net/ice/ice_ethdev.h | 1 +
> > drivers/net/ice/ice_rxtx_vec_common.h | 5 +++++
> > 4 files changed, 28 insertions(+), 1 deletion(-)
> >
> >diff --git a/doc/guides/nics/ice.rst b/doc/guides/nics/ice.rst index
> >1a426438d..d1b80c2fa 100644
> >--- a/doc/guides/nics/ice.rst
> >+++ b/doc/guides/nics/ice.rst
> >@@ -80,6 +80,18 @@ Runtime Config Options
> >
> > -w 80:00.0,pipeline-mode-support=1
> >
> >+- ``Flow Mark Support`` (default ``0``)
> >+
> >+ This is a hint to the driver to select the data path that support
> >+ flow mark extraction
>
> s/support/supports
>
> >+ by default.
> >+ NOTE: This is an experimental devarg, it will be removed when any of
> >+ below conditions is ready.
> >+ 1) all data path support flow mark (currently vPMD does not)
>
> s/path/paths
>
> >+ 2) a new offload like RTE_DEV_RX_OFFLOAD_FLOW_MARK be introduced
> as a standard way to hint.
> >+ Example::
> >+
> >+ -w 80:00.0,flow-mark-support=1
> >+
> > - ``Protocol extraction for per queue``
> >
> > Configure the RX queues to do protocol extraction into mbuf for
> >protocol diff --git a/drivers/net/ice/ice_ethdev.c
> >b/drivers/net/ice/ice_ethdev.c index abf00d404..9f2cb2f40 100644
> >--- a/drivers/net/ice/ice_ethdev.c
> >+++ b/drivers/net/ice/ice_ethdev.c
> >@@ -23,11 +23,13 @@
> > /* devargs */
> > #define ICE_SAFE_MODE_SUPPORT_ARG "safe-mode-support"
> > #define ICE_PIPELINE_MODE_SUPPORT_ARG "pipeline-mode-support"
> >+#define ICE_FLOW_MARK_SUPPORT_ARG "flow-mark-support"
> > #define ICE_PROTO_XTR_ARG "proto_xtr"
> >
> > static const char * const ice_valid_args[] = {
> > ICE_SAFE_MODE_SUPPORT_ARG,
> > ICE_PIPELINE_MODE_SUPPORT_ARG,
> >+ ICE_FLOW_MARK_SUPPORT_ARG,
> > ICE_PROTO_XTR_ARG,
> > NULL
> > };
> >@@ -1987,6 +1989,12 @@ static int ice_parse_devargs(struct rte_eth_dev
> >*dev)
> >
> > ret = rte_kvargs_process(kvlist, ICE_PIPELINE_MODE_SUPPORT_ARG,
> > &parse_bool, &ad->devargs.pipe_mode_support);
> >+ if (ret)
> >+ goto bail;
> >+
> >+ ret = rte_kvargs_process(kvlist, ICE_FLOW_MARK_SUPPORT_ARG,
> >+ &parse_bool, &ad->devargs.flow_mark_support);
>
> Do we need error handling here?
>
> >+ printf("flow_mark = %d\n", ad->devargs.flow_mark_support);
>
> Use PMD_INIT_LOG instead, and in order to provide more informative
> message, what about we directly say "flow_mark is 0, vector path won't be
> selected." or something similar according to the value of flow_makr_support.
>
This is a mistake forgot to remove the debug code.
Thanks
Qi
> >
> > bail:
> > rte_kvargs_free(kvlist);
> >@@ -4571,7 +4579,8 @@ RTE_PMD_REGISTER_KMOD_DEP(net_ice, "*
> igb_uio |
> >uio_pci_generic | vfio-pci"); RTE_PMD_REGISTER_PARAM_STRING(net_ice,
> > ICE_PROTO_XTR_ARG
> "=[queue:]<vlan|ipv4|ipv6|ipv6_flow|tcp>"
> > ICE_SAFE_MODE_SUPPORT_ARG "=<0|1>"
> >- ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>");
> >+ ICE_PIPELINE_MODE_SUPPORT_ARG "=<0|1>"
> >+ ICE_FLOW_MARK_SUPPORT_ARG "=<0|1>");
> >
> > RTE_INIT(ice_init_log)
> > {
> >diff --git a/drivers/net/ice/ice_ethdev.h
> >b/drivers/net/ice/ice_ethdev.h index 4a0d37b32..4d35339a7 100644
> >--- a/drivers/net/ice/ice_ethdev.h
> >+++ b/drivers/net/ice/ice_ethdev.h
> >@@ -391,6 +391,7 @@ struct ice_devargs {
> > int safe_mode_support;
> > uint8_t proto_xtr_dflt;
> > int pipe_mode_support;
> >+ int flow_mark_support;
> > uint8_t proto_xtr[ICE_MAX_QUEUE_NUM]; };
> >
> >diff --git a/drivers/net/ice/ice_rxtx_vec_common.h
> >b/drivers/net/ice/ice_rxtx_vec_common.h
> >index 080ca4175..086428898 100644
> >--- a/drivers/net/ice/ice_rxtx_vec_common.h
> >+++ b/drivers/net/ice/ice_rxtx_vec_common.h
> >@@ -268,6 +268,11 @@ ice_rx_vec_dev_check_default(struct rte_eth_dev
> >*dev) {
> > int i;
> > struct ice_rx_queue *rxq;
> >+ struct ice_adapter *ad =
> >+ ICE_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
> >+
> >+ if (ad->devargs.flow_mark_support)
>
> Or we can add hint info here.
>
> Thanks,
> Xiaolong
>
> >+ return -1;
> >
> > for (i = 0; i < dev->data->nb_rx_queues; i++) {
> > rxq = dev->data->rx_queues[i];
> >--
> >2.13.6
> >
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support
2019-11-19 6:03 ` Zhang, Qi Z
@ 2019-11-19 6:08 ` Ye Xiaolong
0 siblings, 0 replies; 5+ messages in thread
From: Ye Xiaolong @ 2019-11-19 6:08 UTC (permalink / raw)
To: Zhang, Qi Z; +Cc: Yang, Qiming, Cao, Yahui, Xing, Beilei, dev
On 11/19, Zhang, Qi Z wrote:
>
>
>> -----Original Message-----
>> From: Ye, Xiaolong <xiaolong.ye@intel.com>
>> Sent: Tuesday, November 19, 2019 1:53 PM
>> To: Zhang, Qi Z <qi.z.zhang@intel.com>
>> Cc: Yang, Qiming <qiming.yang@intel.com>; Cao, Yahui
>> <yahui.cao@intel.com>; Xing, Beilei <beilei.xing@intel.com>; dev@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support
>>
>> On 11/18, Qi Zhang wrote:
>> >Since not all data paths support flow mark, the driver need
>>
>> s/need/needs
>>
>> >a hint from application to select the correct data path if flow mark is
>> >required. The patch introduce a devarg
>>
>> s/introduce/introduces
>>
>> >"flow-mark-support" as a workaround solution, since a standard way is
>> >still ongoing.
>> >
>> >Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
>> >---
>> >
>> >v3:
>> >- add "experimental notification" in ice.rst
>> >
>> >v2:
>> >- fix typo
>> >
>> > doc/guides/nics/ice.rst | 12 ++++++++++++
>> > drivers/net/ice/ice_ethdev.c | 11 ++++++++++-
>> > drivers/net/ice/ice_ethdev.h | 1 +
>> > drivers/net/ice/ice_rxtx_vec_common.h | 5 +++++
>> > 4 files changed, 28 insertions(+), 1 deletion(-)
>> >
>> >diff --git a/doc/guides/nics/ice.rst b/doc/guides/nics/ice.rst index
>> >1a426438d..d1b80c2fa 100644
>> >--- a/doc/guides/nics/ice.rst
>> >+++ b/doc/guides/nics/ice.rst
>> >@@ -80,6 +80,18 @@ Runtime Config Options
>> >
>> > -w 80:00.0,pipeline-mode-support=1
>> >
>> >+- ``Flow Mark Support`` (default ``0``)
>> >+
>> >+ This is a hint to the driver to select the data path that support
>> >+ flow mark extraction
>>
>> s/support/supports
>>
>> >+ by default.
>> >+ NOTE: This is an experimental devarg, it will be removed when any of
>> >+ below conditions is ready.
>> >+ 1) all data path support flow mark (currently vPMD does not)
>>
>> s/path/paths
>>
>> >+ 2) a new offload like RTE_DEV_RX_OFFLOAD_FLOW_MARK be introduced
>> as a standard way to hint.
>> >+ Example::
>> >+
>> >+ -w 80:00.0,flow-mark-support=1
>> >+
>> > - ``Protocol extraction for per queue``
>> >
>> > Configure the RX queues to do protocol extraction into mbuf for
>> >protocol diff --git a/drivers/net/ice/ice_ethdev.c
>> >b/drivers/net/ice/ice_ethdev.c index abf00d404..9f2cb2f40 100644
>> >--- a/drivers/net/ice/ice_ethdev.c
>> >+++ b/drivers/net/ice/ice_ethdev.c
>> >@@ -23,11 +23,13 @@
>> > /* devargs */
>> > #define ICE_SAFE_MODE_SUPPORT_ARG "safe-mode-support"
>> > #define ICE_PIPELINE_MODE_SUPPORT_ARG "pipeline-mode-support"
>> >+#define ICE_FLOW_MARK_SUPPORT_ARG "flow-mark-support"
>> > #define ICE_PROTO_XTR_ARG "proto_xtr"
>> >
>> > static const char * const ice_valid_args[] = {
>> > ICE_SAFE_MODE_SUPPORT_ARG,
>> > ICE_PIPELINE_MODE_SUPPORT_ARG,
>> >+ ICE_FLOW_MARK_SUPPORT_ARG,
>> > ICE_PROTO_XTR_ARG,
>> > NULL
>> > };
>> >@@ -1987,6 +1989,12 @@ static int ice_parse_devargs(struct rte_eth_dev
>> >*dev)
>> >
>> > ret = rte_kvargs_process(kvlist, ICE_PIPELINE_MODE_SUPPORT_ARG,
>> > &parse_bool, &ad->devargs.pipe_mode_support);
>> >+ if (ret)
>> >+ goto bail;
>> >+
>> >+ ret = rte_kvargs_process(kvlist, ICE_FLOW_MARK_SUPPORT_ARG,
>> >+ &parse_bool, &ad->devargs.flow_mark_support);
>>
>> Do we need error handling here?
>>
>> >+ printf("flow_mark = %d\n", ad->devargs.flow_mark_support);
>>
>> Use PMD_INIT_LOG instead, and in order to provide more informative
>> message, what about we directly say "flow_mark is 0, vector path won't be
>> selected." or something similar according to the value of flow_makr_support.
>>
>
>This is a mistake forgot to remove the debug code.
Ok, I think it's still meaningful to print some hints in ice_rx_vec_dev_check_default
when it return due to flow_mark_support is set.
Thanks,
Xiaolong
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2019-11-19 6:12 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-11-18 0:43 [dpdk-dev] [PATCH v3] net/ice: add flow mark hint support Qi Zhang
2019-11-18 5:57 ` Yang, Qiming
2019-11-19 5:52 ` Ye Xiaolong
2019-11-19 6:03 ` Zhang, Qi Z
2019-11-19 6:08 ` Ye Xiaolong
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).