* Re: [dpdk-dev] [PATCH v1] net/e1000: fix the invalid flow control mode setting
[not found] <20210119065847.66043-1-wenjun1.wu@intel.com>
@ 2021-01-19 7:43 ` Guo, Jia
0 siblings, 0 replies; 5+ messages in thread
From: Guo, Jia @ 2021-01-19 7:43 UTC (permalink / raw)
To: Wu, Wenjun1, dev, Zhang, Qi Z, Zhang, Yuying; +Cc: stable
Hi, Wenjun
> -----Original Message-----
> From: Wu, Wenjun1 <wenjun1.wu@intel.com>
> Sent: Tuesday, January 19, 2021 2:59 PM
> To: dev@dpdk.org; Guo, Jia <jia.guo@intel.com>; Zhang, Qi Z
> <qi.z.zhang@intel.com>; Zhang, Yuying <yuying.zhang@intel.com>
> Cc: Wu, Wenjun1 <wenjun1.wu@intel.com>; stable@dpdk.org
> Subject: [PATCH v1] net/e1000: fix the invalid flow control mode setting
>
> E1000_CTRL register should be updated according to fc_conf->mode's value.
>
> Fixes: af75078fece3 ("first public release")
> Cc: stable@dpdk.org
>
> Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
> ---
> drivers/net/e1000/igb_ethdev.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> diff --git a/drivers/net/e1000/igb_ethdev.c
> b/drivers/net/e1000/igb_ethdev.c index 647aa8d99..390737393 100644
> --- a/drivers/net/e1000/igb_ethdev.c
> +++ b/drivers/net/e1000/igb_ethdev.c
> @@ -3064,6 +3064,7 @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev,
> struct rte_eth_fc_conf *fc_conf)
> uint32_t rx_buf_size;
> uint32_t max_high_water;
> uint32_t rctl;
> + uint32_t ctrl;
>
> hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> if (fc_conf->autoneg != hw->mac.autoneg) @@ -3101,6 +3102,15
> @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev, struct rte_eth_fc_conf
> *fc_conf)
> rctl &= ~E1000_RCTL_PMCF;
>
> E1000_WRITE_REG(hw, E1000_RCTL, rctl);
> +
Some document for ctrl, like rctl setting, would be appreciate.
> + ctrl = E1000_READ_REG(hw, E1000_CTRL);
> + ctrl &= ~(E1000_CTRL_RFCE|E1000_CTRL_TFCE);
> + if (fc_conf->mode == RTE_FC_RX_PAUSE || fc_conf->mode
> == RTE_FC_FULL)
> + ctrl |= E1000_CTRL_RFCE;
> + if (fc_conf->mode == RTE_FC_TX_PAUSE || fc_conf->mode
> == RTE_FC_FULL)
> + ctrl |= E1000_CTRL_TFCE;
What about the case if fc_conf->mode == RTE_FC_NONE? Use a switch would be help for that?
> + E1000_WRITE_REG(hw, E1000_CTRL, ctrl);
> +
> E1000_WRITE_FLUSH(hw);
>
> return 0;
> --
> 2.25.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH v1] net/e1000: fix the invalid flow control mode setting
2021-01-20 7:18 ` Guo, Jia
@ 2021-01-20 23:35 ` Zhang, Qi Z
0 siblings, 0 replies; 5+ messages in thread
From: Zhang, Qi Z @ 2021-01-20 23:35 UTC (permalink / raw)
To: Guo, Jia, Wu, Wenjun1, dev; +Cc: stable
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Guo, Jia
> Sent: Wednesday, January 20, 2021 3:19 PM
> To: Wu, Wenjun1 <wenjun1.wu@intel.com>; dev@dpdk.org
> Cc: stable@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v1] net/e1000: fix the invalid flow control mode
> setting
>
> Acked-by: Jeff Guo <jia.guo@intel.com>
>
> > -----Original Message-----
> > From: Wu, Wenjun1 <wenjun1.wu@intel.com>
> > Sent: Wednesday, January 20, 2021 2:54 PM
> > To: dev@dpdk.org; Guo, Jia <jia.guo@intel.com>
> > Cc: Wu, Wenjun1 <wenjun1.wu@intel.com>; stable@dpdk.org
> > Subject: [PATCH v1] net/e1000: fix the invalid flow control mode
> > setting
> >
> > E1000_CTRL register should be updated according to fc_conf->mode's value.
> >
> > Fixes: af75078fece3 ("first public release")
> > Cc: stable@dpdk.org
> >
> > Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
Applied to dpdk-next-net-intel.
Thanks
Qi
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH v1] net/e1000: fix the invalid flow control mode setting
2021-01-20 6:53 Wenjun Wu
@ 2021-01-20 7:18 ` Guo, Jia
2021-01-20 23:35 ` Zhang, Qi Z
0 siblings, 1 reply; 5+ messages in thread
From: Guo, Jia @ 2021-01-20 7:18 UTC (permalink / raw)
To: Wu, Wenjun1, dev; +Cc: stable
Acked-by: Jeff Guo <jia.guo@intel.com>
> -----Original Message-----
> From: Wu, Wenjun1 <wenjun1.wu@intel.com>
> Sent: Wednesday, January 20, 2021 2:54 PM
> To: dev@dpdk.org; Guo, Jia <jia.guo@intel.com>
> Cc: Wu, Wenjun1 <wenjun1.wu@intel.com>; stable@dpdk.org
> Subject: [PATCH v1] net/e1000: fix the invalid flow control mode setting
>
> E1000_CTRL register should be updated according to fc_conf->mode's value.
>
> Fixes: af75078fece3 ("first public release")
> Cc: stable@dpdk.org
>
> Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
> ---
> drivers/net/e1000/igb_ethdev.c | 34
> ++++++++++++++++++++++++++++++++++
> 1 file changed, 34 insertions(+)
>
> diff --git a/drivers/net/e1000/igb_ethdev.c
> b/drivers/net/e1000/igb_ethdev.c index 647aa8d99..a8fc57d2c 100644
> --- a/drivers/net/e1000/igb_ethdev.c
> +++ b/drivers/net/e1000/igb_ethdev.c
> @@ -3064,6 +3064,7 @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev,
> struct rte_eth_fc_conf *fc_conf)
> uint32_t rx_buf_size;
> uint32_t max_high_water;
> uint32_t rctl;
> + uint32_t ctrl;
>
> hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> if (fc_conf->autoneg != hw->mac.autoneg) @@ -3101,6 +3102,39
> @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev, struct rte_eth_fc_conf
> *fc_conf)
> rctl &= ~E1000_RCTL_PMCF;
>
> E1000_WRITE_REG(hw, E1000_RCTL, rctl);
> +
> + /*
> + * check if we want to change flow control mode - driver
> doesn't have native
> + * capability to do that, so we'll write the registers ourselves
> + */
> + ctrl = E1000_READ_REG(hw, E1000_CTRL);
> +
> + /*
> + * set or clear E1000_CTRL_RFCE and E1000_CTRL_TFCE bits
> depending
> + * on configuration
> + */
> + switch (fc_conf->mode) {
> + case RTE_FC_NONE:
> + ctrl &= ~E1000_CTRL_RFCE & ~E1000_CTRL_TFCE;
> + break;
> + case RTE_FC_RX_PAUSE:
> + ctrl |= E1000_CTRL_RFCE;
> + ctrl &= ~E1000_CTRL_TFCE;
> + break;
> + case RTE_FC_TX_PAUSE:
> + ctrl |= E1000_CTRL_TFCE;
> + ctrl &= ~E1000_CTRL_RFCE;
> + break;
> + case RTE_FC_FULL:
> + ctrl |= E1000_CTRL_RFCE | E1000_CTRL_TFCE;
> + break;
> + default:
> + PMD_INIT_LOG(ERR, "invalid flow control mode");
> + return -EINVAL;
> + }
> +
> + E1000_WRITE_REG(hw, E1000_CTRL, ctrl);
> +
> E1000_WRITE_FLUSH(hw);
>
> return 0;
> --
> 2.25.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* [dpdk-dev] [PATCH v1] net/e1000: fix the invalid flow control mode setting
@ 2021-01-20 6:53 Wenjun Wu
2021-01-20 7:18 ` Guo, Jia
0 siblings, 1 reply; 5+ messages in thread
From: Wenjun Wu @ 2021-01-20 6:53 UTC (permalink / raw)
To: dev, jia.guo; +Cc: Wenjun Wu, stable
E1000_CTRL register should be updated according to fc_conf->mode's value.
Fixes: af75078fece3 ("first public release")
Cc: stable@dpdk.org
Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
---
drivers/net/e1000/igb_ethdev.c | 34 ++++++++++++++++++++++++++++++++++
1 file changed, 34 insertions(+)
diff --git a/drivers/net/e1000/igb_ethdev.c b/drivers/net/e1000/igb_ethdev.c
index 647aa8d99..a8fc57d2c 100644
--- a/drivers/net/e1000/igb_ethdev.c
+++ b/drivers/net/e1000/igb_ethdev.c
@@ -3064,6 +3064,7 @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
uint32_t rx_buf_size;
uint32_t max_high_water;
uint32_t rctl;
+ uint32_t ctrl;
hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
if (fc_conf->autoneg != hw->mac.autoneg)
@@ -3101,6 +3102,39 @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
rctl &= ~E1000_RCTL_PMCF;
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
+
+ /*
+ * check if we want to change flow control mode - driver doesn't have native
+ * capability to do that, so we'll write the registers ourselves
+ */
+ ctrl = E1000_READ_REG(hw, E1000_CTRL);
+
+ /*
+ * set or clear E1000_CTRL_RFCE and E1000_CTRL_TFCE bits depending
+ * on configuration
+ */
+ switch (fc_conf->mode) {
+ case RTE_FC_NONE:
+ ctrl &= ~E1000_CTRL_RFCE & ~E1000_CTRL_TFCE;
+ break;
+ case RTE_FC_RX_PAUSE:
+ ctrl |= E1000_CTRL_RFCE;
+ ctrl &= ~E1000_CTRL_TFCE;
+ break;
+ case RTE_FC_TX_PAUSE:
+ ctrl |= E1000_CTRL_TFCE;
+ ctrl &= ~E1000_CTRL_RFCE;
+ break;
+ case RTE_FC_FULL:
+ ctrl |= E1000_CTRL_RFCE | E1000_CTRL_TFCE;
+ break;
+ default:
+ PMD_INIT_LOG(ERR, "invalid flow control mode");
+ return -EINVAL;
+ }
+
+ E1000_WRITE_REG(hw, E1000_CTRL, ctrl);
+
E1000_WRITE_FLUSH(hw);
return 0;
--
2.25.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH v1] net/e1000: fix the invalid flow control mode setting
[not found] <20210120045859.211284-1-wenjun1.wu@intel.com>
@ 2021-01-20 5:55 ` Guo, Jia
0 siblings, 0 replies; 5+ messages in thread
From: Guo, Jia @ 2021-01-20 5:55 UTC (permalink / raw)
To: Wu, Wenjun1, dev; +Cc: stable
Hi, wenjun
> -----Original Message-----
> From: Wu, Wenjun1 <wenjun1.wu@intel.com>
> Sent: Wednesday, January 20, 2021 12:59 PM
> To: dev@dpdk.org; Guo, Jia <jia.guo@intel.com>
> Cc: Wu, Wenjun1 <wenjun1.wu@intel.com>; stable@dpdk.org
> Subject: [PATCH v1] net/e1000: fix the invalid flow control mode setting
>
> E1000_CTRL register should be updated according to fc_conf->mode's value.
>
> Fixes: af75078fece3 ("first public release")
> Cc: stable@dpdk.org
>
> Signed-off-by: Wenjun Wu <wenjun1.wu@intel.com>
> ---
> drivers/net/e1000/igb_ethdev.c | 32
> ++++++++++++++++++++++++++++++++
> 1 file changed, 32 insertions(+)
>
> diff --git a/drivers/net/e1000/igb_ethdev.c
> b/drivers/net/e1000/igb_ethdev.c index 647aa8d99..8fecd657d 100644
> --- a/drivers/net/e1000/igb_ethdev.c
> +++ b/drivers/net/e1000/igb_ethdev.c
> @@ -3064,6 +3064,7 @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev,
> struct rte_eth_fc_conf *fc_conf)
> uint32_t rx_buf_size;
> uint32_t max_high_water;
> uint32_t rctl;
> + uint32_t ctrl;
>
> hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> if (fc_conf->autoneg != hw->mac.autoneg) @@ -3101,6 +3102,37
> @@ eth_igb_flow_ctrl_set(struct rte_eth_dev *dev, struct rte_eth_fc_conf
> *fc_conf)
> rctl &= ~E1000_RCTL_PMCF;
>
> E1000_WRITE_REG(hw, E1000_RCTL, rctl);
> +
> + /* check if we want to change flow control mode - driver
> doesn't have native
> + * capability to do that, so we'll write the registers ourselves
> */
No need the blank line with the code block which be explain by the doc, and please considerate the format for multiple line, like that:
/*
* XXXXX
* XXXXX
*/
> +
> + ctrl = E1000_READ_REG(hw, E1000_CTRL);
> +
> + /* set or clear E1000_CTRL_RFCE and E1000_CTRL_TFCE bits
> depending
> + * on configuration */
> +
Ditto.
> + switch (fc_conf->mode) {
> + case RTE_FC_NONE:
> + ctrl &= ~E1000_CTRL_RFCE & ~E1000_CTRL_TFCE;
> + break;
> + case RTE_FC_RX_PAUSE:
> + ctrl |= E1000_CTRL_RFCE;
> + ctrl &= ~E1000_CTRL_TFCE;
> + break;
> + case RTE_FC_TX_PAUSE:
> + ctrl |= E1000_CTRL_TFCE;
> + ctrl &= ~E1000_CTRL_RFCE;
> + break;
> + case RTE_FC_FULL:
> + ctrl |= E1000_CTRL_RFCE | E1000_CTRL_TFCE;
> + break;
> + default:
> + PMD_INIT_LOG(ERR, "invalid flow control mode");
> + return -EIO;
Would be better if use "-EINVAL" for the meaning of invalid configure mode?
> + }
> +
> + E1000_WRITE_REG(hw, E1000_CTRL, ctrl);
> +
> E1000_WRITE_FLUSH(hw);
>
> return 0;
> --
> 2.25.1
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2021-01-20 23:35 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <20210119065847.66043-1-wenjun1.wu@intel.com>
2021-01-19 7:43 ` [dpdk-dev] [PATCH v1] net/e1000: fix the invalid flow control mode setting Guo, Jia
[not found] <20210120045859.211284-1-wenjun1.wu@intel.com>
2021-01-20 5:55 ` Guo, Jia
2021-01-20 6:53 Wenjun Wu
2021-01-20 7:18 ` Guo, Jia
2021-01-20 23:35 ` Zhang, Qi Z
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).