diff mbox series

[v3] net/mlx5: add 200G link speed

Message ID 1588767728-5722-1-git-send-email-asafp@mellanox.com (mailing list archive)
State Accepted, archived
Delegated to: Ferruh Yigit
Headers
Series [v3] net/mlx5: add 200G link speed |

Checks

Context Check Description
ci/checkpatch warning coding style issues
ci/iol-intel-Performance success Performance Testing PASS
ci/iol-nxp-Performance success Performance Testing PASS
ci/iol-mellanox-Performance success Performance Testing PASS
ci/travis-robot warning Travis build: failed
ci/Intel-compilation fail Compilation issues
ci/iol-testing success Testing PASS

Commit Message

Asaf Penso May 6, 2020, 12:22 p.m. UTC 
  There is no way to report back a link speed of 200Gbps.

Adding 200G link speed.

Signed-off-by: Asaf Penso <asafp@mellanox.com>
---
v2 - reword commit log and remove 400G

v3 - rebase and fix checkpatch warning
---
 app/test-pmd/cmdline.c                      | 12 +++++++-----
 doc/guides/rel_notes/release_20_05.rst      |  2 ++
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  2 +-
 drivers/net/mlx5/mlx5_ethdev.c              | 22 +++++++++++++++++++++-
 lib/librte_ethdev/rte_ethdev.c              |  2 ++
 lib/librte_ethdev/rte_ethdev.h              |  2 ++
 6 files changed, 35 insertions(+), 7 deletions(-)

Comments

Ferruh Yigit May 6, 2020, 12:59 p.m. UTC | #1 
On 5/6/2020 1:22 PM, Asaf Penso wrote:
> There is no way to report back a link speed of 200Gbps.
> 
> Adding 200G link speed.
> 
> Signed-off-by: Asaf Penso <asafp@mellanox.com>
> ---
> v2 - reword commit log and remove 400G
> 
> v3 - rebase and fix checkpatch warning
> ---
>  app/test-pmd/cmdline.c                      | 12 +++++++-----
>  doc/guides/rel_notes/release_20_05.rst      |  2 ++
>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  2 +-
>  drivers/net/mlx5/mlx5_ethdev.c              | 22 +++++++++++++++++++++-
>  lib/librte_ethdev/rte_ethdev.c              |  2 ++
>  lib/librte_ethdev/rte_ethdev.h              |  2 ++
>  6 files changed, 35 insertions(+), 7 deletions(-)

Hi Asaf,

Patch looks good, but it updates ethdev & testpmd to have 200G defines and mlx5
to use it, so net/mlx5 prefix not really fits to the patch, can you separate
mlx5 changes into another patch?
Thomas Monjalon May 6, 2020, 1:17 p.m. UTC | #2 
06/05/2020 14:59, Ferruh Yigit:
> On 5/6/2020 1:22 PM, Asaf Penso wrote:
> > There is no way to report back a link speed of 200Gbps.
> > 
> > Adding 200G link speed.
> > 
> > Signed-off-by: Asaf Penso <asafp@mellanox.com>
> > ---
> > v2 - reword commit log and remove 400G
> > 
> > v3 - rebase and fix checkpatch warning
> > ---
> >  app/test-pmd/cmdline.c                      | 12 +++++++-----
> >  doc/guides/rel_notes/release_20_05.rst      |  2 ++
> >  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  2 +-
> >  drivers/net/mlx5/mlx5_ethdev.c              | 22 +++++++++++++++++++++-
> >  lib/librte_ethdev/rte_ethdev.c              |  2 ++
> >  lib/librte_ethdev/rte_ethdev.h              |  2 ++
> >  6 files changed, 35 insertions(+), 7 deletions(-)
> 
> Hi Asaf,
> 
> Patch looks good, but it updates ethdev & testpmd to have 200G defines and mlx5
> to use it, so net/mlx5 prefix not really fits to the patch, can you separate
> mlx5 changes into another patch?

I think I was the one advising to squash all.
The reason is that the changes in testpmd and ethdev are
simple and mechanical.
The real change is in mlx5 in my opinion, but I'm fine with splitting as well.
Ferruh Yigit May 6, 2020, 2:01 p.m. UTC | #3 
On 5/6/2020 2:17 PM, Thomas Monjalon wrote:
> 06/05/2020 14:59, Ferruh Yigit:
>> On 5/6/2020 1:22 PM, Asaf Penso wrote:
>>> There is no way to report back a link speed of 200Gbps.
>>>
>>> Adding 200G link speed.
>>>
>>> Signed-off-by: Asaf Penso <asafp@mellanox.com>
>>> ---
>>> v2 - reword commit log and remove 400G
>>>
>>> v3 - rebase and fix checkpatch warning
>>> ---
>>>  app/test-pmd/cmdline.c                      | 12 +++++++-----
>>>  doc/guides/rel_notes/release_20_05.rst      |  2 ++
>>>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  2 +-
>>>  drivers/net/mlx5/mlx5_ethdev.c              | 22 +++++++++++++++++++++-
>>>  lib/librte_ethdev/rte_ethdev.c              |  2 ++
>>>  lib/librte_ethdev/rte_ethdev.h              |  2 ++
>>>  6 files changed, 35 insertions(+), 7 deletions(-)
>>
>> Hi Asaf,
>>
>> Patch looks good, but it updates ethdev & testpmd to have 200G defines and mlx5
>> to use it, so net/mlx5 prefix not really fits to the patch, can you separate
>> mlx5 changes into another patch?
> 
> I think I was the one advising to squash all.
> The reason is that the changes in testpmd and ethdev are
> simple and mechanical.
> The real change is in mlx5 in my opinion, but I'm fine with splitting as well.
> 

Agree the real change is in the PMD and rest is mechanical, my concern was if we
need to refer the ethdev or testpmd change later, we will need to use commit
"net/mlx5: ..." which may be confusing.

I think ethdev, testpmd & doc can be combined into one "ethdev: ..." and PMD
implementation into other patch.

Does it make sense?
Thomas Monjalon May 6, 2020, 2:19 p.m. UTC | #4 
06/05/2020 16:01, Ferruh Yigit:
> On 5/6/2020 2:17 PM, Thomas Monjalon wrote:
> > 06/05/2020 14:59, Ferruh Yigit:
> >> On 5/6/2020 1:22 PM, Asaf Penso wrote:
> >>> There is no way to report back a link speed of 200Gbps.
> >>>
> >>> Adding 200G link speed.
> >>>
> >>> Signed-off-by: Asaf Penso <asafp@mellanox.com>
> >>> ---
> >>> v2 - reword commit log and remove 400G
> >>>
> >>> v3 - rebase and fix checkpatch warning
> >>> ---
> >>>  app/test-pmd/cmdline.c                      | 12 +++++++-----
> >>>  doc/guides/rel_notes/release_20_05.rst      |  2 ++
> >>>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  2 +-
> >>>  drivers/net/mlx5/mlx5_ethdev.c              | 22 +++++++++++++++++++++-
> >>>  lib/librte_ethdev/rte_ethdev.c              |  2 ++
> >>>  lib/librte_ethdev/rte_ethdev.h              |  2 ++
> >>>  6 files changed, 35 insertions(+), 7 deletions(-)
> >>
> >> Hi Asaf,
> >>
> >> Patch looks good, but it updates ethdev & testpmd to have 200G defines and mlx5
> >> to use it, so net/mlx5 prefix not really fits to the patch, can you separate
> >> mlx5 changes into another patch?
> > 
> > I think I was the one advising to squash all.
> > The reason is that the changes in testpmd and ethdev are
> > simple and mechanical.
> > The real change is in mlx5 in my opinion, but I'm fine with splitting as well.
> > 
> 
> Agree the real change is in the PMD and rest is mechanical, my concern was if we
> need to refer the ethdev or testpmd change later, we will need to use commit
> "net/mlx5: ..." which may be confusing.
> 
> I think ethdev, testpmd & doc can be combined into one "ethdev: ..." and PMD
> implementation into other patch.
> 
> Does it make sense?

Yes I understand your concern about future reference.
I am OK with what you propose:
	1/ ethdev: add 200G link speed
	2/ net/mlx5: support 200G link speed
Ferruh Yigit May 6, 2020, 2:22 p.m. UTC | #5 
On 5/6/2020 3:19 PM, Thomas Monjalon wrote:
> 06/05/2020 16:01, Ferruh Yigit:
>> On 5/6/2020 2:17 PM, Thomas Monjalon wrote:
>>> 06/05/2020 14:59, Ferruh Yigit:
>>>> On 5/6/2020 1:22 PM, Asaf Penso wrote:
>>>>> There is no way to report back a link speed of 200Gbps.
>>>>>
>>>>> Adding 200G link speed.
>>>>>
>>>>> Signed-off-by: Asaf Penso <asafp@mellanox.com>
>>>>> ---
>>>>> v2 - reword commit log and remove 400G
>>>>>
>>>>> v3 - rebase and fix checkpatch warning
>>>>> ---
>>>>>  app/test-pmd/cmdline.c                      | 12 +++++++-----
>>>>>  doc/guides/rel_notes/release_20_05.rst      |  2 ++
>>>>>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  2 +-
>>>>>  drivers/net/mlx5/mlx5_ethdev.c              | 22 +++++++++++++++++++++-
>>>>>  lib/librte_ethdev/rte_ethdev.c              |  2 ++
>>>>>  lib/librte_ethdev/rte_ethdev.h              |  2 ++
>>>>>  6 files changed, 35 insertions(+), 7 deletions(-)
>>>>
>>>> Hi Asaf,
>>>>
>>>> Patch looks good, but it updates ethdev & testpmd to have 200G defines and mlx5
>>>> to use it, so net/mlx5 prefix not really fits to the patch, can you separate
>>>> mlx5 changes into another patch?
>>>
>>> I think I was the one advising to squash all.
>>> The reason is that the changes in testpmd and ethdev are
>>> simple and mechanical.
>>> The real change is in mlx5 in my opinion, but I'm fine with splitting as well.
>>>
>>
>> Agree the real change is in the PMD and rest is mechanical, my concern was if we
>> need to refer the ethdev or testpmd change later, we will need to use commit
>> "net/mlx5: ..." which may be confusing.
>>
>> I think ethdev, testpmd & doc can be combined into one "ethdev: ..." and PMD
>> implementation into other patch.
>>
>> Does it make sense?
> 
> Yes I understand your concern about future reference.
> I am OK with what you propose:
> 	1/ ethdev: add 200G link speed
> 	2/ net/mlx5: support 200G link speed
> 

OK I will split while merging.
Asaf Penso May 6, 2020, 2:23 p.m. UTC | #6 
Regards,
Asaf Penso

> -----Original Message-----
> From: Ferruh Yigit <ferruh.yigit@intel.com>
> Sent: Wednesday, May 6, 2020 5:23 PM
> To: Thomas Monjalon <thomas@monjalon.net>
> Cc: dev@dpdk.org; Asaf Penso <asafp@mellanox.com>;
> arybchenko@solarflare.com; Slava Ovsiienko <viacheslavo@mellanox.com>;
> Matan Azrad <matan@mellanox.com>
> Subject: Re: [dpdk-dev] [PATCH v3] net/mlx5: add 200G link speed
> 
> On 5/6/2020 3:19 PM, Thomas Monjalon wrote:
> > 06/05/2020 16:01, Ferruh Yigit:
> >> On 5/6/2020 2:17 PM, Thomas Monjalon wrote:
> >>> 06/05/2020 14:59, Ferruh Yigit:
> >>>> On 5/6/2020 1:22 PM, Asaf Penso wrote:
> >>>>> There is no way to report back a link speed of 200Gbps.
> >>>>>
> >>>>> Adding 200G link speed.
> >>>>>
> >>>>> Signed-off-by: Asaf Penso <asafp@mellanox.com>
> >>>>> ---
> >>>>> v2 - reword commit log and remove 400G
> >>>>>
> >>>>> v3 - rebase and fix checkpatch warning
> >>>>> ---
> >>>>>  app/test-pmd/cmdline.c                      | 12 +++++++-----
> >>>>>  doc/guides/rel_notes/release_20_05.rst      |  2 ++
> >>>>>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  2 +-
> >>>>>  drivers/net/mlx5/mlx5_ethdev.c              | 22
> +++++++++++++++++++++-
> >>>>>  lib/librte_ethdev/rte_ethdev.c              |  2 ++
> >>>>>  lib/librte_ethdev/rte_ethdev.h              |  2 ++
> >>>>>  6 files changed, 35 insertions(+), 7 deletions(-)
> >>>>
> >>>> Hi Asaf,
> >>>>
> >>>> Patch looks good, but it updates ethdev & testpmd to have 200G
> defines and mlx5
> >>>> to use it, so net/mlx5 prefix not really fits to the patch, can you
> separate
> >>>> mlx5 changes into another patch?
> >>>
> >>> I think I was the one advising to squash all.
> >>> The reason is that the changes in testpmd and ethdev are
> >>> simple and mechanical.
> >>> The real change is in mlx5 in my opinion, but I'm fine with splitting as
> well.
> >>>
> >>
> >> Agree the real change is in the PMD and rest is mechanical, my concern
> was if we
> >> need to refer the ethdev or testpmd change later, we will need to use
> commit
> >> "net/mlx5: ..." which may be confusing.
> >>
> >> I think ethdev, testpmd & doc can be combined into one "ethdev: ..." and
> PMD
> >> implementation into other patch.
> >>
> >> Does it make sense?
> >
> > Yes I understand your concern about future reference.
> > I am OK with what you propose:
> > 	1/ ethdev: add 200G link speed
> > 	2/ net/mlx5: support 200G link speed
> >
> 
> OK I will split while merging.

Great, thanks for the assistance.
Ferruh Yigit May 6, 2020, 8:24 p.m. UTC | #7 
On 5/6/2020 1:22 PM, Asaf Penso wrote:
> There is no way to report back a link speed of 200Gbps.
> 
> Adding 200G link speed.
> 
> Signed-off-by: Asaf Penso <asafp@mellanox.com>

The doc warning fixed [1] and patch split into two as discussed [2] while merging.

For ethdev part,
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

Applied to dpdk-next-net/master, thanks.


[1]
WARNING: Inline strong start-string without end-string.

[2]
ethdev: add 200G link speed
net/mlx5: support 200G link speed
diff mbox series

Patch

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 1375f22..e6c6b68 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -769,7 +769,7 @@  static void cmd_help_long_parsed(void *parsed_result,
 			"    Detach physical or virtual dev by port_id\n\n"
 
 			"port config (port_id|all)"
-			" speed (10|100|1000|10000|25000|40000|50000|100000|auto)"
+			" speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto)"
 			" duplex (half|full|auto)\n"
 			"    Set speed and duplex for all ports or port_id\n\n"
 
@@ -1619,6 +1619,8 @@  parse_and_check_speed_duplex(char *speedstr, char *duplexstr, uint32_t *speed)
 			*speed = ETH_LINK_SPEED_50G;
 		} else if (!strcmp(speedstr, "100000")) {
 			*speed = ETH_LINK_SPEED_100G;
+		} else if (!strcmp(speedstr, "200000")) {
+			*speed = ETH_LINK_SPEED_200G;
 		} else if (!strcmp(speedstr, "auto")) {
 			*speed = ETH_LINK_SPEED_AUTONEG;
 		} else {
@@ -1666,7 +1668,7 @@  cmdline_parse_token_string_t cmd_config_speed_all_item1 =
 	TOKEN_STRING_INITIALIZER(struct cmd_config_speed_all, item1, "speed");
 cmdline_parse_token_string_t cmd_config_speed_all_value1 =
 	TOKEN_STRING_INITIALIZER(struct cmd_config_speed_all, value1,
-				"10#100#1000#10000#25000#40000#50000#100000#auto");
+				"10#100#1000#10000#25000#40000#50000#100000#200000#auto");
 cmdline_parse_token_string_t cmd_config_speed_all_item2 =
 	TOKEN_STRING_INITIALIZER(struct cmd_config_speed_all, item2, "duplex");
 cmdline_parse_token_string_t cmd_config_speed_all_value2 =
@@ -1677,7 +1679,7 @@  cmdline_parse_inst_t cmd_config_speed_all = {
 	.f = cmd_config_speed_all_parsed,
 	.data = NULL,
 	.help_str = "port config all speed "
-		"10|100|1000|10000|25000|40000|50000|100000|auto duplex "
+		"10|100|1000|10000|25000|40000|50000|100000|200000|auto duplex "
 							"half|full|auto",
 	.tokens = {
 		(void *)&cmd_config_speed_all_port,
@@ -1741,7 +1743,7 @@  cmdline_parse_token_string_t cmd_config_speed_specific_item1 =
 								"speed");
 cmdline_parse_token_string_t cmd_config_speed_specific_value1 =
 	TOKEN_STRING_INITIALIZER(struct cmd_config_speed_specific, value1,
-				"10#100#1000#10000#25000#40000#50000#100000#auto");
+				"10#100#1000#10000#25000#40000#50000#100000#200000#auto");
 cmdline_parse_token_string_t cmd_config_speed_specific_item2 =
 	TOKEN_STRING_INITIALIZER(struct cmd_config_speed_specific, item2,
 								"duplex");
@@ -1753,7 +1755,7 @@  cmdline_parse_inst_t cmd_config_speed_specific = {
 	.f = cmd_config_speed_specific_parsed,
 	.data = NULL,
 	.help_str = "port config <port_id> speed "
-		"10|100|1000|10000|25000|40000|50000|100000|auto duplex "
+		"10|100|1000|10000|25000|40000|50000|100000|200000|auto duplex "
 							"half|full|auto",
 	.tokens = {
 		(void *)&cmd_config_speed_specific_port,
diff --git a/doc/guides/rel_notes/release_20_05.rst b/doc/guides/rel_notes/release_20_05.rst
index 777d36e..eb8a623 100644
--- a/doc/guides/rel_notes/release_20_05.rst
+++ b/doc/guides/rel_notes/release_20_05.rst
@@ -66,6 +66,8 @@  New Features
   Added tracepoints in ``EAL``, ``ethdev``, ``cryptodev``, ``eventdev`` and
   ``mempool`` libraries for important functions.
 
+* **ethdev: Added a new value to link speed for 200Gbps **
+
 * **Added new API for rte_ring.**
 
   * New synchronization modes for rte_ring.
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index a360ecc..a1cbb2e 100644
--- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
+++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
@@ -2153,7 +2153,7 @@  port config - speed
 
 Set the speed and duplex mode for all ports or a specific port::
 
-   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
+   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
             duplex (half|full|auto)
 
 port config - queues/descriptors
diff --git a/drivers/net/mlx5/mlx5_ethdev.c b/drivers/net/mlx5/mlx5_ethdev.c
index d7d3bc7..47f11b9 100644
--- a/drivers/net/mlx5/mlx5_ethdev.c
+++ b/drivers/net/mlx5/mlx5_ethdev.c
@@ -86,6 +86,7 @@  struct ethtool_link_settings {
 	uint32_t link_mode_masks[];
 };
 
+/* The kernel values can be found in /include/uapi/linux/ethtool.h */
 #define ETHTOOL_GLINKSETTINGS 0x0000004c
 #define ETHTOOL_LINK_MODE_1000baseT_Full_BIT 5
 #define ETHTOOL_LINK_MODE_Autoneg_BIT 6
@@ -119,6 +120,13 @@  struct ethtool_link_settings {
 #define ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT 38
 #define ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT 39
 #endif
+#ifndef HAVE_ETHTOOL_LINK_MODE_200G
+#define ETHTOOL_LINK_MODE_200000baseKR4_Full_BIT 62
+#define ETHTOOL_LINK_MODE_200000baseSR4_Full_BIT 63
+#define ETHTOOL_LINK_MODE_200000baseLR4_ER4_FR4_Full_BIT 0 /* 64 - 64 */
+#define ETHTOOL_LINK_MODE_200000baseDR4_Full_BIT 1 /* 65 - 64 */
+#define ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT 2 /* 66 - 64 */
+#endif
 
 /**
  * Get master interface name from private structure.
@@ -537,7 +545,8 @@  mlx5_set_default_params(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
 	info->default_txportconf.ring_size = 256;
 	info->default_rxportconf.burst_size = MLX5_RX_DEFAULT_BURST;
 	info->default_txportconf.burst_size = MLX5_TX_DEFAULT_BURST;
-	if (priv->link_speed_capa & ETH_LINK_SPEED_100G) {
+	if ((priv->link_speed_capa & ETH_LINK_SPEED_200G) |
+		(priv->link_speed_capa & ETH_LINK_SPEED_100G)) {
 		info->default_rxportconf.nb_queues = 16;
 		info->default_txportconf.nb_queues = 16;
 		if (dev->data->nb_rx_queues > 2 ||
@@ -1028,6 +1037,17 @@  mlx5_link_update_unlocked_gs(struct rte_eth_dev *dev,
 		  MLX5_BITSHIFT(ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT) |
 		  MLX5_BITSHIFT(ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT)))
 		priv->link_speed_capa |= ETH_LINK_SPEED_100G;
+	if (sc & (MLX5_BITSHIFT(ETHTOOL_LINK_MODE_200000baseKR4_Full_BIT) |
+		  MLX5_BITSHIFT(ETHTOOL_LINK_MODE_200000baseSR4_Full_BIT)))
+		priv->link_speed_capa |= ETH_LINK_SPEED_200G;
+
+	sc = ecmd->link_mode_masks[2] |
+		((uint64_t)ecmd->link_mode_masks[3] << 32);
+	if (sc & (MLX5_BITSHIFT(ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT) |
+		  MLX5_BITSHIFT(
+			ETHTOOL_LINK_MODE_200000baseLR4_ER4_FR4_Full_BIT) |
+		  MLX5_BITSHIFT(ETHTOOL_LINK_MODE_200000baseDR4_Full_BIT)))
+		priv->link_speed_capa |= ETH_LINK_SPEED_200G;
 	dev_link.link_duplex = ((ecmd->duplex == DUPLEX_HALF) ?
 				ETH_LINK_HALF_DUPLEX : ETH_LINK_FULL_DUPLEX);
 	dev_link.link_autoneg = !(dev->data->dev_conf.link_speeds &
diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index 72aed59..a64df8f 100644
--- a/lib/librte_ethdev/rte_ethdev.c
+++ b/lib/librte_ethdev/rte_ethdev.c
@@ -1099,6 +1099,8 @@  rte_eth_speed_bitflag(uint32_t speed, int duplex)
 		return ETH_LINK_SPEED_56G;
 	case ETH_SPEED_NUM_100G:
 		return ETH_LINK_SPEED_100G;
+	case ETH_SPEED_NUM_200G:
+		return ETH_LINK_SPEED_200G;
 	default:
 		return 0;
 	}
diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
index 0f6d053..a49242b 100644
--- a/lib/librte_ethdev/rte_ethdev.h
+++ b/lib/librte_ethdev/rte_ethdev.h
@@ -284,6 +284,7 @@  struct rte_eth_stats {
 #define ETH_LINK_SPEED_50G      (1 << 12)  /**<  50 Gbps */
 #define ETH_LINK_SPEED_56G      (1 << 13)  /**<  56 Gbps */
 #define ETH_LINK_SPEED_100G     (1 << 14)  /**< 100 Gbps */
+#define ETH_LINK_SPEED_200G     (1 << 15)  /**< 200 Gbps */
 
 /**
  * Ethernet numeric link speeds in Mbps
@@ -301,6 +302,7 @@  struct rte_eth_stats {
 #define ETH_SPEED_NUM_50G      50000 /**<  50 Gbps */
 #define ETH_SPEED_NUM_56G      56000 /**<  56 Gbps */
 #define ETH_SPEED_NUM_100G    100000 /**< 100 Gbps */
+#define ETH_SPEED_NUM_200G    200000 /**< 200 Gbps */
 
 /**
  * A structure used to retrieve link-level information of an Ethernet port.