diff mbox series

[iproute2,v2] vdpa: Add virtqueue pairs set capacity

Message ID 1649237791-31723-1-git-send-email-qiudayu@archeros.com (mailing list archive)
State Changes Requested
Delegated to: David Ahern
Headers show
Series [iproute2,v2] vdpa: Add virtqueue pairs set capacity | expand

Checks

Context Check Description
netdev/tree_selection success Not a local patch

Commit Message

Michael Qiu April 6, 2022, 9:36 a.m. UTC 
vdpa framework not only support query the max virtqueue pair, but
also for the set action.

This patch enable this capacity, and it is very useful for VMs
 who needs multiqueue support.

After enable this feature, we could simply use below command to
create multi-queue support:

vdpa dev add mgmtdev pci/0000:03:10.3 name foo mac 56:d0:2f:03:c9:6d max_vqp 6

Signed-off-by: Michael Qiu <qiudayu@archeros.com>
---

v2 --> v1:
    rename "max_vq_pairs" to "max_vqp"

    extend the man page for this addition with example
---
 man/man8/vdpa-dev.8 |  9 +++++++++
 vdpa/vdpa.c         | 19 ++++++++++++++++---
 2 files changed, 25 insertions(+), 3 deletions(-)

Comments

Stephen Hemminger April 6, 2022, 3:39 p.m. UTC | #1 
On Wed,  6 Apr 2022 05:36:31 -0400
Michael Qiu <qiudayu@archeros.com> wrote:

> 	if (tb[VDPA_ATTR_DEV_NET_CFG_MAX_VQP]) {
>  		val_u16 = mnl_attr_get_u16(tb[VDPA_ATTR_DEV_NET_CFG_MAX_VQP]);
> -		print_uint(PRINT_ANY, "max_vq_pairs", "max_vq_pairs %d ",
> +		print_uint(PRINT_ANY, "max_vqp", "max_vqp %d ",
>  			     val_u16);

Good, the print should match the set parameters.

Wanted to apply but can not go against main branch, is the targeted for next
Parav Pandit April 6, 2022, 8:56 p.m. UTC | #2 
Hi Michael,

> From: Michael Qiu <qiudayu@archeros.com>
> Sent: Wednesday, April 6, 2022 5:37 AM
> To: Parav Pandit <parav@nvidia.com>; stephen@networkplumber.org
> Cc: netdev@vger.kernel.org; Michael Qiu <qiudayu@archeros.com>
> Subject: [PATCH iproute2 v2] vdpa: Add virtqueue pairs set capacity
> 
> vdpa framework not only support query the max virtqueue pair, but also for
> the set action.
> 
> This patch enable this capacity, and it is very useful for VMs  who needs
> multiqueue support.
> 
> After enable this feature, we could simply use below command to create
> multi-queue support:
> 
> vdpa dev add mgmtdev pci/0000:03:10.3 name foo mac 56:d0:2f:03:c9:6d
> max_vqp 6
> 
> Signed-off-by: Michael Qiu <qiudayu@archeros.com>
> ---
> 
> v2 --> v1:
>     rename "max_vq_pairs" to "max_vqp"
> 
>     extend the man page for this addition with example
> ---
>  man/man8/vdpa-dev.8 |  9 +++++++++
>  vdpa/vdpa.c         | 19 ++++++++++++++++---
>  2 files changed, 25 insertions(+), 3 deletions(-)
> 
> diff --git a/man/man8/vdpa-dev.8 b/man/man8/vdpa-dev.8 index
> aa21ae3..a401115 100644
> --- a/man/man8/vdpa-dev.8
> +++ b/man/man8/vdpa-dev.8
> @@ -74,6 +74,10 @@ This is applicable only for the network type of vdpa
> device. This is optional.
>  - specifies the mtu for the new vdpa device.
>  This is applicable only for the network type of vdpa device. This is optional.
> 
> +.BI max_vqp " MAX_VQP"
> +- specifies the max queue pairs for the new vdpa device.
> +This is applicable only for the network type of vdpa device. This is optional.
> +
>  .SS vdpa dev del - Delete the vdpa device.
> 
>  .PP
> @@ -119,6 +123,11 @@ vdpa dev add name foo mgmtdev vdpa_sim_net
> mac 00:11:22:33:44:55 mtu 9000  Add the vdpa device named foo on the
> management device vdpa_sim_net with mac address of 00:11:22:33:44:55
> and mtu of 9000 bytes.
>  .RE
>  .PP
> +vdpa dev add name foo mgmtdev vdpa_sim_net mac 00:11:22:33:44:55 mtu
> +9000 max_vqp 6 .RS 4 Add the vdpa device named foo on the management
> +device vdpa_sim_net with mac address of 00:11:22:33:44:55, mtu of 9000
> bytes and max virtqueue pairs of 6.
> +.RE
> +.PP
>  vdpa dev del foo
>  .RS 4
>  Delete the vdpa device named foo which was previously created.
> diff --git a/vdpa/vdpa.c b/vdpa/vdpa.c
> index f048e47..104c503 100644
> --- a/vdpa/vdpa.c
> +++ b/vdpa/vdpa.c
> @@ -23,6 +23,7 @@
>  #define VDPA_OPT_VDEV_HANDLE		BIT(3)
>  #define VDPA_OPT_VDEV_MAC		BIT(4)
>  #define VDPA_OPT_VDEV_MTU		BIT(5)
> +#define VDPA_OPT_VDEV_QUEUE_PAIRS	BIT(6)
> 
>  struct vdpa_opts {
>  	uint64_t present; /* flags of present items */ @@ -32,6 +33,7 @@
> struct vdpa_opts {
>  	unsigned int device_id;
>  	char mac[ETH_ALEN];
>  	uint16_t mtu;
> +	uint16_t max_vqp;
>  };
> 
>  struct vdpa {
> @@ -219,6 +221,8 @@ static void vdpa_opts_put(struct nlmsghdr *nlh,
> struct vdpa *vdpa)
>  			     sizeof(opts->mac), opts->mac);
>  	if (opts->present & VDPA_OPT_VDEV_MTU)
>  		mnl_attr_put_u16(nlh, VDPA_ATTR_DEV_NET_CFG_MTU,
> opts->mtu);
> +	if (opts->present & VDPA_OPT_VDEV_QUEUE_PAIRS)
> +		mnl_attr_put_u16(nlh,
> VDPA_ATTR_DEV_NET_CFG_MAX_VQP, opts->max_vqp);
>  }
> 
>  static int vdpa_argv_parse(struct vdpa *vdpa, int argc, char **argv, @@ -
> 287,6 +291,15 @@ static int vdpa_argv_parse(struct vdpa *vdpa, int argc,
> char **argv,
> 
>  			NEXT_ARG_FWD();
>  			o_found |= VDPA_OPT_VDEV_MTU;
> +		} else if ((strcmp(*argv, "max_vqp") == 0) &&
> +			   (o_all & VDPA_OPT_VDEV_QUEUE_PAIRS)) {
> +			NEXT_ARG_FWD();
> +			err = vdpa_argv_u16(vdpa, argc, argv, &opts-
> >max_vqp);
> +			if (err)
> +				return err;
> +
> +			NEXT_ARG_FWD();
> +			o_found |= VDPA_OPT_VDEV_QUEUE_PAIRS;
>  		} else {
>  			fprintf(stderr, "Unknown option \"%s\"\n", *argv);
>  			return -EINVAL;
> @@ -467,7 +480,7 @@ static int cmd_mgmtdev(struct vdpa *vdpa, int argc,
> char **argv)  static void cmd_dev_help(void)  {
>  	fprintf(stderr, "Usage: vdpa dev show [ DEV ]\n");
> -	fprintf(stderr, "       vdpa dev add name NAME mgmtdev
> MANAGEMENTDEV [ mac MACADDR ] [ mtu MTU ]\n");
> +	fprintf(stderr, "       vdpa dev add name NAME mgmtdev
> MANAGEMENTDEV [ mac MACADDR ] [ mtu MTU ] [ max_vqp N ]\n");
>  	fprintf(stderr, "       vdpa dev del DEV\n");
>  	fprintf(stderr, "Usage: vdpa dev config COMMAND [ OPTIONS ]\n");
> } @@ -557,7 +570,7 @@ static int cmd_dev_add(struct vdpa *vdpa, int argc,
> char **argv)
>  					  NLM_F_REQUEST | NLM_F_ACK);
>  	err = vdpa_argv_parse_put(nlh, vdpa, argc, argv,
>  				  VDPA_OPT_VDEV_MGMTDEV_HANDLE |
> VDPA_OPT_VDEV_NAME,
> -				  VDPA_OPT_VDEV_MAC |
> VDPA_OPT_VDEV_MTU);
> +				  VDPA_OPT_VDEV_MAC |
> VDPA_OPT_VDEV_MTU |
> +VDPA_OPT_VDEV_QUEUE_PAIRS);
>  	if (err)
>  		return err;
> 
> @@ -603,7 +616,7 @@ static void pr_out_dev_net_config(struct nlattr **tb)
>  	}
>  	if (tb[VDPA_ATTR_DEV_NET_CFG_MAX_VQP]) {
>  		val_u16 =
> mnl_attr_get_u16(tb[VDPA_ATTR_DEV_NET_CFG_MAX_VQP]);
> -		print_uint(PRINT_ANY, "max_vq_pairs", "max_vq_pairs %d
> ",
> +		print_uint(PRINT_ANY, "max_vqp", "max_vqp %d ",
>  			     val_u16);
>  	}
>  	if (tb[VDPA_ATTR_DEV_NET_CFG_MTU]) {
> --
> 1.8.3.1
> 

You are probably not using right iproute2 tree.

I already see existing code from Eli in commit [1] that supports setting this field.

The iproute2 tree to use for upstream work is [2].

[1] commit 16482fd4d ("vdpa: Support for configuring max VQ pairs for a device")
[2] git://git.kernel.org/pub/scm/network/iproute2/iproute2.git
diff mbox series

Patch

diff --git a/man/man8/vdpa-dev.8 b/man/man8/vdpa-dev.8
index aa21ae3..a401115 100644
--- a/man/man8/vdpa-dev.8
+++ b/man/man8/vdpa-dev.8
@@ -74,6 +74,10 @@  This is applicable only for the network type of vdpa device. This is optional.
 - specifies the mtu for the new vdpa device.
 This is applicable only for the network type of vdpa device. This is optional.
 
+.BI max_vqp " MAX_VQP"
+- specifies the max queue pairs for the new vdpa device.
+This is applicable only for the network type of vdpa device. This is optional.
+
 .SS vdpa dev del - Delete the vdpa device.
 
 .PP
@@ -119,6 +123,11 @@  vdpa dev add name foo mgmtdev vdpa_sim_net mac 00:11:22:33:44:55 mtu 9000
 Add the vdpa device named foo on the management device vdpa_sim_net with mac address of 00:11:22:33:44:55 and mtu of 9000 bytes.
 .RE
 .PP
+vdpa dev add name foo mgmtdev vdpa_sim_net mac 00:11:22:33:44:55 mtu 9000 max_vqp 6
+.RS 4
+Add the vdpa device named foo on the management device vdpa_sim_net with mac address of 00:11:22:33:44:55, mtu of 9000 bytes and max virtqueue pairs of 6.
+.RE
+.PP
 vdpa dev del foo
 .RS 4
 Delete the vdpa device named foo which was previously created.
diff --git a/vdpa/vdpa.c b/vdpa/vdpa.c
index f048e47..104c503 100644
--- a/vdpa/vdpa.c
+++ b/vdpa/vdpa.c
@@ -23,6 +23,7 @@ 
 #define VDPA_OPT_VDEV_HANDLE		BIT(3)
 #define VDPA_OPT_VDEV_MAC		BIT(4)
 #define VDPA_OPT_VDEV_MTU		BIT(5)
+#define VDPA_OPT_VDEV_QUEUE_PAIRS	BIT(6)
 
 struct vdpa_opts {
 	uint64_t present; /* flags of present items */
@@ -32,6 +33,7 @@  struct vdpa_opts {
 	unsigned int device_id;
 	char mac[ETH_ALEN];
 	uint16_t mtu;
+	uint16_t max_vqp;
 };
 
 struct vdpa {
@@ -219,6 +221,8 @@  static void vdpa_opts_put(struct nlmsghdr *nlh, struct vdpa *vdpa)
 			     sizeof(opts->mac), opts->mac);
 	if (opts->present & VDPA_OPT_VDEV_MTU)
 		mnl_attr_put_u16(nlh, VDPA_ATTR_DEV_NET_CFG_MTU, opts->mtu);
+	if (opts->present & VDPA_OPT_VDEV_QUEUE_PAIRS)
+		mnl_attr_put_u16(nlh, VDPA_ATTR_DEV_NET_CFG_MAX_VQP, opts->max_vqp);
 }
 
 static int vdpa_argv_parse(struct vdpa *vdpa, int argc, char **argv,
@@ -287,6 +291,15 @@  static int vdpa_argv_parse(struct vdpa *vdpa, int argc, char **argv,
 
 			NEXT_ARG_FWD();
 			o_found |= VDPA_OPT_VDEV_MTU;
+		} else if ((strcmp(*argv, "max_vqp") == 0) &&
+			   (o_all & VDPA_OPT_VDEV_QUEUE_PAIRS)) {
+			NEXT_ARG_FWD();
+			err = vdpa_argv_u16(vdpa, argc, argv, &opts->max_vqp);
+			if (err)
+				return err;
+
+			NEXT_ARG_FWD();
+			o_found |= VDPA_OPT_VDEV_QUEUE_PAIRS;
 		} else {
 			fprintf(stderr, "Unknown option \"%s\"\n", *argv);
 			return -EINVAL;
@@ -467,7 +480,7 @@  static int cmd_mgmtdev(struct vdpa *vdpa, int argc, char **argv)
 static void cmd_dev_help(void)
 {
 	fprintf(stderr, "Usage: vdpa dev show [ DEV ]\n");
-	fprintf(stderr, "       vdpa dev add name NAME mgmtdev MANAGEMENTDEV [ mac MACADDR ] [ mtu MTU ]\n");
+	fprintf(stderr, "       vdpa dev add name NAME mgmtdev MANAGEMENTDEV [ mac MACADDR ] [ mtu MTU ] [ max_vqp N ]\n");
 	fprintf(stderr, "       vdpa dev del DEV\n");
 	fprintf(stderr, "Usage: vdpa dev config COMMAND [ OPTIONS ]\n");
 }
@@ -557,7 +570,7 @@  static int cmd_dev_add(struct vdpa *vdpa, int argc, char **argv)
 					  NLM_F_REQUEST | NLM_F_ACK);
 	err = vdpa_argv_parse_put(nlh, vdpa, argc, argv,
 				  VDPA_OPT_VDEV_MGMTDEV_HANDLE | VDPA_OPT_VDEV_NAME,
-				  VDPA_OPT_VDEV_MAC | VDPA_OPT_VDEV_MTU);
+				  VDPA_OPT_VDEV_MAC | VDPA_OPT_VDEV_MTU | VDPA_OPT_VDEV_QUEUE_PAIRS);
 	if (err)
 		return err;
 
@@ -603,7 +616,7 @@  static void pr_out_dev_net_config(struct nlattr **tb)
 	}
 	if (tb[VDPA_ATTR_DEV_NET_CFG_MAX_VQP]) {
 		val_u16 = mnl_attr_get_u16(tb[VDPA_ATTR_DEV_NET_CFG_MAX_VQP]);
-		print_uint(PRINT_ANY, "max_vq_pairs", "max_vq_pairs %d ",
+		print_uint(PRINT_ANY, "max_vqp", "max_vqp %d ",
 			     val_u16);
 	}
 	if (tb[VDPA_ATTR_DEV_NET_CFG_MTU]) {