Skip to contentSkip to conversation

net/vdev_netvsc: print warning if Mellanox devices are not configured

PATCH In review Stephen Hemminger 3 participants 11 messages May 23, 2019 – May 27, 2019 Network drivers

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 net/vdev_netvsc: print warning if Mellanox devices are not configured

Several users have run into problems where the MLX drivers were not enabled in their build. And then trying to run their DPDK application on Azure. What happens is that all packets go over the slow path, and failsafe repeatedly probes for never existing sub-device.

Both Mellanox drivers should be checked. MLX4 for current versions, and MLX5 for future upgrades. This code is only called if Hyper-V/Azure is detected.

---
 drivers/net/vdev_netvsc/vdev_netvsc.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/net/vdev_netvsc/vdev_netvsc.c +7 -0
index 801f54c96e01..64f9dbf66e18 100644
@@ -812,6 +812,13 @@ vdev_netvsc_scan_callback(__rte_unused void *arg)
 	struct rte_devargs *devargs;
 	struct rte_bus *vbus = rte_bus_find_by_name("vdev");
 
+#ifndef RTE_LIBRTE_MLX4_PMD
+	DRV_LOG(WARNING, "Mellanox MLX4 not configured.");
+#endif
+#ifndef RTE_LIBRTE_MLX5_PMD
+	DRV_LOG(WARNING, "Mellanox MLX5 not configured.");
+#endif
+
 	RTE_EAL_DEVARGS_FOREACH("vdev", devargs)
 		if (!strncmp(devargs->name, VDEV_NETVSC_DRIVER_NAME,
 			     VDEV_NETVSC_DRIVER_NAME_LEN))
Signature · 1 line
2.20.1
#1 lore raw
Quoted 20 lines · Stephen Hemminger

Several users have run into problems where the MLX drivers were not enabled in their build. And then trying to run their DPDK application on Azure. What happens is that all packets go over the slow path, and failsafe repeatedly probes for never existing sub-device.

Both Mellanox drivers should be checked. MLX4 for current versions, and MLX5 for future upgrades. This code is only called if Hyper-V/Azure is detected.

---
 drivers/net/vdev_netvsc/vdev_netvsc.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/net/vdev_netvsc/vdev_netvsc.c +0 -0
index 801f54c96e01..64f9dbf66e18 100644

drivers/net/vdev_netvsc/vdev_netvsc.c vdev_netvsc_scan_callback(__rte_unused void *arg) · line 820

Quoted 9 lines
drivers/net/vdev_netvsc/vdev_netvsc.c +6 -0
@@ -812,6 +812,13 @@ vdev_netvsc_scan_callback(__rte_unused void *arg)
 	struct rte_devargs *devargs;
 	struct rte_bus *vbus = rte_bus_find_by_name("vdev");
 
+#ifndef RTE_LIBRTE_MLX4_PMD
+	DRV_LOG(WARNING, "Mellanox MLX4 not configured.");
+#endif
+#ifndef RTE_LIBRTE_MLX5_PMD
+	DRV_LOG(WARNING, "Mellanox MLX5 not configured.");
+#endif

Is it OK a virtual PMD being this much aware of another PMD?
Can it be an option to add this check into build system? And if there is direct dependency perhaps even don't compile the 'vdev_netvsc' when none mlx PMD is enabled.

Quoted 5 lines
drivers/net/vdev_netvsc/vdev_netvsc.c +1 -0
+
 	RTE_EAL_DEVARGS_FOREACH("vdev", devargs)
 		if (!strncmp(devargs->name, VDEV_NETVSC_DRIVER_NAME,
 			     VDEV_NETVSC_DRIVER_NAME_LEN))
#2 lore raw
Quoted 24 lines · On Fri, 24 May 2019 14:26:40 +0100 Ferruh Yigit <[email protected]> wrote:
Quoted 23 lines · Stephen Hemminger

Several users have run into problems where the MLX drivers were not enabled in their build. And then trying to run their DPDK application on Azure. What happens is that all packets go over the slow path, and failsafe repeatedly probes for never existing sub-device.

Both Mellanox drivers should be checked. MLX4 for current versions, and MLX5 for future upgrades. This code is only called if Hyper-V/Azure is detected.

---
 drivers/net/vdev_netvsc/vdev_netvsc.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/net/vdev_netvsc/vdev_netvsc.c +0 -0
index 801f54c96e01..64f9dbf66e18 100644
@@ -812,6 +812,13 @@ vdev_netvsc_scan_callback(__rte_unused void *arg)
 	struct rte_devargs *devargs;
 	struct rte_bus *vbus = rte_bus_find_by_name("vdev");
 
Quoted 11 lines

drivers/net/vdev_netvsc/vdev_netvsc.c vdev_netvsc_scan_callback(__rte_unused void *arg) · line 820

Quoted 6 lines
drivers/net/vdev_netvsc/vdev_netvsc.c +6 -0
+#ifndef RTE_LIBRTE_MLX4_PMD
+	DRV_LOG(WARNING, "Mellanox MLX4 not configured.");
+#endif
+#ifndef RTE_LIBRTE_MLX5_PMD
+	DRV_LOG(WARNING, "Mellanox MLX5 not configured.");
+#endif  

Is it OK a virtual PMD being this much aware of another PMD?
Can it be an option to add this check into build system? And if there is direct dependency perhaps even don't compile the 'vdev_netvsc' when none mlx PMD is enabled.

vdev_netvsc is not a device, it is really just a hack to start other device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

#3 lore raw

24/05/2019 18:07, Stephen Hemminger:

Quoted 44 lines
Quoted 24 lines · On Fri, 24 May 2019 14:26:40 +0100 Ferruh Yigit <[email protected]> wrote:
Quoted 23 lines · Stephen Hemminger

Several users have run into problems where the MLX drivers were not enabled in their build. And then trying to run their DPDK application on Azure. What happens is that all packets go over the slow path, and failsafe repeatedly probes for never existing sub-device.

Both Mellanox drivers should be checked. MLX4 for current versions, and MLX5 for future upgrades. This code is only called if Hyper-V/Azure is detected.

---
 drivers/net/vdev_netvsc/vdev_netvsc.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/net/vdev_netvsc/vdev_netvsc.c +0 -0
index 801f54c96e01..64f9dbf66e18 100644
@@ -812,6 +812,13 @@ vdev_netvsc_scan_callback(__rte_unused void *arg)
 	struct rte_devargs *devargs;
 	struct rte_bus *vbus = rte_bus_find_by_name("vdev");
 
Quoted 11 lines

drivers/net/vdev_netvsc/vdev_netvsc.c vdev_netvsc_scan_callback(__rte_unused void *arg) · line 820

Quoted 6 lines
drivers/net/vdev_netvsc/vdev_netvsc.c +6 -0
+#ifndef RTE_LIBRTE_MLX4_PMD
+	DRV_LOG(WARNING, "Mellanox MLX4 not configured.");
+#endif
+#ifndef RTE_LIBRTE_MLX5_PMD
+	DRV_LOG(WARNING, "Mellanox MLX5 not configured.");
+#endif  

Is it OK a virtual PMD being this much aware of another PMD?
Can it be an option to add this check into build system? And if there is direct dependency perhaps even don't compile the 'vdev_netvsc' when none mlx PMD is enabled.

vdev_netvsc is not a device, it is really just a hack to start other device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

Quoted 6 lines

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

#4 lore raw
Quoted 49 lines · On Fri, 24 May 2019 18:38:53 +0200 Thomas Monjalon <[email protected]> wrote:

24/05/2019 18:07, Stephen Hemminger:

Quoted 48 lines
Quoted 24 lines · On Fri, 24 May 2019 14:26:40 +0100 Ferruh Yigit <[email protected]> wrote:
Quoted 23 lines · Stephen Hemminger

Several users have run into problems where the MLX drivers were not enabled in their build. And then trying to run their DPDK application on Azure. What happens is that all packets go over the slow path, and failsafe repeatedly probes for never existing sub-device.

Both Mellanox drivers should be checked. MLX4 for current versions, and MLX5 for future upgrades. This code is only called if Hyper-V/Azure is detected.

---
 drivers/net/vdev_netvsc/vdev_netvsc.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/net/vdev_netvsc/vdev_netvsc.c +0 -0
index 801f54c96e01..64f9dbf66e18 100644
@@ -812,6 +812,13 @@ vdev_netvsc_scan_callback(__rte_unused void *arg)
 	struct rte_devargs *devargs;
 	struct rte_bus *vbus = rte_bus_find_by_name("vdev");
 
Quoted 11 lines

drivers/net/vdev_netvsc/vdev_netvsc.c vdev_netvsc_scan_callback(__rte_unused void *arg) · line 820

Quoted 6 lines
drivers/net/vdev_netvsc/vdev_netvsc.c +6 -0
+#ifndef RTE_LIBRTE_MLX4_PMD
+	DRV_LOG(WARNING, "Mellanox MLX4 not configured.");
+#endif
+#ifndef RTE_LIBRTE_MLX5_PMD
+	DRV_LOG(WARNING, "Mellanox MLX5 not configured.");
+#endif    

Is it OK a virtual PMD being this much aware of another PMD?
Can it be an option to add this check into build system? And if there is direct dependency perhaps even don't compile the 'vdev_netvsc' when none mlx PMD is enabled.

vdev_netvsc is not a device, it is really just a hack to start other device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

Quoted 7 lines
Quoted 2 lines

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

None of the people that ran into this were using meson build. For example, VPP doesn't use meson.

#5 lore raw

24/05/2019 18:48, Stephen Hemminger:

Quoted 55 lines
Quoted 53 lines · On Fri, 24 May 2019 18:38:53 +0200 Thomas Monjalon <[email protected]> wrote:

24/05/2019 18:07, Stephen Hemminger:

Quoted 50 lines
Quoted 24 lines · On Fri, 24 May 2019 14:26:40 +0100 Ferruh Yigit <[email protected]> wrote:
Quoted 23 lines · Stephen Hemminger

Several users have run into problems where the MLX drivers were not enabled in their build. And then trying to run their DPDK application on Azure. What happens is that all packets go over the slow path, and failsafe repeatedly probes for never existing sub-device.

Both Mellanox drivers should be checked. MLX4 for current versions, and MLX5 for future upgrades. This code is only called if Hyper-V/Azure is detected.

---
 drivers/net/vdev_netvsc/vdev_netvsc.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/net/vdev_netvsc/vdev_netvsc.c +0 -0
index 801f54c96e01..64f9dbf66e18 100644
@@ -812,6 +812,13 @@ vdev_netvsc_scan_callback(__rte_unused void *arg)
 	struct rte_devargs *devargs;
 	struct rte_bus *vbus = rte_bus_find_by_name("vdev");
 
Quoted 11 lines

drivers/net/vdev_netvsc/vdev_netvsc.c vdev_netvsc_scan_callback(__rte_unused void *arg) · line 820

Quoted 6 lines
drivers/net/vdev_netvsc/vdev_netvsc.c +6 -0
+#ifndef RTE_LIBRTE_MLX4_PMD
+	DRV_LOG(WARNING, "Mellanox MLX4 not configured.");
+#endif
+#ifndef RTE_LIBRTE_MLX5_PMD
+	DRV_LOG(WARNING, "Mellanox MLX5 not configured.");
+#endif    

Is it OK a virtual PMD being this much aware of another PMD?
Can it be an option to add this check into build system? And if there is direct dependency perhaps even don't compile the 'vdev_netvsc' when none mlx PMD is enabled.

vdev_netvsc is not a device, it is really just a hack to start other device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

Quoted 6 lines
Quoted 3 lines

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

None of the people that ran into this were using meson build. For example, VPP doesn't use meson.

They will use meson when makefile will be removed :)

Anyway, no matter the build system, what we want is a way
to guide users to a proper DPDK installation.
Your solution is to add some very specific logs.
My proposal is to guide the user with a script and some specific parameters so it will fail if a required dependency is not met.

#6 lore raw
Quoted 19 lines · On Fri, 24 May 2019 19:05:20 +0200 Thomas Monjalon <[email protected]> wrote:
Quoted 12 lines
Quoted 12 lines
Quoted 10 lines

device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

Quoted 6 lines
Quoted 3 lines

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

None of the people that ran into this were using meson build. For example, VPP doesn't use meson.

Quoted 7 lines

They will use meson when makefile will be removed :)

Anyway, no matter the build system, what we want is a way
to guide users to a proper DPDK installation.
Your solution is to add some very specific logs.
My proposal is to guide the user with a script and some specific parameters so it will fail if a required dependency is not met.

Not familiar enough with meson magic syntax to make vdev_netvsc not build without MLX. But that would just be pushing the mystery failure further down the road.

Another alternative would be to drop vdev_netvsc from the default Linux build. That way users would have to enable it manually.

My preferred solution would be to just kill vdev_netvsc and go to only netvsc PMD but that is a couple releases away.

#7 lore raw

24/05/2019 19:11, Stephen Hemminger:

Quoted 30 lines
Quoted 19 lines · On Fri, 24 May 2019 19:05:20 +0200 Thomas Monjalon <[email protected]> wrote:
Quoted 12 lines
Quoted 12 lines
Quoted 10 lines

device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

Quoted 6 lines
Quoted 3 lines

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

None of the people that ran into this were using meson build. For example, VPP doesn't use meson.

Quoted 7 lines

They will use meson when makefile will be removed :)

Anyway, no matter the build system, what we want is a way
to guide users to a proper DPDK installation.
Your solution is to add some very specific logs.
My proposal is to guide the user with a script and some specific parameters so it will fail if a required dependency is not met.

Quoted 3 lines

Not familiar enough with meson magic syntax to make vdev_netvsc not build without MLX. But that would just be pushing the mystery failure further down the road.

I think you did not understand my proposal.
Let's take an example. You give this instruction to build DPDK:
buildtools/build-require.sh vdev_netvsc mlx4
If mlx4 is not built then it will fail with this message:
librte_pmd_mlx4 failed to build
So the user knows what went wrong. No mystery.

Quoted 5 lines

Another alternative would be to drop vdev_netvsc from the default Linux build. That way users would have to enable it manually.

My preferred solution would be to just kill vdev_netvsc and go to only netvsc PMD but that is a couple releases away.

#8 lore raw
Quoted 31 lines · On Fri, 24 May 2019 19:32:14 +0200 Thomas Monjalon <[email protected]> wrote:

24/05/2019 19:11, Stephen Hemminger:

Quoted 30 lines
Quoted 26 lines · On Fri, 24 May 2019 19:05:20 +0200 Thomas Monjalon <[email protected]> wrote:
Quoted 12 lines
Quoted 12 lines
Quoted 10 lines

device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

Quoted 6 lines
Quoted 3 lines

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

None of the people that ran into this were using meson build. For example, VPP doesn't use meson.

They will use meson when makefile will be removed :)

Anyway, no matter the build system, what we want is a way
to guide users to a proper DPDK installation.
Your solution is to add some very specific logs.
My proposal is to guide the user with a script and some specific parameters so it will fail if a required dependency is not met.

Quoted 7 lines
Quoted 3 lines

Not familiar enough with meson magic syntax to make vdev_netvsc not build without MLX. But that would just be pushing the mystery failure further down the road.

I think you did not understand my proposal.
Let's take an example. You give this instruction to build DPDK:
buildtools/build-require.sh vdev_netvsc mlx4

There is no build-require.sh now and introducing yet another tool is not going to help.

Quoted 10 lines

If mlx4 is not built then it will fail with this message:
librte_pmd_mlx4 failed to build
So the user knows what went wrong. No mystery.

Quoted 5 lines

Another alternative would be to drop vdev_netvsc from the default Linux build. That way users would have to enable it manually.

My preferred solution would be to just kill vdev_netvsc and go to only netvsc PMD but that is a couple releases away.

Let's just fix the meson.build for now to skip vdev_netvsc unless both mlx4 and mlx5 are present.

The Makefile can just remain untouched.
The people with problem are on older releases anyway. So any attempt to fix the Makefiles won't help those on 18.02, etc.

#9 lore raw

25/05/2019 00:06, Stephen Hemminger:

Quoted 37 lines
Quoted 35 lines · On Fri, 24 May 2019 19:32:14 +0200 Thomas Monjalon <[email protected]> wrote:

24/05/2019 19:11, Stephen Hemminger:

Quoted 33 lines
Quoted 19 lines · On Fri, 24 May 2019 19:05:20 +0200 Thomas Monjalon <[email protected]> wrote:
Quoted 12 lines
Quoted 12 lines
Quoted 10 lines

device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

Quoted 6 lines
Quoted 3 lines

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

None of the people that ran into this were using meson build. For example, VPP doesn't use meson.

Quoted 7 lines

They will use meson when makefile will be removed :)

Anyway, no matter the build system, what we want is a way
to guide users to a proper DPDK installation.
Your solution is to add some very specific logs.
My proposal is to guide the user with a script and some specific parameters so it will fail if a required dependency is not met.

Not familiar enough with meson magic syntax to make vdev_netvsc not build without MLX. But that would just be pushing the mystery failure further down the road.

Quoted 6 lines
Quoted 3 lines

I think you did not understand my proposal.
Let's take an example. You give this instruction to build DPDK:
buildtools/build-require.sh vdev_netvsc mlx4

There is no build-require.sh now and introducing yet another tool is not going to help.

I disagree.
I think we are missing this tool.

Quoted 5 lines
Quoted 5 lines

If mlx4 is not built then it will fail with this message:
librte_pmd_mlx4 failed to build
So the user knows what went wrong. No mystery.

Quoted 7 lines
Quoted 5 lines
Quoted 5 lines

Another alternative would be to drop vdev_netvsc from the default Linux build. That way users would have to enable it manually.

My preferred solution would be to just kill vdev_netvsc and go to only netvsc PMD but that is a couple releases away.

Let's just fix the meson.build for now to skip vdev_netvsc unless both mlx4 and mlx5 are present.

This is not a fix but a hack.
Why vdev_netvsc would need mlx PMDs? Because it is the Azure config?

Quoted 3 lines

The Makefile can just remain untouched.
The people with problem are on older releases anyway. So any attempt to fix the Makefiles won't help those on 18.02, etc.

#10 lore raw
Quoted 56 lines · On Mon, 27 May 2019 10:31:22 +0200 Thomas Monjalon <[email protected]> wrote:

25/05/2019 00:06, Stephen Hemminger:

Quoted 37 lines
Quoted 35 lines · On Fri, 24 May 2019 19:32:14 +0200 Thomas Monjalon <[email protected]> wrote:

24/05/2019 19:11, Stephen Hemminger:

Quoted 33 lines
Quoted 19 lines · On Fri, 24 May 2019 19:05:20 +0200 Thomas Monjalon <[email protected]> wrote:
Quoted 12 lines
Quoted 12 lines
Quoted 10 lines

device drivers on Hyper-V/Azure. If the build system supported dependencies (like Linux kbuild) this would not be necessary. Meson only does dynamic dependencies so that doesn't help.

This is a warning and not fatal only because application will still at least run, and somebody may want to run with SR-IOV with Intel NIC's on Hyper-V.

The warning is just to give users better immediate feedback rather than trying to diagnose poor performance or mystery device not found messages.

Quoted 6 lines
Quoted 3 lines

It really looks strange to me.
What you need is to fail at compilation if requested PMD is not built. I would advise to work on a script to configure meson.

None of the people that ran into this were using meson build. For example, VPP doesn't use meson.

Quoted 7 lines

They will use meson when makefile will be removed :)

Anyway, no matter the build system, what we want is a way
to guide users to a proper DPDK installation.
Your solution is to add some very specific logs.
My proposal is to guide the user with a script and some specific parameters so it will fail if a required dependency is not met.

Not familiar enough with meson magic syntax to make vdev_netvsc not build without MLX. But that would just be pushing the mystery failure further down the road.

Quoted 6 lines
Quoted 3 lines

I think you did not understand my proposal.
Let's take an example. You give this instruction to build DPDK:
buildtools/build-require.sh vdev_netvsc mlx4

There is no build-require.sh now and introducing yet another tool is not going to help.

I disagree.
I think we are missing this tool.

Quoted 8 lines
Quoted 8 lines

If mlx4 is not built then it will fail with this message:
librte_pmd_mlx4 failed to build
So the user knows what went wrong. No mystery.

Quoted 3 lines

Another alternative would be to drop vdev_netvsc from the default Linux build. That way users would have to enable it manually.

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Quoted 2 lines

My preferred solution would be to just kill vdev_netvsc and go to only netvsc PMD but that is a couple releases away.

Let's just fix the meson.build for now to skip vdev_netvsc unless both mlx4 and mlx5 are present.

This is not a fix but a hack.
Why vdev_netvsc would need mlx PMDs? Because it is the Azure config?

MLX is required because no other VF device has been tested. Intel devices barely work (wrong PCI id, and only a single queue). No one has ever tried other vendors.