[PATCH net-next v3 0/4] vsock: Add flags field in the vsock address

vsock enables communication between virtual machines and the host they are
running on. Nested VMs can be setup to use vsock channels, as the multi
transport support has been available in the mainline since the v5.5 Linux kernel
has been released.

Implicitly, if no host->guest vsock transport is loaded, all the vsock packets
are forwarded to the host. This behavior can be used to setup communication
channels between sibling VMs that are running on the same host. One example can
be the vsock channels that can be established within AWS Nitro Enclaves
(see Documentation/virt/ne_overview.rst).

To be able to explicitly mark a connection as being used for a certain use case,
add a flags field in the vsock address data structure. The value of the flags
field is taken into consideration when the vsock transport is assigned. This way
can distinguish between different use cases, such as nested VMs / local
communication and sibling VMs.

The flags field can be set in the user space application connect logic. On the
listen path, the field can be set in the kernel space logic.

Thank you.

Andra

---

Patch Series Changelog

The patch series is built on top of v5.10-rc7.

GitHub repo branch for the latest version of the patch series:

* https://github.com/andraprs/linux/tree/vsock-flag-sibling-comm-v3

v2 -> v3

* Rebase on top of v5.10-rc7.
* Add "svm_flags" as a new field, not reusing "svm_reserved1".
* Update comments to mention when the "VMADDR_FLAG_TO_HOST" flag is set in the
  connect and listen paths.
* Update bitwise check logic to not compare result to the flag value.
* v2: https://lore.kernel.org/lkml/20201204170235.84387-1-andraprs@amazon.com/

v1 -> v2

* Update the vsock flag naming to "VMADDR_FLAG_TO_HOST".
* Use bitwise operators to setup and check the vsock flag.
* Set the vsock flag on the receive path in the vsock transport assignment
  logic.
* Merge the checks for the g2h transport assignment in one "if" block.
* v1: https://lore.kernel.org/lkml/20201201152505.19445-1-andraprs@amazon.com/

---

Andra Paraschiv (4):
  vm_sockets: Add flags field in the vsock address data structure
  vm_sockets: Add VMADDR_FLAG_TO_HOST vsock flag
  af_vsock: Set VMADDR_FLAG_TO_HOST flag on the receive path
  af_vsock: Assign the vsock transport considering the vsock address
    flags

 include/uapi/linux/vm_sockets.h | 25 ++++++++++++++++++++++++-
 net/vmw_vsock/af_vsock.c        | 21 +++++++++++++++++++--
 2 files changed, 43 insertions(+), 3 deletions(-)

-- 
2.20.1 (Apple Git-117)




Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.

vsock enables communication between virtual machines and the host they
are running on. With the multi transport support (guest->host and
host->guest), nested VMs can also use vsock channels for communication.

In addition to this, by default, all the vsock packets are forwarded to
the host, if no host->guest transport is loaded. This behavior can be
implicitly used for enabling vsock communication between sibling VMs.

Add a flags field in the vsock address data structure that can be used
to explicitly mark the vsock connection as being targeted for a certain
type of communication. This way, can distinguish between different use
cases such as nested VMs and sibling VMs.

This field can be set when initializing the vsock address variable used
for the connect() call.

Changelog

v2 -> v3

* Add "svm_flags" as a new field, not reusing "svm_reserved1".

v1 -> v2

* Update the field name to "svm_flags".
* Split the current patch in 2 patches.

Signed-off-by: Andra Paraschiv <redacted>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
---
 include/uapi/linux/vm_sockets.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/vm_sockets.h b/include/uapi/linux/vm_sockets.h
index fd0ed7221645d..619f8e9d55ca4 100644
--- a/include/uapi/linux/vm_sockets.h
+++ b/include/uapi/linux/vm_sockets.h

@@ -148,10 +148,13 @@ struct sockaddr_vm {
 	unsigned short svm_reserved1;
 	unsigned int svm_port;
 	unsigned int svm_cid;
+	unsigned short svm_flags;
 	unsigned char svm_zero[sizeof(struct sockaddr) -
 			       sizeof(sa_family_t) -
 			       sizeof(unsigned short) -
-			       sizeof(unsigned int) - sizeof(unsigned int)];
+			       sizeof(unsigned int) -
+			       sizeof(unsigned int) -
+			       sizeof(unsigned short)];
 };
 
 #define IOCTL_VM_SOCKETS_GET_LOCAL_CID		_IO(7, 0xb9)

-- 
2.20.1 (Apple Git-117)




Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.

The vsock flags can be set during the connect() setup logic, when
initializing the vsock address data structure variable. Then the vsock
transport is assigned, also considering this flags field.

The vsock transport is also assigned on the (listen) receive path. The
flags field needs to be set considering the use case.

Set the value of the vsock flags of the remote address to the one
targeted for packets forwarding to the host, if the following conditions
are met:

* The source CID of the packet is higher than VMADDR_CID_HOST.
* The destination CID of the packet is higher than VMADDR_CID_HOST.

Changelog

v2 -> v3

* No changes.

v1 -> v2

* Set the vsock flag on the receive path in the vsock transport
  assignment logic.
* Use bitwise operator for the vsock flag setup.
* Use the updated "VMADDR_FLAG_TO_HOST" flag naming.

Signed-off-by: Andra Paraschiv <redacted>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
---
 net/vmw_vsock/af_vsock.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index d10916ab45267..83d035eab0b05 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c

@@ -431,6 +431,18 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	unsigned int remote_cid = vsk->remote_addr.svm_cid;
 	int ret;
 
+	/* If the packet is coming with the source and destination CIDs higher
+	 * than VMADDR_CID_HOST, then a vsock channel where all the packets are
+	 * forwarded to the host should be established. Then the host will
+	 * need to forward the packets to the guest.
+	 *
+	 * The flag is set on the (listen) receive path (psk is not NULL). On
+	 * the connect path the flag can be set by the user space application.
+	 */
+	if (psk && vsk->local_addr.svm_cid > VMADDR_CID_HOST &&
+	    vsk->remote_addr.svm_cid > VMADDR_CID_HOST)
+		vsk->remote_addr.svm_flags |= VMADDR_FLAG_TO_HOST;
+
 	switch (sk->sk_type) {
 	case SOCK_DGRAM:
 		new_transport = transport_dgram;

-- 
2.20.1 (Apple Git-117)




Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.

The vsock flags field can be set in the connect path (user space app)
and the (listen) receive path (kernel space logic).

When the vsock transport is assigned, the remote CID is used to
distinguish between types of connection.

Use the vsock flags value (in addition to the CID) from the remote
address to decide which vsock transport to assign. For the sibling VMs
use case, all the vsock packets need to be forwarded to the host, so
always assign the guest->host transport if the VMADDR_FLAG_TO_HOST flag
is set. For the other use cases, the vsock transport assignment logic is
not changed.

Changelog

v2 -> v3

* Update bitwise check logic to not compare result to the flag value.

v1 -> v2

* Use bitwise operator to check the vsock flag.
* Use the updated "VMADDR_FLAG_TO_HOST" flag naming.
* Merge the checks for the g2h transport assignment in one "if" block.

Signed-off-by: Andra Paraschiv <redacted>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
---
 net/vmw_vsock/af_vsock.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 83d035eab0b05..7c306ecf75250 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c

@@ -421,7 +421,8 @@ static void vsock_deassign_transport(struct vsock_sock *vsk)
  * The vsk->remote_addr is used to decide which transport to use:
  *  - remote CID == VMADDR_CID_LOCAL or g2h->local_cid or VMADDR_CID_HOST if
  *    g2h is not loaded, will use local transport;
- *  - remote CID <= VMADDR_CID_HOST will use guest->host transport;
+ *  - remote CID <= VMADDR_CID_HOST or h2g is not loaded or remote flags field
+ *    includes VMADDR_FLAG_TO_HOST flag value, will use guest->host transport;
  *  - remote CID > VMADDR_CID_HOST will use host->guest transport;
  */
 int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)

@@ -429,6 +430,7 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	const struct vsock_transport *new_transport;
 	struct sock *sk = sk_vsock(vsk);
 	unsigned int remote_cid = vsk->remote_addr.svm_cid;
+	unsigned short remote_flags;
 	int ret;
 
 	/* If the packet is coming with the source and destination CIDs higher

@@ -443,6 +445,8 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	    vsk->remote_addr.svm_cid > VMADDR_CID_HOST)
 		vsk->remote_addr.svm_flags |= VMADDR_FLAG_TO_HOST;
 
+	remote_flags = vsk->remote_addr.svm_flags;
+
 	switch (sk->sk_type) {
 	case SOCK_DGRAM:
 		new_transport = transport_dgram;

@@ -450,7 +454,8 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	case SOCK_STREAM:
 		if (vsock_use_local_transport(remote_cid))
 			new_transport = transport_local;
-		else if (remote_cid <= VMADDR_CID_HOST || !transport_h2g)
+		else if (remote_cid <= VMADDR_CID_HOST || !transport_h2g ||
+			 (remote_flags & VMADDR_FLAG_TO_HOST))
 			new_transport = transport_g2h;
 		else
 			new_transport = transport_h2g;

-- 
2.20.1 (Apple Git-117)




Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.

Add VMADDR_FLAG_TO_HOST vsock flag that is used to setup a vsock
connection where all the packets are forwarded to the host.

Then, using this type of vsock channel, vsock communication between
sibling VMs can be built on top of it.

Changelog

v2 -> v3

* Update comments to mention when the flag is set in the connect and
  listen paths.

v1 -> v2

* New patch in v2, it was split from the first patch in the series.
* Remove the default value for the vsock flags field.
* Update the naming for the vsock flag to "VMADDR_FLAG_TO_HOST".

Signed-off-by: Andra Paraschiv <redacted>
---
 include/uapi/linux/vm_sockets.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/include/uapi/linux/vm_sockets.h b/include/uapi/linux/vm_sockets.h
index 619f8e9d55ca4..c99ed29602345 100644
--- a/include/uapi/linux/vm_sockets.h
+++ b/include/uapi/linux/vm_sockets.h

@@ -114,6 +114,26 @@
 
 #define VMADDR_CID_HOST 2
 
+/* The current default use case for the vsock channel is the following:
+ * local vsock communication between guest and host and nested VMs setup.
+ * In addition to this, implicitly, the vsock packets are forwarded to the host
+ * if no host->guest vsock transport is set.
+ *
+ * Set this flag value in the sockaddr_vm corresponding field if the vsock
+ * packets need to be always forwarded to the host. Using this behavior,
+ * vsock communication between sibling VMs can be setup.
+ *
+ * This way can explicitly distinguish between vsock channels created for
+ * different use cases, such as nested VMs (or local communication between
+ * guest and host) and sibling VMs.
+ *
+ * The flag can be set in the connect logic in the user space application flow.
+ * In the listen logic (from kernel space) the flag is set on the remote peer
+ * address. This happens for an incoming connection when it is routed from the
+ * host and comes from the guest (local CID and remote CID > VMADDR_CID_HOST).
+ */
+#define VMADDR_FLAG_TO_HOST 0x0001
+
 /* Invalid vSockets version. */
 
 #define VM_SOCKETS_INVALID_VERSION -1U

-- 
2.20.1 (Apple Git-117)




Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.

On Fri, Dec 11, 2020 at 12:32:38PM +0200, Andra Paraschiv wrote:

vsock enables communication between virtual machines and the host they
are running on. With the multi transport support (guest->host and
host->guest), nested VMs can also use vsock channels for communication.

In addition to this, by default, all the vsock packets are forwarded to
the host, if no host->guest transport is loaded. This behavior can be
implicitly used for enabling vsock communication between sibling VMs.

Add a flags field in the vsock address data structure that can be used
to explicitly mark the vsock connection as being targeted for a certain
type of communication. This way, can distinguish between different use
cases such as nested VMs and sibling VMs.

This field can be set when initializing the vsock address variable used
for the connect() call.

Changelog

v2 -> v3

* Add "svm_flags" as a new field, not reusing "svm_reserved1".

Using the previous 'svn_zero[0]' for the new 'svn_flags' field make sure 
that if an application sets a flag and runs on an older kernel, it will 
receive an error and I think it's perfect, since that kernel is not able 
to handle the flag.

So I think is okay and I confirm my R-b tag ;-)

On Fri, Dec 11, 2020 at 12:32:39PM +0200, Andra Paraschiv wrote:

Add VMADDR_FLAG_TO_HOST vsock flag that is used to setup a vsock
connection where all the packets are forwarded to the host.

Then, using this type of vsock channel, vsock communication between
sibling VMs can be built on top of it.

Changelog

v2 -> v3

* Update comments to mention when the flag is set in the connect and
 listen paths.

v1 -> v2

* New patch in v2, it was split from the first patch in the series.
* Remove the default value for the vsock flags field.
* Update the naming for the vsock flag to "VMADDR_FLAG_TO_HOST".

Signed-off-by: Andra Paraschiv <redacted>
---
include/uapi/linux/vm_sockets.h | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)

Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>

Hi Andra,

On Fri, Dec 11, 2020 at 12:32:37PM +0200, Andra Paraschiv wrote:

vsock enables communication between virtual machines and the host they are
running on. Nested VMs can be setup to use vsock channels, as the multi
transport support has been available in the mainline since the v5.5 Linux kernel
has been released.

Implicitly, if no host->guest vsock transport is loaded, all the vsock packets
are forwarded to the host. This behavior can be used to setup communication
channels between sibling VMs that are running on the same host. One example can
be the vsock channels that can be established within AWS Nitro Enclaves
(see Documentation/virt/ne_overview.rst).

To be able to explicitly mark a connection as being used for a certain use case,
add a flags field in the vsock address data structure. The value of the flags
field is taken into consideration when the vsock transport is assigned. This way
can distinguish between different use cases, such as nested VMs / local
communication and sibling VMs.

The flags field can be set in the user space application connect logic. On the
listen path, the field can be set in the kernel space logic.

I reviewed all the patches and they are in a good shape!

Maybe the last thing to add is a flags check in the 
vsock_addr_validate(), to avoid that flags that we don't know how to 
handle are specified.
For example if in the future we add new flags that this version of the 
kernel is not able to satisfy, we should return an error to the 
application.

I mean something like this:

     diff --git a/net/vmw_vsock/vsock_addr.c b/net/vmw_vsock/vsock_addr.c
     index 909de26cb0e7..73bb1d2fa526 100644
     --- a/net/vmw_vsock/vsock_addr.c
     +++ b/net/vmw_vsock/vsock_addr.c
     @@ -22,6 +22,8 @@ EXPORT_SYMBOL_GPL(vsock_addr_init);
      
      int vsock_addr_validate(const struct sockaddr_vm *addr)
      {
     +       unsigned short svm_valid_flags = VMADDR_FLAG_TO_HOST;
     +
             if (!addr)
                     return -EFAULT;
      
     @@ -31,6 +33,9 @@ int vsock_addr_validate(const struct sockaddr_vm *addr)
             if (addr->svm_zero[0] != 0)
                     return -EINVAL;
      
     +       if (addr->svm_flags & ~svm_valid_flags)
     +               return -EINVAL;
     +
             return 0;
      }
      EXPORT_SYMBOL_GPL(vsock_addr_validate);


Thanks,
Stefano

Thank you.

Andra

---

Patch Series Changelog

The patch series is built on top of v5.10-rc7.

GitHub repo branch for the latest version of the patch series:

* https://github.com/andraprs/linux/tree/vsock-flag-sibling-comm-v3

v2 -> v3

* Rebase on top of v5.10-rc7.
* Add "svm_flags" as a new field, not reusing "svm_reserved1".
* Update comments to mention when the "VMADDR_FLAG_TO_HOST" flag is set in the
 connect and listen paths.
* Update bitwise check logic to not compare result to the flag value.
* v2: https://lore.kernel.org/lkml/20201204170235.84387-1-andraprs@amazon.com/

v1 -> v2

* Update the vsock flag naming to "VMADDR_FLAG_TO_HOST".
* Use bitwise operators to setup and check the vsock flag.
* Set the vsock flag on the receive path in the vsock transport assignment
 logic.
* Merge the checks for the g2h transport assignment in one "if" block.
* v1: https://lore.kernel.org/lkml/20201201152505.19445-1-andraprs@amazon.com/

---

Andra Paraschiv (4):
 vm_sockets: Add flags field in the vsock address data structure
 vm_sockets: Add VMADDR_FLAG_TO_HOST vsock flag
 af_vsock: Set VMADDR_FLAG_TO_HOST flag on the receive path
 af_vsock: Assign the vsock transport considering the vsock address
   flags

include/uapi/linux/vm_sockets.h | 25 ++++++++++++++++++++++++-
net/vmw_vsock/af_vsock.c        | 21 +++++++++++++++++++--
2 files changed, 43 insertions(+), 3 deletions(-)

-- 
2.20.1 (Apple Git-117)




Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.

On 11/12/2020 17:24, Stefano Garzarella wrote:

Hi Andra,

On Fri, Dec 11, 2020 at 12:32:37PM +0200, Andra Paraschiv wrote:

quoted

vsock enables communication between virtual machines and the host 
they are
running on. Nested VMs can be setup to use vsock channels, as the multi
transport support has been available in the mainline since the v5.5 
Linux kernel
has been released.

Implicitly, if no host->guest vsock transport is loaded, all the 
vsock packets
are forwarded to the host. This behavior can be used to setup 
communication
channels between sibling VMs that are running on the same host. One 
example can
be the vsock channels that can be established within AWS Nitro Enclaves
(see Documentation/virt/ne_overview.rst).

To be able to explicitly mark a connection as being used for a 
certain use case,
add a flags field in the vsock address data structure. The value of 
the flags
field is taken into consideration when the vsock transport is 
assigned. This way
can distinguish between different use cases, such as nested VMs / local
communication and sibling VMs.

The flags field can be set in the user space application connect 
logic. On the
listen path, the field can be set in the kernel space logic.

I reviewed all the patches and they are in a good shape!

Hi Stefano,

Thanks for the overall review and for the reconfirmation of the Rb for 
the vsock address data structure changes.

Maybe the last thing to add is a flags check in the
vsock_addr_validate(), to avoid that flags that we don't know how to
handle are specified.

I can add this validation as a new patch in the series, next revision.

Thanks,
Andra

For example if in the future we add new flags that this version of the
kernel is not able to satisfy, we should return an error to the
application.

I mean something like this:

    diff --git a/net/vmw_vsock/vsock_addr.c b/net/vmw_vsock/vsock_addr.c
    index 909de26cb0e7..73bb1d2fa526 100644
    --- a/net/vmw_vsock/vsock_addr.c
    +++ b/net/vmw_vsock/vsock_addr.c
    @@ -22,6 +22,8 @@ EXPORT_SYMBOL_GPL(vsock_addr_init);

     int vsock_addr_validate(const struct sockaddr_vm *addr)
     {
    +       unsigned short svm_valid_flags = VMADDR_FLAG_TO_HOST;
    +
            if (!addr)
                    return -EFAULT;

    @@ -31,6 +33,9 @@ int vsock_addr_validate(const struct sockaddr_vm 
*addr)
            if (addr->svm_zero[0] != 0)
                    return -EINVAL;

    +       if (addr->svm_flags & ~svm_valid_flags)
    +               return -EINVAL;
    +
            return 0;
     }
     EXPORT_SYMBOL_GPL(vsock_addr_validate);


Thanks,
Stefano

quoted

Thank you.

Andra

---

Patch Series Changelog

The patch series is built on top of v5.10-rc7.

GitHub repo branch for the latest version of the patch series:

* https://github.com/andraprs/linux/tree/vsock-flag-sibling-comm-v3

v2 -> v3

* Rebase on top of v5.10-rc7.
* Add "svm_flags" as a new field, not reusing "svm_reserved1".
* Update comments to mention when the "VMADDR_FLAG_TO_HOST" flag is 
set in the
 connect and listen paths.
* Update bitwise check logic to not compare result to the flag value.
* v2: 
https://lore.kernel.org/lkml/20201204170235.84387-1-andraprs@amazon.com/

v1 -> v2

* Update the vsock flag naming to "VMADDR_FLAG_TO_HOST".
* Use bitwise operators to setup and check the vsock flag.
* Set the vsock flag on the receive path in the vsock transport 
assignment
 logic.
* Merge the checks for the g2h transport assignment in one "if" block.
* v1: 
https://lore.kernel.org/lkml/20201201152505.19445-1-andraprs@amazon.com/

---

Andra Paraschiv (4):
 vm_sockets: Add flags field in the vsock address data structure
 vm_sockets: Add VMADDR_FLAG_TO_HOST vsock flag
 af_vsock: Set VMADDR_FLAG_TO_HOST flag on the receive path
 af_vsock: Assign the vsock transport considering the vsock address
   flags

include/uapi/linux/vm_sockets.h | 25 ++++++++++++++++++++++++-
net/vmw_vsock/af_vsock.c        | 21 +++++++++++++++++++--
2 files changed, 43 insertions(+), 3 deletions(-)

-- 
2.20.1 (Apple Git-117)




Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. 
Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. 
Registered in Romania. Registration number J22/2621/2005.

Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.

On Fri, 11 Dec 2020 16:24:13 +0100 Stefano Garzarella wrote:

On Fri, Dec 11, 2020 at 12:32:37PM +0200, Andra Paraschiv wrote:

quoted

vsock enables communication between virtual machines and the host they are
running on. Nested VMs can be setup to use vsock channels, as the multi
transport support has been available in the mainline since the v5.5 Linux kernel
has been released.

Implicitly, if no host->guest vsock transport is loaded, all the vsock packets
are forwarded to the host. This behavior can be used to setup communication
channels between sibling VMs that are running on the same host. One example can
be the vsock channels that can be established within AWS Nitro Enclaves
(see Documentation/virt/ne_overview.rst).

To be able to explicitly mark a connection as being used for a certain use case,
add a flags field in the vsock address data structure. The value of the flags
field is taken into consideration when the vsock transport is assigned. This way
can distinguish between different use cases, such as nested VMs / local
communication and sibling VMs.

The flags field can be set in the user space application connect logic. On the
listen path, the field can be set in the kernel space logic.
 

I reviewed all the patches and they are in a good shape!

Maybe the last thing to add is a flags check in the 
vsock_addr_validate(), to avoid that flags that we don't know how to 
handle are specified.
For example if in the future we add new flags that this version of the 
kernel is not able to satisfy, we should return an error to the 
application.

I mean something like this:

     diff --git a/net/vmw_vsock/vsock_addr.c b/net/vmw_vsock/vsock_addr.c
     index 909de26cb0e7..73bb1d2fa526 100644
     --- a/net/vmw_vsock/vsock_addr.c
     +++ b/net/vmw_vsock/vsock_addr.c
     @@ -22,6 +22,8 @@ EXPORT_SYMBOL_GPL(vsock_addr_init);
      
      int vsock_addr_validate(const struct sockaddr_vm *addr)
      {
     +       unsigned short svm_valid_flags = VMADDR_FLAG_TO_HOST;
     +
             if (!addr)
                     return -EFAULT;
      
     @@ -31,6 +33,9 @@ int vsock_addr_validate(const struct sockaddr_vm *addr)
             if (addr->svm_zero[0] != 0)
                     return -EINVAL;

Strictly speaking this check should be superseded by the check below
(AKA removed). We used to check svm_zero[0], with the new field added
this now checks svm_zero[2]. Old applications may have not initialized
svm_zero[2] (we're talking about binary compatibility here, apps built
with old headers).

     +       if (addr->svm_flags & ~svm_valid_flags)
     +               return -EINVAL;

The flags should also probably be one byte (we can define a "more
flags" flag to unlock further bytes) - otherwise on big endian the 
new flag will fall into svm_zero[1] so the v3 improvements are moot 
for big endian, right?

             return 0;
      }
      EXPORT_SYMBOL_GPL(vsock_addr_validate);

On Sat, Dec 12, 2020 at 09:16:08AM -0800, Jakub Kicinski wrote:

On Fri, 11 Dec 2020 16:24:13 +0100 Stefano Garzarella wrote:

quoted

On Fri, Dec 11, 2020 at 12:32:37PM +0200, Andra Paraschiv wrote:

quoted

vsock enables communication between virtual machines and the host they are
running on. Nested VMs can be setup to use vsock channels, as the multi
transport support has been available in the mainline since the v5.5 Linux kernel
has been released.

Implicitly, if no host->guest vsock transport is loaded, all the vsock packets
are forwarded to the host. This behavior can be used to setup communication
channels between sibling VMs that are running on the same host. One example can
be the vsock channels that can be established within AWS Nitro Enclaves
(see Documentation/virt/ne_overview.rst).

To be able to explicitly mark a connection as being used for a certain use case,
add a flags field in the vsock address data structure. The value of the flags
field is taken into consideration when the vsock transport is assigned. This way
can distinguish between different use cases, such as nested VMs / local
communication and sibling VMs.

The flags field can be set in the user space application connect logic. On the
listen path, the field can be set in the kernel space logic.

I reviewed all the patches and they are in a good shape!

Maybe the last thing to add is a flags check in the
vsock_addr_validate(), to avoid that flags that we don't know how to
handle are specified.
For example if in the future we add new flags that this version of the
kernel is not able to satisfy, we should return an error to the
application.

I mean something like this:

     diff --git a/net/vmw_vsock/vsock_addr.c b/net/vmw_vsock/vsock_addr.c
     index 909de26cb0e7..73bb1d2fa526 100644
     --- a/net/vmw_vsock/vsock_addr.c
     +++ b/net/vmw_vsock/vsock_addr.c
     @@ -22,6 +22,8 @@ EXPORT_SYMBOL_GPL(vsock_addr_init);

      int vsock_addr_validate(const struct sockaddr_vm *addr)
      {
     +       unsigned short svm_valid_flags = VMADDR_FLAG_TO_HOST;
     +
             if (!addr)
                     return -EFAULT;

     @@ -31,6 +33,9 @@ int vsock_addr_validate(const struct sockaddr_vm *addr)
             if (addr->svm_zero[0] != 0)
                     return -EINVAL;

Strictly speaking this check should be superseded by the check below
(AKA removed). We used to check svm_zero[0], with the new field added
this now checks svm_zero[2]. Old applications may have not initialized
svm_zero[2] (we're talking about binary compatibility here, apps built
with old headers).

quoted

     +       if (addr->svm_flags & ~svm_valid_flags)
     +               return -EINVAL;

The flags should also probably be one byte (we can define a "more
flags" flag to unlock further bytes) - otherwise on big endian the
new flag will fall into svm_zero[1] so the v3 improvements are moot
for big endian, right?

Right, I assumed the entire svm_zero[] was zeroed out, but we can't be 
sure.

So, I agree to change the svm_flags to 1 byte (__u8), and remove the 
superseded check that you pointed out.
With these changes we should be fully binary compatibility.

Thanks,
Stefano

quoted

             return 0;
      }
      EXPORT_SYMBOL_GPL(vsock_addr_validate);

On 14/12/2020 10:13, Stefano Garzarella wrote:

On Sat, Dec 12, 2020 at 09:16:08AM -0800, Jakub Kicinski wrote:

quoted

On Fri, 11 Dec 2020 16:24:13 +0100 Stefano Garzarella wrote:

quoted

On Fri, Dec 11, 2020 at 12:32:37PM +0200, Andra Paraschiv wrote:

quoted

vsock enables communication between virtual machines and the host 

they are

quoted

running on. Nested VMs can be setup to use vsock channels, as the 

multi

quoted

transport support has been available in the mainline since the v5.5 

Linux kernel

quoted

has been released.

Implicitly, if no host->guest vsock transport is loaded, all the 

vsock packets

quoted

are forwarded to the host. This behavior can be used to setup 

communication

quoted

channels between sibling VMs that are running on the same host. One 

example can

quoted

be the vsock channels that can be established within AWS Nitro 

Enclaves

quoted

(see Documentation/virt/ne_overview.rst).

To be able to explicitly mark a connection as being used for a 

certain use case,

quoted

add a flags field in the vsock address data structure. The value of 

the flags

quoted

field is taken into consideration when the vsock transport is 

assigned. This way

quoted

can distinguish between different use cases, such as nested VMs / 

local

quoted

communication and sibling VMs.

The flags field can be set in the user space application connect 

logic. On the

quoted

listen path, the field can be set in the kernel space logic.

I reviewed all the patches and they are in a good shape!

Maybe the last thing to add is a flags check in the
vsock_addr_validate(), to avoid that flags that we don't know how to
handle are specified.
For example if in the future we add new flags that this version of the
kernel is not able to satisfy, we should return an error to the
application.

I mean something like this:

     diff --git a/net/vmw_vsock/vsock_addr.c 
b/net/vmw_vsock/vsock_addr.c
     index 909de26cb0e7..73bb1d2fa526 100644
     --- a/net/vmw_vsock/vsock_addr.c
     +++ b/net/vmw_vsock/vsock_addr.c
     @@ -22,6 +22,8 @@ EXPORT_SYMBOL_GPL(vsock_addr_init);

      int vsock_addr_validate(const struct sockaddr_vm *addr)
      {
     +       unsigned short svm_valid_flags = VMADDR_FLAG_TO_HOST;
     +
             if (!addr)
                     return -EFAULT;

     @@ -31,6 +33,9 @@ int vsock_addr_validate(const struct 
sockaddr_vm *addr)
             if (addr->svm_zero[0] != 0)
                     return -EINVAL;

Strictly speaking this check should be superseded by the check below
(AKA removed). We used to check svm_zero[0], with the new field added
this now checks svm_zero[2]. Old applications may have not initialized
svm_zero[2] (we're talking about binary compatibility here, apps built
with old headers).

quoted

     +       if (addr->svm_flags & ~svm_valid_flags)
     +               return -EINVAL;

The flags should also probably be one byte (we can define a "more
flags" flag to unlock further bytes) - otherwise on big endian the
new flag will fall into svm_zero[1] so the v3 improvements are moot
for big endian, right?

Right, I assumed the entire svm_zero[] was zeroed out, but we can't be
sure.

So, I agree to change the svm_flags to 1 byte (__u8), and remove the
superseded check that you pointed out.
With these changes we should be fully binary compatibility.

Here we go, sent out v4:

https://lore.kernel.org/lkml/20201214161122.37717-1-andraprs@amazon.com/

Thank you both.

Andra

quoted

quoted

             return 0;
      }
      EXPORT_SYMBOL_GPL(vsock_addr_validate);

Amazon Development Center (Romania) S.R.L. registered office: 27A Sf. Lazar Street, UBC5, floor 2, Iasi, Iasi County, 700045, Romania. Registered in Romania. Registration number J22/2621/2005.