* [PATCH net-next v2 01/10] doc/netlink: Add a schema for netlink-raw families
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-16 8:41 ` Simon Horman
2023-08-15 19:42 ` [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions Donald Hunter
` (8 subsequent siblings)
9 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
This schema is largely a copy of the genetlink-legacy schema with the
following additions:
- a top-level protonum property, e.g. 0 (for NETLINK_ROUTE)
- add netlink-raw to the list of protocols supported by the schema
- add a value property to mcast-group definitions
This schema is very similar to genetlink-legacy and I considered
making the changes there and symlinking to it. On balance I felt that
might be problematic for accurate schema validation.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
Documentation/netlink/netlink-raw.yaml | 414 +++++++++++++++++++++++++
1 file changed, 414 insertions(+)
create mode 100644 Documentation/netlink/netlink-raw.yaml
diff --git a/Documentation/netlink/netlink-raw.yaml b/Documentation/netlink/netlink-raw.yaml
new file mode 100644
index 000000000000..a5ec6f3e41cc
--- /dev/null
+++ b/Documentation/netlink/netlink-raw.yaml
@@ -0,0 +1,414 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+%YAML 1.2
+---
+$id: http://kernel.org/schemas/netlink/genetlink-legacy.yaml#
+$schema: https://json-schema.org/draft-07/schema
+
+# Common defines
+$defs:
+ uint:
+ type: integer
+ minimum: 0
+ len-or-define:
+ type: [ string, integer ]
+ pattern: ^[0-9A-Za-z_]+( - 1)?$
+ minimum: 0
+
+# Schema for specs
+title: Protocol
+description: Specification of a genetlink protocol
+type: object
+required: [ name, doc, attribute-sets, operations ]
+additionalProperties: False
+properties:
+ name:
+ description: Name of the genetlink family.
+ type: string
+ doc:
+ type: string
+ version:
+ description: Generic Netlink family version. Default is 1.
+ type: integer
+ minimum: 1
+ protocol:
+ description: Schema compatibility level. Default is "genetlink".
+ enum: [ genetlink, genetlink-c, genetlink-legacy, netlink-raw ] # Trim
+ # Start netlink-raw
+ protonum:
+ description: Protocol number to use for netlink-raw
+ type: integer
+ # End netlink-raw
+ uapi-header:
+ description: Path to the uAPI header, default is linux/${family-name}.h
+ type: string
+ # Start genetlink-c
+ c-family-name:
+ description: Name of the define for the family name.
+ type: string
+ c-version-name:
+ description: Name of the define for the verion of the family.
+ type: string
+ max-by-define:
+ description: Makes the number of attributes and commands be specified by a define, not an enum value.
+ type: boolean
+ # End genetlink-c
+ # Start genetlink-legacy
+ kernel-policy:
+ description: |
+ Defines if the input policy in the kernel is global, per-operation, or split per operation type.
+ Default is split.
+ enum: [ split, per-op, global ]
+ # End genetlink-legacy
+
+ definitions:
+ description: List of type and constant definitions (enums, flags, defines).
+ type: array
+ items:
+ type: object
+ required: [ type, name ]
+ additionalProperties: False
+ properties:
+ name:
+ type: string
+ header:
+ description: For C-compatible languages, header which already defines this value.
+ type: string
+ type:
+ enum: [ const, enum, flags, struct ] # Trim
+ doc:
+ type: string
+ # For const
+ value:
+ description: For const - the value.
+ type: [ string, integer ]
+ # For enum and flags
+ value-start:
+ description: For enum or flags the literal initializer for the first value.
+ type: [ string, integer ]
+ entries:
+ description: For enum or flags array of values.
+ type: array
+ items:
+ oneOf:
+ - type: string
+ - type: object
+ required: [ name ]
+ additionalProperties: False
+ properties:
+ name:
+ type: string
+ value:
+ type: integer
+ doc:
+ type: string
+ render-max:
+ description: Render the max members for this enum.
+ type: boolean
+ # Start genetlink-c
+ enum-name:
+ description: Name for enum, if empty no name will be used.
+ type: [ string, "null" ]
+ name-prefix:
+ description: For enum the prefix of the values, optional.
+ type: string
+ # End genetlink-c
+ # Start genetlink-legacy
+ members:
+ description: List of struct members. Only scalars and strings members allowed.
+ type: array
+ items:
+ type: object
+ required: [ name, type ]
+ additionalProperties: False
+ properties:
+ name:
+ type: string
+ type:
+ description: The netlink attribute type
+ enum: [ u8, u16, u32, u64, s8, s16, s32, s64, string, binary ]
+ len:
+ $ref: '#/$defs/len-or-define'
+ byte-order:
+ enum: [ little-endian, big-endian ]
+ doc:
+ description: Documentation for the struct member attribute.
+ type: string
+ enum:
+ description: Name of the enum type used for the attribute.
+ type: string
+ enum-as-flags:
+ description: |
+ Treat the enum as flags. In most cases enum is either used as flags or as values.
+ Sometimes, however, both forms are necessary, in which case header contains the enum
+ form while specific attributes may request to convert the values into a bitfield.
+ type: boolean
+ display-hint: &display-hint
+ description: |
+ Optional format indicator that is intended only for choosing
+ the right formatting mechanism when displaying values of this
+ type.
+ enum: [ hex, mac, fddi, ipv4, ipv6, uuid ]
+ # End genetlink-legacy
+
+ attribute-sets:
+ description: Definition of attribute spaces for this family.
+ type: array
+ items:
+ description: Definition of a single attribute space.
+ type: object
+ required: [ name, attributes ]
+ additionalProperties: False
+ properties:
+ name:
+ description: |
+ Name used when referring to this space in other definitions, not used outside of the spec.
+ type: string
+ name-prefix:
+ description: |
+ Prefix for the C enum name of the attributes. Default family[name]-set[name]-a-
+ type: string
+ enum-name:
+ description: Name for the enum type of the attribute.
+ type: string
+ doc:
+ description: Documentation of the space.
+ type: string
+ subset-of:
+ description: |
+ Name of another space which this is a logical part of. Sub-spaces can be used to define
+ a limited group of attributes which are used in a nest.
+ type: string
+ # Start genetlink-c
+ attr-cnt-name:
+ description: The explicit name for constant holding the count of attributes (last attr + 1).
+ type: string
+ attr-max-name:
+ description: The explicit name for last member of attribute enum.
+ type: string
+ # End genetlink-c
+ attributes:
+ description: List of attributes in the space.
+ type: array
+ items:
+ type: object
+ required: [ name, type ]
+ additionalProperties: False
+ properties:
+ name:
+ type: string
+ type: &attr-type
+ description: The netlink attribute type
+ enum: [ unused, pad, flag, binary, u8, u16, u32, u64, s32, s64,
+ string, nest, array-nest, nest-type-value ]
+ doc:
+ description: Documentation of the attribute.
+ type: string
+ value:
+ description: Value for the enum item representing this attribute in the uAPI.
+ $ref: '#/$defs/uint'
+ type-value:
+ description: Name of the value extracted from the type of a nest-type-value attribute.
+ type: array
+ items:
+ type: string
+ byte-order:
+ enum: [ little-endian, big-endian ]
+ multi-attr:
+ type: boolean
+ nested-attributes:
+ description: Name of the space (sub-space) used inside the attribute.
+ type: string
+ enum:
+ description: Name of the enum type used for the attribute.
+ type: string
+ enum-as-flags:
+ description: |
+ Treat the enum as flags. In most cases enum is either used as flags or as values.
+ Sometimes, however, both forms are necessary, in which case header contains the enum
+ form while specific attributes may request to convert the values into a bitfield.
+ type: boolean
+ checks:
+ description: Kernel input validation.
+ type: object
+ additionalProperties: False
+ properties:
+ flags-mask:
+ description: Name of the flags constant on which to base mask (unsigned scalar types only).
+ type: string
+ min:
+ description: Min value for an integer attribute.
+ type: integer
+ min-len:
+ description: Min length for a binary attribute.
+ $ref: '#/$defs/len-or-define'
+ max-len:
+ description: Max length for a string or a binary attribute.
+ $ref: '#/$defs/len-or-define'
+ sub-type: *attr-type
+ display-hint: *display-hint
+ # Start genetlink-c
+ name-prefix:
+ type: string
+ # End genetlink-c
+ # Start genetlink-legacy
+ struct:
+ description: Name of the struct type used for the attribute.
+ type: string
+ # End genetlink-legacy
+
+ # Make sure name-prefix does not appear in subsets (subsets inherit naming)
+ dependencies:
+ name-prefix:
+ not:
+ required: [ subset-of ]
+ subset-of:
+ not:
+ required: [ name-prefix ]
+
+ operations:
+ description: Operations supported by the protocol.
+ type: object
+ required: [ list ]
+ additionalProperties: False
+ properties:
+ enum-model:
+ description: |
+ The model of assigning values to the operations.
+ "unified" is the recommended model where all message types belong
+ to a single enum.
+ "directional" has the messages sent to the kernel and from the kernel
+ enumerated separately.
+ enum: [ unified, directional ] # Trim
+ name-prefix:
+ description: |
+ Prefix for the C enum name of the command. The name is formed by concatenating
+ the prefix with the upper case name of the command, with dashes replaced by underscores.
+ type: string
+ enum-name:
+ description: Name for the enum type with commands.
+ type: string
+ async-prefix:
+ description: Same as name-prefix but used to render notifications and events to separate enum.
+ type: string
+ async-enum:
+ description: Name for the enum type with notifications/events.
+ type: string
+ # Start genetlink-legacy
+ fixed-header: &fixed-header
+ description: |
+ Name of the structure defining the optional fixed-length protocol
+ header. This header is placed in a message after the netlink and
+ genetlink headers and before any attributes.
+ type: string
+ # End genetlink-legacy
+ list:
+ description: List of commands
+ type: array
+ items:
+ type: object
+ additionalProperties: False
+ required: [ name, doc ]
+ properties:
+ name:
+ description: Name of the operation, also defining its C enum value in uAPI.
+ type: string
+ doc:
+ description: Documentation for the command.
+ type: string
+ value:
+ description: Value for the enum in the uAPI.
+ $ref: '#/$defs/uint'
+ attribute-set:
+ description: |
+ Attribute space from which attributes directly in the requests and replies
+ to this command are defined.
+ type: string
+ flags: &cmd_flags
+ description: Command flags.
+ type: array
+ items:
+ enum: [ admin-perm ]
+ dont-validate:
+ description: Kernel attribute validation flags.
+ type: array
+ items:
+ enum: [ strict, dump ]
+ # Start genetlink-legacy
+ fixed-header: *fixed-header
+ # End genetlink-legacy
+ do: &subop-type
+ description: Main command handler.
+ type: object
+ additionalProperties: False
+ properties:
+ request: &subop-attr-list
+ description: Definition of the request message for a given command.
+ type: object
+ additionalProperties: False
+ properties:
+ attributes:
+ description: |
+ Names of attributes from the attribute-set (not full attribute
+ definitions, just names).
+ type: array
+ items:
+ type: string
+ # Start genetlink-legacy
+ value:
+ description: |
+ ID of this message if value for request and response differ,
+ i.e. requests and responses have different message enums.
+ $ref: '#/$defs/uint'
+ # End genetlink-legacy
+ reply: *subop-attr-list
+ pre:
+ description: Hook for a function to run before the main callback (pre_doit or start).
+ type: string
+ post:
+ description: Hook for a function to run after the main callback (post_doit or done).
+ type: string
+ dump: *subop-type
+ notify:
+ description: Name of the command sharing the reply type with this notification.
+ type: string
+ event:
+ type: object
+ additionalProperties: False
+ properties:
+ attributes:
+ description: Explicit list of the attributes for the notification.
+ type: array
+ items:
+ type: string
+ mcgrp:
+ description: Name of the multicast group generating given notification.
+ type: string
+ mcast-groups:
+ description: List of multicast groups.
+ type: object
+ required: [ list ]
+ additionalProperties: False
+ properties:
+ list:
+ description: List of groups.
+ type: array
+ items:
+ type: object
+ required: [ name ]
+ additionalProperties: False
+ properties:
+ name:
+ description: |
+ The name for the group, used to form the define and the value of the define.
+ type: string
+ # Start genetlink-c
+ c-define-name:
+ description: Override for the name of the define in C uAPI.
+ type: string
+ # End genetlink-c
+ flags: *cmd_flags
+ # Start netlink-raw
+ value:
+ description: Value of the netlink multicast group in the uAPI.
+ type: integer
+ # End netlink-raw
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 01/10] doc/netlink: Add a schema for netlink-raw families
2023-08-15 19:42 ` [PATCH net-next v2 01/10] doc/netlink: Add a schema " Donald Hunter
@ 2023-08-16 8:41 ` Simon Horman
2023-08-16 13:19 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Simon Horman @ 2023-08-16 8:41 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Tue, Aug 15, 2023 at 08:42:45PM +0100, Donald Hunter wrote:
> This schema is largely a copy of the genetlink-legacy schema with the
> following additions:
>
> - a top-level protonum property, e.g. 0 (for NETLINK_ROUTE)
> - add netlink-raw to the list of protocols supported by the schema
> - add a value property to mcast-group definitions
>
> This schema is very similar to genetlink-legacy and I considered
> making the changes there and symlinking to it. On balance I felt that
> might be problematic for accurate schema validation.
>
> Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
> ---
> Documentation/netlink/netlink-raw.yaml | 414 +++++++++++++++++++++++++
> 1 file changed, 414 insertions(+)
> create mode 100644 Documentation/netlink/netlink-raw.yaml
>
> diff --git a/Documentation/netlink/netlink-raw.yaml b/Documentation/netlink/netlink-raw.yaml
> new file mode 100644
> index 000000000000..a5ec6f3e41cc
> --- /dev/null
> +++ b/Documentation/netlink/netlink-raw.yaml
> @@ -0,0 +1,414 @@
> +# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
> +%YAML 1.2
> +---
> +$id: http://kernel.org/schemas/netlink/genetlink-legacy.yaml#
> +$schema: https://json-schema.org/draft-07/schema
> +
> +# Common defines
> +$defs:
> + uint:
> + type: integer
> + minimum: 0
> + len-or-define:
> + type: [ string, integer ]
> + pattern: ^[0-9A-Za-z_]+( - 1)?$
> + minimum: 0
> +
> +# Schema for specs
> +title: Protocol
> +description: Specification of a genetlink protocol
> +type: object
> +required: [ name, doc, attribute-sets, operations ]
> +additionalProperties: False
> +properties:
> + name:
> + description: Name of the genetlink family.
> + type: string
> + doc:
> + type: string
> + version:
> + description: Generic Netlink family version. Default is 1.
> + type: integer
> + minimum: 1
> + protocol:
> + description: Schema compatibility level. Default is "genetlink".
> + enum: [ genetlink, genetlink-c, genetlink-legacy, netlink-raw ] # Trim
> + # Start netlink-raw
> + protonum:
> + description: Protocol number to use for netlink-raw
> + type: integer
> + # End netlink-raw
> + uapi-header:
> + description: Path to the uAPI header, default is linux/${family-name}.h
> + type: string
> + # Start genetlink-c
> + c-family-name:
> + description: Name of the define for the family name.
> + type: string
> + c-version-name:
> + description: Name of the define for the verion of the family.
Hi Donald,
a minor nit from my side: verion -> version
> + type: string
...
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 01/10] doc/netlink: Add a schema for netlink-raw families
2023-08-16 8:41 ` Simon Horman
@ 2023-08-16 13:19 ` Donald Hunter
0 siblings, 0 replies; 29+ messages in thread
From: Donald Hunter @ 2023-08-16 13:19 UTC (permalink / raw)
To: Simon Horman
Cc: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
Simon Horman <horms@kernel.org> writes:
> On Tue, Aug 15, 2023 at 08:42:45PM +0100, Donald Hunter wrote:
>> + c-family-name:
>> + description: Name of the define for the family name.
>> + type: string
>> + c-version-name:
>> + description: Name of the define for the verion of the family.
>
> Hi Donald,
>
> a minor nit from my side: verion -> version
Good catch. This typo was copied over from genetlink-legacy.yaml so I'll
fix it there and in genetlink-c.yaml as well.
Thanks!
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
2023-08-15 19:42 ` [PATCH net-next v2 01/10] doc/netlink: Add a schema " Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-16 2:49 ` Jakub Kicinski
2023-08-15 19:42 ` [PATCH net-next v2 03/10] doc/netlink: Document the netlink-raw " Donald Hunter
` (7 subsequent siblings)
9 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Add description of genetlink-legacy specific attributes to the ynl spec
documentation.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
Documentation/userspace-api/netlink/specs.rst | 47 +++++++++++++++++++
1 file changed, 47 insertions(+)
diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst
index 2e4acde890b7..dde70f9674d4 100644
--- a/Documentation/userspace-api/netlink/specs.rst
+++ b/Documentation/userspace-api/netlink/specs.rst
@@ -443,3 +443,50 @@ nest
Attribute containing other (nested) attributes.
``nested-attributes`` specifies which attribute set is used inside.
+
+genetlink-legacy
+================
+
+The genetlink-legacy schema extends the genetlink schema with some additional
+properties that are needed to support legacy genetlink families.
+
+Globals
+-------
+
+ - ``kernel-policy`` - Specify whether the kernel input policy is ``global``,
+ ``per-op`` or ``split``.
+
+Struct definitions
+------------------
+
+There is a new type of definition called ``struct`` which is used for declaring
+the C struct format of fixed headers and binary attributes.
+
+members
+~~~~~~~
+
+ - ``name`` - The attribute name of the struct member
+ - ``type`` - One of the scalar types ``u8``, ``u16``, ``u32``, ``u64``, ``s8``,
+ ``s16``, ``s32``, ``s64``, ``string`` or ``binary``.
+ - ``byte-order`` - ``big-endian`` or ``little-endian``
+ - ``doc``, ``enum``, ``enum-as-flags``, ``display-hint`` - Same as for
+ attribute definitions.
+
+Attributes
+----------
+
+The genetlink-legacy families can use binary attributes that contain C struct
+data. This is specified using a ``struct`` property containing the name of the
+struct definition.
+
+ - ``struct`` - Name of the struct definition to be used for the attribute.
+
+Operations
+----------
+
+The genetlink-legacy families can use a binary fixed header that contains C
+struct data.
+
+ - ``fixed-header`` - name of the struct definition to be used for the fixed
+ header data. This can be specified as a default for all operations and on a
+ per-operation basis.
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions
2023-08-15 19:42 ` [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions Donald Hunter
@ 2023-08-16 2:49 ` Jakub Kicinski
2023-08-16 8:25 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-16 2:49 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Tue, 15 Aug 2023 20:42:46 +0100 Donald Hunter wrote:
> Add description of genetlink-legacy specific attributes to the ynl spec
> documentation.
>
> Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
> ---
> Documentation/userspace-api/netlink/specs.rst | 47 +++++++++++++++++++
> 1 file changed, 47 insertions(+)
Should we merge this with genetlink-legacy.rst?
> diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst
> index 2e4acde890b7..dde70f9674d4 100644
> --- a/Documentation/userspace-api/netlink/specs.rst
> +++ b/Documentation/userspace-api/netlink/specs.rst
> @@ -443,3 +443,50 @@ nest
>
> Attribute containing other (nested) attributes.
> ``nested-attributes`` specifies which attribute set is used inside.
> +
> +genetlink-legacy
> +================
> +
> +The genetlink-legacy schema extends the genetlink schema with some additional
> +properties that are needed to support legacy genetlink families.
> +
> +Globals
> +-------
> +
> + - ``kernel-policy`` - Specify whether the kernel input policy is ``global``,
> + ``per-op`` or ``split``.
Maybe a few more words:
Specify whether the kernel input policy is ``global`` i.e. the same for
all operation of the family, defined for each operation individually
(``per-op``), or separately for each operation and operation type
(do vs dump) - ``split``.
> + ``per-op`` or ``split``.
> +Struct definitions
> +------------------
> +
> +There is a new type of definition called ``struct`` which is used for declaring
> +the C struct format of fixed headers and binary attributes.
> +
> +members
> +~~~~~~~
> +
> + - ``name`` - The attribute name of the struct member
> + - ``type`` - One of the scalar types ``u8``, ``u16``, ``u32``, ``u64``, ``s8``,
> + ``s16``, ``s32``, ``s64``, ``string`` or ``binary``.
> + - ``byte-order`` - ``big-endian`` or ``little-endian``
> + - ``doc``, ``enum``, ``enum-as-flags``, ``display-hint`` - Same as for
> + attribute definitions.
Hm, genetlink-legacy.rst has this:
https://docs.kernel.org/next/userspace-api/netlink/genetlink-legacy.html#structures
But the larger section is called "Other quirks (todo)"
I guess you have tackled most of the items in this section
so we shouldn't call it "todo" ?
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions
2023-08-16 2:49 ` Jakub Kicinski
@ 2023-08-16 8:25 ` Donald Hunter
2023-08-16 13:16 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-16 8:25 UTC (permalink / raw)
To: Jakub Kicinski
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
Jakub Kicinski <kuba@kernel.org> writes:
> On Tue, 15 Aug 2023 20:42:46 +0100 Donald Hunter wrote:
>> Add description of genetlink-legacy specific attributes to the ynl spec
>> documentation.
>>
>> Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
>> ---
>> Documentation/userspace-api/netlink/specs.rst | 47 +++++++++++++++++++
>> 1 file changed, 47 insertions(+)
>
> Should we merge this with genetlink-legacy.rst?
Sure, I can do that.
>
>> diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst
>> index 2e4acde890b7..dde70f9674d4 100644
>> --- a/Documentation/userspace-api/netlink/specs.rst
>> +++ b/Documentation/userspace-api/netlink/specs.rst
>> @@ -443,3 +443,50 @@ nest
>>
>> Attribute containing other (nested) attributes.
>> ``nested-attributes`` specifies which attribute set is used inside.
>> +
>> +genetlink-legacy
>> +================
>> +
>> +The genetlink-legacy schema extends the genetlink schema with some additional
>> +properties that are needed to support legacy genetlink families.
>> +
>> +Globals
>> +-------
>> +
>> + - ``kernel-policy`` - Specify whether the kernel input policy is ``global``,
>> + ``per-op`` or ``split``.
>
> Maybe a few more words:
>
> Specify whether the kernel input policy is ``global`` i.e. the same for
> all operation of the family, defined for each operation individually
> (``per-op``), or separately for each operation and operation type
> (do vs dump) - ``split``.
Ack. As an aside, what do we mean by "kernel input policy"?
>> + ``per-op`` or ``split``.
>
>> +Struct definitions
>> +------------------
>> +
>> +There is a new type of definition called ``struct`` which is used for declaring
>> +the C struct format of fixed headers and binary attributes.
>> +
>> +members
>> +~~~~~~~
>> +
>> + - ``name`` - The attribute name of the struct member
>> + - ``type`` - One of the scalar types ``u8``, ``u16``, ``u32``, ``u64``, ``s8``,
>> + ``s16``, ``s32``, ``s64``, ``string`` or ``binary``.
>> + - ``byte-order`` - ``big-endian`` or ``little-endian``
>> + - ``doc``, ``enum``, ``enum-as-flags``, ``display-hint`` - Same as for
>> + attribute definitions.
>
> Hm, genetlink-legacy.rst has this:
>
> https://docs.kernel.org/next/userspace-api/netlink/genetlink-legacy.html#structures
>
> But the larger section is called "Other quirks (todo)"
> I guess you have tackled most of the items in this section
> so we shouldn't call it "todo" ?
I'll clean up genetlink-legacy.rst when I merge these changes.
Thanks for the review!
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions
2023-08-16 8:25 ` Donald Hunter
@ 2023-08-16 13:16 ` Donald Hunter
2023-08-16 15:00 ` Jakub Kicinski
0 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-16 13:16 UTC (permalink / raw)
To: Jakub Kicinski
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
Donald Hunter <donald.hunter@gmail.com> writes:
> Jakub Kicinski <kuba@kernel.org> writes:
>
>> On Tue, 15 Aug 2023 20:42:46 +0100 Donald Hunter wrote:
>>> +
>>> +Globals
>>> +-------
>>> +
>>> + - ``kernel-policy`` - Specify whether the kernel input policy is ``global``,
>>> + ``per-op`` or ``split``.
>>
>> Maybe a few more words:
>>
>> Specify whether the kernel input policy is ``global`` i.e. the same for
>> all operation of the family, defined for each operation individually
>> (``per-op``), or separately for each operation and operation type
>> (do vs dump) - ``split``.
>
> Ack. As an aside, what do we mean by "kernel input policy"?
So I've just spotted that kernel-policy is already documented in
core-api/netlink.rst and I guess I shouldn't be documenting it in the
userspace-api at all? I could add a reference to the core-api docs so
that it's easier to find the kernel side docs when reading the
userspace-api?
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions
2023-08-16 13:16 ` Donald Hunter
@ 2023-08-16 15:00 ` Jakub Kicinski
0 siblings, 0 replies; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-16 15:00 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Wed, 16 Aug 2023 14:16:33 +0100 Donald Hunter wrote:
> > Ack. As an aside, what do we mean by "kernel input policy"?
>
> So I've just spotted that kernel-policy is already documented in
> core-api/netlink.rst and I guess I shouldn't be documenting it in the
> userspace-api at all? I could add a reference to the core-api docs so
> that it's easier to find the kernel side docs when reading the
> userspace-api?
Ah, yes, reference sounds good.
But we should also add split to the doc, I only see ``global`` and
``per-op`` described?
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next v2 03/10] doc/netlink: Document the netlink-raw schema extensions
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
2023-08-15 19:42 ` [PATCH net-next v2 01/10] doc/netlink: Add a schema " Donald Hunter
2023-08-15 19:42 ` [PATCH net-next v2 02/10] doc/netlink: Document the genetlink-legacy schema extensions Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-16 2:52 ` Jakub Kicinski
2023-08-15 19:42 ` [PATCH net-next v2 04/10] tools/ynl: Add mcast-group schema parsing to ynl Donald Hunter
` (6 subsequent siblings)
9 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Add description of netlink-raw specific attributes to the ynl spec
documentation and refer to the classic netlink documentation.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
Documentation/userspace-api/netlink/intro.rst | 2 ++
Documentation/userspace-api/netlink/specs.rst | 17 +++++++++++++++++
2 files changed, 19 insertions(+)
diff --git a/Documentation/userspace-api/netlink/intro.rst b/Documentation/userspace-api/netlink/intro.rst
index 0955e9f203d3..3ea70ad53c58 100644
--- a/Documentation/userspace-api/netlink/intro.rst
+++ b/Documentation/userspace-api/netlink/intro.rst
@@ -528,6 +528,8 @@ families may, however, require a larger buffer. 32kB buffer is recommended
for most efficient handling of dumps (larger buffer fits more dumped
objects and therefore fewer recvmsg() calls are needed).
+.. _classic_netlink:
+
Classic Netlink
===============
diff --git a/Documentation/userspace-api/netlink/specs.rst b/Documentation/userspace-api/netlink/specs.rst
index dde70f9674d4..98978a014488 100644
--- a/Documentation/userspace-api/netlink/specs.rst
+++ b/Documentation/userspace-api/netlink/specs.rst
@@ -490,3 +490,20 @@ struct data.
- ``fixed-header`` - name of the struct definition to be used for the fixed
header data. This can be specified as a default for all operations and on a
per-operation basis.
+
+netlink-raw
+===========
+
+The netlink-raw schema extends the genetlink-legacy schema with properties that
+are needed to specify the protocol numbers and multicast IDs used by raw netlink
+families. See :ref:`classic_netlink` for more information.
+
+Globals
+-------
+
+ - ``protonum`` - Protocol number to use when opening a netlink socket.
+
+Multicast group properties
+--------------------------
+
+ - ``value`` - ID to use for multicast group registration.
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 03/10] doc/netlink: Document the netlink-raw schema extensions
2023-08-15 19:42 ` [PATCH net-next v2 03/10] doc/netlink: Document the netlink-raw " Donald Hunter
@ 2023-08-16 2:52 ` Jakub Kicinski
2023-08-16 8:32 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-16 2:52 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Tue, 15 Aug 2023 20:42:47 +0100 Donald Hunter wrote:
> Add description of netlink-raw specific attributes to the ynl spec
> documentation and refer to the classic netlink documentation.
I wonder if we should make this a separate doc, similarly to
genetlink-legacy. Keep the specs.rst focused on newer stuff?
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next v2 03/10] doc/netlink: Document the netlink-raw schema extensions
2023-08-16 2:52 ` Jakub Kicinski
@ 2023-08-16 8:32 ` Donald Hunter
0 siblings, 0 replies; 29+ messages in thread
From: Donald Hunter @ 2023-08-16 8:32 UTC (permalink / raw)
To: Jakub Kicinski
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
Jakub Kicinski <kuba@kernel.org> writes:
> On Tue, 15 Aug 2023 20:42:47 +0100 Donald Hunter wrote:
>> Add description of netlink-raw specific attributes to the ynl spec
>> documentation and refer to the classic netlink documentation.
>
> I wonder if we should make this a separate doc, similarly to
> genetlink-legacy. Keep the specs.rst focused on newer stuff?
Sure. I could also include examples, like genetlink-legacy does.
Thanks!
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next v2 04/10] tools/ynl: Add mcast-group schema parsing to ynl
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (2 preceding siblings ...)
2023-08-15 19:42 ` [PATCH net-next v2 03/10] doc/netlink: Document the netlink-raw " Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-16 15:12 ` Jakub Kicinski
2023-08-15 19:42 ` [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg Donald Hunter
` (5 subsequent siblings)
9 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Add a SpecMcastGroup class to the nlspec lib.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
tools/net/ynl/lib/nlspec.py | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/tools/net/ynl/lib/nlspec.py b/tools/net/ynl/lib/nlspec.py
index 0ff0d18666b2..a41ad89eb369 100644
--- a/tools/net/ynl/lib/nlspec.py
+++ b/tools/net/ynl/lib/nlspec.py
@@ -322,6 +322,21 @@ class SpecOperation(SpecElement):
self.attr_set = self.family.attr_sets[attr_set_name]
+class SpecMcastGroup(SpecElement):
+ """Netlink Multicast Group
+
+ Information about a multicast group.
+
+ Attributes:
+ name name of the mulitcast group
+ value numerical id of this multicast group for netlink-raw
+ yaml raw spec as loaded from the spec file
+ """
+ def __init__(self, family, yaml):
+ super().__init__(family, yaml)
+ self.value = self.yaml.get('value')
+
+
class SpecFamily(SpecElement):
""" Netlink Family Spec class.
@@ -343,6 +358,7 @@ class SpecFamily(SpecElement):
ntfs dict of all async events
consts dict of all constants/enums
fixed_header string, optional name of family default fixed header struct
+ mcast_groups dict of all multicast groups (index by name)
"""
def __init__(self, spec_path, schema_path=None, exclude_ops=None):
with open(spec_path, "r") as stream:
@@ -384,6 +400,7 @@ class SpecFamily(SpecElement):
self.ops = collections.OrderedDict()
self.ntfs = collections.OrderedDict()
self.consts = collections.OrderedDict()
+ self.mcast_groups = collections.OrderedDict()
last_exception = None
while len(self._resolution_list) > 0:
@@ -416,6 +433,9 @@ class SpecFamily(SpecElement):
def new_operation(self, elem, req_val, rsp_val):
return SpecOperation(self, elem, req_val, rsp_val)
+ def new_mcast_group(self, elem):
+ return SpecMcastGroup(self, elem)
+
def add_unresolved(self, elem):
self._resolution_list.append(elem)
@@ -512,3 +532,9 @@ class SpecFamily(SpecElement):
self.ops[op.name] = op
elif op.is_async:
self.ntfs[op.name] = op
+
+ mcgs = self.yaml.get('mcast-groups')
+ if mcgs:
+ for elem in mcgs['list']:
+ mcg = self.new_mcast_group(elem)
+ self.mcast_groups[elem['name']] = mcg
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 04/10] tools/ynl: Add mcast-group schema parsing to ynl
2023-08-15 19:42 ` [PATCH net-next v2 04/10] tools/ynl: Add mcast-group schema parsing to ynl Donald Hunter
@ 2023-08-16 15:12 ` Jakub Kicinski
0 siblings, 0 replies; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-16 15:12 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Tue, 15 Aug 2023 20:42:48 +0100 Donald Hunter wrote:
> +class SpecMcastGroup(SpecElement):
> + """Netlink Multicast Group
> +
> + Information about a multicast group.
I'd add more info about value here. Say something along the lines
of value is specified in the spec only for classic netlink
(netlink-raw) families, genetlink families use dynamic ID allocation
so the ids of multicast groups need to be resolved at runtime.
value will be None for genetlink families.
> +
> + Attributes:
> + name name of the mulitcast group
> + value numerical id of this multicast group for netlink-raw
> + yaml raw spec as loaded from the spec file
> + """
> + def __init__(self, family, yaml):
> + super().__init__(family, yaml)
> + self.value = self.yaml.get('value')
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (3 preceding siblings ...)
2023-08-15 19:42 ` [PATCH net-next v2 04/10] tools/ynl: Add mcast-group schema parsing to ynl Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-16 15:20 ` Jakub Kicinski
2023-08-15 19:42 ` [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (4 subsequent siblings)
9 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Move decode_fixed_header into NlMsg in preparation for adding
netlink-raw support.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
tools/net/ynl/lib/ynl.py | 39 ++++++++++++++++++++++++---------------
1 file changed, 24 insertions(+), 15 deletions(-)
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 3ca28d4bcb18..4fa42a7c5955 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -189,6 +189,7 @@ class NlMsg:
self.error = 0
self.done = 0
+ self.fixed_header_attrs = []
extack_off = None
if self.nl_type == Netlink.NLMSG_ERROR:
@@ -228,6 +229,19 @@ class NlMsg:
desc += f" ({spec['doc']})"
self.extack['miss-type'] = desc
+ def decode_fixed_header(self, ynl, name):
+ fixed_header_members = ynl.consts[name].members
+ self.fixed_header_attrs = dict()
+ offset = 0
+ for m in fixed_header_members:
+ format = NlAttr.get_format(m.type, m.byte_order)
+ [ value ] = format.unpack_from(self.raw, offset)
+ offset += format.size
+ if m.enum:
+ value = ynl._decode_enum(value, m)
+ self.fixed_header_attrs[m.name] = value
+ self.raw = self.raw[offset:]
+
def __repr__(self):
msg = f"nl_len = {self.nl_len} ({len(self.raw)}) nl_flags = 0x{self.nl_flags:x} nl_type = {self.nl_type}\n"
if self.error:
@@ -317,23 +331,18 @@ def _genl_load_families():
class GenlMsg:
- def __init__(self, nl_msg, fixed_header_members=[]):
- self.nl = nl_msg
+ def __init__(self, nl_msg, ynl=None):
+ self.genl_cmd, self.genl_version, _ = struct.unpack_from("BBH", nl_msg.raw, 0)
+ nl_msg.raw = nl_msg.raw[4:]
- self.hdr = nl_msg.raw[0:4]
- offset = 4
-
- self.genl_cmd, self.genl_version, _ = struct.unpack("BBH", self.hdr)
-
- self.fixed_header_attrs = dict()
- for m in fixed_header_members:
- format = NlAttr.get_format(m.type, m.byte_order)
- decoded = format.unpack_from(nl_msg.raw, offset)
- offset += format.size
- self.fixed_header_attrs[m.name] = decoded[0]
+ if ynl:
+ op = ynl.rsp_by_value[self.genl_cmd]
+ if op.fixed_header:
+ nl_msg.decode_fixed_header(ynl, op.fixed_header)
- self.raw = nl_msg.raw[offset:]
+ self.raw = nl_msg.raw
self.raw_attrs = NlAttrs(self.raw)
+ self.fixed_header_attrs = nl_msg.fixed_header_attrs
def __repr__(self):
msg = repr(self.nl)
@@ -596,7 +605,7 @@ class YnlFamily(SpecFamily):
done = True
break
- gm = GenlMsg(nl_msg, fixed_header_members)
+ gm = GenlMsg(nl_msg, self)
# Check if this is a reply to our request
if nl_msg.nl_seq != req_seq or gm.genl_cmd != op.rsp_value:
if gm.genl_cmd in self.async_msg_ids:
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg
2023-08-15 19:42 ` [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg Donald Hunter
@ 2023-08-16 15:20 ` Jakub Kicinski
2023-08-17 15:14 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-16 15:20 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Tue, 15 Aug 2023 20:42:49 +0100 Donald Hunter wrote:
> + def __init__(self, nl_msg, ynl=None):
> + self.genl_cmd, self.genl_version, _ = struct.unpack_from("BBH", nl_msg.raw, 0)
> + nl_msg.raw = nl_msg.raw[4:]
It's a bit of a layering violation that we are futzing with the raw
member of NlMsg inside GenlMsg, no?
Should we add "fixed hdrs len" argument to NlMsg? Either directly or
pass ynl and let get the expected len from ynl? That way NlMsg can
split itself into hdr, userhdrs and attrs without GenlMsg "fixing it
up"?
> - self.hdr = nl_msg.raw[0:4]
> - offset = 4
> -
> - self.genl_cmd, self.genl_version, _ = struct.unpack("BBH", self.hdr)
> -
> - self.fixed_header_attrs = dict()
> - for m in fixed_header_members:
> - format = NlAttr.get_format(m.type, m.byte_order)
> - decoded = format.unpack_from(nl_msg.raw, offset)
> - offset += format.size
> - self.fixed_header_attrs[m.name] = decoded[0]
> + if ynl:
> + op = ynl.rsp_by_value[self.genl_cmd]
> + if op.fixed_header:
> + nl_msg.decode_fixed_header(ynl, op.fixed_header)
>
> - self.raw = nl_msg.raw[offset:]
> + self.raw = nl_msg.raw
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg
2023-08-16 15:20 ` Jakub Kicinski
@ 2023-08-17 15:14 ` Donald Hunter
2023-08-18 1:37 ` Jakub Kicinski
0 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-17 15:14 UTC (permalink / raw)
To: Jakub Kicinski
Cc: Donald Hunter, netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
On Wed, 16 Aug 2023 at 16:20, Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Tue, 15 Aug 2023 20:42:49 +0100 Donald Hunter wrote:
> > + def __init__(self, nl_msg, ynl=None):
> > + self.genl_cmd, self.genl_version, _ = struct.unpack_from("BBH", nl_msg.raw, 0)
> > + nl_msg.raw = nl_msg.raw[4:]
>
> It's a bit of a layering violation that we are futzing with the raw
> member of NlMsg inside GenlMsg, no?
>
> Should we add "fixed hdrs len" argument to NlMsg? Either directly or
> pass ynl and let get the expected len from ynl? That way NlMsg can
> split itself into hdr, userhdrs and attrs without GenlMsg "fixing it
> up"?
I agree, it breaks the layering. The issue is that GenlMsg gets created at
some point after NlMsg, only when we know the nl_msg is suitable for
decoding. The fixed header bit is quite well encapsulated in NlMsg,
it's the genl header that needs pulled out and NlMsg shouldn't know
anything about it. How about I add a take_bytes(length) method or a
generic decode_subheader(format, length) method to NlMsg?
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg
2023-08-17 15:14 ` Donald Hunter
@ 2023-08-18 1:37 ` Jakub Kicinski
2023-08-18 10:21 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-18 1:37 UTC (permalink / raw)
To: Donald Hunter
Cc: Donald Hunter, netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
On Thu, 17 Aug 2023 16:14:35 +0100 Donald Hunter wrote:
> > It's a bit of a layering violation that we are futzing with the raw
> > member of NlMsg inside GenlMsg, no?
> >
> > Should we add "fixed hdrs len" argument to NlMsg? Either directly or
> > pass ynl and let get the expected len from ynl? That way NlMsg can
> > split itself into hdr, userhdrs and attrs without GenlMsg "fixing it
> > up"?
>
> I agree, it breaks the layering. The issue is that GenlMsg gets created at
> some point after NlMsg, only when we know the nl_msg is suitable for
> decoding. The fixed header bit is quite well encapsulated in NlMsg,
> it's the genl header that needs pulled out and NlMsg shouldn't know
> anything about it. How about I add a take_bytes(length) method or a
> generic decode_subheader(format, length) method to NlMsg?
Why do we need to fix up the .raw of NlMsg underlying the GenlMsg
in the first place? GenlMsg by itself didn't need to do that until now.
Another option to consider which would make things more symmetric
between raw and genetlink would be to add a wrapper class for old
families, too. ClassicMsg? CnlMsg? That way we could retain the
separation of NlMsg is just a raw message which could be a NLM_DONE or
some other control thing, and higher level class being used to pull
fixed headers and separate out attrs. Just a thought, not sure it helps.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg
2023-08-18 1:37 ` Jakub Kicinski
@ 2023-08-18 10:21 ` Donald Hunter
0 siblings, 0 replies; 29+ messages in thread
From: Donald Hunter @ 2023-08-18 10:21 UTC (permalink / raw)
To: Jakub Kicinski
Cc: Donald Hunter, netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
On Fri, 18 Aug 2023 at 02:37, Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Thu, 17 Aug 2023 16:14:35 +0100 Donald Hunter wrote:
> > > It's a bit of a layering violation that we are futzing with the raw
> > > member of NlMsg inside GenlMsg, no?
> > >
> > > Should we add "fixed hdrs len" argument to NlMsg? Either directly or
> > > pass ynl and let get the expected len from ynl? That way NlMsg can
> > > split itself into hdr, userhdrs and attrs without GenlMsg "fixing it
> > > up"?
> >
> > I agree, it breaks the layering. The issue is that GenlMsg gets created at
> > some point after NlMsg, only when we know the nl_msg is suitable for
> > decoding. The fixed header bit is quite well encapsulated in NlMsg,
> > it's the genl header that needs pulled out and NlMsg shouldn't know
> > anything about it. How about I add a take_bytes(length) method or a
> > generic decode_subheader(format, length) method to NlMsg?
>
> Why do we need to fix up the .raw of NlMsg underlying the GenlMsg
> in the first place? GenlMsg by itself didn't need to do that until now.
Fair point. I will refactor to leave nl_msg.raw untouched.
> Another option to consider which would make things more symmetric
> between raw and genetlink would be to add a wrapper class for old
> families, too. ClassicMsg? CnlMsg? That way we could retain the
> separation of NlMsg is just a raw message which could be a NLM_DONE or
> some other control thing, and higher level class being used to pull
> fixed headers and separate out attrs. Just a thought, not sure it helps.
I _think_ I can avoid doing this. There's an asymmetry to the way
the NlAttrs get created that I need to fix. When I do that, the rest
should be a bit cleaner.
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (4 preceding siblings ...)
2023-08-15 19:42 ` [PATCH net-next v2 05/10] tools/net/ynl: Refactor decode_fixed_header into NlMsg Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-16 15:29 ` Jakub Kicinski
2023-08-15 19:42 ` [PATCH net-next v2 07/10] tools/net/ynl: Implement nlattr array-nest decoding in ynl Donald Hunter
` (3 subsequent siblings)
9 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Refactor the ynl code to encapsulate protocol specifics into
NetlinkProtocol and GenlProtocol.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
tools/net/ynl/lib/ynl.py | 103 +++++++++++++++++++++++++++++----------
1 file changed, 78 insertions(+), 25 deletions(-)
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 4fa42a7c5955..325dc0d9c5b5 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -25,6 +25,7 @@ class Netlink:
NETLINK_ADD_MEMBERSHIP = 1
NETLINK_CAP_ACK = 10
NETLINK_EXT_ACK = 11
+ NETLINK_GET_STRICT_CHK = 12
# Netlink message
NLMSG_ERROR = 2
@@ -242,6 +243,9 @@ class NlMsg:
self.fixed_header_attrs[m.name] = value
self.raw = self.raw[offset:]
+ def cmd(self):
+ return self.nl_type
+
def __repr__(self):
msg = f"nl_len = {self.nl_len} ({len(self.raw)}) nl_flags = 0x{self.nl_flags:x} nl_type = {self.nl_type}\n"
if self.error:
@@ -344,6 +348,9 @@ class GenlMsg:
self.raw_attrs = NlAttrs(self.raw)
self.fixed_header_attrs = nl_msg.fixed_header_attrs
+ def cmd(self):
+ return self.genl_cmd
+
def __repr__(self):
msg = repr(self.nl)
msg += f"\tgenl_cmd = {self.genl_cmd} genl_ver = {self.genl_version}\n"
@@ -352,9 +359,35 @@ class GenlMsg:
return msg
-class GenlFamily:
- def __init__(self, family_name):
+class NetlinkProtocol:
+ def __init__(self, family_name, proto_num):
self.family_name = family_name
+ self.proto_num = proto_num
+
+ def _message(self, nl_type, nl_flags, seq=None):
+ if seq is None:
+ seq = random.randint(1, 1024)
+ nlmsg = struct.pack("HHII", nl_type, nl_flags, seq, 0)
+ return nlmsg
+
+ def message(self, flags, command, version, seq=None):
+ return self._message(command, flags, seq)
+
+ def decode(self, ynl, nl_msg):
+ op = ynl.rsp_by_value[nl_msg.nl_type]
+ if op.fixed_header:
+ nl_msg.decode_fixed_header(ynl, op.fixed_header)
+ nl_msg.raw_attrs = NlAttrs(nl_msg.raw)
+ return nl_msg
+
+ def get_mcast_id(self, mcast_name, mcast_groups):
+ if mcast_name not in mcast_groups:
+ raise Exception(f'Multicast group "{mcast_name}" not present in the spec')
+ return mcast_groups[mcast_name].id
+
+class GenlProtocol(NetlinkProtocol):
+ def __init__(self, family_name):
+ super().__init__(family_name, Netlink.NETLINK_GENERIC)
global genl_family_name_to_id
if genl_family_name_to_id is None:
@@ -363,6 +396,18 @@ class GenlFamily:
self.genl_family = genl_family_name_to_id[family_name]
self.family_id = genl_family_name_to_id[family_name]['id']
+ def message(self, flags, command, version, seq=None):
+ nlmsg = self._message(self.family_id, flags, seq)
+ genlmsg = struct.pack("BBH", command, version, 0)
+ return nlmsg + genlmsg
+
+ def decode(self, ynl, nl_msg):
+ return GenlMsg(nl_msg, ynl)
+
+ def get_mcast_id(self, mcast_name, mcast_groups):
+ if mcast_name not in self.genl_family['mcast']:
+ raise Exception(f'Multicast group "{mcast_name}" not present in the family')
+ return self.genl_family['mcast'][mcast_name]
#
# YNL implementation details.
@@ -375,9 +420,19 @@ class YnlFamily(SpecFamily):
self.include_raw = False
- self.sock = socket.socket(socket.AF_NETLINK, socket.SOCK_RAW, Netlink.NETLINK_GENERIC)
+ try:
+ if self.proto == "netlink-raw":
+ self.nlproto = NetlinkProtocol(self.yaml['name'],
+ self.yaml['protonum'])
+ else:
+ self.nlproto = GenlProtocol(self.yaml['name'])
+ except KeyError:
+ raise Exception(f"Family '{self.yaml['name']}' not supported by the kernel")
+
+ self.sock = socket.socket(socket.AF_NETLINK, socket.SOCK_RAW, self.nlproto.proto_num)
self.sock.setsockopt(Netlink.SOL_NETLINK, Netlink.NETLINK_CAP_ACK, 1)
self.sock.setsockopt(Netlink.SOL_NETLINK, Netlink.NETLINK_EXT_ACK, 1)
+ self.sock.setsockopt(Netlink.SOL_NETLINK, Netlink.NETLINK_GET_STRICT_CHK, 1)
self.async_msg_ids = set()
self.async_msg_queue = []
@@ -390,18 +445,12 @@ class YnlFamily(SpecFamily):
bound_f = functools.partial(self._op, op_name)
setattr(self, op.ident_name, bound_f)
- try:
- self.family = GenlFamily(self.yaml['name'])
- except KeyError:
- raise Exception(f"Family '{self.yaml['name']}' not supported by the kernel")
def ntf_subscribe(self, mcast_name):
- if mcast_name not in self.family.genl_family['mcast']:
- raise Exception(f'Multicast group "{mcast_name}" not present in the family')
-
+ mcast_id = self.nlproto.get_mcast_id(mcast_name, self.mcast_groups)
self.sock.bind((0, 0))
self.sock.setsockopt(Netlink.SOL_NETLINK, Netlink.NETLINK_ADD_MEMBERSHIP,
- self.family.genl_family['mcast'][mcast_name])
+ mcast_id)
def _add_attr(self, space, name, value):
attr = self.attr_sets[space][name]
@@ -525,9 +574,12 @@ class YnlFamily(SpecFamily):
if self.include_raw:
msg['nlmsg'] = nl_msg
msg['genlmsg'] = genl_msg
- op = self.rsp_by_value[genl_msg.genl_cmd]
+ op = self.rsp_by_value[genl_msg.cmd()]
+ decoded = self._decode(genl_msg.raw_attrs, op.attr_set.name)
+ decoded.update(genl_msg.fixed_header_attrs)
+
msg['name'] = op['name']
- msg['msg'] = self._decode(genl_msg.raw_attrs, op.attr_set.name)
+ msg['msg'] = decoded
self.async_msg_queue.append(msg)
def check_ntf(self):
@@ -547,12 +599,12 @@ class YnlFamily(SpecFamily):
print("Netlink done while checking for ntf!?")
continue
- gm = GenlMsg(nl_msg)
- if gm.genl_cmd not in self.async_msg_ids:
- print("Unexpected msg id done while checking for ntf", gm)
+ decoded = self.nlproto.decode(self, nl_msg)
+ if decoded.cmd() not in self.async_msg_ids:
+ print("Unexpected msg id done while checking for ntf", decoded)
continue
- self.handle_ntf(nl_msg, gm)
+ self.handle_ntf(nl_msg, decoded)
def operation_do_attributes(self, name):
"""
@@ -573,7 +625,7 @@ class YnlFamily(SpecFamily):
nl_flags |= Netlink.NLM_F_DUMP
req_seq = random.randint(1024, 65535)
- msg = _genl_msg(self.family.family_id, nl_flags, op.req_value, 1, req_seq)
+ msg = self.nlproto.message(nl_flags, op.req_value, 1, req_seq)
fixed_header_members = []
if op.fixed_header:
fixed_header_members = self.consts[op.fixed_header].members
@@ -605,18 +657,19 @@ class YnlFamily(SpecFamily):
done = True
break
- gm = GenlMsg(nl_msg, self)
+ decoded = self.nlproto.decode(self, nl_msg)
+
# Check if this is a reply to our request
- if nl_msg.nl_seq != req_seq or gm.genl_cmd != op.rsp_value:
- if gm.genl_cmd in self.async_msg_ids:
- self.handle_ntf(nl_msg, gm)
+ if nl_msg.nl_seq != req_seq or decoded.cmd() != op.rsp_value:
+ if decoded.cmd() in self.async_msg_ids:
+ self.handle_ntf(nl_msg, decoded)
continue
else:
- print('Unexpected message: ' + repr(gm))
+ print('Unexpected message: ' + repr(decoded))
continue
- rsp_msg = self._decode(gm.raw_attrs, op.attr_set.name)
- rsp_msg.update(gm.fixed_header_attrs)
+ rsp_msg = self._decode(decoded.raw_attrs, op.attr_set.name)
+ rsp_msg.update(decoded.fixed_header_attrs)
rsp.append(rsp_msg)
if not rsp:
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families
2023-08-15 19:42 ` [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
@ 2023-08-16 15:29 ` Jakub Kicinski
2023-08-17 9:10 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-16 15:29 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Tue, 15 Aug 2023 20:42:50 +0100 Donald Hunter wrote:
> Refactor the ynl code to encapsulate protocol specifics into
> NetlinkProtocol and GenlProtocol.
Looks good, but do we also need some extra plumbing to decode extack
for classic netlink correctly? Basically shouldn't _decode_extack()
also move to proto? Or we can parameterize it? All we really need there
is to teach it how much of fixed headers parser needs to skip to get to
attributes, really (which, BTW is already kinda buggy for genl families
with fixed headers).
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families
2023-08-16 15:29 ` Jakub Kicinski
@ 2023-08-17 9:10 ` Donald Hunter
2023-08-18 1:18 ` Jakub Kicinski
0 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-17 9:10 UTC (permalink / raw)
To: Jakub Kicinski
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
Jakub Kicinski <kuba@kernel.org> writes:
> On Tue, 15 Aug 2023 20:42:50 +0100 Donald Hunter wrote:
>> Refactor the ynl code to encapsulate protocol specifics into
>> NetlinkProtocol and GenlProtocol.
>
> Looks good, but do we also need some extra plumbing to decode extack
> for classic netlink correctly? Basically shouldn't _decode_extack()
> also move to proto? Or we can parameterize it? All we really need there
> is to teach it how much of fixed headers parser needs to skip to get to
> attributes, really (which, BTW is already kinda buggy for genl families
> with fixed headers).
I have been working on the assumption that extack responses don't
include any fixed headers. I have seen extack messages decoded correctly
for classic netlink, here with RTM_NEWROUTE:
lib.ynl.NlError: Netlink error: Invalid argument
nl_len = 80 (64) nl_flags = 0x300 nl_type = 2
error: -22 extack: {'msg': 'Invalid prefix for given prefix length'}
Is there something I am missing?
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families
2023-08-17 9:10 ` Donald Hunter
@ 2023-08-18 1:18 ` Jakub Kicinski
2023-08-21 14:00 ` Donald Hunter
0 siblings, 1 reply; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-18 1:18 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Thu, 17 Aug 2023 10:10:35 +0100 Donald Hunter wrote:
> > Looks good, but do we also need some extra plumbing to decode extack
> > for classic netlink correctly? Basically shouldn't _decode_extack()
> > also move to proto? Or we can parameterize it? All we really need there
> > is to teach it how much of fixed headers parser needs to skip to get to
> > attributes, really (which, BTW is already kinda buggy for genl families
> > with fixed headers).
>
> I have been working on the assumption that extack responses don't
> include any fixed headers. I have seen extack messages decoded correctly
> for classic netlink, here with RTM_NEWROUTE:
>
> lib.ynl.NlError: Netlink error: Invalid argument
> nl_len = 80 (64) nl_flags = 0x300 nl_type = 2
> error: -22 extack: {'msg': 'Invalid prefix for given prefix length'}
>
> Is there something I am missing?
I'm thinking of extack messages carrying offsets in addition to the
textual error message. NLMSGERR_ATTR_OFFS or NLMSGERR_ATTR_MISS_NEST.
In that case ynl will try to re-parse its own message via
_decode_extack_path() to resolve from the offset to what attribute
was there. See the commit message on a552bfa16:
lib.ynl.NlError: Netlink error: Numerical result out of range
nl_len = 108 (92) nl_flags = 0x300 nl_type = 2
error: -34 extack: {'msg': 'integer out of range',...
'bad-attr': '.ifindex'}
I mean the "bad-attr" thing.
I think it works out of sheer luck here, we happen to skip over
the fixed header because it looks like a 0-length attribute?
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families
2023-08-18 1:18 ` Jakub Kicinski
@ 2023-08-21 14:00 ` Donald Hunter
0 siblings, 0 replies; 29+ messages in thread
From: Donald Hunter @ 2023-08-21 14:00 UTC (permalink / raw)
To: Jakub Kicinski
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
Jakub Kicinski <kuba@kernel.org> writes:
> On Thu, 17 Aug 2023 10:10:35 +0100 Donald Hunter wrote:
>> > Looks good, but do we also need some extra plumbing to decode extack
>> > for classic netlink correctly? Basically shouldn't _decode_extack()
>> > also move to proto? Or we can parameterize it? All we really need there
>> > is to teach it how much of fixed headers parser needs to skip to get to
>> > attributes, really (which, BTW is already kinda buggy for genl families
>> > with fixed headers).
>>
>> I have been working on the assumption that extack responses don't
>> include any fixed headers. I have seen extack messages decoded correctly
>> for classic netlink, here with RTM_NEWROUTE:
>>
>> lib.ynl.NlError: Netlink error: Invalid argument
>> nl_len = 80 (64) nl_flags = 0x300 nl_type = 2
>> error: -22 extack: {'msg': 'Invalid prefix for given prefix length'}
>>
>> Is there something I am missing?
>
> I'm thinking of extack messages carrying offsets in addition to the
> textual error message. NLMSGERR_ATTR_OFFS or NLMSGERR_ATTR_MISS_NEST.
>
> In that case ynl will try to re-parse its own message via
> _decode_extack_path() to resolve from the offset to what attribute
> was there. See the commit message on a552bfa16:
>
> lib.ynl.NlError: Netlink error: Numerical result out of range
> nl_len = 108 (92) nl_flags = 0x300 nl_type = 2
> error: -34 extack: {'msg': 'integer out of range',...
> 'bad-attr': '.ifindex'}
>
> I mean the "bad-attr" thing.
>
> I think it works out of sheer luck here, we happen to skip over
> the fixed header because it looks like a 0-length attribute?
You're right, sheer luck, and maybe only for some values of dp-ifindex.
When I tried to reproduce your test in commit a552bfa16, with a value of
dp-ifindex = 5, then ynl goes into an infinite loop trying to read a
zero length nlattr.
As you say, I'll need to rework the extack handling to account for fixed
headers. At a minimum _decode_extack will need to use nlproto.decode()
and needs to learn to skip the fixed header.
Apologies for being slow to catch up with you on this. Failing to grok
that _decode_extack is decoding the request, not the response.
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next v2 07/10] tools/net/ynl: Implement nlattr array-nest decoding in ynl
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (5 preceding siblings ...)
2023-08-15 19:42 ` [PATCH net-next v2 06/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-16 15:31 ` Jakub Kicinski
2023-08-15 19:42 ` [PATCH net-next v2 08/10] doc/netlink: Add spec for rt addr messages Donald Hunter
` (2 subsequent siblings)
9 siblings, 1 reply; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Add support for the 'array-nest' attribute type that is used by several
netlink-raw families.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
tools/net/ynl/lib/ynl.py | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 325dc0d9c5b5..cd983d1f2ee6 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -504,6 +504,17 @@ class YnlFamily(SpecFamily):
decoded = NlAttr.formatted_string(decoded, attr_spec.display_hint)
return decoded
+ def _decode_array_nest(self, attr, attr_spec):
+ decoded = []
+ offset = 0
+ while offset < len(attr.raw):
+ item = NlAttr(attr.raw, offset)
+ offset += item.full_len
+
+ subattrs = self._decode(NlAttrs(item.raw), attr_spec['nested-attributes'])
+ decoded.append({ item.type: subattrs })
+ return decoded
+
def _decode(self, attrs, space):
attr_space = self.attr_sets[space]
rsp = dict()
@@ -520,6 +531,8 @@ class YnlFamily(SpecFamily):
decoded = True
elif attr_spec["type"] in NlAttr.type_formats:
decoded = attr.as_scalar(attr_spec['type'], attr_spec.byte_order)
+ elif attr_spec["type"] == 'array-nest':
+ decoded = self._decode_array_nest(attr, attr_spec)
else:
raise Exception(f'Unknown {attr_spec["type"]} with name {attr_spec["name"]}')
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH net-next v2 07/10] tools/net/ynl: Implement nlattr array-nest decoding in ynl
2023-08-15 19:42 ` [PATCH net-next v2 07/10] tools/net/ynl: Implement nlattr array-nest decoding in ynl Donald Hunter
@ 2023-08-16 15:31 ` Jakub Kicinski
0 siblings, 0 replies; 29+ messages in thread
From: Jakub Kicinski @ 2023-08-16 15:31 UTC (permalink / raw)
To: Donald Hunter
Cc: netdev, David S. Miller, Eric Dumazet, Paolo Abeni,
Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski, donald.hunter
On Tue, 15 Aug 2023 20:42:51 +0100 Donald Hunter wrote:
> Add support for the 'array-nest' attribute type that is used by several
> netlink-raw families.
>
> Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next v2 08/10] doc/netlink: Add spec for rt addr messages
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (6 preceding siblings ...)
2023-08-15 19:42 ` [PATCH net-next v2 07/10] tools/net/ynl: Implement nlattr array-nest decoding in ynl Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-15 19:42 ` [PATCH net-next v2 09/10] doc/netlink: Add spec for rt link messages Donald Hunter
2023-08-15 19:42 ` [PATCH net-next v2 10/10] doc/netlink: Add spec for rt route messages Donald Hunter
9 siblings, 0 replies; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Add schema for rt addr with support for:
- newaddr, deladdr, getaddr (dump)
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
Documentation/netlink/specs/rt_addr.yaml | 179 +++++++++++++++++++++++
1 file changed, 179 insertions(+)
create mode 100644 Documentation/netlink/specs/rt_addr.yaml
diff --git a/Documentation/netlink/specs/rt_addr.yaml b/Documentation/netlink/specs/rt_addr.yaml
new file mode 100644
index 000000000000..cbee1cedb177
--- /dev/null
+++ b/Documentation/netlink/specs/rt_addr.yaml
@@ -0,0 +1,179 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+
+name: rt-addr
+protocol: netlink-raw
+protonum: 0
+
+doc:
+ Address configuration over rtnetlink.
+
+definitions:
+ -
+ name: ifaddrmsg
+ type: struct
+ members:
+ -
+ name: ifa-family
+ type: u8
+ -
+ name: ifa-prefixlen
+ type: u8
+ -
+ name: ifa-flags
+ type: u8
+ enum: ifa-flags
+ enum-as-flags: true
+ -
+ name: ifa-scope
+ type: u8
+ -
+ name: ifa-index
+ type: u32
+ -
+ name: ifa-cacheinfo
+ type: struct
+ members:
+ -
+ name: ifa-prefered
+ type: u32
+ -
+ name: ifa-valid
+ type: u32
+ -
+ name: cstamp
+ type: u32
+ -
+ name: tstamp
+ type: u32
+
+ -
+ name: ifa-flags
+ type: flags
+ entries:
+ -
+ name: secondary
+ -
+ name: nodad
+ -
+ name: optimistic
+ -
+ name: dadfailed
+ -
+ name: homeaddress
+ -
+ name: deprecated
+ -
+ name: tentative
+ -
+ name: permanent
+ -
+ name: managetempaddr
+ -
+ name: noprefixroute
+ -
+ name: mcautojoin
+ -
+ name: stable-privacy
+
+attribute-sets:
+ -
+ name: addr-attrs
+ attributes:
+ -
+ name: ifa-address
+ type: binary
+ display-hint: ipv4
+ -
+ name: ifa-local
+ type: binary
+ display-hint: ipv4
+ -
+ name: ifa-label
+ type: string
+ -
+ name: ifa-broadcast
+ type: binary
+ display-hint: ipv4
+ -
+ name: ifa-anycast
+ type: binary
+ -
+ name: ifa-cacheinfo
+ type: binary
+ struct: ifa-cacheinfo
+ -
+ name: ifa-multicast
+ type: binary
+ -
+ name: ifa-flags
+ type: u32
+ enum: ifa-flags
+ enum-as-flags: true
+ -
+ name: ifa-rt-priority
+ type: u32
+ -
+ name: ifa-target-netnsid
+ type: binary
+ -
+ name: ifa-proto
+ type: u8
+
+
+operations:
+ fixed-header: ifaddrmsg
+ enum-model: directional
+ list:
+ -
+ name: newaddr
+ doc: Add new address
+ attribute-set: addr-attrs
+ do:
+ request:
+ value: 20
+ attributes: &ifaddr-all
+ - ifa-family
+ - ifa-flags
+ - ifa-prefixlen
+ - ifa-scope
+ - ifa-index
+ - ifa-address
+ - ifa-label
+ - ifa-local
+ - ifa-cacheinfo
+ -
+ name: deladdr
+ doc: Remove address
+ attribute-set: addr-attrs
+ do:
+ request:
+ value: 21
+ attributes:
+ - ifa-family
+ - ifa-flags
+ - ifa-prefixlen
+ - ifa-scope
+ - ifa-index
+ - ifa-address
+ - ifa-local
+ -
+ name: getaddr
+ doc: Dump address information.
+ attribute-set: addr-attrs
+ dump:
+ request:
+ value: 22
+ attributes:
+ - ifa-index
+ reply:
+ value: 20
+ attributes: *ifaddr-all
+
+mcast-groups:
+ list:
+ -
+ name: rtnlgrp-ipv4-ifaddr
+ value: 5
+ -
+ name: rtnlgrp-ipv6-ifaddr
+ value: 9
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH net-next v2 09/10] doc/netlink: Add spec for rt link messages
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (7 preceding siblings ...)
2023-08-15 19:42 ` [PATCH net-next v2 08/10] doc/netlink: Add spec for rt addr messages Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
2023-08-15 19:42 ` [PATCH net-next v2 10/10] doc/netlink: Add spec for rt route messages Donald Hunter
9 siblings, 0 replies; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Add schema for rt link with support for newlink, dellink, getlink,
setlink and getstats.
For example, offload stats can be fetched like this:
./tools/net/ynl/cli.py \
--spec Documentation/netlink/specs/rt_link.yaml \
--dump getstats --json '{ "filter-mask": 8 }'
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
Documentation/netlink/specs/rt_link.yaml | 1374 ++++++++++++++++++++++
1 file changed, 1374 insertions(+)
create mode 100644 Documentation/netlink/specs/rt_link.yaml
diff --git a/Documentation/netlink/specs/rt_link.yaml b/Documentation/netlink/specs/rt_link.yaml
new file mode 100644
index 000000000000..f9b7aedbc0ac
--- /dev/null
+++ b/Documentation/netlink/specs/rt_link.yaml
@@ -0,0 +1,1374 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+
+name: rt-link
+protocol: netlink-raw
+protonum: 0
+
+doc:
+ Link configuration over rtnetlink.
+
+definitions:
+ -
+ name: ifinfo-flags
+ type: flags
+ entries:
+ -
+ name: up
+ -
+ name: broadcast
+ -
+ name: debug
+ -
+ name: loopback
+ -
+ name: point-to-point
+ -
+ name: no-trailers
+ -
+ name: running
+ -
+ name: no-arp
+ -
+ name: promisc
+ -
+ name: all-multi
+ -
+ name: master
+ -
+ name: slave
+ -
+ name: multicast
+ -
+ name: portsel
+ -
+ name: auto-media
+ -
+ name: dynamic
+ -
+ name: lower-up
+ -
+ name: dormant
+ -
+ name: echo
+
+ -
+ name: rtgenmsg
+ type: struct
+ members:
+ -
+ name: family
+ type: u8
+ -
+ name: ifinfomsg
+ type: struct
+ members:
+ -
+ name: ifi-family
+ type: u8
+ -
+ name: padding
+ type: u8
+ -
+ name: ifi-type
+ type: u16
+ -
+ name: ifi-index
+ type: s32
+ -
+ name: ifi-flags
+ type: u32
+ enum: ifinfo-flags
+ enum-as-flags: true
+ -
+ name: ifi-change
+ type: u32
+ -
+ name: ifla-cacheinfo
+ type: struct
+ members:
+ -
+ name: max-reasm-len
+ type: u32
+ -
+ name: tstamp
+ type: u32
+ -
+ name: reachable-time
+ type: s32
+ -
+ name: retrans-time
+ type: u32
+ -
+ name: rtnl-link-stats
+ type: struct
+ members:
+ -
+ name: rx-packets
+ type: u32
+ -
+ name: tx-packets
+ type: u32
+ -
+ name: rx-bytes
+ type: u32
+ -
+ name: tx-bytes
+ type: u32
+ -
+ name: rx-errors
+ type: u32
+ -
+ name: tx-errors
+ type: u32
+ -
+ name: rx-dropped
+ type: u32
+ -
+ name: tx-dropped
+ type: u32
+ -
+ name: multicast
+ type: u32
+ -
+ name: collisions
+ type: u32
+ -
+ name: rx-length-errors
+ type: u32
+ -
+ name: rx-over-errors
+ type: u32
+ -
+ name: rx-crc-errors
+ type: u32
+ -
+ name: rx-frame-errors
+ type: u32
+ -
+ name: rx-fifo-errors
+ type: u32
+ -
+ name: rx-missed-errors
+ type: u32
+ -
+ name: tx-aborted-errors
+ type: u32
+ -
+ name: tx-carrier-errors
+ type: u32
+ -
+ name: tx-fifo-errors
+ type: u32
+ -
+ name: tx-heartbeat-errors
+ type: u32
+ -
+ name: tx-window-errors
+ type: u32
+ -
+ name: rx-compressed
+ type: u32
+ -
+ name: tx-compressed
+ type: u32
+ -
+ name: rx-nohandler
+ type: u32
+ -
+ name: rtnl-link-stats64
+ type: struct
+ members:
+ -
+ name: rx-packets
+ type: u64
+ -
+ name: tx-packets
+ type: u64
+ -
+ name: rx-bytes
+ type: u64
+ -
+ name: tx-bytes
+ type: u64
+ -
+ name: rx-errors
+ type: u64
+ -
+ name: tx-errors
+ type: u64
+ -
+ name: rx-dropped
+ type: u64
+ -
+ name: tx-dropped
+ type: u64
+ -
+ name: multicast
+ type: u64
+ -
+ name: collisions
+ type: u64
+ -
+ name: rx-length-errors
+ type: u64
+ -
+ name: rx-over-errors
+ type: u64
+ -
+ name: rx-crc-errors
+ type: u64
+ -
+ name: rx-frame-errors
+ type: u64
+ -
+ name: rx-fifo-errors
+ type: u64
+ -
+ name: rx-missed-errors
+ type: u64
+ -
+ name: tx-aborted-errors
+ type: u64
+ -
+ name: tx-carrier-errors
+ type: u64
+ -
+ name: tx-fifo-errors
+ type: u64
+ -
+ name: tx-heartbeat-errors
+ type: u64
+ -
+ name: tx-window-errors
+ type: u64
+ -
+ name: rx-compressed
+ type: u64
+ -
+ name: tx-compressed
+ type: u64
+ -
+ name: rx-nohandler
+ type: u64
+ -
+ name: rx-otherhost-dropped
+ type: u64
+ -
+ name: rtnl-link-ifmap
+ type: struct
+ members:
+ -
+ name: mem-start
+ type: u64
+ -
+ name: mem-end
+ type: u64
+ -
+ name: base-addr
+ type: u64
+ -
+ name: irq
+ type: u16
+ -
+ name: dma
+ type: u8
+ -
+ name: port
+ type: u8
+ -
+ name: ipv4-devconf
+ type: struct
+ members:
+ -
+ name: forwarding
+ type: u32
+ -
+ name: mc-forwarding
+ type: u32
+ -
+ name: proxy-arp
+ type: u32
+ -
+ name: accept-redirects
+ type: u32
+ -
+ name: secure-redirects
+ type: u32
+ -
+ name: send-redirects
+ type: u32
+ -
+ name: shared-media
+ type: u32
+ -
+ name: rp-filter
+ type: u32
+ -
+ name: accept-source-route
+ type: u32
+ -
+ name: bootp-relay
+ type: u32
+ -
+ name: log-martians
+ type: u32
+ -
+ name: tag
+ type: u32
+ -
+ name: arpfilter
+ type: u32
+ -
+ name: medium-id
+ type: u32
+ -
+ name: noxfrm
+ type: u32
+ -
+ name: nopolicy
+ type: u32
+ -
+ name: force-igmp-version
+ type: u32
+ -
+ name: arp-announce
+ type: u32
+ -
+ name: arp-ignore
+ type: u32
+ -
+ name: promote-secondaries
+ type: u32
+ -
+ name: arp-accept
+ type: u32
+ -
+ name: arp-notify
+ type: u32
+ -
+ name: accept-local
+ type: u32
+ -
+ name: src-vmark
+ type: u32
+ -
+ name: proxy-arp-pvlan
+ type: u32
+ -
+ name: route-localnet
+ type: u32
+ -
+ name: igmpv2-unsolicited-report-interval
+ type: u32
+ -
+ name: igmpv3-unsolicited-report-interval
+ type: u32
+ -
+ name: ignore-routes-with-linkdown
+ type: u32
+ -
+ name: drop-unicast-in-l2-multicast
+ type: u32
+ -
+ name: drop-gratuitous-arp
+ type: u32
+ -
+ name: bc-forwarding
+ type: u32
+ -
+ name: arp-evict-nocarrier
+ type: u32
+ -
+ name: ipv6-devconf
+ type: struct
+ members:
+ -
+ name: forwarding
+ type: u32
+ -
+ name: hoplimit
+ type: u32
+ -
+ name: mtu6
+ type: u32
+ -
+ name: accept-ra
+ type: u32
+ -
+ name: accept-redirects
+ type: u32
+ -
+ name: autoconf
+ type: u32
+ -
+ name: dad-transmits
+ type: u32
+ -
+ name: rtr-solicits
+ type: u32
+ -
+ name: rtr-solicit-interval
+ type: u32
+ -
+ name: rtr-solicit-delay
+ type: u32
+ -
+ name: use-tempaddr
+ type: u32
+ -
+ name: temp-valid-lft
+ type: u32
+ -
+ name: temp-prefered-lft
+ type: u32
+ -
+ name: regen-max-retry
+ type: u32
+ -
+ name: max-desync-factor
+ type: u32
+ -
+ name: max-addresses
+ type: u32
+ -
+ name: force-mld-version
+ type: u32
+ -
+ name: accept-ra-defrtr
+ type: u32
+ -
+ name: accept-ra-pinfo
+ type: u32
+ -
+ name: accept-ra-rtr-pref
+ type: u32
+ -
+ name: rtr-probe-interval
+ type: u32
+ -
+ name: accept-ra-rt-info-max-plen
+ type: u32
+ -
+ name: proxy-ndp
+ type: u32
+ -
+ name: optimistic-dad
+ type: u32
+ -
+ name: accept-source-route
+ type: u32
+ -
+ name: mc-forwarding
+ type: u32
+ -
+ name: disable-ipv6
+ type: u32
+ -
+ name: accept-dad
+ type: u32
+ -
+ name: force-tllao
+ type: u32
+ -
+ name: ndisc-notify
+ type: u32
+ -
+ name: mldv1-unsolicited-report-interval
+ type: u32
+ -
+ name: mldv2-unsolicited-report-interval
+ type: u32
+ -
+ name: suppress-frag-ndisc
+ type: u32
+ -
+ name: accept-ra-from-local
+ type: u32
+ -
+ name: use-optimistic
+ type: u32
+ -
+ name: accept-ra-mtu
+ type: u32
+ -
+ name: stable-secret
+ type: u32
+ -
+ name: use-oif-addrs-only
+ type: u32
+ -
+ name: accept-ra-min-hop-limit
+ type: u32
+ -
+ name: ignore-routes-with-linkdown
+ type: u32
+ -
+ name: drop-unicast-in-l2-multicast
+ type: u32
+ -
+ name: drop-unsolicited-na
+ type: u32
+ -
+ name: keep-addr-on-down
+ type: u32
+ -
+ name: rtr-solicit-max-interval
+ type: u32
+ -
+ name: seg6-enabled
+ type: u32
+ -
+ name: seg6-require-hmac
+ type: u32
+ -
+ name: enhanced-dad
+ type: u32
+ -
+ name: addr-gen-mode
+ type: u32
+ -
+ name: disable-policy
+ type: u32
+ -
+ name: accept-ra-rt-info-min-plen
+ type: u32
+ -
+ name: ndisc-tclass
+ type: u32
+ -
+ name: rpl-seg-enabled
+ type: u32
+ -
+ name: ra-defrtr-metric
+ type: u32
+ -
+ name: ioam6-enabled
+ type: u32
+ -
+ name: ioam6-id
+ type: u32
+ -
+ name: ioam6-id-wide
+ type: u32
+ -
+ name: ndisc-evict-nocarrier
+ type: u32
+ -
+ name: accept-untracked-na
+ type: u32
+ -
+ name: ifla-icmp6-stats
+ type: struct
+ members:
+ -
+ name: inmsgs
+ type: u64
+ -
+ name: inerrors
+ type: u64
+ -
+ name: outmsgs
+ type: u64
+ -
+ name: outerrors
+ type: u64
+ -
+ name: csumerrors
+ type: u64
+ -
+ name: ratelimithost
+ type: u64
+ -
+ name: ifla-inet6-stats
+ type: struct
+ members:
+ -
+ name: inpkts
+ type: u64
+ -
+ name: inoctets
+ type: u64
+ -
+ name: indelivers
+ type: u64
+ -
+ name: outforwdatagrams
+ type: u64
+ -
+ name: outpkts
+ type: u64
+ -
+ name: outoctets
+ type: u64
+ -
+ name: inhdrerrors
+ type: u64
+ -
+ name: intoobigerrors
+ type: u64
+ -
+ name: innoroutes
+ type: u64
+ -
+ name: inaddrerrors
+ type: u64
+ -
+ name: inunknownprotos
+ type: u64
+ -
+ name: intruncatedpkts
+ type: u64
+ -
+ name: indiscards
+ type: u64
+ -
+ name: outdiscards
+ type: u64
+ -
+ name: outnoroutes
+ type: u64
+ -
+ name: reasmtimeout
+ type: u64
+ -
+ name: reasmreqds
+ type: u64
+ -
+ name: reasmoks
+ type: u64
+ -
+ name: reasmfails
+ type: u64
+ -
+ name: fragoks
+ type: u64
+ -
+ name: fragfails
+ type: u64
+ -
+ name: fragcreates
+ type: u64
+ -
+ name: inmcastpkts
+ type: u64
+ -
+ name: outmcastpkts
+ type: u64
+ -
+ name: inbcastpkts
+ type: u64
+ -
+ name: outbcastpkts
+ type: u64
+ -
+ name: inmcastoctets
+ type: u64
+ -
+ name: outmcastoctets
+ type: u64
+ -
+ name: inbcastoctets
+ type: u64
+ -
+ name: outbcastoctets
+ type: u64
+ -
+ name: csumerrors
+ type: u64
+ -
+ name: noectpkts
+ type: u64
+ -
+ name: ect1-pkts
+ type: u64
+ -
+ name: ect0-pkts
+ type: u64
+ -
+ name: cepkts
+ type: u64
+ -
+ name: reasm-overlaps
+ type: u64
+ - name: br-boolopt-multi
+ type: struct
+ members:
+ -
+ name: optval
+ type: u32
+ -
+ name: optmask
+ type: u32
+ -
+ name: if_stats_msg
+ type: struct
+ members:
+ -
+ name: family
+ type: u8
+ -
+ name: pad1
+ type: u8
+ -
+ name: pad2
+ type: u16
+ -
+ name: ifindex
+ type: u32
+ -
+ name: filter-mask
+ type: u32
+
+
+attribute-sets:
+ -
+ name: link-attrs
+ name-prefix: ifla-
+ attributes:
+ -
+ name: address
+ type: binary
+ display-hint: mac
+ value: 1
+ -
+ name: broadcast
+ type: binary
+ display-hint: mac
+ value: 2
+ -
+ name: ifname
+ type: string
+ value: 3
+ -
+ name: mtu
+ type: u32
+ value: 4
+ -
+ name: link
+ type: u32
+ value: 5
+ -
+ name: qdisc
+ type: string
+ value: 6
+ -
+ name: stats
+ type: binary
+ value: 7
+ struct: rtnl-link-stats
+ -
+ name: cost
+ type: string
+ value: 8
+ -
+ name: priority
+ type: string
+ value: 9
+ -
+ name: master
+ type: u32
+ value: 10
+ -
+ name: wireless
+ type: string
+ value: 11
+ -
+ name: protinfo
+ type: string
+ value: 12
+ -
+ name: txqlen
+ type: u32
+ value: 13
+ -
+ name: map
+ type: binary
+ value: 14
+ struct: rtnl-link-ifmap
+ -
+ name: weight
+ type: u32
+ value: 15
+ -
+ name: operstate
+ type: u8
+ value: 16
+ -
+ name: linkmode
+ type: u8
+ value: 17
+ -
+ name: linkinfo
+ type: nest
+ value: 18
+ nested-attributes: linkinfo-attrs
+ -
+ name: net-ns-pid
+ type: u32
+ value: 19
+ -
+ name: ifalias
+ type: string
+ value: 20
+ -
+ name: num-vf
+ type: u32
+ value: 21
+ -
+ name: vfinfo-list
+ type: nest
+ value: 22
+ nested-attributes: vfinfo-attrs
+ -
+ name: stats64
+ type: binary
+ value: 23
+ struct: rtnl-link-stats64
+ -
+ name: vf-ports
+ type: nest
+ value: 24
+ nested-attributes: vf-ports-attrs
+ -
+ name: port-self
+ type: nest
+ value: 25
+ nested-attributes: port-self-attrs
+ -
+ name: af-spec
+ type: nest
+ value: 26
+ nested-attributes: af-spec-attrs
+ -
+ name: group
+ type: u32
+ value: 27
+ -
+ name: net-ns-fd
+ type: u32
+ value: 28
+ -
+ name: ext-mask
+ type: u32
+ value: 29
+ -
+ name: promiscuity
+ type: u32
+ value: 30
+ -
+ name: num-tx-queues
+ type: u32
+ value: 31
+ -
+ name: num-rx-queues
+ type: u32
+ value: 32
+ -
+ name: carrier
+ type: u8
+ value: 33
+ -
+ name: phys-port-id
+ type: binary
+ value: 34
+ -
+ name: carrier-changes
+ type: u32
+ value: 35
+ -
+ name: phys-switch-id
+ type: binary
+ value: 36
+ -
+ name: link-netnsid
+ type: s32
+ value: 37
+ -
+ name: phys-port-name
+ type: string
+ value: 38
+ -
+ name: proto-down
+ type: u8
+ value: 39
+ -
+ name: gso-max-segs
+ type: u32
+ value: 40
+ -
+ name: gso-max-size
+ type: u32
+ value: 41
+ -
+ name: pad
+ type: string
+ value: 42
+ -
+ name: xdp
+ type: nest
+ value: 43
+ nested-attributes: xdp-attrs
+ -
+ name: event
+ type: u32
+ value: 44
+ -
+ name: new-netnsid
+ type: s32
+ value: 45
+ -
+ name: if-netnsid
+ type: s32
+ value: 46
+ -
+ name: target-netnsid
+ type: s32
+ value: 46
+ -
+ name: carrier-up-count
+ type: u32
+ value: 47
+ -
+ name: carrier-down-count
+ type: u32
+ value: 48
+ -
+ name: new-ifindex
+ type: s32
+ value: 49
+ -
+ name: min-mtu
+ type: u32
+ value: 50
+ -
+ name: max-mtu
+ type: u32
+ value: 51
+ -
+ name: prop-list
+ type: string
+ value: 52
+ -
+ name: alt-ifname
+ type: string
+ value: 53
+ -
+ name: perm-address
+ type: string
+ value: 54
+ -
+ name: proto-down-reason
+ type: string
+ value: 55
+ -
+ name: parent-dev-name
+ type: string
+ value: 56
+ -
+ name: parent-dev-bus-name
+ type: string
+ value: 57
+ -
+ name: gro-max-size
+ type: u32
+ value: 58
+ -
+ name: tso-max-size
+ type: u32
+ value: 59
+ -
+ name: tso-max-segs
+ type: u32
+ value: 60
+ -
+ name: allmulti
+ type: u32
+ -
+ name: devlink-port
+ type: binary
+ -
+ name: gso-ipv4-max-size
+ type: u32
+ -
+ name: gro-ipv4-max-size
+ type: u32
+ -
+ name: af-spec-attrs
+ attributes:
+ -
+ name: "inet"
+ type: nest
+ value: 2
+ nested-attributes: ifla-attrs
+ -
+ name: "inet6"
+ type: nest
+ value: 10
+ nested-attributes: ifla6-attrs
+ -
+ name: "mctp"
+ type: nest
+ value: 45
+ nested-attributes: mctp-attrs
+ -
+ name: vfinfo-attrs
+ attributes: []
+ -
+ name: vf-ports-attrs
+ attributes: []
+ -
+ name: port-self-attrs
+ attributes: []
+ -
+ name: linkinfo-attrs
+ attributes:
+ -
+ name: kind
+ type: string
+ -
+ name: data
+ type: binary
+ nested-attributes: linkinfo-bridge-attrs
+ -
+ name: xstats
+ type: binary
+ -
+ name: slave-kind
+ type: string
+ -
+ name: slave-data
+ type: nest
+ nested-attributes: linkinfo-slave-data-attrs
+ -
+ name: linkinfo-bridge-attrs
+ attributes:
+ -
+ name: forward-delay
+ type: u32
+ -
+ name: hello-time
+ type: u32
+ -
+ name: max-age
+ type: u32
+ -
+ name: ageing-time
+ type: u32
+ -
+ name: stp-state
+ type: u32
+ -
+ name: priority
+ type: u16
+ -
+ name: vlan-filtering
+ type: u8
+ -
+ name: vlan-protocol
+ type: u16
+ -
+ name: group-fwd-mask
+ type: u16
+ -
+ name: root-id
+ type: binary
+ -
+ name: bridge-id
+ type: binary
+ -
+ name: root-port
+ type: u16
+ -
+ name: root-path-cost
+ type: u32
+ -
+ name: topology-change
+ type: u8
+ -
+ name: topology-change-detected
+ type: u8
+ -
+ name: hello-timer
+ type: u64
+ -
+ name: tcn-timer
+ type: u64
+ -
+ name: topology-change-timer
+ type: u64
+ -
+ name: gc-timer
+ type: u64
+ -
+ name: group-addr
+ type: binary
+ -
+ name: fdb-flush
+ type: binary
+ -
+ name: mcast-router
+ type: u8
+ -
+ name: mcast-snooping
+ type: u8
+ -
+ name: mcast-query-use-ifaddr
+ type: u8
+ -
+ name: mcast-querier
+ type: u8
+ -
+ name: mcast-hash-elasticity
+ type: u32
+ -
+ name: mcast-hash-max
+ type: u32
+ -
+ name: mcast-last-member-cnt
+ type: u32
+ -
+ name: mcast-startup-query-cnt
+ type: u32
+ -
+ name: mcast-last-member-intvl
+ type: u64
+ -
+ name: mcast-membership-intvl
+ type: u64
+ -
+ name: mcast-querier-intvl
+ type: u64
+ -
+ name: mcast-query-intvl
+ type: u64
+ -
+ name: mcast-query-response-intvl
+ type: u64
+ -
+ name: mcast-startup-query-intvl
+ type: u64
+ -
+ name: nf-call-iptables
+ type: u8
+ -
+ name: nf-call-ip6-tables
+ type: u8
+ -
+ name: nf-call-arptables
+ type: u8
+ -
+ name: vlan-default-pvid
+ type: u16
+ -
+ name: pad
+ type: u32
+ -
+ name: vlan-stats-enabled
+ type: u8
+ -
+ name: mcast-stats-enabled
+ type: u8
+ -
+ name: mcast-igmp-version
+ type: u8
+ -
+ name: mcast-mld-version
+ type: u8
+ -
+ name: vlan-stats-per-port
+ type: u8
+ -
+ name: multi-boolopt
+ type: binary
+ struct: br-boolopt-multi
+ -
+ name: mcast-querier-state
+ type: binary
+ -
+ name: linkinfo-slave-data-attrs
+ attributes: []
+ -
+ name: xdp-attrs
+ attributes:
+ -
+ name: fd
+ type: s32
+ -
+ name: attached
+ type: u8
+ -
+ name: flags
+ type: u32
+ -
+ name: prog-id
+ type: u32
+ -
+ name: drv-prog-id
+ type: u32
+ -
+ name: skb-prog-id
+ type: u32
+ -
+ name: hw-prog-id
+ type: u32
+ -
+ name: expected-fd
+ type: s32
+ -
+ name: ifla-attrs
+ attributes:
+ -
+ name: conf
+ type: binary
+ struct: ipv4-devconf
+ -
+ name: ifla6-attrs
+ attributes:
+ -
+ name: flags
+ type: u32
+ -
+ name: conf
+ type: binary
+ struct: ipv6-devconf
+ -
+ name: stats
+ type: binary
+ struct: ifla-inet6-stats
+ -
+ name: mcast
+ type: binary
+ -
+ name: cacheinfo
+ type: binary
+ struct: ifla-cacheinfo
+ -
+ name: icmp6-stats
+ type: binary
+ struct: ifla-icmp6-stats
+ -
+ name: token
+ type: binary
+ -
+ name: addr-gen-mode
+ type: u8
+ -
+ name: ra-mtu
+ type: u32
+ -
+ name: mctp-attrs
+ attributes:
+ -
+ name: mctp-net
+ type: u32
+ -
+ name: stats-attrs
+ name-prefix: ifla-stats-
+ attributes:
+ -
+ name: link-64
+ type: binary
+ struct: rtnl-link-stats64
+ -
+ name: link-xstats
+ type: binary
+ -
+ name: link-xstats-slave
+ type: binary
+ -
+ name: link-offload-xstats
+ type: nest
+ nested-attributes: link-offload-xstats
+ -
+ name: af-spec
+ type: binary
+ -
+ name: link-offload-xstats
+ attributes:
+ -
+ name: cpu-hit
+ type: binary
+ -
+ name: hw-s-info
+ type: array-nest
+ nested-attributes: hw-s-info-one
+ -
+ name: l3-stats
+ type: binary
+ -
+ name: hw-s-info-one
+ attributes:
+ -
+ name: request
+ type: u8
+ -
+ name: used
+ type: u8
+
+
+operations:
+ enum-model: directional
+ list:
+ -
+ name: newlink
+ doc: Create a new link.
+ attribute-set: link-attrs
+ fixed-header: ifinfomsg
+ do:
+ request:
+ value: 16
+ attributes:
+ - ifi-index
+ -
+ name: dellink
+ doc: Delete an existing link.
+ attribute-set: link-attrs
+ fixed-header: ifinfomsg
+ do:
+ request:
+ value: 16
+ attributes:
+ - ifi-index
+ -
+ name: getlink
+ doc: Get / dump information about a link.
+ attribute-set: link-attrs
+ fixed-header: ifinfomsg
+ do:
+ request:
+ value: 18
+ attributes:
+ - ifi-index
+ reply:
+ value: 16
+ attributes:
+ - ifi-index
+ -
+ name: setlink
+ doc: Set information about a link.
+ value: 19
+ attribute-set: link-attrs
+ fixed-header: ifinfomsg
+ do:
+ request:
+ attributes:
+ - ifname
+ -
+ name: getstats
+ doc: Get / dump link stats.
+ attribute-set: stats-attrs
+ fixed-header: if_stats_msg
+ dump:
+ request:
+ value: 94
+ attributes:
+ - ifindex
+ reply:
+ value: 92
+
+
+mcast-groups:
+ list:
+ -
+ name: rtnlgrp-link
+ value: 1
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH net-next v2 10/10] doc/netlink: Add spec for rt route messages
2023-08-15 19:42 [PATCH net-next v2 00/10] tools/net/ynl: Add support for netlink-raw families Donald Hunter
` (8 preceding siblings ...)
2023-08-15 19:42 ` [PATCH net-next v2 09/10] doc/netlink: Add spec for rt link messages Donald Hunter
@ 2023-08-15 19:42 ` Donald Hunter
9 siblings, 0 replies; 29+ messages in thread
From: Donald Hunter @ 2023-08-15 19:42 UTC (permalink / raw)
To: netdev, Jakub Kicinski, David S. Miller, Eric Dumazet,
Paolo Abeni, Jonathan Corbet, linux-doc, Stanislav Fomichev,
Arkadiusz Kubalewski
Cc: donald.hunter, Donald Hunter
Add schema for rt route with support for getroute.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
---
Documentation/netlink/specs/rt_route.yaml | 288 ++++++++++++++++++++++
1 file changed, 288 insertions(+)
create mode 100644 Documentation/netlink/specs/rt_route.yaml
diff --git a/Documentation/netlink/specs/rt_route.yaml b/Documentation/netlink/specs/rt_route.yaml
new file mode 100644
index 000000000000..eb4f6bb80100
--- /dev/null
+++ b/Documentation/netlink/specs/rt_route.yaml
@@ -0,0 +1,288 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+
+name: rt-route
+protocol: netlink-raw
+protonum: 0
+
+doc:
+ Route configuration over rtnetlink.
+
+definitions:
+ -
+ name: rtm-type
+ name-prefix: rtn-
+ type: enum
+ entries:
+ - unspec
+ - unicast
+ - local
+ - broadcast
+ - anycast
+ - multicast
+ - blackhole
+ - unreachable
+ - prohibit
+ - throw
+ - nat
+ - xresolve
+ -
+ name: rtmsg
+ type: struct
+ members:
+ -
+ name: rtm-family
+ type: u8
+ -
+ name: rtm-dst-len
+ type: u8
+ -
+ name: rtm-src-len
+ type: u8
+ -
+ name: rtm-tos
+ type: u8
+ -
+ name: rtm-table
+ type: u8
+ -
+ name: rtm-protocol
+ type: u8
+ -
+ name: rtm-scope
+ type: u8
+ -
+ name: rtm-type
+ type: u8
+ enum: rtm-type
+ -
+ name: rtm-flags
+ type: u32
+ -
+ name: rta-cacheinfo
+ type: struct
+ members:
+ -
+ name: rta-clntref
+ type: u32
+ -
+ name: rta-lastuse
+ type: u32
+ -
+ name: rta-expires
+ type: u32
+ -
+ name: rta-error
+ type: u32
+ -
+ name: rta-used
+ type: u32
+
+attribute-sets:
+ -
+ name: route-attrs
+ attributes:
+ -
+ name: rta-dst
+ type: binary
+ display-hint: ipv4
+ -
+ name: rta-src
+ type: binary
+ display-hint: ipv4
+ -
+ name: rta-iif
+ type: u32
+ -
+ name: rta-oif
+ type: u32
+ -
+ name: rta-gateway
+ type: binary
+ display-hint: ipv4
+ -
+ name: rta-priority
+ type: u32
+ -
+ name: rta-prefsrc
+ type: binary
+ display-hint: ipv4
+ -
+ name: rta-metrics
+ type: nest
+ nested-attributes: rta-metrics
+ -
+ name: rta-multipath
+ type: binary
+ -
+ name: rta-protoinfo # not used
+ type: binary
+ -
+ name: rta-flow
+ type: u32
+ -
+ name: rta-cacheinfo
+ type: binary
+ struct: rta-cacheinfo
+ -
+ name: rta-session # not used
+ type: binary
+ -
+ name: rta-mp-algo # not used
+ type: binary
+ -
+ name: rta-table
+ type: u32
+ -
+ name: rta-mark
+ type: u32
+ -
+ name: rta-mfc-stats
+ type: binary
+ -
+ name: rta-via
+ type: binary
+ -
+ name: rta-newdst
+ type: binary
+ -
+ name: rta-pref
+ type: u8
+ -
+ name: rta-encap-type
+ type: u16
+ -
+ name: rta-encap
+ type: binary # tunnel specific nest
+ -
+ name: rta-expires
+ type: u32
+ -
+ name: rta-pad
+ type: binary
+ -
+ name: rta-uid
+ type: u32
+ -
+ name: rta-ttl-propagate
+ type: u8
+ -
+ name: rta-ip-proto
+ type: u8
+ -
+ name: rta-sport
+ type: u16
+ -
+ name: rta-dport
+ type: u16
+ -
+ name: rta-nh-id
+ type: u32
+ -
+ name: rta-metrics
+ attributes:
+ -
+ name: rtax-unspec
+ type: unused
+ value: 0
+ -
+ name: rtax-lock
+ type: u32
+ -
+ name: rtax-mtu
+ type: u32
+ -
+ name: rtax-window
+ type: u32
+ -
+ name: rtax-rtt
+ type: u32
+ -
+ name: rtax-rttvar
+ type: u32
+ -
+ name: rtax-ssthresh
+ type: u32
+ -
+ name: rtax-cwnd
+ type: u32
+ -
+ name: rtax-advmss
+ type: u32
+ -
+ name: rtax-reordering
+ type: u32
+ -
+ name: rtax-hoplimit
+ type: u32
+ -
+ name: rtax-initcwnd
+ type: u32
+ -
+ name: rtax-features
+ type: u32
+ -
+ name: rtax-rto-min
+ type: u32
+ -
+ name: rtax-initrwnd
+ type: u32
+ -
+ name: rtax-quickack
+ type: u32
+ -
+ name: rtax-cc-algo
+ type: string
+ -
+ name: rtax-fastopen-no-cookie
+ type: u32
+
+operations:
+ enum-model: directional
+ list:
+ -
+ name: getroute
+ doc: Dump route information.
+ attribute-set: route-attrs
+ fixed-header: rtmsg
+ dump:
+ request:
+ value: 26
+ reply:
+ value: 24
+ attributes:
+ - rtm-family
+ - rtm-dst-len
+ - rtm-src-len
+ - rtm-tos
+ - rtm-table
+ - rtm-protocol
+ - rtm-scope
+ - rtm-type
+ - rtm-flags
+ - rta-dst
+ - rta-src
+ - rta-iif
+ - rta-oif
+ - rta-gateway
+ - rta-priority
+ - rta-prefsrc
+ - rta-metrics
+ - rta-multipath
+ - rta-flow
+ - rta-cacheinfo
+ - rta-table
+ - rta-mark
+ - rta-mfc-stats
+ - rta-via
+ - rta-newdst
+ - rta-pref
+ - rta-encap-type
+ - rta-encap
+ - rta-expires
+ - rta-pad
+ - rta-uid
+ - rta-ttl-propagate
+ - rta-ip-proto
+ - rta-sport
+ - rta-dport
+ - rta-nh-id
--
2.41.0
^ permalink raw reply related [flat|nested] 29+ messages in thread