[PATCH net-next 0/5] tools: ynl: handful of forward looking updates

[PATCH net-next 0/5] tools: ynl: handful of forward looking updates

[Jakub Kicinski]

Small YNL improvements, mostly for work-in-progress families.

Jakub Kicinski (5):
  tools: ynl: allow passing binary data
  tools: ynl-gen: set length of binary fields
  tools: ynl-gen: fix collecting global policy attrs
  tools: ynl-gen: support empty attribute lists
  netlink: specs: fix indent in fou

 Documentation/netlink/specs/fou.yaml   | 18 +++++++++---------
 tools/net/ynl/generated/ethtool-user.h |  4 ++++
 tools/net/ynl/generated/fou-user.h     |  6 ++++++
 tools/net/ynl/lib/ynl.py               |  7 ++++++-
 tools/net/ynl/ynl-gen-c.py             | 16 +++++++++++++---
 5 files changed, 38 insertions(+), 13 deletions(-)

-- 
2.41.0

[PATCH net-next 1/5] tools: ynl: allow passing binary data

[Jakub Kicinski]

Recent changes made us assume that input for binary data is in hex.
When using YNL as a Python library it's possible to pass in raw bytes.
Bring the ability to do that back.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/ynl.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 6951bcc7efdc..fa4f1c28efc5 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -410,7 +410,12 @@ genl_family_name_to_id = None
         elif attr["type"] == 'string':
             attr_payload = str(value).encode('ascii') + b'\x00'
         elif attr["type"] == 'binary':
-            attr_payload = bytes.fromhex(value)
+            if isinstance(value, bytes):
+                attr_payload = value
+            elif isinstance(value, str):
+                attr_payload = bytes.fromhex(value)
+            else:
+                raise Exception(f'Unknown type for binary attribute, value: {value}')
         elif attr['type'] in NlAttr.type_formats:
             format = NlAttr.get_format(attr['type'], attr.byte_order)
             attr_payload = format.pack(int(value))
-- 
2.41.0

Re: [PATCH net-next 1/5] tools: ynl: allow passing binary data

[Donald Hunter]

Jakub Kicinski <kuba@kernel.org> writes:

> Recent changes made us assume that input for binary data is in hex.
> When using YNL as a Python library it's possible to pass in raw bytes.
> Bring the ability to do that back.
>
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> ---
>  tools/net/ynl/lib/ynl.py | 7 ++++++-
>  1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
> index 6951bcc7efdc..fa4f1c28efc5 100644
> --- a/tools/net/ynl/lib/ynl.py
> +++ b/tools/net/ynl/lib/ynl.py
> @@ -410,7 +410,12 @@ genl_family_name_to_id = None
>          elif attr["type"] == 'string':
>              attr_payload = str(value).encode('ascii') + b'\x00'
>          elif attr["type"] == 'binary':
> -            attr_payload = bytes.fromhex(value)
> +            if isinstance(value, bytes):
> +                attr_payload = value
> +            elif isinstance(value, str):
> +                attr_payload = bytes.fromhex(value)
> +            else:
> +                raise Exception(f'Unknown type for binary attribute, value: {value}')
>          elif attr['type'] in NlAttr.type_formats:
>              format = NlAttr.get_format(attr['type'], attr.byte_order)
>              attr_payload = format.pack(int(value))

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>

[PATCH net-next 2/5] tools: ynl-gen: set length of binary fields

[Jakub Kicinski]

Remember to set the length field in the request setters.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/generated/ethtool-user.h | 4 ++++
 tools/net/ynl/generated/fou-user.h     | 6 ++++++
 tools/net/ynl/ynl-gen-c.py             | 1 +
 3 files changed, 11 insertions(+)

diff --git a/tools/net/ynl/generated/ethtool-user.h b/tools/net/ynl/generated/ethtool-user.h
index d7d4ba855f43..ddc1a5209992 100644
--- a/tools/net/ynl/generated/ethtool-user.h
+++ b/tools/net/ynl/generated/ethtool-user.h
@@ -1422,6 +1422,7 @@ ethtool_wol_set_req_set_sopass(struct ethtool_wol_set_req *req,
 			       const void *sopass, size_t len)
 {
 	free(req->sopass);
+	req->_present.sopass_len = len;
 	req->sopass = malloc(req->_present.sopass_len);
 	memcpy(req->sopass, sopass, req->_present.sopass_len);
 }
@@ -4071,6 +4072,7 @@ ethtool_fec_set_req_set_stats_corrected(struct ethtool_fec_set_req *req,
 					const void *corrected, size_t len)
 {
 	free(req->stats.corrected);
+	req->stats._present.corrected_len = len;
 	req->stats.corrected = malloc(req->stats._present.corrected_len);
 	memcpy(req->stats.corrected, corrected, req->stats._present.corrected_len);
 }
@@ -4079,6 +4081,7 @@ ethtool_fec_set_req_set_stats_uncorr(struct ethtool_fec_set_req *req,
 				     const void *uncorr, size_t len)
 {
 	free(req->stats.uncorr);
+	req->stats._present.uncorr_len = len;
 	req->stats.uncorr = malloc(req->stats._present.uncorr_len);
 	memcpy(req->stats.uncorr, uncorr, req->stats._present.uncorr_len);
 }
@@ -4087,6 +4090,7 @@ ethtool_fec_set_req_set_stats_corr_bits(struct ethtool_fec_set_req *req,
 					const void *corr_bits, size_t len)
 {
 	free(req->stats.corr_bits);
+	req->stats._present.corr_bits_len = len;
 	req->stats.corr_bits = malloc(req->stats._present.corr_bits_len);
 	memcpy(req->stats.corr_bits, corr_bits, req->stats._present.corr_bits_len);
 }
diff --git a/tools/net/ynl/generated/fou-user.h b/tools/net/ynl/generated/fou-user.h
index d8ab50579cd1..a8f860892540 100644
--- a/tools/net/ynl/generated/fou-user.h
+++ b/tools/net/ynl/generated/fou-user.h
@@ -91,6 +91,7 @@ fou_add_req_set_local_v6(struct fou_add_req *req, const void *local_v6,
 			 size_t len)
 {
 	free(req->local_v6);
+	req->_present.local_v6_len = len;
 	req->local_v6 = malloc(req->_present.local_v6_len);
 	memcpy(req->local_v6, local_v6, req->_present.local_v6_len);
 }
@@ -99,6 +100,7 @@ fou_add_req_set_peer_v6(struct fou_add_req *req, const void *peer_v6,
 			size_t len)
 {
 	free(req->peer_v6);
+	req->_present.peer_v6_len = len;
 	req->peer_v6 = malloc(req->_present.peer_v6_len);
 	memcpy(req->peer_v6, peer_v6, req->_present.peer_v6_len);
 }
@@ -192,6 +194,7 @@ fou_del_req_set_local_v6(struct fou_del_req *req, const void *local_v6,
 			 size_t len)
 {
 	free(req->local_v6);
+	req->_present.local_v6_len = len;
 	req->local_v6 = malloc(req->_present.local_v6_len);
 	memcpy(req->local_v6, local_v6, req->_present.local_v6_len);
 }
@@ -200,6 +203,7 @@ fou_del_req_set_peer_v6(struct fou_del_req *req, const void *peer_v6,
 			size_t len)
 {
 	free(req->peer_v6);
+	req->_present.peer_v6_len = len;
 	req->peer_v6 = malloc(req->_present.peer_v6_len);
 	memcpy(req->peer_v6, peer_v6, req->_present.peer_v6_len);
 }
@@ -280,6 +284,7 @@ fou_get_req_set_local_v6(struct fou_get_req *req, const void *local_v6,
 			 size_t len)
 {
 	free(req->local_v6);
+	req->_present.local_v6_len = len;
 	req->local_v6 = malloc(req->_present.local_v6_len);
 	memcpy(req->local_v6, local_v6, req->_present.local_v6_len);
 }
@@ -288,6 +293,7 @@ fou_get_req_set_peer_v6(struct fou_get_req *req, const void *peer_v6,
 			size_t len)
 {
 	free(req->peer_v6);
+	req->_present.peer_v6_len = len;
 	req->peer_v6 = malloc(req->_present.peer_v6_len);
 	memcpy(req->peer_v6, peer_v6, req->_present.peer_v6_len);
 }
diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index bdff8dfc29c9..e27deb199a70 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -428,6 +428,7 @@ from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation, SpecEnumSet, S
 
     def _setter_lines(self, ri, member, presence):
         return [f"free({member});",
+                f"{presence}_len = len;",
                 f"{member} = malloc({presence}_len);",
                 f'memcpy({member}, {self.c_name}, {presence}_len);']
 
-- 
2.41.0

Re: [PATCH net-next 2/5] tools: ynl-gen: set length of binary fields

[Donald Hunter]

Jakub Kicinski <kuba@kernel.org> writes:

> Remember to set the length field in the request setters.
>
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> ---
>  tools/net/ynl/generated/ethtool-user.h | 4 ++++
>  tools/net/ynl/generated/fou-user.h     | 6 ++++++
>  tools/net/ynl/ynl-gen-c.py             | 1 +
>  3 files changed, 11 insertions(+)
>
> diff --git a/tools/net/ynl/generated/ethtool-user.h b/tools/net/ynl/generated/ethtool-user.h
> index d7d4ba855f43..ddc1a5209992 100644
> --- a/tools/net/ynl/generated/ethtool-user.h
> +++ b/tools/net/ynl/generated/ethtool-user.h
> @@ -1422,6 +1422,7 @@ ethtool_wol_set_req_set_sopass(struct ethtool_wol_set_req *req,
>  			       const void *sopass, size_t len)
>  {
>  	free(req->sopass);
> +	req->_present.sopass_len = len;
>  	req->sopass = malloc(req->_present.sopass_len);
>  	memcpy(req->sopass, sopass, req->_present.sopass_len);
>  }
> @@ -4071,6 +4072,7 @@ ethtool_fec_set_req_set_stats_corrected(struct ethtool_fec_set_req *req,
>  					const void *corrected, size_t len)
>  {
>  	free(req->stats.corrected);
> +	req->stats._present.corrected_len = len;
>  	req->stats.corrected = malloc(req->stats._present.corrected_len);
>  	memcpy(req->stats.corrected, corrected, req->stats._present.corrected_len);
>  }
> @@ -4079,6 +4081,7 @@ ethtool_fec_set_req_set_stats_uncorr(struct ethtool_fec_set_req *req,
>  				     const void *uncorr, size_t len)
>  {
>  	free(req->stats.uncorr);
> +	req->stats._present.uncorr_len = len;
>  	req->stats.uncorr = malloc(req->stats._present.uncorr_len);
>  	memcpy(req->stats.uncorr, uncorr, req->stats._present.uncorr_len);
>  }
> @@ -4087,6 +4090,7 @@ ethtool_fec_set_req_set_stats_corr_bits(struct ethtool_fec_set_req *req,
>  					const void *corr_bits, size_t len)
>  {
>  	free(req->stats.corr_bits);
> +	req->stats._present.corr_bits_len = len;
>  	req->stats.corr_bits = malloc(req->stats._present.corr_bits_len);
>  	memcpy(req->stats.corr_bits, corr_bits, req->stats._present.corr_bits_len);
>  }
> diff --git a/tools/net/ynl/generated/fou-user.h b/tools/net/ynl/generated/fou-user.h
> index d8ab50579cd1..a8f860892540 100644
> --- a/tools/net/ynl/generated/fou-user.h
> +++ b/tools/net/ynl/generated/fou-user.h
> @@ -91,6 +91,7 @@ fou_add_req_set_local_v6(struct fou_add_req *req, const void *local_v6,
>  			 size_t len)
>  {
>  	free(req->local_v6);
> +	req->_present.local_v6_len = len;
>  	req->local_v6 = malloc(req->_present.local_v6_len);
>  	memcpy(req->local_v6, local_v6, req->_present.local_v6_len);
>  }
> @@ -99,6 +100,7 @@ fou_add_req_set_peer_v6(struct fou_add_req *req, const void *peer_v6,
>  			size_t len)
>  {
>  	free(req->peer_v6);
> +	req->_present.peer_v6_len = len;
>  	req->peer_v6 = malloc(req->_present.peer_v6_len);
>  	memcpy(req->peer_v6, peer_v6, req->_present.peer_v6_len);
>  }
> @@ -192,6 +194,7 @@ fou_del_req_set_local_v6(struct fou_del_req *req, const void *local_v6,
>  			 size_t len)
>  {
>  	free(req->local_v6);
> +	req->_present.local_v6_len = len;
>  	req->local_v6 = malloc(req->_present.local_v6_len);
>  	memcpy(req->local_v6, local_v6, req->_present.local_v6_len);
>  }
> @@ -200,6 +203,7 @@ fou_del_req_set_peer_v6(struct fou_del_req *req, const void *peer_v6,
>  			size_t len)
>  {
>  	free(req->peer_v6);
> +	req->_present.peer_v6_len = len;
>  	req->peer_v6 = malloc(req->_present.peer_v6_len);
>  	memcpy(req->peer_v6, peer_v6, req->_present.peer_v6_len);
>  }
> @@ -280,6 +284,7 @@ fou_get_req_set_local_v6(struct fou_get_req *req, const void *local_v6,
>  			 size_t len)
>  {
>  	free(req->local_v6);
> +	req->_present.local_v6_len = len;
>  	req->local_v6 = malloc(req->_present.local_v6_len);
>  	memcpy(req->local_v6, local_v6, req->_present.local_v6_len);
>  }
> @@ -288,6 +293,7 @@ fou_get_req_set_peer_v6(struct fou_get_req *req, const void *peer_v6,
>  			size_t len)
>  {
>  	free(req->peer_v6);
> +	req->_present.peer_v6_len = len;
>  	req->peer_v6 = malloc(req->_present.peer_v6_len);
>  	memcpy(req->peer_v6, peer_v6, req->_present.peer_v6_len);
>  }
> diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
> index bdff8dfc29c9..e27deb199a70 100755
> --- a/tools/net/ynl/ynl-gen-c.py
> +++ b/tools/net/ynl/ynl-gen-c.py
> @@ -428,6 +428,7 @@ from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation, SpecEnumSet, S
>  
>      def _setter_lines(self, ri, member, presence):
>          return [f"free({member});",
> +                f"{presence}_len = len;",
>                  f"{member} = malloc({presence}_len);",
>                  f'memcpy({member}, {self.c_name}, {presence}_len);']

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>

[PATCH net-next 3/5] tools: ynl-gen: fix collecting global policy attrs

[Jakub Kicinski]

We look for attributes inside do.request, but there's another
layer of nesting in the spec, look inside do.request.attributes.

This bug had no effect as all global policies we generate (fou)
seem to be full, anyway, and we treat full and empty the same.

Next patch will change the treatment of empty policies.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ynl-gen-c.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index e27deb199a70..13d06931c045 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -978,7 +978,9 @@ from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation, SpecEnumSet, S
 
             for op_mode in ['do', 'dump']:
                 if op_mode in op:
-                    global_set.update(op[op_mode].get('request', []))
+                    req = op[op_mode].get('request')
+                    if req:
+                        global_set.update(req.get('attributes', []))
 
         self.global_policy = []
         self.global_policy_set = attr_set_name
-- 
2.41.0

Re: [PATCH net-next 3/5] tools: ynl-gen: fix collecting global policy attrs

[Donald Hunter]

Jakub Kicinski <kuba@kernel.org> writes:

> We look for attributes inside do.request, but there's another
> layer of nesting in the spec, look inside do.request.attributes.
>
> This bug had no effect as all global policies we generate (fou)
> seem to be full, anyway, and we treat full and empty the same.
>
> Next patch will change the treatment of empty policies.
>
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> ---
>  tools/net/ynl/ynl-gen-c.py | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
> index e27deb199a70..13d06931c045 100755
> --- a/tools/net/ynl/ynl-gen-c.py
> +++ b/tools/net/ynl/ynl-gen-c.py
> @@ -978,7 +978,9 @@ from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation, SpecEnumSet, S
>  
>              for op_mode in ['do', 'dump']:
>                  if op_mode in op:
> -                    global_set.update(op[op_mode].get('request', []))
> +                    req = op[op_mode].get('request')
> +                    if req:
> +                        global_set.update(req.get('attributes', []))
>  
>          self.global_policy = []
>          self.global_policy_set = attr_set_name

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>

[PATCH net-next 4/5] tools: ynl-gen: support empty attribute lists

[Jakub Kicinski]

Differentiate between empty list and None for member lists.
New families may want to create request responses with no attribute.
If we treat those the same as None we end up rendering
a full parsing policy in user space, instead of an empty one.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ynl-gen-c.py | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index 13d06931c045..9209bdcca9c6 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -615,7 +615,7 @@ from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation, SpecEnumSet, S
 
         self.attr_list = []
         self.attrs = dict()
-        if type_list:
+        if type_list is not None:
             for t in type_list:
                 self.attr_list.append((t, self.attr_set[t]),)
         else:
@@ -1543,7 +1543,14 @@ _C_KW = {
 
     ri.cw.write_func_prot('int', f'{op_prefix(ri, "reply", deref=deref)}_parse', func_args)
 
-    _multi_parse(ri, ri.struct["reply"], init_lines, local_vars)
+    if ri.struct["reply"].member_list():
+        _multi_parse(ri, ri.struct["reply"], init_lines, local_vars)
+    else:
+        # Empty reply
+        ri.cw.block_start()
+        ri.cw.p('return MNL_CB_OK;')
+        ri.cw.block_end()
+        ri.cw.nl()
 
 
 def print_req(ri):
-- 
2.41.0

Re: [PATCH net-next 4/5] tools: ynl-gen: support empty attribute lists

[Donald Hunter]

Jakub Kicinski <kuba@kernel.org> writes:

> Differentiate between empty list and None for member lists.
> New families may want to create request responses with no attribute.
> If we treat those the same as None we end up rendering
> a full parsing policy in user space, instead of an empty one.
>
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> ---
>  tools/net/ynl/ynl-gen-c.py | 11 +++++++++--
>  1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
> index 13d06931c045..9209bdcca9c6 100755
> --- a/tools/net/ynl/ynl-gen-c.py
> +++ b/tools/net/ynl/ynl-gen-c.py
> @@ -615,7 +615,7 @@ from lib import SpecFamily, SpecAttrSet, SpecAttr, SpecOperation, SpecEnumSet, S
>  
>          self.attr_list = []
>          self.attrs = dict()
> -        if type_list:
> +        if type_list is not None:
>              for t in type_list:
>                  self.attr_list.append((t, self.attr_set[t]),)
>          else:
> @@ -1543,7 +1543,14 @@ _C_KW = {
>  
>      ri.cw.write_func_prot('int', f'{op_prefix(ri, "reply", deref=deref)}_parse', func_args)
>  
> -    _multi_parse(ri, ri.struct["reply"], init_lines, local_vars)
> +    if ri.struct["reply"].member_list():
> +        _multi_parse(ri, ri.struct["reply"], init_lines, local_vars)
> +    else:
> +        # Empty reply
> +        ri.cw.block_start()
> +        ri.cw.p('return MNL_CB_OK;')
> +        ri.cw.block_end()
> +        ri.cw.nl()
>  
>  
>  def print_req(ri):

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>

[PATCH net-next 5/5] netlink: specs: fix indent in fou

[Jakub Kicinski]

Fix up the indentation. This has no functional effect, AFAICT.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/fou.yaml | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/Documentation/netlink/specs/fou.yaml b/Documentation/netlink/specs/fou.yaml
index 3e13826a3fdf..0af5ab842c04 100644
--- a/Documentation/netlink/specs/fou.yaml
+++ b/Documentation/netlink/specs/fou.yaml
@@ -107,16 +107,16 @@ kernel-policy: global
       flags: [ admin-perm ]
 
       do:
-        request:  &select_attrs
+        request: &select_attrs
           attributes:
-          - af
-          - ifindex
-          - port
-          - peer_port
-          - local_v4
-          - peer_v4
-          - local_v6
-          - peer_v6
+            - af
+            - ifindex
+            - port
+            - peer_port
+            - local_v4
+            - peer_v4
+            - local_v6
+            - peer_v6
 
     -
       name: get
-- 
2.41.0

Re: [PATCH net-next 5/5] netlink: specs: fix indent in fou

[Donald Hunter]

Jakub Kicinski <kuba@kernel.org> writes:

> Fix up the indentation. This has no functional effect, AFAICT.
>
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> ---
>  Documentation/netlink/specs/fou.yaml | 18 +++++++++---------
>  1 file changed, 9 insertions(+), 9 deletions(-)
>
> diff --git a/Documentation/netlink/specs/fou.yaml b/Documentation/netlink/specs/fou.yaml
> index 3e13826a3fdf..0af5ab842c04 100644
> --- a/Documentation/netlink/specs/fou.yaml
> +++ b/Documentation/netlink/specs/fou.yaml
> @@ -107,16 +107,16 @@ kernel-policy: global
>        flags: [ admin-perm ]
>  
>        do:
> -        request:  &select_attrs
> +        request: &select_attrs
>            attributes:
> -          - af
> -          - ifindex
> -          - port
> -          - peer_port
> -          - local_v4
> -          - peer_v4
> -          - local_v6
> -          - peer_v6
> +            - af
> +            - ifindex
> +            - port
> +            - peer_port
> +            - local_v4
> +            - peer_v4
> +            - local_v6
> +            - peer_v6
>  
>      -
>        name: get

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>

Re: [PATCH net-next 0/5] tools: ynl: handful of forward looking updates

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Wed, 23 Aug 2023 17:30:51 -0700 you wrote:
> Small YNL improvements, mostly for work-in-progress families.
> 
> Jakub Kicinski (5):
>   tools: ynl: allow passing binary data
>   tools: ynl-gen: set length of binary fields
>   tools: ynl-gen: fix collecting global policy attrs
>   tools: ynl-gen: support empty attribute lists
>   netlink: specs: fix indent in fou
> 
> [...]

Here is the summary with links:
  - [net-next,1/5] tools: ynl: allow passing binary data
    https://git.kernel.org/netdev/net-next/c/649bde9004ac
  - [net-next,2/5] tools: ynl-gen: set length of binary fields
    https://git.kernel.org/netdev/net-next/c/a149a3a13bbc
  - [net-next,3/5] tools: ynl-gen: fix collecting global policy attrs
    https://git.kernel.org/netdev/net-next/c/dc2ef94d8926
  - [net-next,4/5] tools: ynl-gen: support empty attribute lists
    https://git.kernel.org/netdev/net-next/c/4c8c24e801e6
  - [net-next,5/5] netlink: specs: fix indent in fou
    https://git.kernel.org/netdev/net-next/c/e83d4e9b2d0f

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html