Support wildcard netdev hooks
Phil Sutter
reviewerv5 cover Support wildcard netdev hooks
Changes since v4:
- Adjust code to NFTA_HOOK_PREFIX attribute
Changes since v3:
- Unrelated patch 1 pushed out separately
- Fixed variable types and function prefix in patch 1
- Document interface wildcard support in nft.8
Changes since v2:
- New patch 1 fixing for excessive mnl_attr_nest_end() calls
- Introduce a helper function in patch 2 which also sanity-checks an array index - this is a copy of libnftnl's function for the same purpose, but we rather get rid of the dupicated calling code instead of exposing (and thus maintaining in future) the function from libnftnl
This is the remaining needed code change to support wildcard hook specs. Patch 3 also adds shell test cases to cover the functionality. The flowtable variant is skipped if 'nft list hooks' does not provide flowtable information as this requires NFNL_HOOK_TYPE_NFT_FLOWTABLE in kernel.
Phil Sutter (3):
mnl: Support simple wildcards in netdev hooks
parser_bison: Accept ASTERISK_STRING in flowtable_expr_member
tests: shell: Test ifname-based hooks
doc/nft.txt | 30 +++++++++++-- include/linux/netfilter/nf_tables.h | 2 + src/mnl.c | 26 +++++++++-- src/parser_bison.y | 11 +---- .../features/list_hooks_flowtable_info.sh | 7 +++ .../netdev_chain_name_based_hook_0.json-nft | 34 ++++++++++++++ .../dumps/netdev_chain_name_based_hook_0.nft | 5 +++ .../chains/netdev_chain_name_based_hook_0 | 44 ++++++++++++++++++ .../testcases/flowtable/0016name_based_hook_0 | 45 +++++++++++++++++++ .../dumps/0016name_based_hook_0.json-nft | 32 +++++++++++++ .../flowtable/dumps/0016name_based_hook_0.nft | 6 +++ 11 files changed, 225 insertions(+), 17 deletions(-) create mode 100755 tests/shell/features/list_hooks_flowtable_info.sh create mode 100644 tests/shell/testcases/chains/dumps/netdev_chain_name_based_hook_0.json-nft create mode 100644 tests/shell/testcases/chains/dumps/netdev_chain_name_based_hook_0.nft create mode 100755 tests/shell/testcases/chains/netdev_chain_name_based_hook_0 create mode 100755 tests/shell/testcases/flowtable/0016name_based_hook_0 create mode 100644 tests/shell/testcases/flowtable/dumps/0016name_based_hook_0.json-nft create mode 100644 tests/shell/testcases/flowtable/dumps/0016name_based_hook_0.nft
Signature · 1 line
2.49.0
Phil Sutter
reviewerPATCH v5 1/3 mnl: Support simple wildcards in netdev hooks
When building NFTA_{FLOWTABLE_,}HOOK_DEVS attributes, detect trailing asterisks in interface names and transmit the leading part in a NFTA_DEVICE_PREFIX attribute.
Deserialization (i.e., appending asterisk to interface prefixes returned in NFTA_DEVICE_PREFIX atributes happens in libnftnl.
---
Changes since v4:
- Introduce and use NFTA_DEVICE_PREFIX which contains a NUL-terminated string as well but signals the kernel to interpret it as a prefix to match interfaces on.
- Do not send wildcards in NFTA_HOOK_DEV: On one hand, the kernel can't detect them anymore since they are NUL-terminated as well. On the other, it would defeat the purpose of having NFTA_DEVICE_PREFIX, which is to not crash old user space.
Changes since v3:
- Use uint16_t for 'attr' parameter and size_t for 'len' variable
- Use mnl_nft_ prefix for the helper function
Changes since v2:
- Introduce mnl_attr_put_ifname() to perform the conditional mnl_attr_put() parameter adjustment
- Sanity-check array index in above function to avoid out-of-bounds access
--- include/linux/netfilter/nf_tables.h | 2 ++ src/mnl.c | 26 +++++++++++++++++++++++--- 2 files changed, 25 insertions(+), 3 deletions(-)
include/linux/netfilter/nf_tables.h +2 -0
@@ -1774,10 +1774,12 @@ enum nft_synproxy_attributes {* enum nft_device_attributes - nf_tables device netlink attributes * * @NFTA_DEVICE_NAME: name of this device (NLA_STRING) +* @NFTA_DEVICE_PREFIX: device name prefix, a simple wildcard (NLA_STRING) */ enum nft_devices_attributes {NFTA_DEVICE_UNSPEC ,NFTA_DEVICE_NAME , +NFTA_DEVICE_PREFIX , __NFTA_DEVICE_MAX };#define NFTA_DEVICE_MAX (__NFTA_DEVICE_MAX - 1)
src/mnl.c +23 -3
@@ -795,6 +795,26 @@ static void nft_dev_array_free(const struct nft_dev *dev_array)free_const (dev_array); } +static bool is_wildcard_str (const char *str) +{ +size_t len =strlen (str); + +if (len <1 || str[len -1 ] !='*' ) +return false; +if (len <2 || str[len -2 ] !='\\' ) +return true; +/* XXX: ignore backslash escaping for now */ +return false; +} + +static void mnl_nft_attr_put_ifname (struct nlmsghdr *nlh,const char *ifname) +{ +uint16_t attr =is_wildcard_str (ifname) ? +NFTA_DEVICE_PREFIX :NFTA_DEVICE_NAME ; + +mnl_attr_put_strz (nlh, attr, ifname); +} +static void mnl_nft_chain_devs_build (struct nlmsghdr *nlh,struct cmd *cmd) {const struct expr *dev_expr = cmd->chain->dev_expr;
@@ -803,14 +823,14 @@ static void mnl_nft_chain_devs_build(struct nlmsghdr *nlh, struct cmd *cmd)int i, num_devs =0 ; dev_array =nft_dev_array (dev_expr, &num_devs); -if (num_devs ==1 ) { +if (num_devs ==1 && !is_wildcard_str (dev_array[0 ].ifname)) {cmd_add_loc (cmd, nlh, dev_array[0 ].location);mnl_attr_put_strz (nlh,NFTA_HOOK_DEV , dev_array[0 ].ifname); }else { nest_dev =mnl_attr_nest_start (nlh,NFTA_HOOK_DEVS );for (i =0 ; i < num_devs; i++) {cmd_add_loc (cmd, nlh, dev_array[i].location); -mnl_attr_put_strz (nlh,NFTA_DEVICE_NAME , dev_array[i].ifname); +mnl_nft_attr_put_ifname (nlh, dev_array[i].ifname); }mnl_attr_nest_end (nlh, nest_dev); }
@@ -2091,7 +2111,7 @@ static void mnl_nft_ft_devs_build(struct nlmsghdr *nlh, struct cmd *cmd) nest_dev =mnl_attr_nest_start (nlh,NFTA_FLOWTABLE_HOOK_DEVS );for (i =0 ; i < num_devs; i++) {cmd_add_loc (cmd, nlh, dev_array[i].location); -mnl_attr_put_strz (nlh,NFTA_DEVICE_NAME , dev_array[i].ifname); +mnl_nft_attr_put_ifname (nlh, dev_array[i].ifname); }mnl_attr_nest_end (nlh, nest_dev);
Signature · 1 line
2.49.0
Pablo Neira Ayuso
maintainerRe: [nft PATCH v5 1/3] mnl: Support simple wildcards in netdev hooks
Hi Phil,
NFTA_DEVICE_PREFIX is now available in net.git, let's pick up on this.
Quoted 56 lines · Phil Sutter
When building NFTA_{FLOWTABLE_,}HOOK_DEVS attributes, detect trailing asterisks in interface names and transmit the leading part in a
NFTA_DEVICE_PREFIXattribute.Deserialization (i.e., appending asterisk to interface prefixes returned in
NFTA_DEVICE_PREFIXatributes happens in libnftnl.Signed-off-by Phil Sutter <[email protected]>---
Changes since v4:
- Introduce and useNFTA_DEVICE_PREFIXwhich contains a NUL-terminated string as well but signals the kernel to interpret it as a prefix to match interfaces on.
- Do not send wildcards inNFTA_HOOK_DEV: On one hand, the kernel can't detect them anymore since they are NUL-terminated as well. On the other, it would defeat the purpose of havingNFTA_DEVICE_PREFIX, which is to not crash old user space.Changes since v3:
- Useuint16_tfor 'attr' parameter andsize_tfor 'len' variable
- Usemnl_nft_prefix for the helper functionChanges since v2:
- Introducemnl_attr_put_ifname()to perform the conditionalmnl_attr_put()parameter adjustment
- Sanity-check array index in above function to avoid out-of-bounds access--- include/linux/netfilter/nf_tables.h | 2 ++ src/mnl.c | 26 +++++++++++++++++++++++--- 2 files changed, 25 insertions(+), 3 deletions(-)include/linux/netfilter/nf_tables.h +2
-0@@ -1774,10 +1774,12 @@ enum nft_synproxy_attributes {* enum nft_device_attributes - nf_tables device netlink attributes * * @NFTA_DEVICE_NAME: name of this device (NLA_STRING) +* @NFTA_DEVICE_PREFIX: device name prefix, a simple wildcard (NLA_STRING) */ enum nft_devices_attributes {NFTA_DEVICE_UNSPEC ,NFTA_DEVICE_NAME , +NFTA_DEVICE_PREFIX , __NFTA_DEVICE_MAX };#define NFTA_DEVICE_MAX (__NFTA_DEVICE_MAX - 1)src/mnl.c +0
-0@@ -795,6 +795,26 @@ static void nft_dev_array_free(const struct nft_dev *dev_array)free_const (dev_array); }
src/mnl.c static void nft_dev_array_free(const struct nft_dev *dev_array) · line 806
Quoted 10 lines
src/mnl.c +9
-0+static bool is_wildcard_str (const char *str) +{ +size_t len =strlen (str); + +if (len <1 || str[len -1 ] !='*' ) +return false; +if (len <2 || str[len -2 ] !='\\' ) +return true; +/* XXX: ignore backslash escaping for now */
Is this comment here still valid?
Quoted 43 lines
src/mnl.c +14
-3+return false; +} + +static void mnl_nft_attr_put_ifname (struct nlmsghdr *nlh,const char *ifname) +{ +uint16_t attr =is_wildcard_str (ifname) ? +NFTA_DEVICE_PREFIX :NFTA_DEVICE_NAME ; + +mnl_attr_put_strz (nlh, attr, ifname); +} +static void mnl_nft_chain_devs_build (struct nlmsghdr *nlh,struct cmd *cmd) {const struct expr *dev_expr = cmd->chain->dev_expr;@@ -803,14 +823,14 @@ static void mnl_nft_chain_devs_build(struct nlmsghdr *nlh, struct cmd *cmd)int i, num_devs =0 ; dev_array =nft_dev_array (dev_expr, &num_devs); -if (num_devs ==1 ) { +if (num_devs ==1 && !is_wildcard_str (dev_array[0 ].ifname)) {cmd_add_loc (cmd, nlh, dev_array[0 ].location);mnl_attr_put_strz (nlh,NFTA_HOOK_DEV , dev_array[0 ].ifname); }else { nest_dev =mnl_attr_nest_start (nlh,NFTA_HOOK_DEVS );for (i =0 ; i < num_devs; i++) {cmd_add_loc (cmd, nlh, dev_array[i].location); -mnl_attr_put_strz (nlh,NFTA_DEVICE_NAME , dev_array[i].ifname); +mnl_nft_attr_put_ifname (nlh, dev_array[i].ifname); }mnl_attr_nest_end (nlh, nest_dev); }@@ -2091,7 +2111,7 @@ static void mnl_nft_ft_devs_build(struct nlmsghdr *nlh, struct cmd *cmd) nest_dev =mnl_attr_nest_start (nlh,NFTA_FLOWTABLE_HOOK_DEVS );for (i =0 ; i < num_devs; i++) {cmd_add_loc (cmd, nlh, dev_array[i].location); -mnl_attr_put_strz (nlh,NFTA_DEVICE_NAME , dev_array[i].ifname); +mnl_nft_attr_put_ifname (nlh, dev_array[i].ifname); }mnl_attr_nest_end (nlh, nest_dev);Signature · 1 line
2.49.0
Phil Sutter
reviewerRe: [nft PATCH v5 1/3] mnl: Support simple wildcards in netdev hooks
Quoted 62 lines · Pablo Neira Ayuso
Hi Phil,
NFTA_DEVICE_PREFIXis now available in net.git, let's pick up on this.Quoted 57 lines · Phil Sutter
When building NFTA_{FLOWTABLE_,}HOOK_DEVS attributes, detect trailing asterisks in interface names and transmit the leading part in a
NFTA_DEVICE_PREFIXattribute.Deserialization (i.e., appending asterisk to interface prefixes returned in
NFTA_DEVICE_PREFIXatributes happens in libnftnl.Signed-off-by Phil Sutter <[email protected]>---
Changes since v4:
- Introduce and useNFTA_DEVICE_PREFIXwhich contains a NUL-terminated string as well but signals the kernel to interpret it as a prefix to match interfaces on.
- Do not send wildcards inNFTA_HOOK_DEV: On one hand, the kernel can't detect them anymore since they are NUL-terminated as well. On the other, it would defeat the purpose of havingNFTA_DEVICE_PREFIX, which is to not crash old user space.Changes since v3:
- Useuint16_tfor 'attr' parameter andsize_tfor 'len' variable
- Usemnl_nft_prefix for the helper functionChanges since v2:
- Introducemnl_attr_put_ifname()to perform the conditionalmnl_attr_put()parameter adjustment
- Sanity-check array index in above function to avoid out-of-bounds access--- include/linux/netfilter/nf_tables.h | 2 ++ src/mnl.c | 26 +++++++++++++++++++++++--- 2 files changed, 25 insertions(+), 3 deletions(-)include/linux/netfilter/nf_tables.h +2
-0@@ -1774,10 +1774,12 @@ enum nft_synproxy_attributes {* enum nft_device_attributes - nf_tables device netlink attributes * * @NFTA_DEVICE_NAME: name of this device (NLA_STRING) +* @NFTA_DEVICE_PREFIX: device name prefix, a simple wildcard (NLA_STRING) */ enum nft_devices_attributes {NFTA_DEVICE_UNSPEC ,NFTA_DEVICE_NAME , +NFTA_DEVICE_PREFIX , __NFTA_DEVICE_MAX };#define NFTA_DEVICE_MAX (__NFTA_DEVICE_MAX - 1)src/mnl.c +0
-0@@ -795,6 +795,26 @@ static void nft_dev_array_free(const struct nft_dev *dev_array)free_const (dev_array); }
Quoted 11 lines
src/mnl.c static void nft_dev_array_free(const struct nft_dev *dev_array) · line 806
Quoted 9 lines
src/mnl.c +9
-0+static bool is_wildcard_str (const char *str) +{ +size_t len =strlen (str); + +if (len <1 || str[len -1 ] !='*' ) +return false; +if (len <2 || str[len -2 ] !='\\' ) +return true; +/* XXX: ignore backslash escaping for now */ Is this comment here still valid?
Yes, sadly. The above covers for eth* and eth\* but not for eth\\* since a proper solution didn't quickly come to mind which avoids playing whack-a-mole. (E.g., does eth\\\\\\* escape the wildcard or not?)
Guess I could just count the number of backslashes immediately preceding the asterisk and return true if the sum is odd?
Cheers, Phil
Re: [nft PATCH v5 1/3] mnl: Support simple wildcards in netdev hooks
Quoted 75 lines · Phil Sutter
Quoted 62 lines · Pablo Neira Ayuso
Hi Phil,
NFTA_DEVICE_PREFIXis now available in net.git, let's pick up on this.Quoted 57 lines · Phil Sutter
When building NFTA_{FLOWTABLE_,}HOOK_DEVS attributes, detect trailing asterisks in interface names and transmit the leading part in a
NFTA_DEVICE_PREFIXattribute.Deserialization (i.e., appending asterisk to interface prefixes returned in
NFTA_DEVICE_PREFIXatributes happens in libnftnl.Signed-off-by Phil Sutter <[email protected]>---
Changes since v4:
- Introduce and useNFTA_DEVICE_PREFIXwhich contains a NUL-terminated string as well but signals the kernel to interpret it as a prefix to match interfaces on.
- Do not send wildcards inNFTA_HOOK_DEV: On one hand, the kernel can't detect them anymore since they are NUL-terminated as well. On the other, it would defeat the purpose of havingNFTA_DEVICE_PREFIX, which is to not crash old user space.Changes since v3:
- Useuint16_tfor 'attr' parameter andsize_tfor 'len' variable
- Usemnl_nft_prefix for the helper functionChanges since v2:
- Introducemnl_attr_put_ifname()to perform the conditionalmnl_attr_put()parameter adjustment
- Sanity-check array index in above function to avoid out-of-bounds access--- include/linux/netfilter/nf_tables.h | 2 ++ src/mnl.c | 26 +++++++++++++++++++++++--- 2 files changed, 25 insertions(+), 3 deletions(-)include/linux/netfilter/nf_tables.h +2
-0@@ -1774,10 +1774,12 @@ enum nft_synproxy_attributes {* enum nft_device_attributes - nf_tables device netlink attributes * * @NFTA_DEVICE_NAME: name of this device (NLA_STRING) +* @NFTA_DEVICE_PREFIX: device name prefix, a simple wildcard (NLA_STRING) */ enum nft_devices_attributes {NFTA_DEVICE_UNSPEC ,NFTA_DEVICE_NAME , +NFTA_DEVICE_PREFIX , __NFTA_DEVICE_MAX };#define NFTA_DEVICE_MAX (__NFTA_DEVICE_MAX - 1)src/mnl.c +0
-0@@ -795,6 +795,26 @@ static void nft_dev_array_free(const struct nft_dev *dev_array)free_const (dev_array); }Quoted 11 lines
src/mnl.c static void nft_dev_array_free(const struct nft_dev *dev_array) · line 806
Quoted 9 lines
src/mnl.c +9
-0+static bool is_wildcard_str (const char *str) +{ +size_t len =strlen (str); + +if (len <1 || str[len -1 ] !='*' ) +return false; +if (len <2 || str[len -2 ] !='\\' ) +return true; +/* XXX: ignore backslash escaping for now */ Is this comment here still valid?
Quoted 6 lines
Yes, sadly. The above covers for eth* and eth\* but not for eth\\* since a proper solution didn't quickly come to mind which avoids playing whack-a-mole. (E.g., does eth\\\\\\* escape the wildcard or not?)
Guess I could just count the number of backslashes immediately preceding the asterisk and return true if the sum is odd?
Thanks for explaining the comment, this supports eth* and eth\* which is sufficient at this stage.
For this series:
Phil Sutter
reviewerRe: [nft PATCH v5 1/3] mnl: Support simple wildcards in netdev hooks
Quoted 83 lines · Pablo Neira Ayuso
Quoted 75 lines · Phil Sutter
Quoted 62 lines · Pablo Neira Ayuso
Hi Phil,
NFTA_DEVICE_PREFIXis now available in net.git, let's pick up on this.Quoted 57 lines · Phil Sutter
When building NFTA_{FLOWTABLE_,}HOOK_DEVS attributes, detect trailing asterisks in interface names and transmit the leading part in a
NFTA_DEVICE_PREFIXattribute.Deserialization (i.e., appending asterisk to interface prefixes returned in
NFTA_DEVICE_PREFIXatributes happens in libnftnl.Signed-off-by Phil Sutter <[email protected]>---
Changes since v4:
- Introduce and useNFTA_DEVICE_PREFIXwhich contains a NUL-terminated string as well but signals the kernel to interpret it as a prefix to match interfaces on.
- Do not send wildcards inNFTA_HOOK_DEV: On one hand, the kernel can't detect them anymore since they are NUL-terminated as well. On the other, it would defeat the purpose of havingNFTA_DEVICE_PREFIX, which is to not crash old user space.Changes since v3:
- Useuint16_tfor 'attr' parameter andsize_tfor 'len' variable
- Usemnl_nft_prefix for the helper functionChanges since v2:
- Introducemnl_attr_put_ifname()to perform the conditionalmnl_attr_put()parameter adjustment
- Sanity-check array index in above function to avoid out-of-bounds access--- include/linux/netfilter/nf_tables.h | 2 ++ src/mnl.c | 26 +++++++++++++++++++++++--- 2 files changed, 25 insertions(+), 3 deletions(-)include/linux/netfilter/nf_tables.h +2
-0@@ -1774,10 +1774,12 @@ enum nft_synproxy_attributes {* enum nft_device_attributes - nf_tables device netlink attributes * * @NFTA_DEVICE_NAME: name of this device (NLA_STRING) +* @NFTA_DEVICE_PREFIX: device name prefix, a simple wildcard (NLA_STRING) */ enum nft_devices_attributes {NFTA_DEVICE_UNSPEC ,NFTA_DEVICE_NAME , +NFTA_DEVICE_PREFIX , __NFTA_DEVICE_MAX };#define NFTA_DEVICE_MAX (__NFTA_DEVICE_MAX - 1)src/mnl.c +0
-0@@ -795,6 +795,26 @@ static void nft_dev_array_free(const struct nft_dev *dev_array)free_const (dev_array); }Quoted 11 lines
src/mnl.c static void nft_dev_array_free(const struct nft_dev *dev_array) · line 806
Quoted 9 lines
src/mnl.c +9
-0+static bool is_wildcard_str (const char *str) +{ +size_t len =strlen (str); + +if (len <1 || str[len -1 ] !='*' ) +return false; +if (len <2 || str[len -2 ] !='\\' ) +return true; +/* XXX: ignore backslash escaping for now */ Is this comment here still valid?
Quoted 6 lines
Yes, sadly. The above covers for eth* and eth\* but not for eth\\* since a proper solution didn't quickly come to mind which avoids playing whack-a-mole. (E.g., does eth\\\\\\* escape the wildcard or not?)
Guess I could just count the number of backslashes immediately preceding the asterisk and return true if the sum is odd?
Quoted 6 lines
Thanks for explaining the comment, this supports eth* and eth\* which is sufficient at this stage.
For this series:
Reviewed-by Pablo Neira Ayuso <[email protected]>
Series applied, thanks!
Phil Sutter
reviewerPATCH v5 2/3 parser_bison: Accept ASTERISK_STRING in flowtable_expr_member
In reply to Phil SutterAll clauses are identical, so instead of adding a third one for ASTERISK_STRING, use a single one for 'string' (which combines all three variants).
---
Changes since v3:
- Cover interface wildcards in nft.8
--- doc/nft.txt | 30 ++++++++++++++++++++++++++---- src/parser_bison.y | 11 +---------- 2 files changed, 27 insertions(+), 14 deletions(-)
doc/nft.txt +26 -4
@@ -387,13 +387,19 @@ add table inet mytable CHAINS ------ [verse] -{*add* | *create*} *chain* ['family'] 'table' 'chain' [*{ type* 'type' *hook* 'hook' [*device* 'device'] *priority* 'priority' *;* [*policy* 'policy' *;*] [*comment* 'comment' *;*] *}*] +____ +{*add* | *create*} *chain* ['family'] 'table' 'chain' [*{ type* 'type' *hook* 'hook' ['DEVICE'] *priority* 'priority' *;* [*policy* 'policy' *;*] [*comment* 'comment' *;*] *}*] {*delete* | *destroy* | *list* | *flush*} *chain* ['family'] 'table' 'chain' *list chains* ['family'] *delete chain* ['family'] 'table' *handle* 'handle' *destroy chain* ['family'] 'table' *handle* 'handle' *rename chain* ['family'] 'table' 'chain' 'newname' +'DEVICE' := {*device* 'DEVICE_NAME' | *devices = {* 'DEVICE_LIST' *}*} +'DEVICE_LIST' := 'DEVICE_NAME' [*,* 'DEVICE_LIST'] +'DEVICE_NAME' := 'string' | 'string'*** +____ + Chains are containers for rules. They exist in two kinds, base chains and regular chains. A base chain is an entry point for packets from the networking stack, a regular chain may be used as jump target and is used for better rule
@@ -436,7 +442,7 @@ Apart from the special cases illustrated above (e.g. *nat* type not supporting * The netdev family supports merely two combinations, namely *filter* type with *ingress* hook and *filter* type with *egress* hook. Base chains in this - family also require the *device* parameter to be present since they exist per + family also require the 'DEVICE' parameter to be present since they exist per interface only. * The arp family supports only the *input* and *output* hooks, both in chains of type *filter*.
@@ -449,7 +455,13 @@ Apart from the special cases illustrated above (e.g. *nat* type not supporting The *device* parameter accepts a network interface name as a string, and is required when adding a base chain that filters traffic on the ingress or egress hooks. Any ingress or egress chains will only filter traffic from the -interface specified in the *device* parameter. +interface specified in the *device* parameter. The same base chain may be used +for multiple devices by using the *devices* parameter instead. + +With newer kernels there is also basic support for wildcards in 'DEVICE_NAME' +by specifying an asterisk suffix. The chain will apply to all interfaces +matching the given prefix. Use the *list hooks* command to see the current +status. The *priority* parameter accepts a signed integer value or a standard priority name which specifies the order in which chains with the same *hook* value are
@@ -763,11 +775,16 @@ per element comment field FLOWTABLES ----------- [verse] -{*add* | *create*} *flowtable* ['family'] 'table' 'flowtable' *{ hook* 'hook' *priority* 'priority' *; devices = {* 'device'[*,* ...] *} ; }* +____ +{*add* | *create*} *flowtable* ['family'] 'table' 'flowtable' *{ hook* 'hook' *priority* 'priority' *; devices = {* 'DEVICE_LIST' *} ; }* *list flowtables* ['family'] ['table'] {*delete* | *destroy* | *list*} *flowtable* ['family'] 'table' 'flowtable' *delete* *flowtable* ['family'] 'table' *handle* 'handle' +'DEVICE_LIST' := 'DEVICE_NAME' [*,* 'DEVICE_LIST'] +'DEVICE_NAME' := 'string' | 'string'*** +____ + Flowtables allow you to accelerate packet forwarding in software. Flowtables entries are represented through a tuple that is composed of the input interface, source and destination address, source and destination port; and layer 3/4
@@ -786,6 +803,11 @@ The *priority* can be a signed integer or *filter* which stands for 0. Addition and subtraction can be used to set relative priority, e.g. filter + 5 equals to 5. +With newer kernels there is basic support for wildcards in 'DEVICE_LIST' by +specifying an asterisk suffix. The flowtable will apply to all interfaces +matching the given prefix. Use the *list hooks* command to see the current +status. + [horizontal] *add*:: Add a new flowtable for the given family with the given name. *delete*:: Delete the specified flowtable.
src/parser_bison.y +1 -10
@@ -2473,16 +2473,7 @@ flowtable_list_expr : flowtable_expr_member | flowtable_list_expr COMMA opt_newline ; -flowtable_expr_member : QUOTED_STRING - { - struct expr *expr = ifname_expr_alloc(&@$, state->msgs, $1); - - if (!expr) - YYERROR; - - $$ = expr; - } - | STRING +flowtable_expr_member : string { struct expr *expr = ifname_expr_alloc(&@$, state->msgs, $1);
Signature · 1 line
2.49.0
Pablo Neira Ayuso
maintainerRe: [nft PATCH v5 2/3] parser_bison: Accept ASTERISK_STRING in flowtable_expr_member
Hi Phil,
Quoted 49 lines · Phil Sutter
All clauses are identical, so instead of adding a third one for
ASTERISK_STRING, use a single one for 'string' (which combines all three variants).Signed-off-by Phil Sutter <[email protected]>---
Changes since v3:
- Cover interface wildcards in nft.8--- doc/nft.txt | 30 ++++++++++++++++++++++++++---- src/parser_bison.y | 11 +---------- 2 files changed, 27 insertions(+), 14 deletions(-)doc/nft.txt +8
-2@@ -387,13 +387,19 @@ add table inet mytable CHAINS ------ [verse] -{*add* | *create*} *chain* ['family'] 'table' 'chain' [*{ type* 'type' *hook* 'hook' [*device* 'device'] *priority* 'priority' *;* [*policy* 'policy' *;*] [*comment* 'comment' *;*] *}*] +____ +{*add* | *create*} *chain* ['family'] 'table' 'chain' [*{ type* 'type' *hook* 'hook' ['DEVICE'] *priority* 'priority' *;* [*policy* 'policy' *;*] [*comment* 'comment' *;*] *}*] {*delete* | *destroy* | *list* | *flush*} *chain* ['family'] 'table' 'chain' *list chains* ['family'] *delete chain* ['family'] 'table' *handle* 'handle' *destroy chain* ['family'] 'table' *handle* 'handle' *rename chain* ['family'] 'table' 'chain' 'newname' +'DEVICE' := {*device* 'DEVICE_NAME' | *devices = {* 'DEVICE_LIST' *}*} +'DEVICE_LIST' := 'DEVICE_NAME' [*,* 'DEVICE_LIST'] +'DEVICE_NAME' := 'string' | 'string'*** +____ + Chains are containers for rules. They exist in two kinds, base chains and regular chains. A base chain is an entry point for packets from the networking stack, a regular chain may be used as jump target and is used for better rule@@ -436,7 +442,7 @@ Apart from the special cases illustrated above (e.g. *nat* type not supporting * The netdev family supports merely two combinations, namely *filter* type with *ingress* hook and *filter* type with *egress* hook. Base chains in this - family also require the *device* parameter to be present since they exist per + family also require the 'DEVICE' parameter to be present since they exist per interface only. * The arp family supports only the *input* and *output* hooks, both in chains of type *filter*.@@ -449,7 +455,13 @@ Apart from the special cases illustrated above (e.g. *nat* type not supporting The *device* parameter accepts a network interface name as a string, and is
doc/nft.txt Apart from the special cases illustrated above (e.g. *nat* type not supporting · line 464
Quoted 10 lines
doc/nft.txt +7
-1required when adding a base chain that filters traffic on the ingress or egress hooks. Any ingress or egress chains will only filter traffic from the -interface specified in the *device* parameter. +interface specified in the *device* parameter. The same base chain may be used +for multiple devices by using the *devices* parameter instead. + +With newer kernels there is also basic support for wildcards in 'DEVICE_NAME' +by specifying an asterisk suffix. The chain will apply to all interfaces +matching the given prefix. Use the *list hooks* command to see the current +status.
Maybe explain here too that newer kernels also allow to pre-register a match on unexisting devices (specify version), while old kernel fail with ENOENT?
Quoted 21 lines
doc/nft.txt +6
-1The *priority* parameter accepts a signed integer value or a standard priority name which specifies the order in which chains with the same *hook* value are@@ -763,11 +775,16 @@ per element comment field FLOWTABLES ----------- [verse] -{*add* | *create*} *flowtable* ['family'] 'table' 'flowtable' *{ hook* 'hook' *priority* 'priority' *; devices = {* 'device'[*,* ...] *} ; }* +____ +{*add* | *create*} *flowtable* ['family'] 'table' 'flowtable' *{ hook* 'hook' *priority* 'priority' *; devices = {* 'DEVICE_LIST' *} ; }* *list flowtables* ['family'] ['table'] {*delete* | *destroy* | *list*} *flowtable* ['family'] 'table' 'flowtable' *delete* *flowtable* ['family'] 'table' *handle* 'handle' +'DEVICE_LIST' := 'DEVICE_NAME' [*,* 'DEVICE_LIST'] +'DEVICE_NAME' := 'string' | 'string'*** +____ + Flowtables allow you to accelerate packet forwarding in software. Flowtables entries are represented through a tuple that is composed of the input interface, source and destination address, source and destination port; and layer 3/4
doc/nft.txt The *priority* can be a signed integer or *filter* which stands for 0. Addition · line 809
Quoted 7 lines
doc/nft.txt +4
-0@@ -786,6 +803,11 @@ The *priority* can be a signed integer or *filter* which stands for 0. Addition and subtraction can be used to set relative priority, e.g. filter + 5 equals to 5. +With newer kernels there is basic support for wildcards in 'DEVICE_LIST' by +specifying an asterisk suffix. The flowtable will apply to all interfaces +matching the given prefix. Use the *list hooks* command to see the current +status.
... to see the
hooks that are that registered per device that match on the wildcard device.
Phil Sutter
reviewerRe: [nft PATCH v5 2/3] parser_bison: Accept ASTERISK_STRING in flowtable_expr_member
Quoted 58 lines · Pablo Neira Ayuso
Hi Phil,
Quoted 55 lines · Phil Sutter
All clauses are identical, so instead of adding a third one for
ASTERISK_STRING, use a single one for 'string' (which combines all three variants).Signed-off-by Phil Sutter <[email protected]>---
Changes since v3:
- Cover interface wildcards in nft.8--- doc/nft.txt | 30 ++++++++++++++++++++++++++---- src/parser_bison.y | 11 +---------- 2 files changed, 27 insertions(+), 14 deletions(-)doc/nft.txt +11
-3@@ -387,13 +387,19 @@ add table inet mytable CHAINS ------ [verse] -{*add* | *create*} *chain* ['family'] 'table' 'chain' [*{ type* 'type' *hook* 'hook' [*device* 'device'] *priority* 'priority' *;* [*policy* 'policy' *;*] [*comment* 'comment' *;*] *}*] +____ +{*add* | *create*} *chain* ['family'] 'table' 'chain' [*{ type* 'type' *hook* 'hook' ['DEVICE'] *priority* 'priority' *;* [*policy* 'policy' *;*] [*comment* 'comment' *;*] *}*] {*delete* | *destroy* | *list* | *flush*} *chain* ['family'] 'table' 'chain' *list chains* ['family'] *delete chain* ['family'] 'table' *handle* 'handle' *destroy chain* ['family'] 'table' *handle* 'handle' *rename chain* ['family'] 'table' 'chain' 'newname' +'DEVICE' := {*device* 'DEVICE_NAME' | *devices = {* 'DEVICE_LIST' *}*} +'DEVICE_LIST' := 'DEVICE_NAME' [*,* 'DEVICE_LIST'] +'DEVICE_NAME' := 'string' | 'string'*** +____ + Chains are containers for rules. They exist in two kinds, base chains and regular chains. A base chain is an entry point for packets from the networking stack, a regular chain may be used as jump target and is used for better rule@@ -436,7 +442,7 @@ Apart from the special cases illustrated above (e.g. *nat* type not supporting * The netdev family supports merely two combinations, namely *filter* type with *ingress* hook and *filter* type with *egress* hook. Base chains in this - family also require the *device* parameter to be present since they exist per + family also require the 'DEVICE' parameter to be present since they exist per interface only. * The arp family supports only the *input* and *output* hooks, both in chains of type *filter*.@@ -449,7 +455,13 @@ Apart from the special cases illustrated above (e.g. *nat* type not supporting The *device* parameter accepts a network interface name as a string, and is required when adding a base chain that filters traffic on the ingress or egress hooks. Any ingress or egress chains will only filter traffic from the -interface specified in the *device* parameter. +interface specified in the *device* parameter. The same base chain may be used +for multiple devices by using the *devices* parameter instead. +
Quoted 8 lines
doc/nft.txt Apart from the special cases illustrated above (e.g. *nat* type not supporting · line 464
Quoted 4 lines
doc/nft.txt +4
-0+With newer kernels there is also basic support for wildcards in 'DEVICE_NAME' +by specifying an asterisk suffix. The chain will apply to all interfaces +matching the given prefix. Use the *list hooks* command to see the current +status.Maybe explain here too that newer kernels also allow to pre-register a match on unexisting devices (specify version), while old kernel fail with ENOENT?
ACK, will do!
Quoted 24 lines
Quoted 24 lines
doc/nft.txt +6
-1The *priority* parameter accepts a signed integer value or a standard priority name which specifies the order in which chains with the same *hook* value are@@ -763,11 +775,16 @@ per element comment field FLOWTABLES ----------- [verse] -{*add* | *create*} *flowtable* ['family'] 'table' 'flowtable' *{ hook* 'hook' *priority* 'priority' *; devices = {* 'device'[*,* ...] *} ; }* +____ +{*add* | *create*} *flowtable* ['family'] 'table' 'flowtable' *{ hook* 'hook' *priority* 'priority' *; devices = {* 'DEVICE_LIST' *} ; }* *list flowtables* ['family'] ['table'] {*delete* | *destroy* | *list*} *flowtable* ['family'] 'table' 'flowtable' *delete* *flowtable* ['family'] 'table' *handle* 'handle' +'DEVICE_LIST' := 'DEVICE_NAME' [*,* 'DEVICE_LIST'] +'DEVICE_NAME' := 'string' | 'string'*** +____ + Flowtables allow you to accelerate packet forwarding in software. Flowtables entries are represented through a tuple that is composed of the input interface, source and destination address, source and destination port; and layer 3/4@@ -786,6 +803,11 @@ The *priority* can be a signed integer or *filter* which stands for 0. Addition and subtraction can be used to set relative priority, e.g. filter + 5 equals to 5.
Quoted 8 lines
doc/nft.txt The *priority* can be a signed integer or *filter* which stands for 0. Addition · line 809
Quoted 4 lines
doc/nft.txt +4
-0+With newer kernels there is basic support for wildcards in 'DEVICE_LIST' by +specifying an asterisk suffix. The flowtable will apply to all interfaces +matching the given prefix. Use the *list hooks* command to see the current +status.... to see the
hooks that are that registered per device that match on the wildcard device.
LGTM!
Thanks, Phil
Phil Sutter
reviewerPATCH v5 3/3 tests: shell: Test ifname-based hooks
In reply to Phil SutterAssert that:
- Non-matching interface specs are accepted
- Existing interfaces are hooked into upon flowtable/chain creation
- A new device matching the spec is hooked into immediately
- No stale hooks remain in 'nft list hooks' output
- Wildcard hooks basically work
--- .../features/list_hooks_flowtable_info.sh | 7 +++ .../netdev_chain_name_based_hook_0.json-nft | 34 ++++++++++++++ .../dumps/netdev_chain_name_based_hook_0.nft | 5 +++ .../chains/netdev_chain_name_based_hook_0 | 44 ++++++++++++++++++ .../testcases/flowtable/0016name_based_hook_0 | 45 +++++++++++++++++++ .../dumps/0016name_based_hook_0.json-nft | 32 +++++++++++++ .../flowtable/dumps/0016name_based_hook_0.nft | 6 +++ 7 files changed, 173 insertions(+) create mode 100755 tests/shell/features/list_hooks_flowtable_info.sh create mode 100644 tests/shell/testcases/chains/dumps/netdev_chain_name_based_hook_0.json-nft create mode 100644 tests/shell/testcases/chains/dumps/netdev_chain_name_based_hook_0.nft create mode 100755 tests/shell/testcases/chains/netdev_chain_name_based_hook_0 create mode 100755 tests/shell/testcases/flowtable/0016name_based_hook_0 create mode 100644 tests/shell/testcases/flowtable/dumps/0016name_based_hook_0.json-nft create mode 100644 tests/shell/testcases/flowtable/dumps/0016name_based_hook_0.nft
tests/shell/features/list_hooks_flowtable_info.sh new file +7 -0
@@ -0,0 +1,7 @@ +#!/bin/sh + +# check for flowtable info in 'list hooks' output + +unshare -n bash -c " \ +$NFT \"table inet t { flowtable ft { hook ingress priority 0; devices = { lo }; }; }\"; \ +$NFT list hooks netdev device lo | grep -q flowtable\ inet\ t\ ft"
tests/shell/testcases/chains/dumps/netdev_chain_name_based_hook_0.json-nft new file +34 -0
@@ -0,0 +1,34 @@ +{ + "nftables": [ + { + "metainfo": { + "version": "VERSION", + "release_name": "RELEASE_NAME", + "json_schema_version": 1 + } + }, + { + "table": { + "family": "netdev", + "name": "t", + "handle": 0 + } + }, + { + "chain": { + "family": "netdev", + "table": "t", + "name": "c", + "handle": 0, + "dev": [ + "foo*", + "lo" + ], + "type": "filter", + "hook": "ingress", + "prio": 0, + "policy": "accept" + } + } + ] +}
tests/shell/testcases/chains/dumps/netdev_chain_name_based_hook_0.nft new file +5 -0
@@ -0,0 +1,5 @@ +table netdev t { + chain c { + type filter hook ingress devices = { "foo*", "lo" } priority filter; policy accept; + } +}
tests/shell/testcases/chains/netdev_chain_name_based_hook_0 new file +44 -0
44 lines not on this page · raw message
tests/shell/testcases/flowtable/0016name_based_hook_0 new file +45 -0
45 lines not on this page · raw message
tests/shell/testcases/flowtable/dumps/0016name_based_hook_0.json-nft new file +32 -0
@@ -0,0 +1,32 @@ +{ + "nftables": [ + { + "metainfo": { + "version": "VERSION", + "release_name": "RELEASE_NAME", + "json_schema_version": 1 + } + }, + { + "table": { + "family": "ip", + "name": "t", + "handle": 0 + } + }, + { + "flowtable": { + "family": "ip", + "name": "ft", + "table": "t", + "handle": 0, + "hook": "ingress", + "prio": 0, + "dev": [ + "foo*", + "lo" + ] + } + } + ] +}
tests/shell/testcases/flowtable/dumps/0016name_based_hook_0.nft new file +6 -0
@@ -0,0 +1,6 @@ +table ip t { + flowtable ft { + hook ingress priority filter + devices = { "foo*", "lo" } + } +}
Signature · 1 line
2.49.0