nft: Fix add_bitwise_u16() on Big Endian
Phil Sutter
reviewerPATCH 1/1 nft: Fix add_bitwise_u16() on Big Endian
Type used for 'mask' and 'xor' parameters was wrong, 'int' is four bytes on 32 or 64 bit architectures. After casting a uint16_t to int, on Big Endian the first two bytes of data are (the leading) zero which libnftnl then copies instead of the actual value.
This problem was noticed when using '--fragment' option:
| # iptables-nft -A FORWARD --fragment -j ACCEPT | # nft list ruleset | grep frag-off | ip frag-off & 0 != 0 counter packets 0 bytes 0 accept
With this fix in place, the resulting nft rule is correct:
| ip frag-off & 8191 != 0 counter packets 0 bytes 0 accept
--- iptables/nft-shared.c | 2 +- iptables/nft-shared.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
iptables/nft-shared.c +1 -1
@@ -69,7 +69,7 @@ void add_payload(struct nftnl_rule *r, int offset, int len, uint32_t base) }/* bitwise operation is = sreg & mask ^ xor */ -void add_bitwise_u16 (struct nftnl_rule *r,int mask,int xor) +void add_bitwise_u16 (struct nftnl_rule *r,uint16_t mask,uint16_t xor) {struct nftnl_expr *expr;
iptables/nft-shared.h +1 -1
@@ -111,7 +111,7 @@ struct nft_family_ops {void add_meta (struct nftnl_rule *r,uint32_t key);void add_payload (struct nftnl_rule *r,int offset,int len,uint32_t base);void add_bitwise (struct nftnl_rule *r,uint8_t *mask,size_t len); -void add_bitwise_u16 (struct nftnl_rule *r,int mask,int xor); +void add_bitwise_u16 (struct nftnl_rule *r,uint16_t mask,uint16_t xor);void add_cmp_ptr (struct nftnl_rule *r,uint32_t op,void *data,size_t len);void add_cmp_u8 (struct nftnl_rule *r,uint8_t val,uint32_t op);void add_cmp_u16 (struct nftnl_rule *r,uint16_t val,uint32_t op);
Signature · 1 line
2.23.0
Quoted 11 lines · Phil Sutter
Type used for 'mask' and 'xor' parameters was wrong, 'int' is four bytes on 32 or 64 bit architectures. After casting a
uint16_tto int, on Big Endian the first two bytes of data are (the leading) zero which libnftnl then copies instead of the actual value.This problem was noticed when using '--fragment' option:
| # iptables-nft -A FORWARD --fragment -j ACCEPT | # nft list ruleset | grep frag-off | ip frag-off & 0 != 0 counter packets 0 bytes 0 accept
Quoted 6 lines
With this fix in place, the resulting nft rule is correct:
| ip frag-off & 8191 != 0 counter packets 0 bytes 0 accept
Fixes 2f1fbab671576 ("iptables: nft: add -f support")Signed-off-by Phil Sutter <[email protected]>