diff options
| author |   Benoît Ganne <bganne@cisco.com> | 2020-08-31 18:59:34 +0200 |
|---|---|---|
| committer |   Damjan Marion <dmarion@me.com> | 2020-09-01 12:03:27 +0000 |
| commit | 6e334e3e77bb156a9317a37500077a218a04f7a3 (patch) | |
| tree | fa2caf364ba7d99ca7e3895c37bb1047c6be8c0f /src/vnet/ip/ip4_format.c | |
| parent | 1d104c5ecdce37301fdfea0e62a533a2e5342ee0 (diff) | |
ip: fix ip zero checksum verification
In one's complement, there are two representations of zero: the all
zero and the all one bit values, often referred to as +0 and -0. See
RFC 1624 section 3 for more details.
This used to be taken care of in ip4_header_checksum(), but it is no
longer the case. The check ip->checksum == ip4_header_checksum (ip) is
no longer correct in the -0 case.
Always use ip4_header_checksum_is_valid() instead (which behaves
correctly since 9a79a1ab931c3b5a7ae07d6f0fcfef7c4368a2c4).
Type: fix
Fixes: e5f0050c7a5d411f96af6401797529d58825e2af
Change-Id: Iacc6b60645a834287b085aecb9e3fdb4554cf0cf
Signed-off-by: Benoît Ganne <bganne@cisco.com>
Diffstat (limited to 'src/vnet/ip/ip4_format.c')
| -rw-r--r-- | src/vnet/ip/ip4_format.c | 7 |
1 files changed, 4 insertions, 3 deletions
diff --git a/src/vnet/ip/ip4_format.c b/src/vnet/ip/ip4_format.c index 786a01d396b..c6639b20716 100644 --- a/src/vnet/ip/ip4_format.c +++ b/src/vnet/ip/ip4_format.c @@ -150,9 +150,10 @@ format_ip4_header (u8 * s, va_list * args) /* Check and report invalid checksums. */ { - u16 c = ip4_header_checksum (ip); - if (c != ip->checksum) - s = format (s, " (should be 0x%04x)", clib_net_to_host_u16 (c)); + if (!ip4_header_checksum_is_valid (ip)) + s = + format (s, " (should be 0x%04x)", + clib_net_to_host_u16 (ip4_header_checksum (ip))); } s = format (s, " dscp %U ecn %U", |