Hi Huichao, > According to RFC791,the options may appear or not in datagrams. > They must be implemented by all IP modules (host and gateways). > What is optional is their transmission in any particular datagram, > not their implementation.So we have to deal with it during the > fragmenting process.Add some test data for the IPv4 header optional > field fragmenting. > > Signed-off-by: Huichao Cai <chcch...@163.com> > --- > app/test/test_ipfrag.c | 319 > ++++++++++++++++++++++++++++++++--- > lib/ip_frag/rte_ip_frag.h | 6 + > lib/ip_frag/rte_ipv4_fragmentation.c | 70 +++++++- > 3 files changed, 372 insertions(+), 23 deletions(-) > > diff --git a/app/test/test_ipfrag.c b/app/test/test_ipfrag.c > index 1ced25a..f6ff2d0 100644 > --- a/app/test/test_ipfrag.c > +++ b/app/test/test_ipfrag.c > @@ -18,10 +18,109 @@ > #define NUM_MBUFS 128 > #define BURST 32 > > +#define IPV4_IPOPT_MANUAL > + > +#ifdef IPV4_IPOPT_MANUAL
Could you explain why do we need that define at all? As I can read the code, right now IPV4_IPOPT_MANUAL is always defined, so all '#else' blocks are simply dead code. Is there any reason to keep it? If so, then the code probably need to be re-ordered somehow, to make '#else' part to be enabled and executed: let say a separate test-case(s), and/or separate function or extra parameter for test_get_ipv4_opt(). Konstantin
