On 11/03/2021 13:13, Xueming Li wrote: > This patch updates kvargs parser to support value of multiple lists or > ranges: > k1=v[1,2]v[3-5] > > Signed-off-by: Xueming Li <xuemi...@nvidia.com> > Acked-by: Viacheslav Ovsiienko <viachesl...@nvidia.com> > Acked-by: Thomas Monjalon <tho...@monjalon.net> > --- > app/test/test_kvargs.c | 46 +++++++++++++-- > lib/librte_kvargs/rte_kvargs.c | 101 +++++++++++++++++++++++---------- > 2 files changed, 112 insertions(+), 35 deletions(-) > Hi folks, This is essentially an FYI. This change introduced a weak functional ABI regression into DPDK. When I test librte_kvargs using the v20.11 unit test binary I get the following. EAL: Detected 80 lcore(s) EAL: Detected 2 NUMA nodes EAL: Detected shared linkage of DPDK EAL: Multi-process socket /var/run/dpdk/rte/mp_socket EAL: Selected IOVA mode 'PA' EAL: 4090 hugepages of size 2097152 reserved, but no mounted hugetlbfs found for that size APP: HPET is not enabled, using TSC as default timer RTE>>kvargs_autotest == test valid case == == test invalid case == rte_kvargs_parse() returned 0 (but should not) while processing <foo=1,foo=> using valid_keys=<foo,check> Test Failed RTE>> The reason this is failing in v20.11 and passing at the HEAD, is that "no value" test case (foo=1,foo=) were removed from test_invalid_kvargs, and was added to test_valid_kvargs. So tokens that were definitely invalid in v20.11, are now is valid in v21.11. My 2c is that as long as the valid values in v20.11 are still OK, there is no regression. Thanks, Ray K
