On Sun, 10 Jun 2018 16:46:42 +0530 Jerin Jacob <jerin.ja...@caviumnetworks.com> wrote:
> -----Original Message----- > > Date: Fri, 8 Jun 2018 11:11:54 -0700 > > From: Stephen Hemminger <step...@networkplumber.org> > > To: "Wiles, Keith" <keith.wi...@intel.com> > > Cc: "dev@dpdk.org" <dev@dpdk.org>, Stephen Hemminger > > <sthem...@microsoft.com> > > Subject: Re: [dpdk-dev] [PATCH v10 1/5] eal: add rte_uuid support > > > > On Fri, 8 Jun 2018 17:15:04 +0000 > > "Wiles, Keith" <keith.wi...@intel.com> wrote: > > > > > > On Jun 8, 2018, at 9:59 AM, Stephen Hemminger > > > > <step...@networkplumber.org> wrote: > > > > > > > > Since uuid functions may not be available everywhere, implement > > > > uuid functions in DPDK. These are based off the BSD licensed > > > > libuuid in util-link. > > > > > > > > Signed-off-by: Stephen Hemminger <sthem...@microsoft.com> > > > > --- > > > > lib/librte_eal/bsdapp/eal/Makefile | 1 + > > > > lib/librte_eal/common/Makefile | 2 +- > > > > lib/librte_eal/common/eal_common_uuid.c | 193 +++++++++++++++++++++++ > > > > lib/librte_eal/common/include/rte_uuid.h | 129 +++++++++++++++ > > > > lib/librte_eal/common/meson.build | 2 + > > > > lib/librte_eal/linuxapp/eal/Makefile | 1 + > > > > lib/librte_eal/rte_eal_version.map | 9 ++ > > > > 7 files changed, 336 insertions(+), 1 deletion(-) > > > > create mode 100644 lib/librte_eal/common/eal_common_uuid.c > > > > create mode 100644 lib/librte_eal/common/include/rte_uuid.h > > > > > > > > > > Hi Stephen, Why does this need to be in EAl/Common would this be better > > > in the lib directory for utils or string routines? Does the EAl use the > > > feature for something? > > > > > > Regards, > > > Keith > > > > > > > It could be anywhere. Not tied to being in EAL common but that is where the > > PCI parsing code is. > > Still not 100% sure we need our own version of this simple code. It is more > > about BSD and eventually Windows support. > > > IMO, It is useful to add it in common EAL as some drivers can also use it. > But, I think, we must add unit test case for this new API. > I don't see much point in going through the effort of writing unit test since it is just a straight copy of Ted's uuid code which has been around for a long time.
