[AMD Official Use Only - AMD Internal Distribution Only] <snipped>
> > On Wed, 11 Sep 2024 03:13:14 +0000 > "Varghese, Vipin" <vipin.vargh...@amd.com> wrote: > > > > Agreed. This one of those cases where the existing project hwloc > > > which is part of open-mpi is more complete and well supported. It > > > supports multiple OS's and can deal with more quirks. > > > > Thank you Stephen for the inputs, last year when checked hwloc for distros > there were anomalies for NUMA and Physical socket Identification on AMD > EPYC Soc. > > I will recheck the distros version of hwloc, if these work out fine I will > > re- > work with hwloc libraries making it OS independent too. > > DPDK doesn't exist to resolve problems with upstreaming hardware support in > other packages. Stephen, thank you for your comment. You have asked in earlier request to try using hwloc library. I have mentioned at least till last year popular distros were still using older version for hwloc and I will recheck if it has changed. This I assumed for mitigating the `sysfs` and bringing in `OS` portability. If DPDK does supports something only because it is harder, > slower, more painful to deal with another project; then you create long term > technical debt. Now I really confused, for any application or end user which want to use the right set of lcores for performance DPDK provides abstraction and ease of use. Maybe I am misreading the above idea, so let me back off a bit and share version-2 of the RFC.
