[dpdk-dev] [dpdk-stable] [PATCH] test/eal: do not scan PCI devices for memory tests

David Marchand david.marchand at redhat.com
Thu Aug 1 15:26:33 CEST 2019


On Thu, Aug 1, 2019 at 3:24 PM David Marchand <david.marchand at redhat.com> wrote:
>
> Hey Gaëtan,
>
> On Thu, Aug 1, 2019 at 3:22 PM Gaëtan Rivet <gaetan.rivet at 6wind.com> wrote:
> >
> > On Thu, Aug 01, 2019 at 02:29:21PM +0200, David Marchand wrote:
> > > On Thu, Aug 1, 2019 at 2:28 PM David Marchand <david.marchand at redhat.com> wrote:
> > > >
> > > > The memory tests currently check that, for normal mode (not legacy mode),
> > > > there is no memory left behind when exiting.
> > > >
> > > > The problem is that if a ethdev port is allocated when scanning pci
> > > > devices (even if the driver probe fails like when you have a virtio
> > > > management interface attached to the kernel), on exit, dpdk won't free
> > > > the associated memory since ethdev never frees the ethdev memzone.
> > > >
> > > > Workaround this by disabling pci scan.
> > >
> > > Not entirely happy with this patch.
> > > I am open to suggestions :-)
> > >
> >> Why not cleanup on .fini the ethdev subsystem?
>
> I had this in mind, tried it quickly, but it still failed.
> So I suppose .fini is executed after rte_eal_cleanup (where the
> freeing of the hugepages happens).

Or we move rte_eal_cleanup in a .fini and play with priorities... too
dangerous for 19.08, from my pov.


-- 
David Marchand


More information about the dev mailing list