> -----Original Message----- > From: Stephen Hemminger [mailto:[email protected]] > Sent: Wednesday, January 11, 2017 5:36 PM > To: Ananyev, Konstantin <[email protected]> > Cc: Sergey Vyazmitinov <[email protected]>; [email protected]; > Yigit, Ferruh <[email protected]>; [email protected] > Subject: Re: [dpdk-dev] [PATCH] kni: use bulk functions to allocate and free > mbufs > > On Wed, 11 Jan 2017 17:28:21 +0000 > "Ananyev, Konstantin" <[email protected]> wrote: > > > > -----Original Message----- > > > From: dev [mailto:[email protected]] On Behalf Of Stephen Hemminger > > > Sent: Wednesday, January 11, 2017 4:18 PM > > > To: Sergey Vyazmitinov <[email protected]> > > > Cc: [email protected]; Yigit, Ferruh <[email protected]>; > > > [email protected] > > > Subject: Re: [dpdk-dev] [PATCH] kni: use bulk functions to allocate and > > > free mbufs > > > > > > On Fri, 30 Dec 2016 04:50:16 +0700 > > > Sergey Vyazmitinov <[email protected]> wrote: > > > > > > > /** > > > > + * Free n packets mbuf back into its original mempool. > > > > + * > > > > + * Free each mbuf, and all its segments in case of chained buffers. > > > > Each > > > > + * segment is added back into its original mempool. > > > > + * > > > > + * @param mp > > > > + * The packets mempool. > > > > + * @param mbufs > > > > + * The packets mbufs array to be freed. > > > > + * @param n > > > > + * Number of packets. > > > > + */ > > > > +static inline void rte_pktmbuf_free_bulk(struct rte_mempool *mp, > > > > + struct rte_mbuf **mbufs, unsigned n) > > > > +{ > > > > + struct rte_mbuf *mbuf, *m_next; > > > > + unsigned i; > > > > + for (i = 0; i < n; ++i) { > > > > + mbuf = mbufs[i]; > > > > + __rte_mbuf_sanity_check(mbuf, 1); > > > > + > > > > + mbuf = mbuf->next; > > > > + while (mbuf != NULL) { > > > > + m_next = mbuf->next; > > > > + rte_pktmbuf_free_seg(mbuf); > > > > + mbuf = m_next; > > > > + } > > > > + } > > > > + rte_mempool_put_bulk(mp, (void * const *)mbufs, n); > > > > +} > > > > > > The mbufs may come from different pools. You need to handle that. > > > > I suppose both stituations are possible: > > 1) user knows off-hand that all mbufs in the group are from the same mempool > > 2) user can't guarantee that all mbufs in the group are from same mempool. > > > > As I understand that patch is for case 1) only. > > For 2) it could be a separate function and separate patch. > > > > Konstantin > > > > > > Please don't make unnecessary assumptions in pursuit of minor optimizations.
I don't suggest to make *any* assumptions. What I am saying we can have 2 functions for two different cases. Obviously we'll have to document it properly. Konstantin > It is trivial to write a correct free bulk that handles pool changing. > Also the free_seg could be bulked as well.
