> -----Original Message----- > From: Jakub Kicinski <k...@kernel.org> > Sent: Tuesday, July 21, 2020 11:05 AM > To: Nguyen, Anthony L <anthony.l.ngu...@intel.com> > Cc: da...@davemloft.net; Michael, Alice <alice.mich...@intel.com>; > netdev@vger.kernel.org; nhor...@redhat.com; sassm...@redhat.com; > Kirsher, Jeffrey T <jeffrey.t.kirs...@intel.com>; Brady, Alan > <alan.br...@intel.com>; Burra, Phani R <phani.r.bu...@intel.com>; Hay, > Joshua A <joshua.a....@intel.com>; Chittim, Madhu > <madhu.chit...@intel.com>; Linga, Pavan Kumar > <pavan.kumar.li...@intel.com>; Skidmore, Donald C > <donald.c.skidm...@intel.com>; Brandeburg, Jesse > <jesse.brandeb...@intel.com>; Samudrala, Sridhar > <sridhar.samudr...@intel.com> > Subject: Re: [net-next v4 06/15] iecm: Implement mailbox functionality > > On Mon, 20 Jul 2020 17:38:01 -0700 Tony Nguyen wrote: > > + (cq->next_to_use)++; > > + if (cq->next_to_use == cq->ring_size) > > + cq->next_to_use = 0; > > + } > > + > > + /* Force memory write to complete before letting hardware > > + * know that there are new descriptors to fetch. > > + */ > > + iecm_wmb(); > > dma_wmb() would probably be sufficient here? > > > + wr32(hw, cq->reg.tail, cq->next_to_use);
We will investigate and see if dma_wmb is sufficient. We do have a test for triggering the weak-ordered issue so we should be able to tell for certain. The reasoning makes sense dma_wmb is sufficient but we want to test to be sure. Alan
