Hi, Yes , Matan and I need to work together and revisit this patch in light of the split patch series and remove any references to RoCE v2...
Thanks for the feedback Jason and apologies for the oversight, we should have worked this out internally before sending out V5 Regards Som > -----Original Message----- > From: Jason Gunthorpe [mailto:[email protected]] > Sent: Thursday, June 11, 2015 9:41 AM > To: Matan Barak > Cc: Doug Ledford; Or Gerlitz; Moni Shoua; Sean Hefty; Somnath Kotur; linux- > [email protected]; Somnath Kotur; Devesh Sharma > Subject: Re: [PATCH for-next V5 12/12] RDMA/ocrdma: Changes in driver to > incorporate the moving of GID Table mgmt to IB/Core. > > On Mon, Jun 08, 2015 at 05:12:15PM +0300, Matan Barak wrote: > > From: Somnath Kotur <[email protected]> > > > > 1.Check and set port capability flags to indicate RoCEV2 support. > > ??? This series has nothing to with rocev2 now, what is this about? > > > mutex_init(&dev->dev_lock); > > - dev->sgid_tbl = kzalloc(sizeof(union ib_gid) * > > - OCRDMA_MAX_SGID, GFP_KERNEL); > > Should sgid_tbl be dropped from the structure? > > > +int ocrdma_modify_gid(struct ib_device *ibdev, u8 port_num, unsigned > int index, > > + const union ib_gid *gid, const struct ib_gid_attr *attr, > > + void **context) > > +{ > > + struct ocrdma_dev *dev; > > + > > + dev = get_ocrdma_dev(ibdev); > > > > return 0; > > } > > Empty modify gid? Shouldn't it be completely empty? > > This is correct? This HW sends the full SGID in the WQE? > > > +enum { > > + OCRDMA_L3_TYPE_IB_GRH = 0x00, > > + OCRDMA_L3_TYPE_IPV4 = 0x01, > > + OCRDMA_L3_TYPE_IPV6 = 0x02 > > +}; > > These added constants are not used? Probably others as well? > > Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
