On Wed, Jan 24, 2018 at 01:51:41PM +0200, Leon Romanovsky wrote: > On Wed, Jan 24, 2018 at 06:35:20PM +0800, Liuyixian (Eason) wrote: > > > > > > On 2018/1/23 23:04, Jason Gunthorpe wrote: > > > On Tue, Jan 23, 2018 at 08:51:09PM +0800, Liuyixian (Eason) wrote: > > >> Could you please give me further comments on that email? If you don't > > >> agree with what I said in that email, can you give some suggestions > > >> to resolve compatibility issue? > > > > > > You can't have a not compatible in your table. > > > > > > You need to add some kind of mechanism for userspace to enable the new > > > features conditionally. > > > > > > Jason > > > > > Thanks, I will consider your suggestion in next version patch set. > > ABI_VERSION cam help you to achieve that. ABI_VERSION is mostly totally broken in verbs and breaks compatability with. It is a last emergency resort. Probably in this case you just need to lengthen the response udata struct and the kernel can detect if the longer struct is provided to enable the feature and userspace can detect !0 to determine if the kernel supports it. Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
