-----"Jason Gunthorpe" <jgg@xxxxxxxx> wrote: ----- >To: "Bernard Metzler" <bmt@xxxxxxxxxxxxxx> >From: "Jason Gunthorpe" <jgg@xxxxxxxx> >Date: 06/03/2019 07:50PM >Cc: linux-rdma@xxxxxxxxxxxxxxx >Subject: [EXTERNAL] Re: [PATCH for-next v1 00/12] SIW: Software iWarp >RDMA (siw) driver > >On Sun, May 26, 2019 at 01:41:44PM +0200, Bernard Metzler wrote: >> This patch set contributes the SoftiWarp driver rebased for >> Kernel 5.2-rc1. SoftiWarp (siw) implements the iWarp RDMA >> protocol over kernel TCP sockets. The driver integrates with >> the linux-rdma framework. >> >> With this new driver version, the following things where >> changed, compared to the v8 RFC of siw: >> >> o Rebased to 5.2-rc1 >> >> o All IDR code got removed. >> >> o Both MR and QP deallocation verbs now synchronously >> free the resources referenced by the RDMA mid-layer. >> >> o IPv6 support was added. >> >> o For compatibility with Chelsio iWarp hardware, the RX >> path was slightly reworked. It now allows packet intersection >> between tagged and untagged RDMAP operations. While not >> a defined behavior as of IETF RFC 5040/5041, some RDMA hardware >> may intersect an ongoing outbound (large) tagged message, such >> as an multisegment RDMA Read Response with sending an untagged >> message, such as an RDMA Send frame. This behavior was only >> detected in an NVMeF setup, where siw was used at target side, >> and RDMA hardware at client side (during file write). siw now >> implements two input paths for tagged and untagged messages each, >> and allows the intersected placement of both messages. >> >> o The siw kernel abi file got renamed from siw_user.h to siw-abi.h. >> >> Many thanks for reviewing and testing the driver, especially to >> Steve, Leon, Jason, Doug, Olga, Dennis, Gal. You all helped to >> significantly improve the siw driver over the last year. It is >> very much appreciated. > >You need to open a PR for rdma-core before this can be merged with >the >userspace part. > >Jason > > OK I created PR #536, which adds the siw user library to rdma-core. Unfortunately, when uploading, travis brought up many issues with atomics etc. Is there a good way to have the very same strict checking locally, since local build was always successful... In any case, sorry for abusing the PR procedure for code cleanup (amending commits and force push cycles)! Many thanks, Bernard.
