Hi Bernard, > Much appreciated! > These are quite some patches, and I will need some time > to go through. Would bee nice if those would be broken > down into smaller bundles (introduce non-blocking connect, > _siw_cep_close() subroutine, fixing cep reference counting, > smp_mb() after STag invalidation, ..). They mostly fall out naturally getting one step further with each commit. So most of them depend on each other. I'll see if I can reorder some of them, but I'm not sure it's really worth the effort. > Anyway, many thanks for the effort, Thanks a lot for the review. > it will improve the driver! Yes! On top I have some code to support MPA rev1 in peer_to_peer mode in order to interoperate with a Chelcio T404-BT card running under Windows. In preparation I've code that moves the currently hardcoded values (which where module params before) into a per device structure, some like 'sdev->options.crc_strict'. With that we only need to find a good way to pass these parameters from userspace to the device. I guess that should be done somehow via the 'rdma link add' command, or via files similar to /proc/sys/net/ipv4/conf/*. Here's my branch with all (partly incomplete) siw changes: https://git.samba.org/?p=metze/linux/wip.git;a=shortlog;h=refs/heads/rdma-next-siw > First comments: > > A non blocking connect does really makes sense as you > are pointing out. I hope it doesn't complicate the CM > code even further. > > I think we agreed upon not using BUG() and BUG_ON(), > so please don't introduce it. Ok. > 'I hit a lot of bugs' is not very helpful, but just > a statement. More details are in the individual commit messages, should I double them in the cover letter next time? Currently I'm quite busy with other stuff... I hope to find some time in the next weeks to comment more detailed and post a new revision. metze
