Our paper about rr has been accepted to appear at the Usenix Annual Technical Conference. Thanks to Dawn for suggesting that conference, and to the ATC program committee for accepting it :-). I'm scheduled to present it on July 13. The paper is similar to the version we previously uploaded to arXiv.
Some of the reviewers requested more material: additional technical details, explanations of some of our design choices compared to alternatives, and reflection on our "real world" experience with rr. There wasn't space for that within the conference page limits, so our shepherd suggested publishing a tech-report version of the paper with the extra content. I've done that and uploaded "Engineering Record And Replay For Deployability: Extended Technical Report". I hope some people find it interesting and useful.