On Oct 17, 2013, at 8:07 AM, bram.speecka...@gmail.com wrote: > I've already narrowed down the second issue back in February by compiling > revisions myself (as much as I could, many of the revisions in between won't > build) and reported my findings in bug 812871.
Aside that I was going to mail off-list, but it might help others: thanks for doing that! Narrowing regression ranges is a pain, but extremely helpful in finding problems. Doing it by building on each bisect is extra-exciting, and I appreciate your work, there. If you (or others reading this) are helping to hunt down regression ranges in the future, consider this my regularly scheduled pitch for mozregression. http://harthur.github.io/mozregression/ It takes a minute to learn how it works, but it will bisect through nightlies and let you say "yes, good" or "no, bad" until it finds a one-day range, and then give you an hg-link to the range itself. Awfully helpful, and yet I'm often surprised to find that people don't know about it. J --- Johnathan Nightingale VP Firefox @johnath _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
