On Tue, Jan 13, 2015 at 10:13:34AM +0900, Mike Hommey wrote: > Hi, > > I just landed bug 762449 on inbound, which enables Jemalloc 3 on the > nightly channel. This puts us a step forward to removing mozjemalloc, > our highly patched fork of jemalloc 0.9. More importantly, it will allow > us to more closely work with upstream, by proposing patches that do > apply there (which is not the case every time we do changes to > mozjemalloc), or taking improvements from upstream (which, for some, we > backported to mozjemalloc, with pain). > > For now, the change is such that it won't ride the trains, because there > are a few known "regressions" that need to be addressed first, but I'd > expect them to be fixed by next merge-day. > > From a developer point of view, this shouldn't change anything. > > Thanks to Guilherme Gonçalves, Eric Rahm, and Emanuel Hoogeveen for the > hard work.
By now, we're at Jemalloc 4, and I just turned it on again for the third time (not riding the trains yet, like back then). There are known performance regressions, let's see how they turn out in reality (comparing talos results on try is not entirely reliable yet to get the full picture) and see where we go from there. Third try's the charm? Mike _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
