On Thu, 20 Jun 2013, Chris Mason wrote: > Quoting Sage Weil (2013-06-20 21:00:21) > > On Thu, 20 Jun 2013, Chris Mason wrote: > > > Awesome, thanks for getting the traces for us. Looks like this one has > > > been around since v3.7, so I'm not going to try and sneak it into the > > > 3.10 final. I'll have it in the next merge window and for stable. > > > > Weird, these same tests have been running on it nightly for ages and it > > seems like these failures just started with 3.9. Perhaps some other > > change made it hang when it didn't before? > > > > It's always possible, there are a ton of moving pieces here. The > wait_event you were hung on was waiting for crcs to finish, and that > part at least isn't new.
K. There was also a shift of writes to leveldb (which does the mmap thing), so that may explain the change in behavior. > Somewhat unrelated, but are you still using notreelog? Nope, just noatime. Thanks- sage -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html