On Tue, Oct 28, 2014 at 11:07 AM, Andy Stormont < astorm...@racktopsystems.com> wrote:
> The import code in libzfs could also use some work. If I remember > correctly the code that examines disk slices is O(n * n). Though it’s > broken it seems to work okay on machines with <= 32 cores but with more > than that you’ll likely run into the assert at the end of zpool_open_func - > which happens when a thread examining a slice realises another thread has > already marked the slice as not containing a pool after it’s started. > > I had a go at rewriting the code earlier this year to fix those issues > which you can find here: http://cr.illumos.org/~webrev/andy_js/zpool/ > I’m not sure if this code works or even if it’s the latest version but if > there’s interest I can start looking at it again. > Yes, it's a problem if the zpool import doesn't work. Have you tested your fix now? It looks like you are fixing 3 issues: 1. enabling new warnings 2. race condition when doing zpool import 3. a different method of discovering devices Can you break up your review into 3 reviews, one for each issue? --matt
_______________________________________________ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer
