Folks, Here's a couple of fixes for regressions reported in the past couple of days. They've passed xfstests smoke testing here. I'm not really certain whether the direct Io deadlock fix is the best way to fix the problem, but I can't think of any other way of doing it at this point because we have no way of passing a transaction context to the direct IO allocation callback path. If anyone has any bright ideas, I'm all ears.... Cheers, Dave. _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
