On Wed, 2011-04-27 at 12:50 -0500, James Bottomley wrote: > To test the theory, Chris asked me to try with data=ordered. > Unfortunately, the deadlock still shows up. This is what I get. As another data point: I'm trying the same kernel with CONFIG_PREEMPT enabled. This time the deadlock doesn't happen. Instead, kswapd0 gets pegged at 99% CPU for much of the untar, but it does eventually complete. James -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>