On Wed, Jun 15, 2016 at 02:50:07PM +0000, Trond Myklebust wrote: > Fair enough. So it sounds as if you are suggesting we can just drop using inode_dio_*() altogether? Maybe. The big question is how a direct write vs truncate race is handled. Either way I think a patch like this deserves a detailed analysis documented in the changelog and/or comments in the code. -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html