On Tue, 2007-09-25 at 20:59 +0800, Yan Zheng wrote: > Hello, > > I made a mistake in previous reply. Inline data are convert into > extent after a file is truncated , so there is no need to check > inode->i_size in btrfs_get_extent when encounter inline data. The > remaining advantage of truncating inline data is that inline data > doesn't leak on crash. I don't know whether it's still worth a change. > If it's worth it, here is the patch. >
Thanks! I'll put this in the queue after my large blocksize work is done. -chris _______________________________________________ Btrfs-devel mailing list [email protected] http://oss.oracle.com/mailman/listinfo/btrfs-devel
