Hi folks, This patch set builds on 3.2.0-alpha1, and brings full v5 filesystem support to xfs_db, including xfs_metadump. xfs_check is still not supported, though that is not a major issue at this point as it is deprecated. The first part of the patch series fixes a couple of minor bugs, followed by syncing up with the kernel code to match the series I just posted for 3.13. This is necessary for CRC support in xfs_db. The rest of the series is converting xfs_db to libxfs for it's IO engine and sorting out all the issues that this conversion uncovers. It then enables CRC verification and calculation, and with that done re-enables write support for v5 filesystems. The final patches enable large inode clusters on v5 filesytems, and remove the experimental warning from the v5 superblock verification. This last change was made so that we no longer have noise hiding test failures and is based on the premise that when we release 3.2.0 the v5 filesystem support in xfsprogs will no longer be experimental - it will be fully supported for production systems. Killing the output now allows us to see real failures instead of noise.... Comments, thoughts, and flames welcome. Cheers, Dave. _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs