On 10/18/13 7:43 AM, Thavatchai Makphaibulchoke wrote: > On 10/03/2013 09:22 PM, Andreas Dilger wrote: > >> >> This is a pretty significant price to pay for SELinux in any case. >> I guess it is probably lower overhead with 256-byte inodes, but >> anything that adds 5-25% overhead shouldn't be taken for granted. >> >> Cheers, Andreas >> > > So far we have determined that, > > - With SELinux enabled a ramdisk filesytem, with default node size of 128 bytes, extended xattr is generated > - In one of the aim7 workload, mbcache has a hit ratio of about 65% > > Seems like mbcache itself and the mbcache lock optimization attempt > by the patch could actually improve performance for a real word > system where extended xattr is employed. I believe the patch should > go in if there isn't any concern. But the 128-byte inode size is not common, so in general, mbcache isn't exercised that often. (still, it is there, and if there's a nice scalability patch for it, it's probably worth it). -Eric > Please let me know if there is any comment or concern about the patch. > > Thanks, > Mak. > > > -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
