![]() | |||||||||||||
Messages in this thread |
Hi Ted, On Thu, Jun 15, 2006 at 01:49:31AM -0400, Theodore Tso wrote: > On Thu, Jun 15, 2006 at 07:50:19AM +1000, Nathan Scott wrote: > > As to whether a new inode operation is useful/needed - *shrug* - not > > really my call, I was saying we can work with whatever ends up being > > the final solution, provided it keeps per-inode granularity. > > XFS should be return a per-inode value for st_blksize by simply setting > kstat->st_blksize in linvfs_getattr() found in fs/xfs/linux-2.6/xfs_iops.c. Hmm, you're looking at an older kernel there I guess - that routine doesn't exist anymore. Look at -mm, its the most recent version. > in the Irix/Linux compatibility layer (yuck, I still can't believe > this got into mainline), There is no IRIX/Linux compatibility layer, you're misunderstanding the code (which is understandable, its erm a bit crufty in places). There's a fair bit of history there too - *shrug* - its on the improve but its a complex body of code and takes time to mutate. We've tended to focus on real problems and needed features in the past, moreso than cosmetics, but thats getting attention nowadays too. > I'll let the XFS maintainers decide how put in per-inode st_blksize > values --- but it is definitely doable. Yep, agreed. cheers. -- Nathan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||||
| Last update: 2006-06-15 09:04 [from the cache] ©2003-2008 | |||||||||||||