On Nov 4, 2013, at 2:57 PM, Dave Chinner <david@xxxxxxxxxxxxx> wrote: > On Sun, Nov 03, 2013 at 06:42:00PM -0500, Theodore Ts'o wrote: >> On Mon, Nov 04, 2013 at 10:14:42AM +1100, Dave Chinner wrote: >>> >>> FIEMAP is not the correct interface for data modifying operations. >>> It is an interface that returns information about file metadata (i.e >>> the layout of a file) - it is not an interface for modifying the >>> contents of the file. >> >> Well, it's been argued that FIEMAP was designed to be extensible, and >> we should use it for other operations. Where the bounds of that >> stretches to is certainly an arguable point, and I can understand your >> observation that something which causes a change to the contents of >> the file might not be best choice. > > Yes, I agree it is extensible, but it was never intended to be > extended into a data modification ioctl. It's for querying metadata > related to file data, not for modifying file data. I’d agree with Dave here - there is nothing about FIEMAP that indicates it is an interface for data modification, only for reporting out the underlying layout of a file. I think falloc() is already used in this regard for hole punching, and it makes more sense to me to extend that for the discard interface as well. Cheers, Andreas -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
