On Fri, Oct 16, 2020 at 04:58:53PM +0530, Chandan Babu R wrote: > On Friday 16 October 2020 12:34:48 PM IST Christoph Hellwig wrote: > > On Thu, Oct 15, 2020 at 03:31:26PM +0530, Chandan Babu R wrote: > > > How about following the traits of XFS_IEXT_WRITE_UNWRITTEN_CNT (writing > > > to unwritten extent) and XFS_IEXT_REFLINK_END_COW_CNT (moving an extent > > > from cow fork to data fork) and setting XFS_IEXT_REFLINK_REMAP_CNT to a > > > worst case value of 2? A write spanning the entirety of an unwritten extent > > > does not change the extent count. Similarly, If there are no extents in the > > > data fork spanning the file range mapped by an extent in the cow > > > fork, moving the extent from cow fork to data fork increases the extent count > > > by just 1 and not by the worst case count of 2. > > > > No, I think the dynamic value is perfectly fine, as we have all the > > information trivially available. I just don't think having a separate > > macro and the comment explaining it away from the actual functionality > > is helpful. > > > > Darrick, I think using the macros approach is more suitable. But I can go > ahead and implement the approach decided by the community. Please let me know > your opinion. The macro only gets used in one place anyway, so I don't see as strong a need for it as the other places. I think this one could be open-coded next to the places where we decide the values of smap_real and dmap_written. (i.e. what Christoph is suggesting) --D > -- > chandan > > >
