On Tue, May 18, 2010 at 09:40:28PM +0800, Yan, Zheng wrote:
> On Sat, Mar 20, 2010 at 12:24 PM, Sean Bartell
> <wingedtachikoma@xxxxxxxxx> wrote:
> > An extent_io_tree is used for all free space information. This allows
> > removal of ext2_alloc_block and ext2_free_block, and makes
> > create_ext2_image less ext2-specific.
> > + ret = ext2_cache_free_extents(ext2_fs, &orig_free_tree);
> > + if (ret) {
> > + fprintf(stderr, "error during cache_free_extents %d\n", ret);
> > + goto fail;
> > + }
> > + /* preserve first 64KiB, just in case */
> > + clear_extent_dirty(&orig_free_tree, 0, BTRFS_SUPER_INFO_OFFSET - 1, 0);
> > +
> > + ret = copy_dirtiness(&free_tree, &orig_free_tree);
> > + if (ret) {
> > + fprintf(stderr, "error during copy_dirtiness %d\n", ret);
> > + goto fail;
> > + }
> extent_io_tree is not very space efficient. caching free space in
> several places is not good. I prefer adding a function that checks
> if a given block is used to the 'convert_fs' structure.
Good point. I'll change cache_free_extents to something like
int (*iterate_used_extents)(struct convert_fs *fs, u64 start, u64 end,
void *priv, int (*cb)(u64 start, u64 end))
create_image_file_range and do_convert should work well with a callback.
This also opens up the possibility of finding free extents
incrementally: call iterate_used_extents on the first GB, then
custom_alloc_extent will call it on the next GB once free space runs
out.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
