Source
committed 5cd054dd5c1 if((ret_value = H5B_iterate(idx_info->f, idx_info->dxpl_id, H5B_BTREE, idx_info->layout->u.chunk.u.btree.addr, H5D_btree_idx_iterate_cb, &udata)) < 0)/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Copyright by The HDF Group. * * Copyright by the Board of Trustees of the University of Illinois. * * All rights reserved. * * * * This file is part of HDF5. The full HDF5 copyright notice, including * * terms governing use, modification, and redistribution, is contained in * * the files COPYING and Copyright.html. COPYING can be found at the root * * of the source code distribution tree; Copyright.html can be found at the * * root level of an installed copy of the electronic HDF5 document set and * * is linked from the top-level documents page. It can also be found at * * http://hdfgroup.org/HDF5/doc/Copyright.html. If you do not have * * access to either file, you may request a copy from help@hdfgroup.org. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *//* Programmer: Robb Matzke <matzke@llnl.gov> * Wednesday, October 8, 1997 * * Purpose: v1 B-tree indexed (chunked) I/O functions. The logical * multi-dimensional data space is regularly partitioned into * same-sized "chunks", the first of which is aligned with the * logical origin. The chunks are given a multi-dimensional * index which is used as a lookup key in a B-tree that maps * chunk index to disk address. Each chunk can be compressed * independently and the chunks may move around in the file as * their storage requirements change. * * Cache: Disk I/O is performed in units of chunks and H5MF_alloc() * contains code to optionally align chunks on disk block * boundaries for performance. * * The chunk cache is an extendible hash indexed by a function * of storage B-tree address and chunk N-dimensional offset * within the dataset. Collisions are not resolved -- one of * the two chunks competing for the hash slot must be preempted * from the cache. All entries in the hash also participate in * a doubly-linked list and entries are penalized by moving them * toward the front of the list. When a new chunk is about to * be added to the cache the heap is pruned by preempting * entries near the front of the list to make room for the new * entry which is added to the end of the list. *//****************//* Module Setup *//****************/