/*-
 *  modified for Lites 1.1
 *
 *  Aug 1995, Godmar Back (gback@cs.utah.edu)
 *  University of Utah, Department of Computer Science
 */
/*-
 * SPDX-License-Identifier: BSD-3-Clause
 *
 * Copyright (c) 1982, 1986, 1989, 1993
 *      The Regents of the University of California.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *      @(#)ffs_balloc.c        8.4 (Berkeley) 9/23/93
 * $FreeBSD$
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/endian.h>
#include <sys/bio.h>
#include <sys/buf2.h>
#include <sys/limits.h>
#include <sys/lock.h>
#include <sys/mount.h>
#include <sys/vnode.h>
#include <sys/mutex2.h>

#include <vfs/ext2fs/fs.h>
#include <vfs/ext2fs/inode.h>
#include <vfs/ext2fs/ext2fs.h>
#include <vfs/ext2fs/ext2_dinode.h>
#include <vfs/ext2fs/ext2_extern.h>
#include <vfs/ext2fs/ext2_mount.h>

static int
ext2_ext_balloc(struct inode *ip, uint32_t lbn, int size,
    struct ucred *cred, struct buf **bpp, int flags)
{
        return (EINVAL);
}

/*
 * Balloc defines the structure of filesystem storage
 * by allocating the physical blocks on a device given
 * the inode and the logical block number in a file.
 */
int
ext2_balloc(struct inode *ip, e2fs_lbn_t lbn, int size, struct ucred *cred,
    struct buf **bpp, int flags)
{
        struct m_ext2fs *fs;
        struct ext2mount *ump;
        struct buf *bp, *nbp;
        struct vnode *vp = ITOV(ip);
        struct indir indirs[EXT2_NIADDR + 2];
        e4fs_daddr_t nb, newb;
        e2fs_daddr_t *bap, pref;
        int num, i, error;

        *bpp = NULL;
        if (lbn < 0)
                return (EFBIG);
        fs = ip->i_e2fs;
        ump = ip->i_ump;

        /*
         * check if this is a sequential block allocation.
         * If so, increment next_alloc fields to allow ext2_blkpref
         * to make a good guess
         */
        if (lbn == ip->i_next_alloc_block + 1) {
                ip->i_next_alloc_block++;
                ip->i_next_alloc_goal++;
        }

        if (ip->i_flag & IN_E4EXTENTS)
                return (ext2_ext_balloc(ip, lbn, size, cred, bpp, flags));

        /*
         * The first EXT2_NDADDR blocks are direct blocks
         */
        if (lbn < EXT2_NDADDR) {
                nb = ip->i_db[lbn];
                /*
                 * no new block is to be allocated, and no need to expand
                 * the file
                 */
                if (nb != 0) {
                        error = bread(vp, lblktodoff(fs, lbn), fs->e2fs_bsize,
                            &bp);
                        if (error) {
                                brelse(bp);
                                return (error);
                        }
                        bp->b_bio2.bio_offset = fsbtodoff(fs, nb);
                        if (ip->i_size >= (lbn + 1) * fs->e2fs_bsize) {
                                *bpp = bp;
                                return (0);
                        }
                } else {
                        EXT2_LOCK(ump);
                        error = ext2_alloc(ip, lbn,
                            ext2_blkpref(ip, lbn, (int)lbn, &ip->i_db[0], 0),
                            fs->e2fs_bsize, cred, &newb);
                        if (error)
                                return (error);
                        /*
                         * If the newly allocated block exceeds 32-bit limit,
                         * we can not use it in file block maps.
                         */
                        if (newb > UINT_MAX)
                                return (EFBIG);
                        bp = getblk(vp, lblktodoff(fs, lbn), fs->e2fs_bsize, 0, 0);
                        bp->b_bio2.bio_offset = fsbtodoff(fs, newb);
                        if (flags & BA_CLRBUF)
                                vfs_bio_clrbuf(bp);
                }
                ip->i_db[lbn] = dofftofsb(fs, bp->b_bio2.bio_offset);
                ip->i_flag |= IN_CHANGE | IN_UPDATE;
                *bpp = bp;
                return (0);
        }
        /*
         * Determine the number of levels of indirection.
         */
        pref = 0;
        if ((error = ext2_getlbns(vp, lbn, indirs, &num)) != 0)
                return (error);
#ifdef INVARIANTS
        if (num < 1)
                panic("ext2_balloc: ext2_getlbns returned indirect block");
#endif
        /*
         * Fetch the first indirect block allocating if necessary.
         */
        --num;
        nb = ip->i_ib[indirs[0].in_off];
        if (nb == 0) {
                EXT2_LOCK(ump);
                pref = ext2_blkpref(ip, lbn, indirs[0].in_off +
                    EXT2_NDIR_BLOCKS, &ip->i_db[0], 0);
                if ((error = ext2_alloc(ip, lbn, pref, fs->e2fs_bsize, cred,
                    &newb)))
                        return (error);
                if (newb > UINT_MAX)
                        return (EFBIG);
                nb = newb;
                bp = getblk(vp, lblktodoff(fs, indirs[1].in_lbn),
                    fs->e2fs_bsize, 0, 0);
                bp->b_bio2.bio_offset = fsbtodoff(fs, newb);
                vfs_bio_clrbuf(bp);
                /*
                 * Write synchronously so that indirect blocks
                 * never point at garbage.
                 */
                if ((error = bwrite(bp)) != 0) {
                        ext2_blkfree(ip, nb, fs->e2fs_bsize);
                        return (error);
                }
                ip->i_ib[indirs[0].in_off] = newb;
                ip->i_flag |= IN_CHANGE | IN_UPDATE;
        }
        /*
         * Fetch through the indirect blocks, allocating as necessary.
         */
        for (i = 1;;) {
                error = bread(vp, lblktodoff(fs, indirs[i].in_lbn),
                    (int)fs->e2fs_bsize, &bp);
                if (error) {
                        brelse(bp);
                        return (error);
                }
                bap = (e2fs_daddr_t *)bp->b_data;
                nb = le32toh(bap[indirs[i].in_off]);
                if (i == num)
                        break;
                i += 1;
                if (nb != 0) {
                        bqrelse(bp);
                        continue;
                }
                EXT2_LOCK(ump);
                if (pref == 0)
                        pref = ext2_blkpref(ip, lbn, indirs[i].in_off, bap,
                            lblkno(fs, bp->b_loffset));
                error = ext2_alloc(ip, lbn, pref, (int)fs->e2fs_bsize, cred, &newb);
                if (error) {
                        brelse(bp);
                        return (error);
                }
                if (newb > UINT_MAX) {
                        brelse(bp);
                        return (EFBIG);
                }
                nb = newb;
                nbp = getblk(vp, lblktodoff(fs, indirs[i].in_lbn),
                    fs->e2fs_bsize, 0, 0);
                nbp->b_bio2.bio_offset = fsbtodoff(fs, nb);
                vfs_bio_clrbuf(nbp);
                /*
                 * Write synchronously so that indirect blocks
                 * never point at garbage.
                 */
                if ((error = bwrite(nbp)) != 0) {
                        ext2_blkfree(ip, nb, fs->e2fs_bsize);
                        brelse(bp);
                        return (error);
                }
                bap[indirs[i - 1].in_off] = htole32(nb);
                /*
                 * If required, write synchronously, otherwise use
                 * delayed write.
                 */
                if (flags & IO_SYNC) {
                        bwrite(bp);
                } else {
                        if (bp->b_bufsize == fs->e2fs_bsize)
                                bp->b_flags |= B_CLUSTEROK;
                        bdwrite(bp);
                }
        }
        /*
         * Get the data block, allocating if necessary.
         */
        if (nb == 0) {
                EXT2_LOCK(ump);
                pref = ext2_blkpref(ip, lbn, indirs[i].in_off, &bap[0],
                    lblkno(fs, bp->b_loffset));
                if ((error = ext2_alloc(ip,
                    lbn, pref, (int)fs->e2fs_bsize, cred, &newb)) != 0) {
                        brelse(bp);
                        return (error);
                }
                if (newb > UINT_MAX) {
                        brelse(bp);
                        return (EFBIG);
                }
                nb = newb;
                nbp = getblk(vp, lblktodoff(fs, lbn), fs->e2fs_bsize, 0, 0);
                nbp->b_bio2.bio_offset = fsbtodoff(fs, nb);
                if (flags & BA_CLRBUF)
                        vfs_bio_clrbuf(nbp);
                bap[indirs[i].in_off] = htole32(nb);
                /*
                 * If required, write synchronously, otherwise use
                 * delayed write.
                 */
                if (flags & IO_SYNC) {
                        bwrite(bp);
                } else {
                        if (bp->b_bufsize == fs->e2fs_bsize)
                                bp->b_flags |= B_CLUSTEROK;
                        bdwrite(bp);
                }
                *bpp = nbp;
                return (0);
        }
        brelse(bp);
        if (flags & BA_CLRBUF) {
                int seqcount = (flags & BA_SEQMASK) >> BA_SEQSHIFT;

                if (seqcount && (vp->v_mount->mnt_flag & MNT_NOCLUSTERR) == 0) {
                        error = cluster_read(vp, ip->i_size,
                            lblktodoff(fs, lbn), (int)fs->e2fs_bsize,
                            MAXBSIZE, (size_t)seqcount, &nbp);
                } else {
                        error = bread(vp, lblktodoff(fs, lbn),
                            (int)fs->e2fs_bsize, &nbp);
                }
                if (error) {
                        brelse(nbp);
                        return (error);
                }
        } else {
                nbp = getblk(vp, lblktodoff(fs, lbn), fs->e2fs_bsize, 0, 0);
                nbp->b_bio2.bio_offset = fsbtodoff(fs, nb);
        }
        *bpp = nbp;
        return (0);
}