On Wed, Apr 16, 2014 at 06:37:08PM -0400, Brandon Mercer wrote:
> As I mentioned previously, this does more than just allow you to boot
> from TB disks. When I configure a 100GB disk in qemu-kvm it takes
> forever to format. On small flash installations, formatting is faster
> there as well. You don't need a 1TB disk to benefit from ffs2.
But bear in mind that ffs2 has more overhead in terms of metadata.
IMO, making it the default is not a good idea.
-Otto
>
> On Wed, Apr 16, 2014 at 3:57 PM, Brandon Mercer
> <[email protected]> wrote:
> > The other day I was doing an install in qemu-kvm and newfs was taking
> > forever, to the tune of hours. This is similar to formatting on arm
> > boards. In my quest to track down why, I discovered that ffs2 takes far
> > less time to format than ffs1 (about 30 seconds for the entire disk).
> >
> > I've put together a diff that updates the boot blocks on amd64 to be
> > able to boot ffs2. From there it's a one line change to make newfs
> > format ffs2 by default. Obviously this would need to happen for other
> > architectures as well and I'd be glad to tackle that if others see
> > this as worthwhile. Please let me know your thoughts.
> >
> > Brandon
> >
> >
> > Index: sbin/newfs/newfs.c
> > ===================================================================
> > RCS file: /cvs/src/sbin/newfs/newfs.c,v
> > retrieving revision 1.95
> > diff -u -p -u -r1.95 newfs.c
> > --- sbin/newfs/newfs.c 22 Nov 2013 04:12:48 -0000 1.95
> > +++ sbin/newfs/newfs.c 16 Apr 2014 17:47:02 -0000
> > @@ -112,7 +112,7 @@ u_short dkcksum(struct disklabel *);
> >
> > int mfs; /* run as the memory based filesystem */
> > int Nflag; /* run without writing file system */
> > -int Oflag = 1; /* 0 = 4.3BSD ffs, 1 = 4.4BSD ffs, 2 = ffs2
> > */
> > +int Oflag = 2; /* 0 = 4.3BSD ffs, 1 = 4.4BSD ffs, 2 = ffs2
> > */
> > daddr_t fssize; /* file system size in 512-byte
> > blocks */
> > long long sectorsize; /* bytes/sector */
> > int fsize = 0; /* fragment size */
> >
> > Index: arch/amd64/stand/biosboot/biosboot.S
> > ===================================================================
> > RCS file: /cvs/src/sys/arch/amd64/stand/biosboot/biosboot.S,v
> > retrieving revision 1.7
> > diff -u -p -r1.7 biosboot.S
> > --- arch/amd64/stand/biosboot/biosboot.S 5 Jul 2011 17:38:54 -0000
> > 1.7
> > +++ arch/amd64/stand/biosboot/biosboot.S 16 Apr 2014 17:22:25 -0000
> > @@ -108,6 +108,9 @@
> > * While this can be calculated as
> > * howmany(di_size, fs_bsize) it takes us too
> > * many code bytes to do it.
> > + * blkskew uint8t the skew used to parse di_db[]. this is set to four
> > by
> > + * installboot for ffs2 (due to 64-bit blocks) and
> > should
> > + * be zero for ffs1.
> > *
> > * All of these are patched directly into the code where they are used
> > * (once only, each), to save space.
> > @@ -121,7 +124,7 @@
> > */
> >
> > .globl inodeblk, inodedbl, fs_bsize_p, fsbtodb, p_offset, nblocks
> > - .globl fs_bsize_s, force_chs
> > + .globl fs_bsize_s, force_chs, blkskew
> > .type inodeblk, @function
> > .type inodedbl, @function
> > .type fs_bsize_p, @function
> > @@ -130,6 +133,7 @@
> > .type p_offset, @function
> > .type nblocks, @function
> > .type force_chs, @function
> > + .type blkskew, @function
> >
> >
> > /* Clobbers %ax, maybe more */
> > @@ -460,7 +464,8 @@ load_blocks:
> >
> > /* Get the next filesystem block number into %eax */
> > lodsl /* %eax = *(%si++), make sure 0x66 0xad */
> > -
> > +blkskew = .+2
> > + addw $0x90, %si /* adjust %si if needed (for ffs2) */
> > pushal /* Save all 32-bit registers */
> >
> > /*
> > Index: arch/amd64/stand/boot/Makefile
> > ===================================================================
> > RCS file: /cvs/src/sys/arch/amd64/stand/boot/Makefile,v
> > retrieving revision 1.26
> > diff -u -p -r1.26 Makefile
> > --- arch/amd64/stand/boot/Makefile 28 Dec 2013 15:16:28 -0000 1.26
> > +++ arch/amd64/stand/boot/Makefile 16 Apr 2014 17:22:51 -0000
> > @@ -38,7 +38,7 @@ SRCS+= alloc.c ctime.c exit.c memcmp.c m
> > SRCS+= close.c closeall.c cons.c cread.c dev.c disklabel.c dkcksum.c
> > fstat.c \
> > lseek.c open.c read.c readdir.c stat.c
> > SRCS+= elf32.c elf64.c loadfile.c
> > -SRCS+= ufs.c
> > +SRCS+= ufs.c ufs2.c
> > .if ${SOFTRAID:L} == "yes"
> > SRCS+= aes_xts.c explicit_bzero.c hmac_sha1.c pbkdf2.c rijndael.c sha1.c
> > .endif
> > Index: arch/amd64/stand/boot/conf.c
> > ===================================================================
> > RCS file: /cvs/src/sys/arch/amd64/stand/boot/conf.c,v
> > retrieving revision 1.31
> > diff -u -p -r1.31 conf.c
> > --- arch/amd64/stand/boot/conf.c 18 Feb 2014 13:56:02 -0000 1.31
> > +++ arch/amd64/stand/boot/conf.c 16 Apr 2014 17:25:17 -0000
> > @@ -31,6 +31,7 @@
> > #include <netinet/in.h>
> > #include <libsa.h>
> > #include <lib/libsa/ufs.h>
> > +#include <lib/libsa/ufs2.h>
> > #ifdef notdef
> > #include <lib/libsa/cd9660.h>
> > #include <lib/libsa/fat.h>
> > @@ -66,6 +67,8 @@ int nibprobes = nitems(probe_list);
> > struct fs_ops file_system[] = {
> > { ufs_open, ufs_close, ufs_read, ufs_write, ufs_seek,
> > ufs_stat, ufs_readdir },
> > + { ufs2_open, ufs2_close, ufs2_read, ufs2_write,
> > ufs2_seek,
> > + ufs2_stat, ufs2_readdir },
> > #ifdef notdef
> > { fat_open, fat_close, fat_read, fat_write, fat_seek,
> > fat_stat, fat_readdir },
> > Index: arch/amd64/stand/installboot/installboot.c
> > ===================================================================
> > RCS file: /cvs/src/sys/arch/amd64/stand/installboot/installboot.c,v
> > retrieving revision 1.27
> > diff -u -p -r1.27 installboot.c
> > --- arch/amd64/stand/installboot/installboot.c 13 Nov 2013 04:11:34 -0000
> > 1.27
> > +++ arch/amd64/stand/installboot/installboot.c 16 Apr 2014 17:38:00 -0000
> > @@ -2,6 +2,7 @@
> > /* $NetBSD: installboot.c,v 1.5 1995/11/17 23:23:50 gwr Exp $ */
> >
> > /*
> > + * Copyright (c) 2013 Pedro Martelletto
> > * Copyright (c) 2011 Joel Sing <[email protected]>
> > * Copyright (c) 2010 Otto Moerbeek <[email protected]>
> > * Copyright (c) 2003 Tom Cosgrove <[email protected]>
> > @@ -91,6 +92,7 @@ struct sym_data pbr_symbols[] = {
> > {"_inodeblk", 4},
> > {"_inodedbl", 4},
> > {"_nblocks", 2},
> > + {"_blkskew", 1},
> > {NULL}
> > };
> >
> > @@ -104,6 +106,8 @@ struct sym_data pbr_symbols[] = {
> > static char *loadproto(char *, long *);
> > static int getbootparams(char *, int, struct disklabel *);
> > static void devread(int, void *, daddr_t, size_t, char *);
> > +static int sbchk(struct fs *, daddr_t);
> > +static void sbread(int, daddr_t, struct fs **);
> > static void sym_set_value(struct sym_data *, char *, u_int32_t);
> > static void pbr_set_symbols(char *, char *, struct sym_data *);
> > static void usage(void);
> > @@ -412,6 +416,76 @@ devread(int fd, void *buf, daddr_t blk,
> > }
> >
> > static char sblock[SBSIZE];
> > +static const daddr_t sbtry[] = SBLOCKSEARCH;
> > +
> > +static int
> > +sbchk(struct fs *fs, daddr_t sbloc)
> > +{
> > + if (verbose)
> > + warnx("looking for superblock at %lld", sbloc);
> > +
> > + if (fs->fs_magic != FS_UFS2_MAGIC && fs->fs_magic != FS_UFS1_MAGIC)
> > {
> > + if (verbose)
> > + warnx("bad superblock magic 0x%x", fs->fs_magic);
> > + return (0);
> > + }
> > +
> > + /*
> > + * Looking for an FFS1 file system at SBLOCK_UFS2 will find the
> > + * wrong superblock for file systems with 64k block size.
> > + */
> > + if (fs->fs_magic == FS_UFS1_MAGIC && sbloc == SBLOCK_UFS2) {
> > + if (verbose)
> > + warnx("skipping ffs1 superblock at %lld", sbloc);
> > + return (0);
> > + }
> > +
> > + if (fs->fs_bsize <= 0 ||
> > + fs->fs_bsize < sizeof(struct fs) ||
> > + fs->fs_bsize > MAXBSIZE) {
> > + if (verbose)
> > + warnx("invalid superblock block size %d",
> > + fs->fs_bsize);
> > + return (0);
> > + }
> > +
> > + if (fs->fs_sbsize <= 0 || fs->fs_sbsize > SBSIZE) {
> > + if (verbose)
> > + warnx("invalid superblock size %d", fs->fs_sbsize);
> > + return (0);
> > + }
> > +
> > + if (fs->fs_inopb <= 0) {
> > + if (verbose)
> > + warnx("invalid superblock inodes/block %d",
> > + fs->fs_inopb);
> > + return (0);
> > + }
> > +
> > + if (verbose)
> > + warnx("found valid %s superblock",
> > + fs->fs_magic == FS_UFS2_MAGIC ? "ffs2" : "ffs1");
> > +
> > + return (1);
> > +}
> > +
> > +static void
> > +sbread(int fd, daddr_t poffset, struct fs **fs)
> > +{
> > + int i;
> > + daddr_t sboff;
> > +
> > + for (i = 0; sbtry[i] != -1; i++) {
> > + sboff = sbtry[i] / DEV_BSIZE;
> > + devread(fd, sblock, poffset + sboff, SBSIZE, "superblock");
> > + *fs = (struct fs *)sblock;
> > + if (sbchk(*fs, sbtry[i]))
> > + break;
> > + }
> > +
> > + if (sbtry[i] == -1)
> > + errx(1, "couldn't find ffs superblock");
> > +}
> >
> > /*
> > * Read information about /boot's inode, then put this and filesystem
> > @@ -427,11 +501,13 @@ getbootparams(char *boot, int devfd, str
> > struct fs *fs;
> > char *buf;
> > u_int blk, *ap;
> > - struct ufs1_dinode *ip;
> > + struct ufs1_dinode *ip1;
> > + struct ufs2_dinode *ip2;
> > int ndb;
> > int mib[3];
> > size_t size;
> > dev_t dev;
> > + int skew;
> >
> > /*
> > * Open 2nd-level boot program and record enough details about
> > @@ -488,16 +564,7 @@ getbootparams(char *boot, int devfd, str
> > pp = &dl->d_partitions[DISKPART(statbuf.st_dev)];
> > close(fd);
> >
> > - /* Read superblock. */
> > - devread(devfd, sblock, DL_SECTOBLK(dl, DL_GETPOFFSET(pp)) + SBLOCK,
> > - SBSIZE, "superblock");
> > - fs = (struct fs *)sblock;
> > -
> > - /* Sanity-check super-block. */
> > - if (fs->fs_magic != FS_MAGIC)
> > - errx(1, "Bad magic number in superblock");
> > - if (fs->fs_inopb <= 0)
> > - err(1, "Bad inopb=%d in superblock", fs->fs_inopb);
> > + sbread(devfd, DL_SECTOBLK(dl, pp->p_offset), &fs);
> >
> > /* Read inode. */
> > if ((buf = malloc(fs->fs_bsize)) == NULL)
> > @@ -505,15 +572,25 @@ getbootparams(char *boot, int devfd, str
> >
> > blk = fsbtodb(fs, ino_to_fsba(fs, statbuf.st_ino));
> >
> > - devread(devfd, buf, DL_SECTOBLK(dl, DL_GETPOFFSET(pp)) + blk,
> > - fs->fs_bsize, "inode");
> > - ip = (struct ufs1_dinode *)(buf) + ino_to_fsbo(fs, statbuf.st_ino);
> > -
> > /*
> > * Have the inode. Figure out how many filesystem blocks (not disk
> > * sectors) there are for biosboot to load.
> > */
> > - ndb = howmany(ip->di_size, fs->fs_bsize);
> > + devread(devfd, buf, DL_SECTOBLK(dl, pp->p_offset) + blk,
> > + fs->fs_bsize, "inode");
> > + if (fs->fs_magic == FS_UFS2_MAGIC) {
> > + ip2 = (struct ufs2_dinode *)(buf) +
> > + ino_to_fsbo(fs, statbuf.st_ino);
> > + ndb = howmany(ip2->di_size, fs->fs_bsize);
> > + ap = (u_int *)ip2->di_db;
> > + skew = sizeof(u_int32_t);
> > + } else {
> > + ip1 = (struct ufs1_dinode *)(buf) +
> > + ino_to_fsbo(fs, statbuf.st_ino);
> > + ndb = howmany(ip1->di_size, fs->fs_bsize);
> > + ap = (u_int *)ip1->di_db;
> > + skew = 0;
> > + }
> > if (ndb <= 0)
> > errx(1, "No blocks to load");
> >
> > @@ -542,10 +619,10 @@ getbootparams(char *boot, int devfd, str
> > sym_set_value(pbr_symbols, "_p_offset", pp->p_offset);
> > sym_set_value(pbr_symbols, "_inodeblk",
> > ino_to_fsba(fs, statbuf.st_ino));
> > - ap = ip->di_db;
> > sym_set_value(pbr_symbols, "_inodedbl",
> > ((((char *)ap) - buf) + INODEOFF));
> > sym_set_value(pbr_symbols, "_nblocks", ndb);
> > + sym_set_value(pbr_symbols, "_blkskew", skew);
> >
> > if (verbose) {
> > fprintf(stderr, "%s is %d blocks x %d bytes\n",
> > @@ -555,6 +632,8 @@ getbootparams(char *boot, int devfd, str
> > ffs(fs->fs_fsize / dl->d_secsize) - 1,
> > DL_GETPOFFSET(pp), ino_to_fsba(fs, statbuf.st_ino),
> > (unsigned int)((((char *)ap) - buf) + INODEOFF));
> > + fprintf(stderr, "expecting %d-bit fs blocks (skew %d)\n",
> > + skew ? 64 : 32, skew);
> > }
> >
> > return 0;
> > @@ -762,6 +841,7 @@ sr_installboot(int devfd)
> > sym_set_value(pbr_symbols, "_inodeblk", inodeblk);
> > sym_set_value(pbr_symbols, "_inodedbl", inodedbl);
> > sym_set_value(pbr_symbols, "_nblocks", nblocks);
> > + sym_set_value(pbr_symbols, "_blkskew", 0);
> >
> > if (verbose)
> > fprintf(stderr, "%s is %d blocks x %d bytes\n",
> > Index: lib/libsa/ufs2.c
> > ===================================================================
> > RCS file: lib/libsa/ufs2.c
> > diff -N lib/libsa/ufs2.c
> > --- /dev/null 1 Jan 1970 00:00:00 -0000
> > +++ lib/libsa/ufs2.c 16 Apr 2014 19:21:00 -0000
> > @@ -0,0 +1,713 @@
> > +/*-
> > + * Copyright (c) 1993
> > + * The Regents of the University of California. All rights reserved.
> > + *
> > + * This code is derived from software contributed to Berkeley by
> > + * The Mach Operating System project at Carnegie-Mellon University.
> > + *
> > + * 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.
> > + *
> > + *
> > + * Copyright (c) 1990, 1991 Carnegie Mellon University
> > + * All Rights Reserved.
> > + *
> > + * Author: David Golub
> > + *
> > + * Permission to use, copy, modify and distribute this software and its
> > + * documentation is hereby granted, provided that both the copyright
> > + * notice and this permission notice appear in all copies of the
> > + * software, derivative works or modified versions, and any portions
> > + * thereof, and that both notices appear in supporting documentation.
> > + *
> > + * CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
> > + * CONDITION. CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND FOR
> > + * ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
> > + *
> > + * Carnegie Mellon requests users of this software to return to
> > + *
> > + * Software Distribution Coordinator or [email protected]
> > + * School of Computer Science
> > + * Carnegie Mellon University
> > + * Pittsburgh PA 15213-3890
> > + *
> > + * any improvements or extensions that they make and grant Carnegie the
> > + * rights to redistribute these changes.
> > + */
> > +
> > +/*
> > + * Stand-alone file reading package.
> > + */
> > +
> > +#include <sys/param.h>
> > +#include <sys/time.h>
> > +#include <sys/stat.h>
> > +#include <ufs/ffs/fs.h>
> > +#include <ufs/ufs/dinode.h>
> > +#include <ufs/ufs/dir.h>
> > +#include <lib/libkern/libkern.h>
> > +
> > +#include "stand.h"
> > +#include "ufs2.h"
> > +
> > +/*
> > + * In-core open file.
> > + */
> > +struct file {
> > + off_t f_seekp; /* seek pointer */
> > + struct fs *f_fs; /* pointer to super-block */
> > + struct ufs2_dinode f_di; /* copy of on-disk inode */
> > + int f_nindir[NIADDR];
> > + /* number of blocks mapped by
> > + indirect block at level i */
> > + char *f_blk[NIADDR]; /* buffer for indirect block at
> > + level i */
> > + size_t f_blksize[NIADDR];
> > + /* size of buffer */
> > + daddr_t f_blkno[NIADDR];/* disk address of block in buffer
> > */
> > + char *f_buf; /* buffer for data block */
> > + size_t f_buf_size; /* size of data block */
> > + daddr_t f_buf_blkno; /* block number of data block */
> > +};
> > +
> > +static int read_inode(ufsino_t, struct open_file *);
> > +static int block_map(struct open_file *, daddr_t, daddr_t *);
> > +static int buf_read_file(struct open_file *, char **, size_t *);
> > +static int search_directory(char *, struct open_file *, ufsino_t *);
> > +static int ufs2_close_internal(struct file *);
> > +#ifdef COMPAT_UFS
> > +static void ffs_oldfscompat(struct fs *);
> > +#endif
> > +
> > +/*
> > + * Read a new inode into a file structure.
> > + */
> > +static int
> > +read_inode(ufsino_t inumber, struct open_file *f)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > + struct fs *fs = fp->f_fs;
> > + char *buf;
> > + size_t rsize;
> > + int rc;
> > +
> > + /*
> > + * Read inode and save it.
> > + */
> > + buf = alloc(fs->fs_bsize);
> > + twiddle();
> > + rc = (f->f_dev->dv_strategy)(f->f_devdata, F_READ,
> > + fsbtodb(fs, ino_to_fsba(fs, inumber)), fs->fs_bsize, buf,
> > &rsize);
> > + if (rc)
> > + goto out;
> > + if (rsize != (size_t)fs->fs_bsize) {
> > + rc = EIO;
> > + goto out;
> > + }
> > +
> > + {
> > + struct ufs2_dinode *dp;
> > +
> > + dp = (struct ufs2_dinode *)buf;
> > + fp->f_di = dp[ino_to_fsbo(fs, inumber)];
> > + }
> > +
> > + /*
> > + * Clear out the old buffers
> > + */
> > + {
> > + int level;
> > +
> > + for (level = 0; level < NIADDR; level++)
> > + fp->f_blkno[level] = -1;
> > + fp->f_buf_blkno = -1;
> > + fp->f_seekp = 0;
> > + }
> > +out:
> > + free(buf, fs->fs_bsize);
> > + return (rc);
> > +}
> > +
> > +/*
> > + * Given an offset in a file, find the disk block number that
> > + * contains that block.
> > + */
> > +static int
> > +block_map(struct open_file *f, daddr_t file_block, daddr_t *disk_block_p)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > + daddr_t ind_block_num;
> > + char *ind_p;
> > + struct fs *fs = fp->f_fs;
> > + int level, idx, rc;
> > +
> > + /*
> > + * Index structure of an inode:
> > + *
> > + * di_db[0..NDADDR-1] hold block numbers for blocks
> > + * 0..NDADDR-1
> > + *
> > + * di_ib[0] index block 0 is the single indirect block
> > + * holds block numbers for blocks
> > + * NDADDR .. NDADDR + NINDIR(fs)-1
> > + *
> > + * di_ib[1] index block 1 is the double indirect block
> > + * holds block numbers for INDEX blocks for
> > blocks
> > + * NDADDR + NINDIR(fs) ..
> > + * NDADDR + NINDIR(fs) + NINDIR(fs)**2 - 1
> > + *
> > + * di_ib[2] index block 2 is the triple indirect block
> > + * holds block numbers for double-indirect
> > + * blocks for blocks
> > + * NDADDR + NINDIR(fs) + NINDIR(fs)**2 ..
> > + * NDADDR + NINDIR(fs) + NINDIR(fs)**2
> > + * + NINDIR(fs)**3 - 1
> > + */
> > +
> > + if (file_block < NDADDR) {
> > + /* Direct block. */
> > + *disk_block_p = fp->f_di.di_db[file_block];
> > + return (0);
> > + }
> > +
> > + file_block -= NDADDR;
> > +
> > + /*
> > + * nindir[0] = NINDIR
> > + * nindir[1] = NINDIR**2
> > + * nindir[2] = NINDIR**3
> > + * etc
> > + */
> > + for (level = 0; level < NIADDR; level++) {
> > + if (file_block < fp->f_nindir[level])
> > + break;
> > + file_block -= fp->f_nindir[level];
> > + }
> > + if (level == NIADDR) {
> > + /* Block number too high */
> > + return (EFBIG);
> > + }
> > +
> > + ind_block_num = fp->f_di.di_ib[level];
> > +
> > + for (; level >= 0; level--) {
> > + if (ind_block_num == 0) {
> > + *disk_block_p = 0; /* missing */
> > + return (0);
> > + }
> > +
> > + if (fp->f_blkno[level] != ind_block_num) {
> > + if (fp->f_blk[level] == (char *)0)
> > + fp->f_blk[level] =
> > + alloc(fs->fs_bsize);
> > + twiddle();
> > + rc = (f->f_dev->dv_strategy)(f->f_devdata, F_READ,
> > + fsbtodb(fp->f_fs, ind_block_num), fs->fs_bsize,
> > + fp->f_blk[level], &fp->f_blksize[level]);
> > + if (rc)
> > + return (rc);
> > + if (fp->f_blksize[level] != (size_t)fs->fs_bsize)
> > + return (EIO);
> > + fp->f_blkno[level] = ind_block_num;
> > + }
> > +
> > + ind_p = fp->f_blk[level];
> > +
> > + if (level > 0) {
> > + idx = file_block / fp->f_nindir[level - 1];
> > + file_block %= fp->f_nindir[level - 1];
> > + } else
> > + idx = file_block;
> > +
> > + ind_block_num = ind_p[idx];
> > + }
> > +
> > + *disk_block_p = ind_block_num;
> > + return (0);
> > +}
> > +
> > +/*
> > + * Read a portion of a file into an internal buffer. Return
> > + * the location in the buffer and the amount in the buffer.
> > + */
> > +static int
> > +buf_read_file(struct open_file *f, char **buf_p, size_t *size_p)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > + struct fs *fs = fp->f_fs;
> > + daddr_t file_block, disk_block;
> > + size_t block_size;
> > + long off;
> > + int rc;
> > +
> > + off = blkoff(fs, fp->f_seekp);
> > + file_block = lblkno(fs, fp->f_seekp);
> > + block_size = dblksize(fs, &fp->f_di, file_block);
> > +
> > + if (file_block != fp->f_buf_blkno) {
> > + rc = block_map(f, file_block, &disk_block);
> > + if (rc)
> > + return (rc);
> > +
> > + if (fp->f_buf == (char *)0)
> > + fp->f_buf = alloc(fs->fs_bsize);
> > +
> > + if (disk_block == 0) {
> > + bzero(fp->f_buf, block_size);
> > + fp->f_buf_size = block_size;
> > + } else {
> > + twiddle();
> > + rc = (f->f_dev->dv_strategy)(f->f_devdata, F_READ,
> > + fsbtodb(fs, disk_block),
> > + block_size, fp->f_buf, &fp->f_buf_size);
> > + if (rc)
> > + return (rc);
> > + }
> > +
> > + fp->f_buf_blkno = file_block;
> > + }
> > +
> > + /*
> > + * Return address of byte in buffer corresponding to
> > + * offset, and size of remainder of buffer after that
> > + * byte.
> > + */
> > + *buf_p = fp->f_buf + off;
> > + *size_p = block_size - off;
> > +
> > + /*
> > + * But truncate buffer at end of file.
> > + */
> > + if (*size_p > fp->f_di.di_size - fp->f_seekp)
> > + *size_p = fp->f_di.di_size - fp->f_seekp;
> > +
> > + return (0);
> > +}
> > +
> > +/*
> > + * Search a directory for a name and return its
> > + * i_number.
> > + */
> > +static int
> > +search_directory(char *name, struct open_file *f, ufsino_t *inumber_p)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > + int namlen, length, rc;
> > + struct direct *dp, *edp;
> > + size_t buf_size;
> > + char *buf;
> > +
> > + length = strlen(name);
> > +
> > + fp->f_seekp = 0;
> > + while (fp->f_seekp < fp->f_di.di_size) {
> > + rc = buf_read_file(f, &buf, &buf_size);
> > + if (rc)
> > + return (rc);
> > +
> > + dp = (struct direct *)buf;
> > + edp = (struct direct *)(buf + buf_size);
> > + while (dp < edp) {
> > + if (dp->d_ino == 0)
> > + goto next;
> > +#if BYTE_ORDER == LITTLE_ENDIAN
> > + if (fp->f_fs->fs_maxsymlinklen <= 0)
> > + namlen = dp->d_type;
> > + else
> > +#endif
> > + namlen = dp->d_namlen;
> > + if (namlen == length &&
> > + !strcmp(name, dp->d_name)) {
> > + /* found entry */
> > + *inumber_p = dp->d_ino;
> > + return (0);
> > + }
> > + next:
> > + dp = (struct direct *)((char *)dp + dp->d_reclen);
> > + }
> > + fp->f_seekp += buf_size;
> > + }
> > + return (ENOENT);
> > +}
> > +
> > +/*
> > + * Open a file.
> > + */
> > +int
> > +ufs2_open(char *path, struct open_file *f)
> > +{
> > + char namebuf[MAXPATHLEN+1], *cp, *ncp, *buf = NULL;
> > + ufsino_t inumber, parent_inumber;
> > + int rc, c, nlinks = 0;
> > + struct file *fp;
> > + size_t buf_size;
> > + struct fs *fs;
> > +
> > + /* allocate file system specific data structure */
> > + fp = alloc(sizeof(struct file));
> > + bzero(fp, sizeof(struct file));
> > + f->f_fsdata = (void *)fp;
> > +
> > + /* allocate space and read super block */
> > + fs = alloc(SBSIZE);
> > + fp->f_fs = fs;
> > + twiddle();
> > + rc = (f->f_dev->dv_strategy)(f->f_devdata, F_READ,
> > + SBLOCK_UFS2 / DEV_BSIZE, SBSIZE, (char *)fs, &buf_size);
> > + if (rc)
> > + goto out;
> > +
> > + if (buf_size != SBSIZE || fs->fs_magic != FS_UFS2_MAGIC ||
> > + fs->fs_bsize > MAXBSIZE || fs->fs_bsize < sizeof(struct fs)) {
> > + rc = EINVAL;
> > + goto out;
> > + }
> > +#ifdef COMPAT_UFS
> > + ffs_oldfscompat(fs);
> > +#endif
> > +
> > + /*
> > + * Calculate indirect block levels.
> > + */
> > + {
> > + int mult;
> > + int level;
> > +
> > + mult = 1;
> > + for (level = 0; level < NIADDR; level++) {
> > + mult *= NINDIR(fs);
> > + fp->f_nindir[level] = mult;
> > + }
> > + }
> > +
> > + inumber = ROOTINO;
> > + if ((rc = read_inode(inumber, f)) != 0)
> > + goto out;
> > +
> > + cp = path;
> > + while (*cp) {
> > +
> > + /*
> > + * Remove extra separators
> > + */
> > + while (*cp == '/')
> > + cp++;
> > + if (*cp == '\0')
> > + break;
> > +
> > + /*
> > + * Check that current node is a directory.
> > + */
> > + if ((fp->f_di.di_mode & IFMT) != IFDIR) {
> > + rc = ENOTDIR;
> > + goto out;
> > + }
> > +
> > + /*
> > + * Get next component of path name.
> > + */
> > + {
> > + int len = 0;
> > +
> > + ncp = cp;
> > + while ((c = *cp) != '\0' && c != '/') {
> > + if (++len > MAXNAMLEN) {
> > + rc = ENOENT;
> > + goto out;
> > + }
> > + cp++;
> > + }
> > + *cp = '\0';
> > + }
> > +
> > + /*
> > + * Look up component in current directory.
> > + * Save directory inumber in case we find a
> > + * symbolic link.
> > + */
> > + parent_inumber = inumber;
> > + rc = search_directory(ncp, f, &inumber);
> > + *cp = c;
> > + if (rc)
> > + goto out;
> > +
> > + /*
> > + * Open next component.
> > + */
> > + if ((rc = read_inode(inumber, f)) != 0)
> > + goto out;
> > +
> > + /*
> > + * Check for symbolic link.
> > + */
> > + if ((fp->f_di.di_mode & IFMT) == IFLNK) {
> > + int link_len = fp->f_di.di_size;
> > + int len;
> > +
> > + len = strlen(cp);
> > +
> > + if (link_len + len > MAXPATHLEN ||
> > + ++nlinks > MAXSYMLINKS) {
> > + rc = ENOENT;
> > + goto out;
> > + }
> > +
> > + bcopy(cp, &namebuf[link_len], len + 1);
> > +
> > + if (link_len < fs->fs_maxsymlinklen) {
> > + bcopy(fp->f_di.di_shortlink, namebuf,
> > + (unsigned) link_len);
> > + } else {
> > + /*
> > + * Read file for symbolic link
> > + */
> > + size_t buf_size;
> > + daddr_t disk_block;
> > + struct fs *fs = fp->f_fs;
> > +
> > + if (!buf)
> > + buf = alloc(fs->fs_bsize);
> > + rc = block_map(f, 0, &disk_block);
> > + if (rc)
> > + goto out;
> > +
> > + twiddle();
> > + rc = (f->f_dev->dv_strategy)(f->f_devdata,
> > + F_READ, fsbtodb(fs, disk_block),
> > + fs->fs_bsize, buf, &buf_size);
> > + if (rc)
> > + goto out;
> > +
> > + bcopy((char *)buf, namebuf,
> > (unsigned)link_len);
> > + }
> > +
> > + /*
> > + * If relative pathname, restart at parent
> > directory.
> > + * If absolute pathname, restart at root.
> > + */
> > + cp = namebuf;
> > + if (*cp != '/')
> > + inumber = parent_inumber;
> > + else
> > + inumber = ROOTINO;
> > +
> > + if ((rc = read_inode(inumber, f)) != 0)
> > + goto out;
> > + }
> > + }
> > +
> > + /*
> > + * Found terminal component.
> > + */
> > + rc = 0;
> > +out:
> > + if (buf)
> > + free(buf, fs->fs_bsize);
> > + if (rc)
> > + (void)ufs2_close_internal(fp);
> > +
> > + return (rc);
> > +}
> > +
> > +int
> > +ufs2_close(struct open_file *f)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > +
> > + f->f_fsdata = (void *)0;
> > + if (fp == (struct file *)0)
> > + return (0);
> > +
> > + return (ufs2_close_internal(fp));
> > +}
> > +
> > +static int
> > +ufs2_close_internal(struct file *fp)
> > +{
> > + int level;
> > +
> > + for (level = 0; level < NIADDR; level++) {
> > + if (fp->f_blk[level])
> > + free(fp->f_blk[level], fp->f_fs->fs_bsize);
> > + }
> > + if (fp->f_buf)
> > + free(fp->f_buf, fp->f_fs->fs_bsize);
> > + free(fp->f_fs, SBSIZE);
> > + free(fp, sizeof(struct file));
> > + return (0);
> > +}
> > +
> > +/*
> > + * Copy a portion of a file into kernel memory.
> > + * Cross block boundaries when necessary.
> > + */
> > +int
> > +ufs2_read(struct open_file *f, void *start, size_t size, size_t *resid)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > + char *buf, *addr = start;
> > + size_t csize, buf_size;
> > + int rc = 0;
> > +
> > + while (size != 0) {
> > + if (fp->f_seekp >= fp->f_di.di_size)
> > + break;
> > +
> > + rc = buf_read_file(f, &buf, &buf_size);
> > + if (rc)
> > + break;
> > +
> > + csize = size;
> > + if (csize > buf_size)
> > + csize = buf_size;
> > +
> > + bcopy(buf, addr, csize);
> > +
> > + fp->f_seekp += csize;
> > + addr += csize;
> > + size -= csize;
> > + }
> > + if (resid)
> > + *resid = size;
> > + return (rc);
> > +}
> > +
> > +/*
> > + * Not implemented.
> > + */
> > +int
> > +ufs2_write(struct open_file *f, void *start, size_t size, size_t *resid)
> > +{
> > +
> > + return (EROFS);
> > +}
> > +
> > +off_t
> > +ufs2_seek(struct open_file *f, off_t offset, int where)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > +
> > + switch (where) {
> > + case SEEK_SET:
> > + fp->f_seekp = offset;
> > + break;
> > + case SEEK_CUR:
> > + fp->f_seekp += offset;
> > + break;
> > + case SEEK_END:
> > + fp->f_seekp = fp->f_di.di_size - offset;
> > + break;
> > + default:
> > + return (-1);
> > + }
> > + return (fp->f_seekp);
> > +}
> > +
> > +int
> > +ufs2_stat(struct open_file *f, struct stat *sb)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > +
> > + /* only important stuff */
> > + sb->st_mode = fp->f_di.di_mode;
> > + sb->st_uid = fp->f_di.di_uid;
> > + sb->st_gid = fp->f_di.di_gid;
> > + sb->st_size = fp->f_di.di_size;
> > + return (0);
> > +}
> > +
> > +#ifndef NO_READDIR
> > +int
> > +ufs2_readdir(struct open_file *f, char *name)
> > +{
> > + struct file *fp = (struct file *)f->f_fsdata;
> > + struct direct *dp, *edp;
> > + size_t buf_size;
> > + int rc, namlen;
> > + char *buf;
> > +
> > + if (name == NULL)
> > + fp->f_seekp = 0;
> > + else {
> > + /* end of dir */
> > + if (fp->f_seekp >= fp->f_di.di_size) {
> > + *name = '\0';
> > + return -1;
> > + }
> > +
> > + do {
> > + if ((rc = buf_read_file(f, &buf, &buf_size)) != 0)
> > + return rc;
> > +
> > + dp = (struct direct *)buf;
> > + edp = (struct direct *)(buf + buf_size);
> > + while (dp < edp && dp->d_ino == 0)
> > + dp = (struct direct *)((char *)dp +
> > dp->d_reclen);
> > + fp->f_seekp += buf_size -
> > + ((u_int8_t *)edp - (u_int8_t *)dp);
> > + } while (dp >= edp);
> > +
> > +#if BYTE_ORDER == LITTLE_ENDIAN
> > + if (fp->f_fs->fs_maxsymlinklen <= 0)
> > + namlen = dp->d_type;
> > + else
> > +#endif
> > + namlen = dp->d_namlen;
> > + strncpy(name, dp->d_name, namlen + 1);
> > +
> > + fp->f_seekp += dp->d_reclen;
> > + }
> > +
> > + return 0;
> > +}
> > +#endif
> > +
> > +#ifdef COMPAT_UFS
> > +/*
> > + * Sanity checks for old file systems.
> > + *
> > + * XXX - goes away some day.
> > + */
> > +static void
> > +ffs_oldfscompat(struct fs *fs)
> > +{
> > + int i;
> > +
> > + fs->fs_npsect = max(fs->fs_npsect, fs->fs_nsect); /* XXX */
> > + fs->fs_interleave = max(fs->fs_interleave, 1); /* XXX */
> > + if (fs->fs_postblformat == FS_42POSTBLFMT) /* XXX */
> > + fs->fs_nrpos = 8; /* XXX */
> > + if (fs->fs_inodefmt < FS_44INODEFMT) { /* XXX */
> > + quad_t sizepb = fs->fs_bsize; /* XXX */
> > + /* XXX */
> > + fs->fs_maxfilesize = fs->fs_bsize * NDADDR - 1; /* XXX */
> > + for (i = 0; i < NIADDR; i++) { /* XXX */
> > + sizepb *= NINDIR(fs); /* XXX */
> > + fs->fs_maxfilesize += sizepb; /* XXX */
> > + } /* XXX */
> > + fs->fs_qbmask = ~fs->fs_bmask; /* XXX */
> > + fs->fs_qfmask = ~fs->fs_fmask; /* XXX */
> > + } /* XXX */
> > +}
> > +#endif
> > Index: lib/libsa/ufs2.h
> > ===================================================================
> > RCS file: lib/libsa/ufs2.h
> > diff -N lib/libsa/ufs2.h
> > --- /dev/null 1 Jan 1970 00:00:00 -0000
> > +++ lib/libsa/ufs2.h 16 Apr 2014 17:39:44 -0000
> > @@ -0,0 +1,38 @@
> > +/*-
> > + * Copyright (c) 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.
> > + *
> > + * @(#)ufs.h 8.1 (Berkeley) 6/11/93
> > + */
> > +
> > +int ufs2_open(char *, struct open_file *);
> > +int ufs2_close(struct open_file *);
> > +int ufs2_read(struct open_file *, void *, size_t, size_t *);
> > +int ufs2_write(struct open_file *, void *, size_t, size_t *);
> > +int ufs2_stat(struct open_file *, struct stat *);
> > +int ufs2_readdir(struct open_file *, char *);
> > +off_t ufs2_seek(struct open_file *, off_t, int);
