2 * modified for Lites 1.1
4 * Aug 1995, Godmar Back (gback@cs.utah.edu)
5 * University of Utah, Department of Computer Science
9 * The Regents of the University of California. All rights reserved.
11 * Redistribution and use in source and binary forms, with or without
12 * modification, are permitted provided that the following conditions
14 * 1. Redistributions of source code must retain the above copyright
15 * notice, this list of conditions and the following disclaimer.
16 * 2. Redistributions in binary form must reproduce the above copyright
17 * notice, this list of conditions and the following disclaimer in the
18 * documentation and/or other materials provided with the distribution.
19 * 4. Neither the name of the University nor the names of its contributors
20 * may be used to endorse or promote products derived from this software
21 * without specific prior written permission.
23 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
24 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
25 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
26 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
27 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
28 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
29 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
30 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
31 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
32 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
35 * @(#)ufs_readwrite.c 8.7 (Berkeley) 1/21/94
39 #define BLKSIZE(a, b, c) blksize(a, b, c)
40 #define FS struct ext2_sb_info
42 #define READ ext2_read
43 #define READ_S "ext2_read"
44 #define WRITE ext2_write
45 #define WRITE_S "ext2_write"
48 * Vnode op for reading.
53 struct vop_read_args /* {
67 long size, xfersize, blkoffset;
68 int error, orig_resid;
69 int seqcount = ap->a_ioflag >> IO_SEQSHIFT;
78 if (uio->uio_rw != UIO_READ)
79 panic("%s: mode", READ_S);
81 if (vp->v_type == VLNK) {
82 if ((int)ip->i_size < vp->v_mount->mnt_maxsymlinklen)
83 panic("%s: short symlink", READ_S);
84 } else if (vp->v_type != VREG && vp->v_type != VDIR)
85 panic("%s: type %d", READ_S, vp->v_type);
88 if ((uoff_t)uio->uio_offset > fs->fs_maxfilesize)
91 orig_resid = uio->uio_resid;
92 for (error = 0, bp = NULL; uio->uio_resid > 0; bp = NULL) {
93 if ((bytesinfile = ip->i_size - uio->uio_offset) <= 0)
95 lbn = lblkno(fs, uio->uio_offset);
97 size = BLKSIZE(fs, ip, lbn);
98 blkoffset = blkoff(fs, uio->uio_offset);
100 xfersize = fs->s_frag_size - blkoffset;
101 if (uio->uio_resid < xfersize)
102 xfersize = uio->uio_resid;
103 if (bytesinfile < xfersize)
104 xfersize = bytesinfile;
106 if (lblktosize(fs, nextlbn) >= ip->i_size)
107 error = bread(vp, lbn, size, NOCRED, &bp);
108 else if ((vp->v_mount->mnt_flag & MNT_NOCLUSTERR) == 0)
109 error = cluster_read(vp,
110 ip->i_size, lbn, size, NOCRED,
111 uio->uio_resid, (ap->a_ioflag >> IO_SEQSHIFT), &bp);
112 else if (seqcount > 1) {
113 int nextsize = BLKSIZE(fs, ip, nextlbn);
114 error = breadn(vp, lbn,
115 size, &nextlbn, &nextsize, 1, NOCRED, &bp);
117 error = bread(vp, lbn, size, NOCRED, &bp);
125 * We should only get non-zero b_resid when an I/O error
126 * has occurred, which should cause us to break above.
127 * However, if the short read did not cause an error,
128 * then we want to ensure that we do not uiomove bad
129 * or uninitialized data.
132 if (size < xfersize) {
138 uiomove((char *)bp->b_data + blkoffset, (int)xfersize, uio);
146 if (orig_resid > 0 && (error == 0 || uio->uio_resid != orig_resid) &&
147 (vp->v_mount->mnt_flag & MNT_NOATIME) == 0)
148 ip->i_flag |= IN_ACCESS;
153 * Vnode op for writing.
157 struct vop_write_args /* {
161 struct ucred *a_cred;
173 int blkoffset, error, flags, ioflag, resid, size, xfersize;
175 ioflag = ap->a_ioflag;
176 seqcount = ap->a_ioflag >> IO_SEQSHIFT;
182 if (uio->uio_rw != UIO_WRITE)
183 panic("%s: mode", WRITE_S);
186 switch (vp->v_type) {
188 if (ioflag & IO_APPEND)
189 uio->uio_offset = ip->i_size;
190 if ((ip->i_flags & APPEND) && uio->uio_offset != ip->i_size)
196 if ((ioflag & IO_SYNC) == 0)
197 panic("%s: nonsync dir write", WRITE_S);
200 panic("%s: type", WRITE_S);
204 if (uio->uio_offset < 0 ||
205 (uoff_t)uio->uio_offset + uio->uio_resid > fs->fs_maxfilesize)
208 * Maybe this should be above the vnode op call, but so long as
209 * file servers have no limits, I don't think it matters.
212 if (vp->v_type == VREG && td != NULL) {
213 PROC_LOCK(td->td_proc);
214 if (uio->uio_offset + uio->uio_resid >
215 lim_cur(td->td_proc, RLIMIT_FSIZE)) {
216 psignal(td->td_proc, SIGXFSZ);
217 PROC_UNLOCK(td->td_proc);
220 PROC_UNLOCK(td->td_proc);
223 resid = uio->uio_resid;
225 flags = ioflag & IO_SYNC ? B_SYNC : 0;
227 for (error = 0; uio->uio_resid > 0;) {
228 lbn = lblkno(fs, uio->uio_offset);
229 blkoffset = blkoff(fs, uio->uio_offset);
230 xfersize = fs->s_frag_size - blkoffset;
231 if (uio->uio_resid < xfersize)
232 xfersize = uio->uio_resid;
234 if (uio->uio_offset + xfersize > ip->i_size)
235 vnode_pager_setsize(vp, uio->uio_offset + xfersize);
238 * Avoid a data-consistency race between write() and mmap()
239 * by ensuring that newly allocated blocks are zerod. The
240 * race can occur even in the case where the write covers
245 if (fs->s_frag_size > xfersize)
251 error = ext2_balloc(ip,
252 lbn, blkoffset + xfersize, ap->a_cred, &bp, flags);
256 if (uio->uio_offset + xfersize > ip->i_size) {
257 ip->i_size = uio->uio_offset + xfersize;
260 size = BLKSIZE(fs, ip, lbn) - bp->b_resid;
265 uiomove((char *)bp->b_data + blkoffset, (int)xfersize, uio);
266 if ((ioflag & IO_VMIO) &&
267 (LIST_FIRST(&bp->b_dep) == NULL)) /* in ext2fs? */
268 bp->b_flags |= B_RELBUF;
270 if (ioflag & IO_SYNC) {
272 } else if (xfersize + blkoffset == fs->s_frag_size) {
273 if ((vp->v_mount->mnt_flag & MNT_NOCLUSTERW) == 0) {
274 bp->b_flags |= B_CLUSTEROK;
275 cluster_write(vp, bp, ip->i_size, seqcount);
280 bp->b_flags |= B_CLUSTEROK;
283 if (error || xfersize == 0)
285 ip->i_flag |= IN_CHANGE | IN_UPDATE;
288 * If we successfully wrote any data, and we are not the superuser
289 * we clear the setuid and setgid bits as a precaution against
292 if (resid > uio->uio_resid && ap->a_cred && ap->a_cred->cr_uid != 0)
293 ip->i_mode &= ~(ISUID | ISGID);
295 if (ioflag & IO_UNIT) {
296 (void)ext2_truncate(vp, osize,
297 ioflag & IO_SYNC, ap->a_cred, uio->uio_td);
298 uio->uio_offset -= resid - uio->uio_resid;
299 uio->uio_resid = resid;
301 } else if (resid > uio->uio_resid && (ioflag & IO_SYNC))
302 error = ext2_update(vp, 1);