/*
* This routine is called to handle zeroing any space in the last
* block of the file that is beyond the EOF. We do this since the
* size is being increased without writing anything to that block
* and we don't want anyone to read the garbage on the disk.
*/
STATIC int /* error (positive) */
xfs_zero_last_block(
xfs_inode_t *ip,
xfs_fsize_t offset,
xfs_fsize_t isize)
{
xfs_fileoff_t last_fsb;
xfs_mount_t *mp = ip->i_mount;
int nimaps;
int zero_offset;
int zero_len;
int error = 0;
xfs_bmbt_irec_t imap;
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
zero_offset = XFS_B_FSB_OFFSET(mp, isize);
if (zero_offset == 0) {
/*
* There are no extra bytes in the last block on disk to
* zero, so return.
*/
return 0;
}
last_fsb = XFS_B_TO_FSBT(mp, isize);
nimaps = 1;
error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap,
&nimaps, NULL, NULL);
if (error) {
return error;
}
ASSERT(nimaps > 0);
/*
* If the block underlying isize is just a hole, then there
* is nothing to zero.
*/
if (imap.br_startblock == HOLESTARTBLOCK) {
return 0;
}
/*
* Zero the part of the last block beyond the EOF, and write it
* out sync. We need to drop the ilock while we do this so we
* don't deadlock when the buffer cache calls back to us.
*/
xfs_iunlock(ip, XFS_ILOCK_EXCL);
zero_len = mp->m_sb.sb_blocksize - zero_offset;
if (isize + zero_len > offset)
zero_len = offset - isize;
error = xfs_iozero(ip, isize, zero_len);
xfs_ilock(ip, XFS_ILOCK_EXCL);
ASSERT(error >= 0);
return error;
}
/*
* If the caller is doing a write at the end of the file, then extend the
* allocation out to the file system's write iosize. We clean up any extra
* space left over when the file is closed in xfs_inactive().
*/
STATIC int
xfs_iomap_eof_want_preallocate(
xfs_mount_t *mp,
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int ioflag,
xfs_bmbt_irec_t *imap,
int nimaps,
int *prealloc)
{
xfs_fileoff_t start_fsb;
xfs_filblks_t count_fsb;
xfs_fsblock_t firstblock;
int n, error, imaps;
*prealloc = 0;
if ((offset + count) <= ip->i_size)
return 0;
/*
* If there are any real blocks past eof, then don't
* do any speculative allocation.
*/
start_fsb = XFS_B_TO_FSBT(mp, ((xfs_ufsize_t)(offset + count - 1)));
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_MAXIOFFSET(mp));
while (count_fsb > 0) {
imaps = nimaps;
firstblock = NULLFSBLOCK;
error = xfs_bmapi(NULL, ip, start_fsb, count_fsb, 0,
&firstblock, 0, imap, &imaps, NULL, NULL);
if (error)
return error;
for (n = 0; n < imaps; n++) {
if ((imap[n].br_startblock != HOLESTARTBLOCK) &&
(imap[n].br_startblock != DELAYSTARTBLOCK))
return 0;
start_fsb += imap[n].br_blockcount;
count_fsb -= imap[n].br_blockcount;
}
}
*prealloc = 1;
return 0;
}
/*
* Add a block to the directory.
* This routine is for data and free blocks, not leaf/node blocks
* which are handled by xfs_da_grow_inode.
*/
int
xfs_dir2_grow_inode(
xfs_da_args_t *args,
int space, /* v2 dir's space XFS_DIR2_xxx_SPACE */
xfs_dir2_db_t *dbp) /* out: block number added */
{
xfs_fileoff_t bno; /* directory offset of new block */
int count; /* count of filesystem blocks */
xfs_inode_t *dp; /* incore directory inode */
int error;
int got; /* blocks actually mapped */
int i;
xfs_bmbt_irec_t map; /* single structure for bmap */
int mapi; /* mapping index */
xfs_bmbt_irec_t *mapp; /* bmap mapping structure(s) */
xfs_mount_t *mp;
int nmap; /* number of bmap entries */
xfs_trans_t *tp;
xfs_drfsbno_t nblks;
trace_xfs_dir2_grow_inode(args, space);
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
nblks = dp->i_d.di_nblocks;
/*
* Set lowest possible block in the space requested.
*/
bno = XFS_B_TO_FSBT(mp, space * XFS_DIR2_SPACE_SIZE);
count = mp->m_dirblkfsbs;
/*
* Find the first hole for our block.
*/
if ((error = xfs_bmap_first_unused(tp, dp, count, &bno, XFS_DATA_FORK)))
return error;
nmap = 1;
ASSERT(args->firstblock != NULL);
/*
* Try mapping the new block contiguously (one extent).
*/
if ((error = xfs_bmapi(tp, dp, bno, count,
XFS_BMAPI_WRITE|XFS_BMAPI_METADATA|XFS_BMAPI_CONTIG,
args->firstblock, args->total, &map, &nmap,
args->flist)))
return error;
ASSERT(nmap <= 1);
if (nmap == 1) {
mapp = ↦
mapi = 1;
}
/*
* Didn't work and this is a multiple-fsb directory block.
* Try again with contiguous flag turned on.
*/
else if (nmap == 0 && count > 1) {
xfs_fileoff_t b; /* current file offset */
/*
* Space for maximum number of mappings.
*/
mapp = kmem_alloc(sizeof(*mapp) * count, KM_SLEEP);
/*
* Iterate until we get to the end of our block.
*/
for (b = bno, mapi = 0; b < bno + count; ) {
int c; /* current fsb count */
/*
* Can't map more than MAX_NMAP at once.
*/
nmap = MIN(XFS_BMAP_MAX_NMAP, count);
c = (int)(bno + count - b);
if ((error = xfs_bmapi(tp, dp, b, c,
XFS_BMAPI_WRITE|XFS_BMAPI_METADATA,
args->firstblock, args->total,
&mapp[mapi], &nmap, args->flist))) {
kmem_free(mapp);
return error;
}
if (nmap < 1)
break;
/*
* Add this bunch into our table, go to the next offset.
*/
mapi += nmap;
b = mapp[mapi - 1].br_startoff +
mapp[mapi - 1].br_blockcount;
}
}
/*
* Didn't work.
*/
else {
mapi = 0;
mapp = NULL;
}
/*
* See how many fsb's we got.
*/
for (i = 0, got = 0; i < mapi; i++)
got += mapp[i].br_blockcount;
/*
* Didn't get enough fsb's, or the first/last block's are wrong.
*/
if (got != count || mapp[0].br_startoff != bno ||
mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount !=
bno + count) {
if (mapp != &map)
kmem_free(mapp);
return XFS_ERROR(ENOSPC);
}
/*
* Done with the temporary mapping table.
*/
if (mapp != &map)
kmem_free(mapp);
/* account for newly allocated blocks in reserved blocks total */
args->total -= dp->i_d.di_nblocks - nblks;
*dbp = xfs_dir2_da_to_db(mp, (xfs_dablk_t)bno);
/*
* Update file's size if this is the data space and it grew.
*/
if (space == XFS_DIR2_DATA_SPACE) {
xfs_fsize_t size; /* directory file (data) size */
size = XFS_FSB_TO_B(mp, bno + count);
if (size > dp->i_d.di_size) {
dp->i_d.di_size = size;
xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
}
}
return 0;
}
int
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_filblks_t count_fsb;
xfs_filblks_t numblks_fsb;
xfs_fsblock_t firstfsb;
int nimaps;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint resblks;
int committed;
int error;
xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN, ip, offset, count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
/*
* Reserve enough blocks in this transaction for two complete extent
* btree splits. We may be converting the middle part of an unwritten
* extent and in this case we will insert two new extents in the btree
* each of which could cause a full split.
*
* This reservation amount will be used in the first call to
* xfs_bmbt_split() to select an AG with enough space to satisfy the
* rest of the operation.
*/
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
do {
/*
* set up a transaction to convert the range of extents
* from unwritten to real. Do allocations in a loop until
* we have covered the range passed in.
*
* Note that we open code the transaction allocation here
* to pass KM_NOFS--we can't risk to recursing back into
* the filesystem here as we might be asked to write out
* the same inode that we complete here and might deadlock
* on the iolock.
*/
xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
tp = _xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE, KM_NOFS);
tp->t_flags |= XFS_TRANS_RESERVE;
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
/*
* Modify the unwritten extent state of the buffer.
*/
xfs_bmap_init(&free_list, &firstfsb);
nimaps = 1;
error = xfs_bmapi(tp, ip, offset_fsb, count_fsb,
XFS_BMAPI_WRITE|XFS_BMAPI_CONVERT, &firstfsb,
1, &imap, &nimaps, &free_list, NULL);
if (error)
goto error_on_bmapi_transaction;
error = xfs_bmap_finish(&(tp), &(free_list), &committed);
if (error)
goto error_on_bmapi_transaction;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return XFS_ERROR(error);
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((numblks_fsb = imap.br_blockcount) == 0) {
/*
* The numblks_fsb value should always get
* smaller, otherwise the loop is stuck.
*/
ASSERT(imap.br_blockcount);
break;
}
offset_fsb += numblks_fsb;
count_fsb -= numblks_fsb;
} while (count_fsb > 0);
return 0;
error_on_bmapi_transaction:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
/*
* Pass in a delayed allocate extent, convert it to real extents;
* return to the caller the extent we create which maps on top of
* the originating callers request.
*
* Called without a lock on the inode.
*
* We no longer bother to look at the incoming map - all we have to
* guarantee is that whatever we allocate fills the required range.
*/
int
xfs_iomap_write_allocate(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
xfs_bmbt_irec_t *map,
int *retmap)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb, last_block;
xfs_fileoff_t end_fsb, map_start_fsb;
xfs_fsblock_t first_block;
xfs_bmap_free_t free_list;
xfs_filblks_t count_fsb;
xfs_bmbt_irec_t imap;
xfs_trans_t *tp;
int nimaps, committed;
int error = 0;
int nres;
*retmap = 0;
/*
* Make sure that the dquots are there.
*/
error = xfs_qm_dqattach(ip, 0);
if (error)
return XFS_ERROR(error);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = map->br_blockcount;
map_start_fsb = map->br_startoff;
XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
while (count_fsb != 0) {
/*
* Set up a transaction with which to allocate the
* backing store for the file. Do allocations in a
* loop until we get some space in the range we are
* interested in. The other space that might be allocated
* is in the delayed allocation extent on which we sit
* but before our buffer starts.
*/
nimaps = 0;
while (nimaps == 0) {
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
tp->t_flags |= XFS_TRANS_RESERVE;
nres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK);
error = xfs_trans_reserve(tp, nres,
XFS_WRITE_LOG_RES(mp),
0, XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
xfs_bmap_init(&free_list, &first_block);
/*
* it is possible that the extents have changed since
* we did the read call as we dropped the ilock for a
* while. We have to be careful about truncates or hole
* punchs here - we are not allowed to allocate
* non-delalloc blocks here.
*
* The only protection against truncation is the pages
* for the range we are being asked to convert are
* locked and hence a truncate will block on them
* first.
*
* As a result, if we go beyond the range we really
* need and hit an delalloc extent boundary followed by
* a hole while we have excess blocks in the map, we
* will fill the hole incorrectly and overrun the
* transaction reservation.
*
* Using a single map prevents this as we are forced to
* check each map we look for overlap with the desired
* range and abort as soon as we find it. Also, given
* that we only return a single map, having one beyond
* what we can return is probably a bit silly.
*
* We also need to check that we don't go beyond EOF;
* this is a truncate optimisation as a truncate sets
* the new file size before block on the pages we
* currently have locked under writeback. Because they
* are about to be tossed, we don't need to write them
* back....
*/
nimaps = 1;
end_fsb = XFS_B_TO_FSB(mp, ip->i_size);
error = xfs_bmap_last_offset(NULL, ip, &last_block,
XFS_DATA_FORK);
if (error)
goto trans_cancel;
last_block = XFS_FILEOFF_MAX(last_block, end_fsb);
if ((map_start_fsb + count_fsb) > last_block) {
count_fsb = last_block - map_start_fsb;
if (count_fsb == 0) {
error = EAGAIN;
goto trans_cancel;
}
}
/* Go get the actual blocks */
error = xfs_bmapi(tp, ip, map_start_fsb, count_fsb,
XFS_BMAPI_WRITE, &first_block, 1,
&imap, &nimaps, &free_list, NULL);
if (error)
goto trans_cancel;
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error)
goto trans_cancel;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error0;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
/*
* See if we were able to allocate an extent that
* covers at least part of the callers request
*/
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((offset_fsb >= imap.br_startoff) &&
(offset_fsb < (imap.br_startoff +
imap.br_blockcount))) {
*map = imap;
*retmap = 1;
XFS_STATS_INC(xs_xstrat_quick);
return 0;
}
/*
* So far we have not mapped the requested part of the
* file, just surrounding data, try again.
*/
count_fsb -= imap.br_blockcount;
map_start_fsb = imap.br_startoff + imap.br_blockcount;
}
trans_cancel:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
error0:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
int
xfs_iomap_write_delay(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int ioflag,
xfs_bmbt_irec_t *ret_imap,
int *nmaps)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_fileoff_t last_fsb;
xfs_off_t aligned_offset;
xfs_fileoff_t ioalign;
xfs_fsblock_t firstblock;
xfs_extlen_t extsz;
int nimaps;
xfs_bmbt_irec_t imap[XFS_WRITE_IMAPS];
int prealloc, flushed = 0;
int error;
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
/*
* Make sure that the dquots are there. This doesn't hold
* the ilock across a disk read.
*/
error = xfs_qm_dqattach_locked(ip, 0);
if (error)
return XFS_ERROR(error);
extsz = xfs_get_extsz_hint(ip);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
error = xfs_iomap_eof_want_preallocate(mp, ip, offset, count,
ioflag, imap, XFS_WRITE_IMAPS, &prealloc);
if (error)
return error;
retry:
if (prealloc) {
aligned_offset = XFS_WRITEIO_ALIGN(mp, (offset + count - 1));
ioalign = XFS_B_TO_FSBT(mp, aligned_offset);
last_fsb = ioalign + mp->m_writeio_blocks;
} else {
last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
}
if (prealloc || extsz) {
error = xfs_iomap_eof_align_last_fsb(mp, ip, extsz, &last_fsb);
if (error)
return error;
}
nimaps = XFS_WRITE_IMAPS;
firstblock = NULLFSBLOCK;
error = xfs_bmapi(NULL, ip, offset_fsb,
(xfs_filblks_t)(last_fsb - offset_fsb),
XFS_BMAPI_DELAY | XFS_BMAPI_WRITE |
XFS_BMAPI_ENTIRE, &firstblock, 1, imap,
&nimaps, NULL, NULL);
if (error && (error != ENOSPC))
return XFS_ERROR(error);
/*
* If bmapi returned us nothing, and if we didn't get back EDQUOT,
* then we must have run out of space - flush all other inodes with
* delalloc blocks and retry without EOF preallocation.
*/
if (nimaps == 0) {
xfs_iomap_enter_trace(XFS_IOMAP_WRITE_NOSPACE,
ip, offset, count);
if (flushed)
return XFS_ERROR(ENOSPC);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_flush_inodes(ip);
xfs_ilock(ip, XFS_ILOCK_EXCL);
flushed = 1;
error = 0;
prealloc = 0;
goto retry;
}
if (!(imap[0].br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_cmn_err_fsblock_zero(ip, &imap[0]);
*ret_imap = imap[0];
*nmaps = 1;
return 0;
}
int
xfs_iomap_write_delay(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int ioflag,
xfs_bmbt_irec_t *ret_imap,
int *nmaps)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb;
xfs_fileoff_t last_fsb;
xfs_fsize_t isize;
xfs_fsblock_t firstblock;
int nimaps;
int error;
xfs_bmbt_irec_t imap[XFS_WRITE_IMAPS];
int aeof;
int fsynced = 0;
ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE) != 0);
/*
* Make sure that the dquots are there. This doesn't hold
* the ilock across a disk read.
*/
error = XFS_QM_DQATTACH(mp, ip, XFS_QMOPT_ILOCKED);
if (error)
return XFS_ERROR(error);
retry:
isize = ip->i_d.di_size;
if (io->io_new_size > isize) {
isize = io->io_new_size;
}
aeof = 0;
offset_fsb = XFS_B_TO_FSBT(mp, offset);
last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
/*
* If the caller is doing a write at the end of the file,
* then extend the allocation (and the buffer used for the write)
* out to the file system's write iosize. We clean up any extra
* space left over when the file is closed in xfs_inactive().
*
* For sync writes, we are flushing delayed allocate space to
* try to make additional space available for allocation near
* the filesystem full boundary - preallocation hurts in that
* situation, of course.
*/
if (!(ioflag & BMAPI_SYNC) && ((offset + count) > ip->i_d.di_size)) {
xfs_off_t aligned_offset;
xfs_filblks_t count_fsb;
unsigned int iosize;
xfs_fileoff_t ioalign;
int n;
xfs_fileoff_t start_fsb;
/*
* If there are any real blocks past eof, then don't
* do any speculative allocation.
*/
start_fsb = XFS_B_TO_FSBT(mp,
((xfs_ufsize_t)(offset + count - 1)));
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_MAXIOFFSET(mp));
while (count_fsb > 0) {
nimaps = XFS_WRITE_IMAPS;
error = XFS_BMAPI(mp, NULL, io, start_fsb, count_fsb,
0, &firstblock, 0, imap, &nimaps, NULL);
if (error) {
return error;
}
for (n = 0; n < nimaps; n++) {
if ( !(io->io_flags & XFS_IOCORE_RT) &&
!imap[n].br_startblock) {
cmn_err(CE_PANIC,"Access to block "
"zero: fs <%s> inode: %lld "
"start_block : %llx start_off "
": %llx blkcnt : %llx "
"extent-state : %x \n",
(ip->i_mount)->m_fsname,
(long long)ip->i_ino,
imap[n].br_startblock,
imap[n].br_startoff,
imap[n].br_blockcount,
imap[n].br_state);
}
if ((imap[n].br_startblock != HOLESTARTBLOCK) &&
(imap[n].br_startblock != DELAYSTARTBLOCK)) {
goto write_map;
}
start_fsb += imap[n].br_blockcount;
count_fsb -= imap[n].br_blockcount;
}
}
iosize = mp->m_writeio_blocks;
aligned_offset = XFS_WRITEIO_ALIGN(mp, (offset + count - 1));
ioalign = XFS_B_TO_FSBT(mp, aligned_offset);
last_fsb = ioalign + iosize;
aeof = 1;
}
write_map:
nimaps = XFS_WRITE_IMAPS;
firstblock = NULLFSBLOCK;
/*
* If mounted with the "-o swalloc" option, roundup the allocation
* request to a stripe width boundary if the file size is >=
* stripe width and we are allocating past the allocation eof.
*/
if (!(io->io_flags & XFS_IOCORE_RT) && mp->m_swidth
&& (mp->m_flags & XFS_MOUNT_SWALLOC)
&& (isize >= XFS_FSB_TO_B(mp, mp->m_swidth)) && aeof) {
int eof;
xfs_fileoff_t new_last_fsb;
new_last_fsb = roundup_64(last_fsb, mp->m_swidth);
error = xfs_bmap_eof(ip, new_last_fsb, XFS_DATA_FORK, &eof);
if (error) {
return error;
}
if (eof) {
last_fsb = new_last_fsb;
}
/*
* Roundup the allocation request to a stripe unit (m_dalign) boundary
* if the file size is >= stripe unit size, and we are allocating past
* the allocation eof.
*/
} else if (!(io->io_flags & XFS_IOCORE_RT) && mp->m_dalign &&
(isize >= XFS_FSB_TO_B(mp, mp->m_dalign)) && aeof) {
int eof;
xfs_fileoff_t new_last_fsb;
new_last_fsb = roundup_64(last_fsb, mp->m_dalign);
error = xfs_bmap_eof(ip, new_last_fsb, XFS_DATA_FORK, &eof);
if (error) {
return error;
}
if (eof) {
last_fsb = new_last_fsb;
}
/*
* Round up the allocation request to a real-time extent boundary
* if the file is on the real-time subvolume.
*/
} else if (io->io_flags & XFS_IOCORE_RT && aeof) {
int eof;
xfs_fileoff_t new_last_fsb;
new_last_fsb = roundup_64(last_fsb, mp->m_sb.sb_rextsize);
error = XFS_BMAP_EOF(mp, io, new_last_fsb, XFS_DATA_FORK, &eof);
if (error) {
return error;
}
if (eof)
last_fsb = new_last_fsb;
}
error = xfs_bmapi(NULL, ip, offset_fsb,
(xfs_filblks_t)(last_fsb - offset_fsb),
XFS_BMAPI_DELAY | XFS_BMAPI_WRITE |
XFS_BMAPI_ENTIRE, &firstblock, 1, imap,
&nimaps, NULL);
/*
* This can be EDQUOT, if nimaps == 0
*/
if (error && (error != ENOSPC)) {
return XFS_ERROR(error);
}
/*
* If bmapi returned us nothing, and if we didn't get back EDQUOT,
* then we must have run out of space.
*/
if (nimaps == 0) {
xfs_iomap_enter_trace(XFS_IOMAP_WRITE_NOSPACE,
io, offset, count);
if (xfs_flush_space(ip, &fsynced, &ioflag))
return XFS_ERROR(ENOSPC);
error = 0;
goto retry;
}
*ret_imap = imap[0];
*nmaps = 1;
if ( !(io->io_flags & XFS_IOCORE_RT) && !ret_imap->br_startblock) {
cmn_err(CE_PANIC,"Access to block zero: fs <%s> inode: %lld "
"start_block : %llx start_off : %llx blkcnt : %llx "
"extent-state : %x \n",
(ip->i_mount)->m_fsname,
(long long)ip->i_ino,
ret_imap->br_startblock, ret_imap->br_startoff,
ret_imap->br_blockcount,ret_imap->br_state);
}
return 0;
}
/*
* Allocate a block and fill it with dquots.
* This is called when the bmapi finds a hole.
*/
STATIC int
xfs_qm_dqalloc(
xfs_trans_t **tpp,
xfs_mount_t *mp,
xfs_dquot_t *dqp,
xfs_inode_t *quotip,
xfs_fileoff_t offset_fsb,
xfs_buf_t **O_bpp)
{
xfs_fsblock_t firstblock;
xfs_bmap_free_t flist;
xfs_bmbt_irec_t map;
int nmaps, error, committed;
xfs_buf_t *bp;
xfs_trans_t *tp = *tpp;
ASSERT(tp != NULL);
xfs_dqtrace_entry(dqp, "DQALLOC");
/*
* Initialize the bmap freelist prior to calling bmapi code.
*/
XFS_BMAP_INIT(&flist, &firstblock);
xfs_ilock(quotip, XFS_ILOCK_EXCL);
/*
* Return if this type of quotas is turned off while we didn't
* have an inode lock
*/
if (XFS_IS_THIS_QUOTA_OFF(dqp)) {
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return (ESRCH);
}
/*
* xfs_trans_commit normally decrements the vnode ref count
* when it unlocks the inode. Since we want to keep the quota
* inode around, we bump the vnode ref count now.
*/
VN_HOLD(XFS_ITOV(quotip));
xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
nmaps = 1;
if ((error = xfs_bmapi(tp, quotip,
offset_fsb, XFS_DQUOT_CLUSTER_SIZE_FSB,
XFS_BMAPI_METADATA | XFS_BMAPI_WRITE,
&firstblock,
XFS_QM_DQALLOC_SPACE_RES(mp),
&map, &nmaps, &flist))) {
goto error0;
}
ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
ASSERT(nmaps == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
(map.br_startblock != HOLESTARTBLOCK));
/*
* Keep track of the blkno to save a lookup later
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
/* now we can just get the buffer (there's nothing to read yet) */
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
dqp->q_blkno,
XFS_QI_DQCHUNKLEN(mp),
0);
if (!bp || (error = XFS_BUF_GETERROR(bp)))
goto error1;
/*
* Make a chunk of dquots out of this buffer and log
* the entire thing.
*/
xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id),
dqp->dq_flags & XFS_DQ_ALLTYPES, bp);
/*
* xfs_bmap_finish() may commit the current transaction and
* start a second transaction if the freelist is not empty.
*
* Since we still want to modify this buffer, we need to
* ensure that the buffer is not released on commit of
* the first transaction and ensure the buffer is added to the
* second transaction.
*
* If there is only one transaction then don't stop the buffer
* from being released when it commits later on.
*/
xfs_trans_bhold(tp, bp);
if ((error = xfs_bmap_finish(tpp, &flist, firstblock, &committed))) {
goto error1;
}
if (committed) {
tp = *tpp;
xfs_trans_bjoin(tp, bp);
} else {
xfs_trans_bhold_release(tp, bp);
}
*O_bpp = bp;
return 0;
error1:
xfs_bmap_cancel(&flist);
error0:
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return (error);
}
int
xfs_iomap(
xfs_inode_t *ip,
xfs_off_t offset,
ssize_t count,
int flags,
xfs_iomap_t *iomapp,
int *niomaps)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb, end_fsb;
int error = 0;
int lockmode = 0;
xfs_bmbt_irec_t imap;
int nimaps = 1;
int bmapi_flags = 0;
int iomap_flags = 0;
ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
switch (flags & (BMAPI_READ | BMAPI_WRITE | BMAPI_ALLOCATE)) {
case BMAPI_READ:
xfs_iomap_enter_trace(XFS_IOMAP_READ_ENTER, ip, offset, count);
lockmode = xfs_ilock_map_shared(ip);
bmapi_flags = XFS_BMAPI_ENTIRE;
break;
case BMAPI_WRITE:
xfs_iomap_enter_trace(XFS_IOMAP_WRITE_ENTER, ip, offset, count);
lockmode = XFS_ILOCK_EXCL;
if (flags & BMAPI_IGNSTATE)
bmapi_flags |= XFS_BMAPI_IGSTATE|XFS_BMAPI_ENTIRE;
xfs_ilock(ip, lockmode);
break;
case BMAPI_ALLOCATE:
xfs_iomap_enter_trace(XFS_IOMAP_ALLOC_ENTER, ip, offset, count);
lockmode = XFS_ILOCK_SHARED;
bmapi_flags = XFS_BMAPI_ENTIRE;
/* Attempt non-blocking lock */
if (flags & BMAPI_TRYLOCK) {
if (!xfs_ilock_nowait(ip, lockmode))
return XFS_ERROR(EAGAIN);
} else {
xfs_ilock(ip, lockmode);
}
break;
default:
BUG();
}
ASSERT(offset <= mp->m_maxioffset);
if ((xfs_fsize_t)offset + count > mp->m_maxioffset)
count = mp->m_maxioffset - offset;
end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
error = xfs_bmapi(NULL, ip, offset_fsb,
(xfs_filblks_t)(end_fsb - offset_fsb),
bmapi_flags, NULL, 0, &imap,
&nimaps, NULL, NULL);
if (error)
goto out;
switch (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)) {
case BMAPI_WRITE:
/* If we found an extent, return it */
if (nimaps &&
(imap.br_startblock != HOLESTARTBLOCK) &&
(imap.br_startblock != DELAYSTARTBLOCK)) {
xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip,
offset, count, iomapp, &imap, flags);
break;
}
if (flags & (BMAPI_DIRECT|BMAPI_MMAP)) {
error = xfs_iomap_write_direct(ip, offset, count, flags,
&imap, &nimaps, nimaps);
} else {
error = xfs_iomap_write_delay(ip, offset, count, flags,
&imap, &nimaps);
}
if (!error) {
xfs_iomap_map_trace(XFS_IOMAP_ALLOC_MAP, ip,
offset, count, iomapp, &imap, flags);
}
iomap_flags = IOMAP_NEW;
break;
case BMAPI_ALLOCATE:
/* If we found an extent, return it */
xfs_iunlock(ip, lockmode);
lockmode = 0;
if (nimaps && !isnullstartblock(imap.br_startblock)) {
xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip,
offset, count, iomapp, &imap, flags);
break;
}
error = xfs_iomap_write_allocate(ip, offset, count,
&imap, &nimaps);
break;
}
if (nimaps) {
*niomaps = xfs_imap_to_bmap(ip, offset, &imap,
iomapp, nimaps, *niomaps, iomap_flags);
} else if (niomaps) {
*niomaps = 0;
}
out:
if (lockmode)
xfs_iunlock(ip, lockmode);
return XFS_ERROR(error);
}
/*
* Allocate a block and fill it with dquots.
* This is called when the bmapi finds a hole.
*/
STATIC int
xfs_qm_dqalloc(
xfs_trans_t *tp,
xfs_mount_t *mp,
xfs_dquot_t *dqp,
xfs_inode_t *quotip,
xfs_fileoff_t offset_fsb,
xfs_buf_t **O_bpp)
{
xfs_fsblock_t firstblock;
xfs_bmap_free_t flist;
xfs_bmbt_irec_t map;
int nmaps, error, committed;
xfs_buf_t *bp;
ASSERT(tp != NULL);
xfs_dqtrace_entry(dqp, "DQALLOC");
/*
* Initialize the bmap freelist prior to calling bmapi code.
*/
XFS_BMAP_INIT(&flist, &firstblock);
xfs_ilock(quotip, XFS_ILOCK_EXCL);
/*
* Return if this type of quotas is turned off while we didn't
* have an inode lock
*/
if (XFS_IS_THIS_QUOTA_OFF(dqp)) {
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return (ESRCH);
}
/*
* xfs_trans_commit normally decrements the vnode ref count
* when it unlocks the inode. Since we want to keep the quota
* inode around, we bump the vnode ref count now.
*/
VN_HOLD(XFS_ITOV(quotip));
xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
nmaps = 1;
if ((error = xfs_bmapi(tp, quotip,
offset_fsb, XFS_DQUOT_CLUSTER_SIZE_FSB,
XFS_BMAPI_METADATA | XFS_BMAPI_WRITE,
&firstblock,
XFS_QM_DQALLOC_SPACE_RES(mp),
&map, &nmaps, &flist))) {
goto error0;
}
ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
ASSERT(nmaps == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
(map.br_startblock != HOLESTARTBLOCK));
/*
* Keep track of the blkno to save a lookup later
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
/* now we can just get the buffer (there's nothing to read yet) */
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
dqp->q_blkno,
XFS_QI_DQCHUNKLEN(mp),
0);
if (!bp || (error = XFS_BUF_GETERROR(bp)))
goto error1;
/*
* Make a chunk of dquots out of this buffer and log
* the entire thing.
*/
xfs_qm_init_dquot_blk(tp, mp, INT_GET(dqp->q_core.d_id, ARCH_CONVERT),
dqp->dq_flags & (XFS_DQ_USER|XFS_DQ_GROUP),
bp);
if ((error = xfs_bmap_finish(&tp, &flist, firstblock, &committed))) {
goto error1;
}
*O_bpp = bp;
return 0;
error1:
xfs_bmap_cancel(&flist);
error0:
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return (error);
}
/*
* Getdents (readdir) for leaf and node directories.
* This reads the data blocks only, so is the same for both forms.
*/
int /* error */
xfs_dir2_leaf_getdents(
xfs_inode_t *dp, /* incore directory inode */
void *dirent,
size_t bufsize,
xfs_off_t *offset,
filldir_t filldir)
{
xfs_dabuf_t *bp; /* data block buffer */
int byteoff; /* offset in current block */
xfs_dir2_db_t curdb; /* db for current block */
xfs_dir2_off_t curoff; /* current overall offset */
xfs_dir2_data_t *data; /* data block structure */
xfs_dir2_data_entry_t *dep; /* data entry */
xfs_dir2_data_unused_t *dup; /* unused entry */
int error = 0; /* error return value */
int i; /* temporary loop index */
int j; /* temporary loop index */
int length; /* temporary length value */
xfs_bmbt_irec_t *map; /* map vector for blocks */
xfs_extlen_t map_blocks; /* number of fsbs in map */
xfs_dablk_t map_off; /* last mapped file offset */
int map_size; /* total entries in *map */
int map_valid; /* valid entries in *map */
xfs_mount_t *mp; /* filesystem mount point */
xfs_dir2_off_t newoff; /* new curoff after new blk */
int nmap; /* mappings to ask xfs_bmapi */
char *ptr = NULL; /* pointer to current data */
int ra_current; /* number of read-ahead blks */
int ra_index; /* *map index for read-ahead */
int ra_offset; /* map entry offset for ra */
int ra_want; /* readahead count wanted */
/*
* If the offset is at or past the largest allowed value,
* give up right away.
*/
if (*offset >= XFS_DIR2_MAX_DATAPTR)
return 0;
mp = dp->i_mount;
/*
* Set up to bmap a number of blocks based on the caller's
* buffer size, the directory block size, and the filesystem
* block size.
*/
map_size = howmany(bufsize + mp->m_dirblksize, mp->m_sb.sb_blocksize);
map = kmem_alloc(map_size * sizeof(*map), KM_SLEEP);
map_valid = ra_index = ra_offset = ra_current = map_blocks = 0;
bp = NULL;
/*
* Inside the loop we keep the main offset value as a byte offset
* in the directory file.
*/
curoff = xfs_dir2_dataptr_to_byte(mp, *offset);
/*
* Force this conversion through db so we truncate the offset
* down to get the start of the data block.
*/
map_off = xfs_dir2_db_to_da(mp, xfs_dir2_byte_to_db(mp, curoff));
/*
* Loop over directory entries until we reach the end offset.
* Get more blocks and readahead as necessary.
*/
while (curoff < XFS_DIR2_LEAF_OFFSET) {
/*
* If we have no buffer, or we're off the end of the
* current buffer, need to get another one.
*/
if (!bp || ptr >= (char *)bp->data + mp->m_dirblksize) {
/*
* If we have a buffer, we need to release it and
* take it out of the mapping.
*/
if (bp) {
xfs_da_brelse(NULL, bp);
bp = NULL;
map_blocks -= mp->m_dirblkfsbs;
/*
* Loop to get rid of the extents for the
* directory block.
*/
for (i = mp->m_dirblkfsbs; i > 0; ) {
j = MIN((int)map->br_blockcount, i);
map->br_blockcount -= j;
map->br_startblock += j;
map->br_startoff += j;
/*
* If mapping is done, pitch it from
* the table.
*/
if (!map->br_blockcount && --map_valid)
memmove(&map[0], &map[1],
sizeof(map[0]) *
map_valid);
i -= j;
}
}
/*
* Recalculate the readahead blocks wanted.
*/
ra_want = howmany(bufsize + mp->m_dirblksize,
mp->m_sb.sb_blocksize) - 1;
ASSERT(ra_want >= 0);
/*
* If we don't have as many as we want, and we haven't
* run out of data blocks, get some more mappings.
*/
if (1 + ra_want > map_blocks &&
map_off <
xfs_dir2_byte_to_da(mp, XFS_DIR2_LEAF_OFFSET)) {
/*
* Get more bmaps, fill in after the ones
* we already have in the table.
*/
nmap = map_size - map_valid;
error = xfs_bmapi(NULL, dp,
map_off,
xfs_dir2_byte_to_da(mp,
XFS_DIR2_LEAF_OFFSET) - map_off,
XFS_BMAPI_METADATA, NULL, 0,
&map[map_valid], &nmap, NULL, NULL);
/*
* Don't know if we should ignore this or
* try to return an error.
* The trouble with returning errors
* is that readdir will just stop without
* actually passing the error through.
*/
if (error)
break; /* XXX */
/*
* If we got all the mappings we asked for,
* set the final map offset based on the
* last bmap value received.
* Otherwise, we've reached the end.
*/
if (nmap == map_size - map_valid)
map_off =
map[map_valid + nmap - 1].br_startoff +
map[map_valid + nmap - 1].br_blockcount;
else
map_off =
xfs_dir2_byte_to_da(mp,
XFS_DIR2_LEAF_OFFSET);
/*
* Look for holes in the mapping, and
* eliminate them. Count up the valid blocks.
*/
for (i = map_valid; i < map_valid + nmap; ) {
if (map[i].br_startblock ==
HOLESTARTBLOCK) {
nmap--;
length = map_valid + nmap - i;
if (length)
memmove(&map[i],
&map[i + 1],
sizeof(map[i]) *
length);
} else {
map_blocks +=
map[i].br_blockcount;
i++;
}
}
map_valid += nmap;
}
/*
* No valid mappings, so no more data blocks.
*/
if (!map_valid) {
curoff = xfs_dir2_da_to_byte(mp, map_off);
break;
}
/*
* Read the directory block starting at the first
* mapping.
*/
curdb = xfs_dir2_da_to_db(mp, map->br_startoff);
error = xfs_da_read_buf(NULL, dp, map->br_startoff,
map->br_blockcount >= mp->m_dirblkfsbs ?
XFS_FSB_TO_DADDR(mp, map->br_startblock) :
-1,
&bp, XFS_DATA_FORK);
/*
* Should just skip over the data block instead
* of giving up.
*/
if (error)
break; /* XXX */
/*
* Adjust the current amount of read-ahead: we just
* read a block that was previously ra.
*/
if (ra_current)
ra_current -= mp->m_dirblkfsbs;
/*
* Do we need more readahead?
*/
for (ra_index = ra_offset = i = 0;
ra_want > ra_current && i < map_blocks;
i += mp->m_dirblkfsbs) {
ASSERT(ra_index < map_valid);
/*
* Read-ahead a contiguous directory block.
*/
if (i > ra_current &&
map[ra_index].br_blockcount >=
mp->m_dirblkfsbs) {
//TODO: make sure this is right, should readahead be
//replaced by readbuf?
//xfs_buf_readahead(mp->m_ddev_targp,
// XFS_FSB_TO_DADDR(mp,
// map[ra_index].br_startblock +
// ra_offset),
// (int)BTOBB(mp->m_dirblksize));
//TODO: figure out flags, flags =0
//libxfs_readbuf(mp->m_dev,
// XFS_FSB_TO_DADDR(mp,
// map[ra_index].br_startblock +
// ra_offset),
// (int)BTOBB(mp->m_dirblksize), 0);
ra_current = i;
}
/*
* Read-ahead a non-contiguous directory block.
* This doesn't use our mapping, but this
* is a very rare case.
*/
else if (i > ra_current) {
(void)xfs_da_reada_buf(NULL, dp,
map[ra_index].br_startoff +
ra_offset, XFS_DATA_FORK);
ra_current = i;
}
/*
* Advance offset through the mapping table.
*/
for (j = 0; j < mp->m_dirblkfsbs; j++) {
/*
* The rest of this extent but not
* more than a dir block.
*/
length = MIN(mp->m_dirblkfsbs,
(int)(map[ra_index].br_blockcount -
ra_offset));
j += length;
ra_offset += length;
/*
* Advance to the next mapping if
* this one is used up.
*/
if (ra_offset ==
map[ra_index].br_blockcount) {
ra_offset = 0;
ra_index++;
}
}
}
/*
* Having done a read, we need to set a new offset.
*/
newoff = xfs_dir2_db_off_to_byte(mp, curdb, 0);
/*
* Start of the current block.
*/
if (curoff < newoff)
curoff = newoff;
/*
* Make sure we're in the right block.
*/
else if (curoff > newoff)
ASSERT(xfs_dir2_byte_to_db(mp, curoff) ==
curdb);
data = bp->data;
xfs_dir2_data_check(dp, bp);
/*
* Find our position in the block.
*/
ptr = (char *)&data->u;
byteoff = xfs_dir2_byte_to_off(mp, curoff);
/*
* Skip past the header.
*/
if (byteoff == 0)
curoff += (uint)sizeof(data->hdr);
/*
* Skip past entries until we reach our offset.
*/
else {
while ((char *)ptr - (char *)data < byteoff) {
dup = (xfs_dir2_data_unused_t *)ptr;
if (be16_to_cpu(dup->freetag)
== XFS_DIR2_DATA_FREE_TAG) {
length = be16_to_cpu(dup->length);
ptr += length;
continue;
}
dep = (xfs_dir2_data_entry_t *)ptr;
length =
xfs_dir2_data_entsize(dep->namelen);
ptr += length;
}
/*
* Now set our real offset.
*/
curoff =
xfs_dir2_db_off_to_byte(mp,
xfs_dir2_byte_to_db(mp, curoff),
(char *)ptr - (char *)data);
if (ptr >= (char *)data + mp->m_dirblksize) {
continue;
}
}
}
/*
* We have a pointer to an entry.
* Is it a live one?
*/
dup = (xfs_dir2_data_unused_t *)ptr;
/*
* No, it's unused, skip over it.
*/
if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
length = be16_to_cpu(dup->length);
ptr += length;
curoff += length;
continue;
}
dep = (xfs_dir2_data_entry_t *)ptr;
length = xfs_dir2_data_entsize(dep->namelen);
if (filldir(dirent, (char *)dep->name, dep->namelen,
xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff,
be64_to_cpu(dep->inumber), DT_UNKNOWN))
break;
/*
* Advance to next entry in the block.
*/
ptr += length;
curoff += length;
/* bufsize may have just been a guess; don't go negative */
bufsize = bufsize > length ? bufsize - length : 0;
}
/*
* All done. Set output offset value to current offset.
*/
if (curoff > xfs_dir2_dataptr_to_byte(mp, XFS_DIR2_MAX_DATAPTR))
*offset = XFS_DIR2_MAX_DATAPTR & 0x7fffffff;
else
*offset = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff;
kmem_free(map);
if (bp)
xfs_da_brelse(NULL, bp);
return error;
}
/*
* This routine allocates disk space for the given file.
* Originally derived from xfs_alloc_file_space().
*/
int
libxfs_alloc_file_space(
xfs_inode_t *ip,
xfs_off_t offset,
xfs_off_t len,
int alloc_type,
int attr_flags)
{
xfs_mount_t *mp;
xfs_off_t count;
xfs_filblks_t datablocks;
xfs_filblks_t allocated_fsb;
xfs_filblks_t allocatesize_fsb;
xfs_fsblock_t firstfsb;
xfs_bmap_free_t free_list;
xfs_bmbt_irec_t *imapp;
xfs_bmbt_irec_t imaps[1];
int reccount;
uint resblks;
xfs_fileoff_t startoffset_fsb;
xfs_trans_t *tp;
int xfs_bmapi_flags;
int committed;
int error;
if (len <= 0)
return EINVAL;
count = len;
error = 0;
imapp = &imaps[0];
reccount = 1;
xfs_bmapi_flags = XFS_BMAPI_WRITE | (alloc_type ? XFS_BMAPI_PREALLOC : 0);
mp = ip->i_mount;
startoffset_fsb = XFS_B_TO_FSBT(mp, offset);
allocatesize_fsb = XFS_B_TO_FSB(mp, count);
/* allocate file space until done or until there is an error */
while (allocatesize_fsb && !error) {
datablocks = allocatesize_fsb;
tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
resblks = (uint)XFS_DIOSTRAT_SPACE_RES(mp, datablocks);
error = xfs_trans_reserve(tp, resblks, 0, 0, 0, 0);
if (error)
break;
xfs_trans_ijoin(tp, ip, 0);
xfs_trans_ihold(tp, ip);
XFS_BMAP_INIT(&free_list, &firstfsb);
error = xfs_bmapi(tp, ip, startoffset_fsb, allocatesize_fsb,
xfs_bmapi_flags, &firstfsb, 0, imapp,
&reccount, &free_list);
if (error)
break;
/* complete the transaction */
error = xfs_bmap_finish(&tp, &free_list, firstfsb, &committed);
if (error)
break;
error = xfs_trans_commit(tp, 0, NULL);
if (error)
break;
allocated_fsb = imapp->br_blockcount;
if (reccount == 0)
return ENOSPC;
startoffset_fsb += allocated_fsb;
allocatesize_fsb -= allocated_fsb;
}
return error;
}
int
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_trans_t *tp;
xfs_fileoff_t offset_fsb;
xfs_filblks_t count_fsb;
xfs_filblks_t numblks_fsb;
xfs_bmbt_irec_t imap;
int committed;
int error;
int nres;
int nimaps;
xfs_fsblock_t firstfsb;
xfs_bmap_free_t free_list;
xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN,
&ip->i_iocore, offset, count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
do {
nres = XFS_DIOSTRAT_SPACE_RES(mp, 0);
/*
* set up a transaction to convert the range of extents
* from unwritten to real. Do allocations in a loop until
* we have covered the range passed in.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
error = xfs_trans_reserve(tp, nres,
XFS_WRITE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
goto error0;
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
/*
* Modify the unwritten extent state of the buffer.
*/
XFS_BMAP_INIT(&free_list, &firstfsb);
nimaps = 1;
error = xfs_bmapi(tp, ip, offset_fsb, count_fsb,
XFS_BMAPI_WRITE, &firstfsb,
1, &imap, &nimaps, &free_list);
if (error)
goto error_on_bmapi_transaction;
error = xfs_bmap_finish(&(tp), &(free_list),
firstfsb, &committed);
if (error)
goto error_on_bmapi_transaction;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES, NULL);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
goto error0;
if ( !(io->io_flags & XFS_IOCORE_RT) && !imap.br_startblock) {
cmn_err(CE_PANIC,"Access to block zero: fs <%s> "
"inode: %lld start_block : %llx start_off : "
"%llx blkcnt : %llx extent-state : %x \n",
(ip->i_mount)->m_fsname,
(long long)ip->i_ino,
imap.br_startblock,imap.br_startoff,
imap.br_blockcount,imap.br_state);
}
if ((numblks_fsb = imap.br_blockcount) == 0) {
/*
* The numblks_fsb value should always get
* smaller, otherwise the loop is stuck.
*/
ASSERT(imap.br_blockcount);
break;
}
offset_fsb += numblks_fsb;
count_fsb -= numblks_fsb;
} while (count_fsb > 0);
return 0;
error_on_bmapi_transaction:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
xfs_iunlock(ip, XFS_ILOCK_EXCL);
error0:
return XFS_ERROR(error);
}
/*
* Pass in a delayed allocate extent, convert it to real extents;
* return to the caller the extent we create which maps on top of
* the originating callers request.
*
* Called without a lock on the inode.
*/
int
xfs_iomap_write_allocate(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
xfs_bmbt_irec_t *map,
int *retmap)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb, last_block;
xfs_fileoff_t end_fsb, map_start_fsb;
xfs_fsblock_t first_block;
xfs_bmap_free_t free_list;
xfs_filblks_t count_fsb;
xfs_bmbt_irec_t imap[XFS_STRAT_WRITE_IMAPS];
xfs_trans_t *tp;
int i, nimaps, committed;
int error = 0;
int nres;
*retmap = 0;
/*
* Make sure that the dquots are there.
*/
if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
return XFS_ERROR(error);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = map->br_blockcount;
map_start_fsb = map->br_startoff;
XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
while (count_fsb != 0) {
/*
* Set up a transaction with which to allocate the
* backing store for the file. Do allocations in a
* loop until we get some space in the range we are
* interested in. The other space that might be allocated
* is in the delayed allocation extent on which we sit
* but before our buffer starts.
*/
nimaps = 0;
while (nimaps == 0) {
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
nres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK);
error = xfs_trans_reserve(tp, nres,
XFS_WRITE_LOG_RES(mp),
0, XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error == ENOSPC) {
error = xfs_trans_reserve(tp, 0,
XFS_WRITE_LOG_RES(mp),
0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
}
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
XFS_BMAP_INIT(&free_list, &first_block);
nimaps = XFS_STRAT_WRITE_IMAPS;
/*
* Ensure we don't go beyond eof - it is possible
* the extents changed since we did the read call,
* we dropped the ilock in the interim.
*/
end_fsb = XFS_B_TO_FSB(mp, ip->i_d.di_size);
xfs_bmap_last_offset(NULL, ip, &last_block,
XFS_DATA_FORK);
last_block = XFS_FILEOFF_MAX(last_block, end_fsb);
if ((map_start_fsb + count_fsb) > last_block) {
count_fsb = last_block - map_start_fsb;
if (count_fsb == 0) {
error = EAGAIN;
goto trans_cancel;
}
}
/* Go get the actual blocks */
error = xfs_bmapi(tp, ip, map_start_fsb, count_fsb,
XFS_BMAPI_WRITE, &first_block, 1,
imap, &nimaps, &free_list);
if (error)
goto trans_cancel;
error = xfs_bmap_finish(&tp, &free_list,
first_block, &committed);
if (error)
goto trans_cancel;
error = xfs_trans_commit(tp,
XFS_TRANS_RELEASE_LOG_RES, NULL);
if (error)
goto error0;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
/*
* See if we were able to allocate an extent that
* covers at least part of the callers request
*/
for (i = 0; i < nimaps; i++) {
if ( !(io->io_flags & XFS_IOCORE_RT) &&
!imap[i].br_startblock) {
cmn_err(CE_PANIC,"Access to block zero: "
"fs <%s> inode: %lld "
"start_block : %llx start_off : %llx "
"blkcnt : %llx extent-state : %x \n",
(ip->i_mount)->m_fsname,
(long long)ip->i_ino,
imap[i].br_startblock,
imap[i].br_startoff,
imap[i].br_blockcount,imap[i].br_state);
}
if ((offset_fsb >= imap[i].br_startoff) &&
(offset_fsb < (imap[i].br_startoff +
imap[i].br_blockcount))) {
*map = imap[i];
*retmap = 1;
XFS_STATS_INC(xs_xstrat_quick);
return 0;
}
count_fsb -= imap[i].br_blockcount;
}
/* So far we have not mapped the requested part of the
* file, just surrounding data, try again.
*/
nimaps--;
map_start_fsb = imap[nimaps].br_startoff +
imap[nimaps].br_blockcount;
}
trans_cancel:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
error0:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
int /* error (positive) */
xfs_zero_eof(
xfs_inode_t *ip,
xfs_off_t offset, /* starting I/O offset */
xfs_fsize_t isize) /* current inode size */
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t start_zero_fsb;
xfs_fileoff_t end_zero_fsb;
xfs_fileoff_t zero_count_fsb;
xfs_fileoff_t last_fsb;
xfs_fileoff_t zero_off;
xfs_fsize_t zero_len;
int nimaps;
int error = 0;
xfs_bmbt_irec_t imap;
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
ASSERT(offset > isize);
/*
* First handle zeroing the block on which isize resides.
* We only zero a part of that block so it is handled specially.
*/
error = xfs_zero_last_block(ip, offset, isize);
if (error) {
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
return error;
}
/*
* Calculate the range between the new size and the old
* where blocks needing to be zeroed may exist. To get the
* block where the last byte in the file currently resides,
* we need to subtract one from the size and truncate back
* to a block boundary. We subtract 1 in case the size is
* exactly on a block boundary.
*/
last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
if (last_fsb == end_zero_fsb) {
/*
* The size was only incremented on its last block.
* We took care of that above, so just return.
*/
return 0;
}
ASSERT(start_zero_fsb <= end_zero_fsb);
while (start_zero_fsb <= end_zero_fsb) {
nimaps = 1;
zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb,
0, NULL, 0, &imap, &nimaps, NULL, NULL);
if (error) {
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
return error;
}
ASSERT(nimaps > 0);
if (imap.br_state == XFS_EXT_UNWRITTEN ||
imap.br_startblock == HOLESTARTBLOCK) {
/*
* This loop handles initializing pages that were
* partially initialized by the code below this
* loop. It basically zeroes the part of the page
* that sits on a hole and sets the page as P_HOLE
* and calls remapf if it is a mapped file.
*/
start_zero_fsb = imap.br_startoff + imap.br_blockcount;
ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
continue;
}
/*
* There are blocks we need to zero.
* Drop the inode lock while we're doing the I/O.
* We'll still have the iolock to protect us.
*/
xfs_iunlock(ip, XFS_ILOCK_EXCL);
zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);
if ((zero_off + zero_len) > offset)
zero_len = offset - zero_off;
error = xfs_iozero(ip, zero_off, zero_len);
if (error) {
goto out_lock;
}
start_zero_fsb = imap.br_startoff + imap.br_blockcount;
ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
xfs_ilock(ip, XFS_ILOCK_EXCL);
}
return 0;
out_lock:
xfs_ilock(ip, XFS_ILOCK_EXCL);
ASSERT(error >= 0);
return error;
}
STATIC int
xfs_map_blocks(
struct inode *inode,
loff_t offset,
struct xfs_bmbt_irec *imap,
int type,
int nonblocking)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
ssize_t count = 1 << inode->i_blkbits;
xfs_fileoff_t offset_fsb, end_fsb;
int error = 0;
int bmapi_flags = XFS_BMAPI_ENTIRE;
int nimaps = 1;
if (XFS_FORCED_SHUTDOWN(mp))
return -XFS_ERROR(EIO);
if (type == IO_UNWRITTEN)
bmapi_flags |= XFS_BMAPI_IGSTATE;
if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) {
if (nonblocking)
return -XFS_ERROR(EAGAIN);
xfs_ilock(ip, XFS_ILOCK_SHARED);
}
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
(ip->i_df.if_flags & XFS_IFEXTENTS));
ASSERT(offset <= mp->m_maxioffset);
if (offset + count > mp->m_maxioffset)
count = mp->m_maxioffset - offset;
end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
error = xfs_bmapi(NULL, ip, offset_fsb, end_fsb - offset_fsb,
bmapi_flags, NULL, 0, imap, &nimaps, NULL);
xfs_iunlock(ip, XFS_ILOCK_SHARED);
if (error)
return -XFS_ERROR(error);
if (type == IO_DELALLOC &&
(!nimaps || isnullstartblock(imap->br_startblock))) {
error = xfs_iomap_write_allocate(ip, offset, count, imap);
if (!error)
trace_xfs_map_blocks_alloc(ip, offset, count, type, imap);
return -XFS_ERROR(error);
}
#ifdef DEBUG
if (type == IO_UNWRITTEN) {
ASSERT(nimaps);
ASSERT(imap->br_startblock != HOLESTARTBLOCK);
ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
}
#endif
if (nimaps)
trace_xfs_map_blocks_found(ip, offset, count, type, imap);
return 0;
}
int
xfs_iomap_write_direct(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int flags,
xfs_bmbt_irec_t *ret_imap,
int *nmaps,
int found)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_fileoff_t last_fsb;
xfs_filblks_t count_fsb, resaligned;
xfs_fsblock_t firstfsb;
xfs_extlen_t extsz, temp;
int nimaps;
int bmapi_flag;
int quota_flag;
int rt;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint qblocks, resblks, resrtextents;
int committed;
int error;
/*
* Make sure that the dquots are there. This doesn't hold
* the ilock across a disk read.
*/
error = xfs_qm_dqattach_locked(ip, 0);
if (error)
return XFS_ERROR(error);
rt = XFS_IS_REALTIME_INODE(ip);
extsz = xfs_get_extsz_hint(ip);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
if ((offset + count) > ip->i_size) {
error = xfs_iomap_eof_align_last_fsb(mp, ip, extsz, &last_fsb);
if (error)
goto error_out;
} else {
if (found && (ret_imap->br_startblock == HOLESTARTBLOCK))
last_fsb = MIN(last_fsb, (xfs_fileoff_t)
ret_imap->br_blockcount +
ret_imap->br_startoff);
}
count_fsb = last_fsb - offset_fsb;
ASSERT(count_fsb > 0);
resaligned = count_fsb;
if (unlikely(extsz)) {
if ((temp = do_mod(offset_fsb, extsz)))
resaligned += temp;
if ((temp = do_mod(resaligned, extsz)))
resaligned += extsz - temp;
}
if (unlikely(rt)) {
resrtextents = qblocks = resaligned;
resrtextents /= mp->m_sb.sb_rextsize;
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
quota_flag = XFS_QMOPT_RES_RTBLKS;
} else {
resrtextents = 0;
resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned);
quota_flag = XFS_QMOPT_RES_REGBLKS;
}
/*
* Allocate and setup the transaction
*/
xfs_iunlock(ip, XFS_ILOCK_EXCL);
tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), resrtextents,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
/*
* Check for running out of space, note: need lock to return
*/
if (error)
xfs_trans_cancel(tp, 0);
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (error)
goto error_out;
error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
if (error)
goto error1;
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
bmapi_flag = XFS_BMAPI_WRITE;
if ((flags & BMAPI_DIRECT) && (offset < ip->i_size || extsz))
bmapi_flag |= XFS_BMAPI_PREALLOC;
/*
* Issue the xfs_bmapi() call to allocate the blocks
*/
xfs_bmap_init(&free_list, &firstfsb);
nimaps = 1;
error = xfs_bmapi(tp, ip, offset_fsb, count_fsb, bmapi_flag,
&firstfsb, 0, &imap, &nimaps, &free_list, NULL);
if (error)
goto error0;
/*
* Complete the transaction
*/
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error)
goto error0;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error_out;
/*
* Copy any maps to caller's array and return any error.
*/
if (nimaps == 0) {
error = ENOSPC;
goto error_out;
}
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip))) {
error = xfs_cmn_err_fsblock_zero(ip, &imap);
goto error_out;
}
*ret_imap = imap;
*nmaps = 1;
return 0;
error0: /* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
xfs_bmap_cancel(&free_list);
xfs_trans_unreserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
error1: /* Just cancel transaction */
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
*nmaps = 0; /* nothing set-up here */
error_out:
return XFS_ERROR(error);
}
/*
* Maps a dquot to the buffer containing its on-disk version.
* This returns a ptr to the buffer containing the on-disk dquot
* in the bpp param, and a ptr to the on-disk dquot within that buffer
*/
STATIC int
xfs_qm_dqtobp(
xfs_trans_t **tpp,
xfs_dquot_t *dqp,
xfs_disk_dquot_t **O_ddpp,
xfs_buf_t **O_bpp,
uint flags)
{
xfs_bmbt_irec_t map;
int nmaps, error;
xfs_buf_t *bp;
xfs_inode_t *quotip;
xfs_mount_t *mp;
xfs_disk_dquot_t *ddq;
xfs_dqid_t id;
boolean_t newdquot;
xfs_trans_t *tp = (tpp ? *tpp : NULL);
mp = dqp->q_mount;
id = be32_to_cpu(dqp->q_core.d_id);
nmaps = 1;
newdquot = B_FALSE;
/*
* If we don't know where the dquot lives, find out.
*/
if (dqp->q_blkno == (xfs_daddr_t) 0) {
/* We use the id as an index */
dqp->q_fileoffset = (xfs_fileoff_t)id / XFS_QM_DQPERBLK(mp);
nmaps = 1;
quotip = XFS_DQ_TO_QIP(dqp);
xfs_ilock(quotip, XFS_ILOCK_SHARED);
/*
* Return if this type of quotas is turned off while we didn't
* have an inode lock
*/
if (XFS_IS_THIS_QUOTA_OFF(dqp)) {
xfs_iunlock(quotip, XFS_ILOCK_SHARED);
return (ESRCH);
}
/*
* Find the block map; no allocations yet
*/
error = xfs_bmapi(NULL, quotip, dqp->q_fileoffset,
XFS_DQUOT_CLUSTER_SIZE_FSB,
XFS_BMAPI_METADATA,
NULL, 0, &map, &nmaps, NULL);
xfs_iunlock(quotip, XFS_ILOCK_SHARED);
if (error)
return (error);
ASSERT(nmaps == 1);
ASSERT(map.br_blockcount == 1);
/*
* offset of dquot in the (fixed sized) dquot chunk.
*/
dqp->q_bufoffset = (id % XFS_QM_DQPERBLK(mp)) *
sizeof(xfs_dqblk_t);
if (map.br_startblock == HOLESTARTBLOCK) {
/*
* We don't allocate unless we're asked to
*/
if (!(flags & XFS_QMOPT_DQALLOC))
return (ENOENT);
ASSERT(tp);
if ((error = xfs_qm_dqalloc(tpp, mp, dqp, quotip,
dqp->q_fileoffset, &bp)))
return (error);
tp = *tpp;
newdquot = B_TRUE;
} else {
/*
* store the blkno etc so that we don't have to do the
* mapping all the time
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
}
}
ASSERT(dqp->q_blkno != DELAYSTARTBLOCK);
ASSERT(dqp->q_blkno != HOLESTARTBLOCK);
/*
* Read in the buffer, unless we've just done the allocation
* (in which case we already have the buf).
*/
if (! newdquot) {
xfs_dqtrace_entry(dqp, "DQTOBP READBUF");
if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
dqp->q_blkno,
XFS_QI_DQCHUNKLEN(mp),
0, &bp))) {
return (error);
}
if (error || !bp)
return XFS_ERROR(error);
}
ASSERT(XFS_BUF_ISBUSY(bp));
ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
/*
* calculate the location of the dquot inside the buffer.
*/
ddq = (xfs_disk_dquot_t *)((char *)XFS_BUF_PTR(bp) + dqp->q_bufoffset);
/*
* A simple sanity check in case we got a corrupted dquot...
*/
if (xfs_qm_dqcheck(ddq, id, dqp->dq_flags & XFS_DQ_ALLTYPES,
flags & (XFS_QMOPT_DQREPAIR|XFS_QMOPT_DOWARN),
"dqtobp")) {
if (!(flags & XFS_QMOPT_DQREPAIR)) {
xfs_trans_brelse(tp, bp);
return XFS_ERROR(EIO);
}
XFS_BUF_BUSY(bp); /* We dirtied this */
}
*O_bpp = bp;
*O_ddpp = ddq;
return (0);
}
int
xfs_iomap_write_direct(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int flags,
xfs_bmbt_irec_t *ret_imap,
int *nmaps,
int found)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb;
xfs_fileoff_t last_fsb;
xfs_filblks_t count_fsb;
xfs_fsblock_t firstfsb;
int nimaps;
int error;
int bmapi_flag;
int quota_flag;
int rt;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
xfs_filblks_t qblocks, resblks;
int committed;
int resrtextents;
/*
* Make sure that the dquots are there. This doesn't hold
* the ilock across a disk read.
*/
error = XFS_QM_DQATTACH(ip->i_mount, ip, XFS_QMOPT_ILOCKED);
if (error)
return XFS_ERROR(error);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
count_fsb = last_fsb - offset_fsb;
if (found && (ret_imap->br_startblock == HOLESTARTBLOCK)) {
xfs_fileoff_t map_last_fsb;
map_last_fsb = ret_imap->br_blockcount + ret_imap->br_startoff;
if (map_last_fsb < last_fsb) {
last_fsb = map_last_fsb;
count_fsb = last_fsb - offset_fsb;
}
ASSERT(count_fsb > 0);
}
/*
* Determine if reserving space on the data or realtime partition.
*/
if ((rt = XFS_IS_REALTIME_INODE(ip))) {
xfs_extlen_t extsz;
if (!(extsz = ip->i_d.di_extsize))
extsz = mp->m_sb.sb_rextsize;
resrtextents = qblocks = (count_fsb + extsz - 1);
do_div(resrtextents, mp->m_sb.sb_rextsize);
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
quota_flag = XFS_QMOPT_RES_RTBLKS;
} else {
resrtextents = 0;
resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, count_fsb);
quota_flag = XFS_QMOPT_RES_REGBLKS;
}
/*
* Allocate and setup the transaction
*/
xfs_iunlock(ip, XFS_ILOCK_EXCL);
tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), resrtextents,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
/*
* Check for running out of space, note: need lock to return
*/
if (error)
xfs_trans_cancel(tp, 0);
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (error)
goto error_out;
if (XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, tp, ip, qblocks, 0, quota_flag)) {
error = (EDQUOT);
goto error1;
}
bmapi_flag = XFS_BMAPI_WRITE;
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
if (!(flags & BMAPI_MMAP) && (offset < ip->i_d.di_size || rt))
bmapi_flag |= XFS_BMAPI_PREALLOC;
/*
* Issue the bmapi() call to allocate the blocks
*/
XFS_BMAP_INIT(&free_list, &firstfsb);
nimaps = 1;
error = xfs_bmapi(tp, ip, offset_fsb, count_fsb,
bmapi_flag, &firstfsb, 0, &imap, &nimaps, &free_list);
if (error)
goto error0;
/*
* Complete the transaction
*/
error = xfs_bmap_finish(&tp, &free_list, firstfsb, &committed);
if (error)
goto error0;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES, NULL);
if (error)
goto error_out;
/*
* Copy any maps to caller's array and return any error.
*/
if (nimaps == 0) {
error = (ENOSPC);
goto error_out;
}
*ret_imap = imap;
*nmaps = 1;
if ( !(io->io_flags & XFS_IOCORE_RT) && !ret_imap->br_startblock) {
cmn_err(CE_PANIC,"Access to block zero: fs <%s> inode: %lld "
"start_block : %llx start_off : %llx blkcnt : %llx "
"extent-state : %x \n",
(ip->i_mount)->m_fsname,
(long long)ip->i_ino,
ret_imap->br_startblock, ret_imap->br_startoff,
ret_imap->br_blockcount,ret_imap->br_state);
}
return 0;
error0: /* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
xfs_bmap_cancel(&free_list);
XFS_TRANS_UNRESERVE_QUOTA_NBLKS(mp, tp, ip, qblocks, 0, quota_flag);
error1: /* Just cancel transaction */
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
*nmaps = 0; /* nothing set-up here */
error_out:
return XFS_ERROR(error);
}
