/*
- * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2000-2006 Silicon Graphics, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
#include "xfs_trans_priv.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
-#include "xfs_dir.h"
#include "xfs_dir2.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
#include "xfs_bmap_btree.h"
#include "xfs_alloc_btree.h"
#include "xfs_ialloc_btree.h"
-#include "xfs_dir_sf.h"
#include "xfs_dir2_sf.h"
#include "xfs_attr_sf.h"
#include "xfs_dinode.h"
xfs_dinode_t *dip;
/*
- * Call the space managment code to find the location of the
+ * Call the space management code to find the location of the
* inode on disk.
*/
imap.im_blkno = 0;
xfs_inode_t *ip,
xfs_dinode_t **dipp,
xfs_buf_t **bpp,
- xfs_daddr_t bno)
+ xfs_daddr_t bno,
+ uint imap_flags)
{
+ xfs_imap_t imap;
xfs_buf_t *bp;
int error;
- xfs_imap_t imap;
-#ifdef __KERNEL__
int i;
int ni;
-#endif
if (ip->i_blkno == (xfs_daddr_t)0) {
/*
* inode on disk.
*/
imap.im_blkno = bno;
- error = xfs_imap(mp, tp, ip->i_ino, &imap, XFS_IMAP_LOOKUP);
- if (error != 0) {
+ if ((error = xfs_imap(mp, tp, ip->i_ino, &imap,
+ XFS_IMAP_LOOKUP | imap_flags)))
return error;
- }
/*
* If the inode number maps to a block outside the bounds
*/
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, imap.im_blkno,
(int)imap.im_len, XFS_BUF_LOCK, &bp);
-
if (error) {
#ifdef DEBUG
xfs_fs_cmn_err(CE_ALERT, mp, "xfs_itobp: "
#endif /* DEBUG */
return error;
}
-#ifdef __KERNEL__
+
/*
* Validate the magic number and version of every inode in the buffer
* (if DEBUG kernel) or the first inode in the buffer, otherwise.
+ * No validation is done here in userspace (xfs_repair).
*/
-#ifdef DEBUG
- ni = BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog;
-#else
- ni = 1;
+#if !defined(__KERNEL__)
+ ni = 0;
+#elif defined(DEBUG)
+ ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 :
+ (BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog);
+#else /* usual case */
+ ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 : 1;
#endif
+
for (i = 0; i < ni; i++) {
int di_ok;
xfs_dinode_t *dip;
if (unlikely(XFS_TEST_ERROR(!di_ok, mp, XFS_ERRTAG_ITOBP_INOTOBP,
XFS_RANDOM_ITOBP_INOTOBP))) {
#ifdef DEBUG
- prdev("bad inode magic/vsn daddr %lld #%d (magic=%x)",
- mp->m_ddev_targp,
+ if (!(imap_flags & XFS_IMAP_BULKSTAT))
+ cmn_err(CE_ALERT,
+ "Device %s - bad inode magic/vsn "
+ "daddr %lld #%d (magic=%x)",
+ XFS_BUFTARG_NAME(mp->m_ddev_targp),
(unsigned long long)imap.im_blkno, i,
INT_GET(dip->di_core.di_magic, ARCH_CONVERT));
#endif
return XFS_ERROR(EFSCORRUPTED);
}
}
-#endif /* __KERNEL__ */
xfs_inobp_check(mp, bp);
switch (INT_GET(dip->di_core.di_aformat, ARCH_CONVERT)) {
case XFS_DINODE_FMT_LOCAL:
atp = (xfs_attr_shortform_t *)XFS_DFORK_APTR(dip);
- size = (int)INT_GET(atp->hdr.totsize, ARCH_CONVERT);
+ size = be16_to_cpu(atp->hdr.totsize);
error = xfs_iformat_local(ip, dip, XFS_ATTR_FORK, size);
break;
case XFS_DINODE_FMT_EXTENTS:
STATIC uint
_xfs_dic2xflags(
- xfs_dinode_core_t *dic,
__uint16_t di_flags)
{
uint flags = 0;
flags |= XFS_XFLAG_EXTSIZE;
if (di_flags & XFS_DIFLAG_EXTSZINHERIT)
flags |= XFS_XFLAG_EXTSZINHERIT;
+ if (di_flags & XFS_DIFLAG_NODEFRAG)
+ flags |= XFS_XFLAG_NODEFRAG;
}
return flags;
{
xfs_dinode_core_t *dic = &ip->i_d;
- return _xfs_dic2xflags(dic, dic->di_flags) |
- (XFS_CFORK_Q(dic) ? XFS_XFLAG_HASATTR : 0);
+ return _xfs_dic2xflags(dic->di_flags) |
+ (XFS_CFORK_Q(dic) ? XFS_XFLAG_HASATTR : 0);
}
uint
xfs_dic2xflags(
xfs_dinode_core_t *dic)
{
- return _xfs_dic2xflags(dic, INT_GET(dic->di_flags, ARCH_CONVERT)) |
- (XFS_CFORK_Q_DISK(dic) ? XFS_XFLAG_HASATTR : 0);
+ return _xfs_dic2xflags(INT_GET(dic->di_flags, ARCH_CONVERT)) |
+ (XFS_CFORK_Q_DISK(dic) ? XFS_XFLAG_HASATTR : 0);
}
/*
* Given a mount structure and an inode number, return a pointer
- * to a newly allocated in-core inode coresponding to the given
+ * to a newly allocated in-core inode corresponding to the given
* inode number.
*
* Initialize the inode's attributes and extent pointers if it
* return NULL as well. Set i_blkno to 0 so that xfs_itobp() will
* know that this is a new incore inode.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &bp, bno);
-
- if (error != 0) {
+ error = xfs_itobp(mp, tp, ip, &dip, &bp, bno, 0);
+ if (error) {
kmem_zone_free(xfs_inode_zone, ip);
return error;
}
{
xfs_ino_t ino;
xfs_inode_t *ip;
- vnode_t *vp;
+ bhv_vnode_t *vp;
uint flags;
int error;
di_flags |= XFS_DIFLAG_NOSYMLINKS;
if (pip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
di_flags |= XFS_DIFLAG_PROJINHERIT;
+ if ((pip->i_d.di_flags & XFS_DIFLAG_NODEFRAG) &&
+ xfs_inherit_nodefrag)
+ di_flags |= XFS_DIFLAG_NODEFRAG;
ip->i_d.di_flags |= di_flags;
}
/* FALLTHROUGH */
*/
xfs_trans_log_inode(tp, ip, flags);
- /* now that we have an i_mode we can set Linux inode ops (& unlock) */
- VFS_INIT_VNODE(XFS_MTOVFS(tp->t_mountp), vp, XFS_ITOBHV(ip), 1);
+ /* now that we have an i_mode we can setup inode ops and unlock */
+ bhv_vfs_init_vnode(XFS_MTOVFS(tp->t_mountp), vp, XFS_ITOBHV(ip), 1);
*ipp = ip;
return 0;
(xfs_ufsize_t)XFS_MAXIOFFSET(mp)) -
map_first),
XFS_BMAPI_ENTIRE, NULL, 0, imaps, &nimaps,
- NULL))
+ NULL, NULL))
return;
ASSERT(nimaps == 1);
ASSERT(imaps[0].br_startblock == HOLESTARTBLOCK);
(void*)(unsigned long)((toss_finish >> 32) & 0xffffffff),
(void*)(unsigned long)(toss_finish & 0xffffffff),
(void*)(unsigned long)current_cpu(),
- (void*)0,
- (void*)0,
- (void*)0,
- (void*)0);
+ (void*)(unsigned long)current_pid(),
+ (void*)NULL,
+ (void*)NULL,
+ (void*)NULL);
}
#else
#define xfs_itrunc_trace(tag, ip, flag, new_size, toss_start, toss_finish)
* calling into the buffer/page cache code and we can't hold the
* inode lock when we do so.
*
+ * We need to wait for any direct I/Os in flight to complete before we
+ * proceed with the truncate. This is needed to prevent the extents
+ * being read or written by the direct I/Os from being removed while the
+ * I/O is in flight as there is no other method of synchronising
+ * direct I/O with the truncate operation. Also, because we hold
+ * the IOLOCK in exclusive mode, we prevent new direct I/Os from being
+ * started until the truncate completes and drops the lock. Essentially,
+ * the vn_iowait() call forms an I/O barrier that provides strict ordering
+ * between direct I/Os and the truncate operation.
+ *
* The flags parameter can have either the value XFS_ITRUNC_DEFINITE
* or XFS_ITRUNC_MAYBE. The XFS_ITRUNC_MAYBE value should be used
* in the case that the caller is locking things out of order and
xfs_fsize_t last_byte;
xfs_off_t toss_start;
xfs_mount_t *mp;
- vnode_t *vp;
+ bhv_vnode_t *vp;
ASSERT(ismrlocked(&ip->i_iolock, MR_UPDATE) != 0);
ASSERT((new_size == 0) || (new_size <= ip->i_d.di_size));
mp = ip->i_mount;
vp = XFS_ITOV(ip);
+
+ vn_iowait(vp); /* wait for the completion of any pending DIOs */
+
/*
- * Call VOP_TOSS_PAGES() or VOP_FLUSHINVAL_PAGES() to get rid of pages and buffers
+ * Call toss_pages or flushinval_pages to get rid of pages
* overlapping the region being removed. We have to use
- * the less efficient VOP_FLUSHINVAL_PAGES() in the case that the
+ * the less efficient flushinval_pages in the case that the
* caller may not be able to finish the truncate without
* dropping the inode's I/O lock. Make sure
* to catch any pages brought in by buffers overlapping
* so that we don't toss things on the same block as
* new_size but before it.
*
- * Before calling VOP_TOSS_PAGES() or VOP_FLUSHINVAL_PAGES(), make sure to
+ * Before calling toss_page or flushinval_pages, make sure to
* call remapf() over the same region if the file is mapped.
* This frees up mapped file references to the pages in the
- * given range and for the VOP_FLUSHINVAL_PAGES() case it ensures
+ * given range and for the flushinval_pages case it ensures
* that we get the latest mapped changes flushed out.
*/
toss_start = XFS_B_TO_FSB(mp, (xfs_ufsize_t)new_size);
last_byte);
if (last_byte > toss_start) {
if (flags & XFS_ITRUNC_DEFINITE) {
- VOP_TOSS_PAGES(vp, toss_start, -1, FI_REMAPF_LOCKED);
+ bhv_vop_toss_pages(vp, toss_start, -1, FI_REMAPF_LOCKED);
} else {
- VOP_FLUSHINVAL_PAGES(vp, toss_start, -1, FI_REMAPF_LOCKED);
+ bhv_vop_flushinval_pages(vp, toss_start, -1, FI_REMAPF_LOCKED);
}
}
* runs.
*/
XFS_BMAP_INIT(&free_list, &first_block);
- error = xfs_bunmapi(ntp, ip, first_unmap_block,
- unmap_len,
+ error = XFS_BUNMAPI(mp, ntp, &ip->i_iocore,
+ first_unmap_block, unmap_len,
XFS_BMAPI_AFLAG(fork) |
(sync ? 0 : XFS_BMAPI_ASYNC),
XFS_ITRUNC_MAX_EXTENTS,
- &first_block, &free_list, &done);
+ &first_block, &free_list,
+ NULL, &done);
if (error) {
/*
* If the bunmapi call encounters an error,
* Here we put the head pointer into our next pointer,
* and then we fall through to point the head at us.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
if (error) {
return error;
}
* of dealing with the buffer when there is no need to
* change it.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
* Now last_ibp points to the buffer previous to us on
* the unlinked list. Pull us from the list.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
/*
* Decrement the pin count of the given inode, and wake up
* anyone in xfs_iwait_unpin() if the count goes to 0. The
- * inode must have been previoulsy pinned with a call to xfs_ipin().
+ * inode must have been previously pinned with a call to xfs_ipin().
*/
void
xfs_iunpin(
ASSERT(atomic_read(&ip->i_pincount) > 0);
if (atomic_dec_and_test(&ip->i_pincount)) {
- vnode_t *vp = XFS_ITOV_NULL(ip);
+ /*
+ * If the inode is currently being reclaimed, the
+ * linux inode _and_ the xfs vnode may have been
+ * freed so we cannot reference either of them safely.
+ * Hence we should not try to do anything to them
+ * if the xfs inode is currently in the reclaim
+ * path.
+ *
+ * However, we still need to issue the unpin wakeup
+ * call as the inode reclaim may be blocked waiting for
+ * the inode to become unpinned.
+ */
+ if (!(ip->i_flags & (XFS_IRECLAIM|XFS_IRECLAIMABLE))) {
+ bhv_vnode_t *vp = XFS_ITOV_NULL(ip);
- /* make sync come back and flush this inode */
- if (vp) {
- struct inode *inode = LINVFS_GET_IP(vp);
+ /* make sync come back and flush this inode */
+ if (vp) {
+ struct inode *inode = vn_to_inode(vp);
- if (!(inode->i_state & I_NEW))
- mark_inode_dirty_sync(inode);
+ if (!(inode->i_state &
+ (I_NEW|I_FREEING|I_CLEAR)))
+ mark_inode_dirty_sync(inode);
+ }
}
-
wake_up(&ip->i_ipin_wait);
}
}
ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork));
memcpy(cp, ifp->if_u1.if_data, ifp->if_bytes);
}
- if (whichfork == XFS_DATA_FORK) {
- if (unlikely(XFS_DIR_SHORTFORM_VALIDATE_ONDISK(mp, dip))) {
- XFS_ERROR_REPORT("xfs_iflush_fork",
- XFS_ERRLEVEL_LOW, mp);
- return XFS_ERROR(EFSCORRUPTED);
- }
- }
break;
case XFS_DINODE_FMT_EXTENTS:
XFS_STATS_INC(xs_iflush_count);
ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE|MR_ACCESS));
- ASSERT(valusema(&ip->i_flock) <= 0);
+ ASSERT(issemalocked(&(ip->i_flock)));
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
ip->i_d.di_nextents > ip->i_df.if_ext_max);
/*
* Get the buffer containing the on-disk inode.
*/
- error = xfs_itobp(mp, NULL, ip, &dip, &bp, 0);
- if (error != 0) {
+ error = xfs_itobp(mp, NULL, ip, &dip, &bp, 0, 0);
+ if (error) {
xfs_ifunlock(ip);
return error;
}
corrupt_out:
xfs_buf_relse(bp);
- xfs_force_shutdown(mp, XFS_CORRUPT_INCORE);
+ xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
xfs_iflush_abort(ip);
/*
* Unlocks the flush lock
xfs_buf_relse(bp);
}
- xfs_force_shutdown(mp, XFS_CORRUPT_INCORE);
+ xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
if(!bufwasdelwri) {
/*
SPLDECL(s);
ASSERT(ismrlocked(&ip->i_lock, MR_UPDATE|MR_ACCESS));
- ASSERT(valusema(&ip->i_flock) <= 0);
+ ASSERT(issemalocked(&(ip->i_flock)));
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
ip->i_d.di_nextents > ip->i_df.if_ext_max);
xfs_mount_t *mp)
{
xfs_inode_t *ip;
- vnode_t *vp;
+ bhv_vnode_t *vp;
again:
XFS_MOUNT_ILOCK(mp);
{
int error;
mode_t orgmode = mode;
- struct inode *inode = LINVFS_GET_IP(XFS_ITOV(ip));
+ struct inode *inode = vn_to_inode(XFS_ITOV(ip));
if (mode & S_IWUSR) {
umode_t imode = inode->i_mode;
xfs_iext_add(
xfs_ifork_t *ifp, /* inode fork pointer */
xfs_extnum_t idx, /* index to begin adding exts */
- int ext_diff) /* nubmer of extents to add */
+ int ext_diff) /* number of extents to add */
{
int byte_diff; /* new bytes being added */
int new_size; /* size of extents after adding */
xfs_extnum_t ext_diff; /* extents to remove in current list */
xfs_extnum_t nex1; /* number of extents before idx */
xfs_extnum_t nex2; /* extents after idx + count */
- int nlists; /* entries in indirecton array */
+ int nlists; /* entries in indirection array */
int page_idx = idx; /* index in target extent list */
ASSERT(ifp->if_flags & XFS_IFEXTIREC);
*/
memcpy(ifp->if_u2.if_inline_ext, ifp->if_u1.if_extents,
nextents * sizeof(xfs_bmbt_rec_t));
- kmem_free(ifp->if_u1.if_extents, KM_SLEEP);
+ kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes);
ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext;
ifp->if_real_bytes = 0;
}
ifp->if_bytes = 0;
}
+/*
+ * Return a pointer to the extent record for file system block bno.
+ */
+xfs_bmbt_rec_t * /* pointer to found extent record */
+xfs_iext_bno_to_ext(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_fileoff_t bno, /* block number to search for */
+ xfs_extnum_t *idxp) /* index of target extent */
+{
+ xfs_bmbt_rec_t *base; /* pointer to first extent */
+ xfs_filblks_t blockcount = 0; /* number of blocks in extent */
+ xfs_bmbt_rec_t *ep = NULL; /* pointer to target extent */
+ xfs_ext_irec_t *erp = NULL; /* indirection array pointer */
+ int high; /* upper boundary in search */
+ xfs_extnum_t idx = 0; /* index of target extent */
+ int low; /* lower boundary in search */
+ xfs_extnum_t nextents; /* number of file extents */
+ xfs_fileoff_t startoff = 0; /* start offset of extent */
+
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ if (nextents == 0) {
+ *idxp = 0;
+ return NULL;
+ }
+ low = 0;
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ /* Find target extent list */
+ int erp_idx = 0;
+ erp = xfs_iext_bno_to_irec(ifp, bno, &erp_idx);
+ base = erp->er_extbuf;
+ high = erp->er_extcount - 1;
+ } else {
+ base = ifp->if_u1.if_extents;
+ high = nextents - 1;
+ }
+ /* Binary search extent records */
+ while (low <= high) {
+ idx = (low + high) >> 1;
+ ep = base + idx;
+ startoff = xfs_bmbt_get_startoff(ep);
+ blockcount = xfs_bmbt_get_blockcount(ep);
+ if (bno < startoff) {
+ high = idx - 1;
+ } else if (bno >= startoff + blockcount) {
+ low = idx + 1;
+ } else {
+ /* Convert back to file-based extent index */
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ idx += erp->er_extoff;
+ }
+ *idxp = idx;
+ return ep;
+ }
+ }
+ /* Convert back to file-based extent index */
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ idx += erp->er_extoff;
+ }
+ if (bno >= startoff + blockcount) {
+ if (++idx == nextents) {
+ ep = NULL;
+ } else {
+ ep = xfs_iext_get_ext(ifp, idx);
+ }
+ }
+ *idxp = idx;
+ return ep;
+}
+
/*
* Return a pointer to the indirection array entry containing the
* extent record for filesystem block bno. Store the index of the
* target irec in *erp_idxp.
*/
-xfs_ext_irec_t *
+xfs_ext_irec_t * /* pointer to found extent record */
xfs_iext_bno_to_irec(
xfs_ifork_t *ifp, /* inode fork pointer */
xfs_fileoff_t bno, /* block number to search for */
{
xfs_ext_irec_t *erp = NULL; /* indirection array pointer */
xfs_ext_irec_t *erp_next; /* next indirection array entry */
- xfs_extnum_t erp_idx; /* indirection array index */
+ int erp_idx; /* indirection array index */
int nlists; /* number of extent irec's (lists) */
int high; /* binary search upper limit */
int low; /* binary search lower limit */