7 files changed, 65 insertions, 45 deletions

diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog
index c7e9237379c2..83f3322765cd 100644
--- a/fs/ntfs/ChangeLog
+++ b/fs/ntfs/ChangeLog
@@ -29,7 +29,8 @@ ToDo/Notes:
           The Windows boot will run chkdsk and then reboot.  The user can then
           immediately boot into Linux rather than having to do a full Windows
           boot first before rebooting into Linux and we will recognize such a
-          journal and empty it as it is clean by definition.
+          journal and empty it as it is clean by definition.  Note, this only
+          works if chkdsk left the journal in an obviously clean state.
         - Support journals ($LogFile) with only one restart page as well as
           journals with two different restart pages.  We sanity check both and
           either use the only sane one or the more recent one of the two in the
@@ -94,6 +95,13 @@ ToDo/Notes:
           my ways.
         - Fix various bugs in the runlist merging code.  (Based on libntfs
           changes by Richard Russon.)
+        - Fix sparse warnings that have crept in over time.
+        - Change ntfs_cluster_free() to require a write locked runlist on entry
+          since we otherwise get into a lock reversal deadlock if a read locked
+          runlist is passed in. In the process also change it to take an ntfs
+          inode instead of a vfs inode as parameter.
+        - Fix the definition of the CHKD ntfs record magic.  It had an off by
+          two error causing it to be CHKB instead of CHKD.
 
 2.1.23 - Implement extension of resident files and make writing safe as well as
          many bug fixes, cleanups, and enhancements...
diff --git a/fs/ntfs/layout.h b/fs/ntfs/layout.h
index 609ad1728ce4..01f2dfa39cec 100644
--- a/fs/ntfs/layout.h
+++ b/fs/ntfs/layout.h
@@ -123,7 +123,7 @@ enum {
         magic_RCRD = const_cpu_to_le32(0x44524352), /* Log record page. */
 
         /* Found in $LogFile/$DATA.  (May be found in $MFT/$DATA, also?) */
-        magic_CHKD = const_cpu_to_le32(0x424b4843), /* Modified by chkdsk. */
+        magic_CHKD = const_cpu_to_le32(0x444b4843), /* Modified by chkdsk. */
 
         /* Found in all ntfs record containing records. */
         magic_BAAD = const_cpu_to_le32(0x44414142), /* Failed multi sector
@@ -308,10 +308,8 @@ typedef le16 MFT_RECORD_FLAGS;
  * The _LE versions are to be applied on little endian MFT_REFs.
  * Note: The _LE versions will return a CPU endian formatted value!
  */
-typedef enum {
-        MFT_REF_MASK_CPU        = 0x0000ffffffffffffULL,
-        MFT_REF_MASK_LE         = const_cpu_to_le64(0x0000ffffffffffffULL),
-} MFT_REF_CONSTS;
+#define MFT_REF_MASK_CPU 0x0000ffffffffffffULL
+#define MFT_REF_MASK_LE const_cpu_to_le64(0x0000ffffffffffffULL)
 
 typedef u64 MFT_REF;
 typedef le64 leMFT_REF;
diff --git a/fs/ntfs/lcnalloc.c b/fs/ntfs/lcnalloc.c
index 7b5934290685..5af3bf0b7eee 100644
--- a/fs/ntfs/lcnalloc.c
+++ b/fs/ntfs/lcnalloc.c
@@ -779,14 +779,13 @@ out:
 
 /**
  * __ntfs_cluster_free - free clusters on an ntfs volume
- * @vi:         vfs inode whose runlist describes the clusters to free
- * @start_vcn:  vcn in the runlist of @vi at which to start freeing clusters
+ * @ni:         ntfs inode whose runlist describes the clusters to free
+ * @start_vcn:  vcn in the runlist of @ni at which to start freeing clusters
  * @count:      number of clusters to free or -1 for all clusters
- * @write_locked:       true if the runlist is locked for writing
  * @is_rollback:        true if this is a rollback operation
  *
  * Free @count clusters starting at the cluster @start_vcn in the runlist
- * described by the vfs inode @vi.
+ * described by the vfs inode @ni.
  *
  * If @count is -1, all clusters from @start_vcn to the end of the runlist are
  * deallocated.  Thus, to completely free all clusters in a runlist, use
@@ -801,31 +800,28 @@ out:
  * Return the number of deallocated clusters (not counting sparse ones) on
  * success and -errno on error.
  *
- * Locking: - The runlist described by @vi must be locked on entry and is
- *            locked on return.  Note if the runlist is locked for reading the
- *            lock may be dropped and reacquired.  Note the runlist may be
- *            modified when needed runlist fragments need to be mapped.
+ * Locking: - The runlist described by @ni must be locked for writing on entry
+ *            and is locked on return.  Note the runlist may be modified when
+ *            needed runlist fragments need to be mapped.
  *          - The volume lcn bitmap must be unlocked on entry and is unlocked
  *            on return.
  *          - This function takes the volume lcn bitmap lock for writing and
  *            modifies the bitmap contents.
  */
-s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
-                const BOOL write_locked, const BOOL is_rollback)
+s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn, s64 count,
+                const BOOL is_rollback)
 {
         s64 delta, to_free, total_freed, real_freed;
-        ntfs_inode *ni;
         ntfs_volume *vol;
         struct inode *lcnbmp_vi;
         runlist_element *rl;
         int err;
 
-        BUG_ON(!vi);
+        BUG_ON(!ni);
         ntfs_debug("Entering for i_ino 0x%lx, start_vcn 0x%llx, count "
-                        "0x%llx.%s", vi->i_ino, (unsigned long long)start_vcn,
+                        "0x%llx.%s", ni->mft_no, (unsigned long long)start_vcn,
                         (unsigned long long)count,
                         is_rollback ? " (rollback)" : "");
-        ni = NTFS_I(vi);
         vol = ni->vol;
         lcnbmp_vi = vol->lcnbmp_ino;
         BUG_ON(!lcnbmp_vi);
@@ -843,7 +839,7 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
 
         total_freed = real_freed = 0;
 
-        rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, write_locked);
+        rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, TRUE);
         if (IS_ERR(rl)) {
                 if (!is_rollback)
                         ntfs_error(vol->sb, "Failed to find first runlist "
@@ -897,7 +893,7 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
 
                         /* Attempt to map runlist. */
                         vcn = rl->vcn;
-                        rl = ntfs_attr_find_vcn_nolock(ni, vcn, write_locked);
+                        rl = ntfs_attr_find_vcn_nolock(ni, vcn, TRUE);
                         if (IS_ERR(rl)) {
                                 err = PTR_ERR(rl);
                                 if (!is_rollback)
@@ -965,8 +961,7 @@ err_out:
          * If rollback fails, set the volume errors flag, emit an error
          * message, and return the error code.
          */
-        delta = __ntfs_cluster_free(vi, start_vcn, total_freed, write_locked,
-                        TRUE);
+        delta = __ntfs_cluster_free(ni, start_vcn, total_freed, TRUE);
         if (delta < 0) {
                 ntfs_error(vol->sb, "Failed to rollback (error %i).  Leaving "
                                 "inconsistent metadata!  Unmount and run "
diff --git a/fs/ntfs/lcnalloc.h b/fs/ntfs/lcnalloc.h
index e4d7fb98d685..a6a8827882e7 100644
--- a/fs/ntfs/lcnalloc.h
+++ b/fs/ntfs/lcnalloc.h
@@ -2,7 +2,7 @@
  * lcnalloc.h - Exports for NTFS kernel cluster (de)allocation.  Part of the
  *              Linux-NTFS project.
  *
- * Copyright (c) 2004 Anton Altaparmakov
+ * Copyright (c) 2004-2005 Anton Altaparmakov
  *
  * This program/include file is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License as published
@@ -28,6 +28,7 @@
 #include <linux/fs.h>
 
 #include "types.h"
+#include "inode.h"
 #include "runlist.h"
 #include "volume.h"
 
@@ -42,18 +43,17 @@ extern runlist_element *ntfs_cluster_alloc(ntfs_volume *vol,
                 const VCN start_vcn, const s64 count, const LCN start_lcn,
                 const NTFS_CLUSTER_ALLOCATION_ZONES zone);
 
-extern s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn,
-                s64 count, const BOOL write_locked, const BOOL is_rollback);
+extern s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn,
+                s64 count, const BOOL is_rollback);
 
 /**
  * ntfs_cluster_free - free clusters on an ntfs volume
- * @vi:         vfs inode whose runlist describes the clusters to free
- * @start_vcn:  vcn in the runlist of @vi at which to start freeing clusters
+ * @ni:         ntfs inode whose runlist describes the clusters to free
+ * @start_vcn:  vcn in the runlist of @ni at which to start freeing clusters
  * @count:      number of clusters to free or -1 for all clusters
- * @write_locked:       true if the runlist is locked for writing
  *
  * Free @count clusters starting at the cluster @start_vcn in the runlist
- * described by the vfs inode @vi.
+ * described by the ntfs inode @ni.
  *
  * If @count is -1, all clusters from @start_vcn to the end of the runlist are
  * deallocated.  Thus, to completely free all clusters in a runlist, use
@@ -65,19 +65,18 @@ extern s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn,
  * Return the number of deallocated clusters (not counting sparse ones) on
  * success and -errno on error.
  *
- * Locking: - The runlist described by @vi must be locked on entry and is
- *            locked on return.  Note if the runlist is locked for reading the
- *            lock may be dropped and reacquired.  Note the runlist may be
- *            modified when needed runlist fragments need to be mapped.
+ * Locking: - The runlist described by @ni must be locked for writing on entry
+ *            and is locked on return.  Note the runlist may be modified when
+ *            needed runlist fragments need to be mapped.
  *          - The volume lcn bitmap must be unlocked on entry and is unlocked
  *            on return.
  *          - This function takes the volume lcn bitmap lock for writing and
  *            modifies the bitmap contents.
  */
-static inline s64 ntfs_cluster_free(struct inode *vi, const VCN start_vcn,
-                s64 count, const BOOL write_locked)
+static inline s64 ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn,
+                s64 count)
 {
-        return __ntfs_cluster_free(vi, start_vcn, count, write_locked, FALSE);
+        return __ntfs_cluster_free(ni, start_vcn, count, FALSE);
 }
 
 extern int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
diff --git a/fs/ntfs/logfile.c b/fs/ntfs/logfile.c
index 0173e95500d9..0fd70295cca6 100644
--- a/fs/ntfs/logfile.c
+++ b/fs/ntfs/logfile.c
@@ -51,7 +51,8 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi,
                 RESTART_PAGE_HEADER *rp, s64 pos)
 {
         u32 logfile_system_page_size, logfile_log_page_size;
-        u16 usa_count, usa_ofs, usa_end, ra_ofs;
+        u16 ra_ofs, usa_count, usa_ofs, usa_end = 0;
+        BOOL have_usa = TRUE;
 
         ntfs_debug("Entering.");
         /*
@@ -86,6 +87,14 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi,
                                 (int)sle16_to_cpu(rp->minor_ver));
                 return FALSE;
         }
+        /*
+         * If chkdsk has been run the restart page may not be protected by an
+         * update sequence array.
+         */
+        if (ntfs_is_chkd_record(rp->magic) && !le16_to_cpu(rp->usa_count)) {
+                have_usa = FALSE;
+                goto skip_usa_checks;
+        }
         /* Verify the size of the update sequence array. */
         usa_count = 1 + (logfile_system_page_size >> NTFS_BLOCK_SIZE_BITS);
         if (usa_count != le16_to_cpu(rp->usa_count)) {
@@ -102,6 +111,7 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi,
                                 "inconsistent update sequence array offset.");
                 return FALSE;
         }
+skip_usa_checks:
         /*
          * Verify the position of the restart area.  It must be:
          *      - aligned to 8-byte boundary,
@@ -109,7 +119,8 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi,
          *      - within the system page size.
          */
         ra_ofs = le16_to_cpu(rp->restart_area_offset);
-        if (ra_ofs & 7 || ra_ofs < usa_end ||
+        if (ra_ofs & 7 || (have_usa ? ra_ofs < usa_end :
+                        ra_ofs < sizeof(RESTART_PAGE_HEADER)) ||
                         ra_ofs > logfile_system_page_size) {
                 ntfs_error(vi->i_sb, "$LogFile restart page specifies "
                                 "inconsistent restart area offset.");
@@ -402,8 +413,12 @@ static int ntfs_check_and_load_restart_page(struct inode *vi,
                         idx++;
                 } while (to_read > 0);
         }
-        /* Perform the multi sector transfer deprotection on the buffer. */
-        if (post_read_mst_fixup((NTFS_RECORD*)trp,
+        /*
+         * Perform the multi sector transfer deprotection on the buffer if the
+         * restart page is protected.
+         */
+        if ((!ntfs_is_chkd_record(trp->magic) || le16_to_cpu(trp->usa_count))
+                        && post_read_mst_fixup((NTFS_RECORD*)trp,
                         le32_to_cpu(rp->system_page_size))) {
                 /*
                  * A multi sector tranfer error was detected.  We only need to
@@ -615,11 +630,16 @@ is_empty:
                  * Otherwise just throw it away.
                  */
                 if (rstr2_lsn > rstr1_lsn) {
+                        ntfs_debug("Using second restart page as it is more "
+                                        "recent.");
                         ntfs_free(rstr1_ph);
                         rstr1_ph = rstr2_ph;
                         /* rstr1_lsn = rstr2_lsn; */
-                } else
+                } else {
+                        ntfs_debug("Using first restart page as it is more "
+                                        "recent.");
                         ntfs_free(rstr2_ph);
+                }
                 rstr2_ph = NULL;
         }
         /* All consistency checks passed. */
diff --git a/fs/ntfs/logfile.h b/fs/ntfs/logfile.h
index 42388f95ea6d..a51f3dd0e9eb 100644
--- a/fs/ntfs/logfile.h
+++ b/fs/ntfs/logfile.h
@@ -113,7 +113,7 @@ typedef struct {
  */
 enum {
         RESTART_VOLUME_IS_CLEAN = const_cpu_to_le16(0x0002),
-        RESTART_SPACE_FILLER    = 0xffff, /* gcc: Force enum bit width to 16. */
+        RESTART_SPACE_FILLER    = const_cpu_to_le16(0xffff), /* gcc: Force enum bit width to 16. */
 } __attribute__ ((__packed__));
 
 typedef le16 RESTART_AREA_FLAGS;
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index 2c32b84385a8..247586d1d5dc 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -1953,7 +1953,7 @@ restore_undo_alloc:
         a = ctx->attr;
         a->data.non_resident.highest_vcn = cpu_to_sle64(old_last_vcn - 1);
 undo_alloc:
-        if (ntfs_cluster_free(vol->mft_ino, old_last_vcn, -1, TRUE) < 0) {
+        if (ntfs_cluster_free(mft_ni, old_last_vcn, -1) < 0) {
                 ntfs_error(vol->sb, "Failed to free clusters from mft data "
                                 "attribute.%s", es);
                 NVolSetErrors(vol);