3  * NTFS kernel mft record operations.
21  * ntfs_mft_record_check - Check the consistency of an MFT record
35 		ntfs_error(sb, "Record %llu has no FILE magic (0x%x)\n",  in ntfs_mft_record_check()
43 		ntfs_error(sb, "Record %llu has corrupt fix-up values fields\n",  in ntfs_mft_record_check()
49 		ntfs_error(sb, "Record %llu has corrupt allocation size (%u <> %u)\n",  in ntfs_mft_record_check()
56 		ntfs_error(sb, "Record %llu has corrupt in-use size (%u > %u)\n",  in ntfs_mft_record_check()
63 		ntfs_error(sb, "Attributes badly aligned in record %llu\n",  in ntfs_mft_record_check()
70 		ntfs_error(sb, "Record %llu is corrupt\n", mft_no);  in ntfs_mft_record_check()
81  * map_mft_record_folio - map the folio in which a specific mft record resides
82  * @ni:		ntfs inode whose mft record page to map
84  * This maps the folio in which the mft record of the ntfs inode @ni is
87  * This allocates a new buffer (@ni->mrec), copies the MFT record data from
94  * Return: A pointer to the allocated and fixed-up mft record (@ni->mrec).
110 	 * page of the wanted mft record.  in map_mft_record_folio()
119 	/* If the wanted index is out of bounds the mft record doesn't exist. */  in map_mft_record_folio()
125 …"Attempt to read mft record 0x%llx, which is beyond the end of the mft. This is probably a bug in …  in map_mft_record_folio()
168  * map_mft_record - map and pin an mft record
169  * @ni:		ntfs inode whose MFT record to map
171  * This function ensures the MFT record for the given inode is mapped and
174  * It increments the reference count of the ntfs inode. If the record is
175  * already mapped (@ni->folio is set), it returns the cached record
180  * record data.
182  * Return: A pointer to the mft record. You need to check the returned
210  * unmap_mft_record - release a reference to a mapped mft record
211  * @ni:		ntfs inode whose MFT record to unmap
220  * NOTE: If caller has modified the mft record, it is imperative to set the mft
221  * record dirty BEFORE calling unmap_mft_record().
244  * Load the extent mft record @mref and attach it to its base inode @base_ni.
245  * Return the mapped extent mft record if IS_ERR(result) is false.  Otherwise
262 	ntfs_debug("Mapping extent mft record 0x%llx (base mft record 0x%llx).",  in map_extent_mft_record()
287 		/* We found the record; just have to map and return it. */  in map_extent_mft_record()
305 				"Failed to map extent mft record, error code %ld.",  in map_extent_mft_record()
311 	/* Record wasn't there. Get a new ntfs inode and initialize it. */  in map_extent_mft_record()
321 	/* Now map the record. */  in map_extent_mft_record()
387  * @ni:		ntfs inode describing the mapped mft record
421  * ntfs_bio_end_io - bio completion callback for MFT record writes
436  * ntfs_sync_mft_mirror - synchronize an mft record to the mft mirror
437  * @vol:	ntfs volume on which the mft record to synchronize resides
438  * @mft_no:	mft record number of mft record to synchronize
439  * @m:		mapped, mst protected (extent) mft record to synchronize
441  * Write the mapped, mst protected (extent) mft record @m with mft record
465 	/* Get the page containing the mirror copy of the mft record @m. */  in ntfs_sync_mft_mirror()
476 	/* Offset of the mft mirror record inside the page. */  in ntfs_sync_mft_mirror()
478 	/* The address in the page of the mirror copy of the mft record @m. */  in ntfs_sync_mft_mirror()
480 	/* Copy the mst protected mft record to the mirror. */  in ntfs_sync_mft_mirror()
511 		ntfs_error(vol->sb, "I/O error while writing mft mirror record 0x%llx!", mft_no);  in ntfs_sync_mft_mirror()
522  * write_mft_record_nolock - write out a mapped (extent) mft record
523  * @ni:		ntfs inode describing the mapped (extent) mft record
524  * @m:		mapped (extent) mft record to write
527  * Write the mapped (extent) mft record @m described by the (regular or extent)
528  * ntfs inode @ni to backing store.  If the mft record @m has a counterpart in
531  * We only write the mft record if the ntfs inode @ni is dirty.
533  * On success, clean the mft record and return 0.
534  * On error (specifically ENOMEM), we redirty the record so it can be retried.
556 	 * for the mft record @m and the page it is in.  in write_mft_record_nolock()
609 …"I/O error while writing mft record 0x%llx!  Marking base inode as bad.  You should unmount the vo…  in write_mft_record_nolock()
627 			"Not enough memory to write mft record. Redirtying so the write is retried later.");  in write_mft_record_nolock()
669  * ntfs_may_write_mft_record - check if an mft record may be written out
670  * @vol:	[IN]  ntfs volume on which the mft record to check resides
671  * @mft_no:	[IN]  mft record number of the mft record to check
672  * @m:		[IN]  mapped mft record to check
676  * Check if the mapped (base or extent) mft record @m with mft record number
678  * and possible the ntfs inode of the mft record is locked and the base vfs
688  * Return 'true' if the mft record may be written out and 'false' if not.
696  * If the inode is found in icache we know the mft record must be a base mft
697  * record.  If it is dirty, we do not write it and return 'false' as the vfs
699  * cause the base mft record to be redirtied and written out again.
701  * If the inode is in icache and not dirty, we attempt to lock the mft record
703  * record and we return 'false'.
705  * If we manage to obtain the lock we have exclusive access to the mft record,
706  * which also allows us safe writeout of the mft record.  We then set
709  * Note we cannot just lock the mft record and sleep while waiting for the lock
714  * If the mft record is not a FILE record or it is a base mft record, we can
717  * We now know the mft record is an extent mft record.  We check if the inode
718  * corresponding to its base mft record is in icache. If it is not, we cannot
721  * We now have the base inode for the extent mft record.  We check if it has an
722  * ntfs inode for the extent mft record attached. If not, it is safe to write
723  * the extent mft record and we return 'true'.
728  * If it is not dirty, we attempt to lock the extent mft record. If the lock
732  * record. We set @locked_ni to the now locked ntfs inode and return 'true'.
753 	 * Check if the inode corresponding to this mft record is in the VFS  in ntfs_may_write_mft_record()
783 		/* If the inode is dirty, do not write this record. */  in ntfs_may_write_mft_record()
792 		/* The inode is not dirty, try to take the mft record lock. */  in ntfs_may_write_mft_record()
794 			ntfs_debug("Mft record 0x%llx is already locked, do not write it.", mft_no);  in ntfs_may_write_mft_record()
799 		ntfs_debug("Managed to lock mft record 0x%llx, write it.",  in ntfs_may_write_mft_record()
802 		 * The write has to occur while we hold the mft record lock so  in ntfs_may_write_mft_record()
810 	/* Write the record if it is not a mft record (type "FILE"). */  in ntfs_may_write_mft_record()
812 		ntfs_debug("Mft record 0x%llx is not a FILE record, write it.",  in ntfs_may_write_mft_record()
816 	/* Write the mft record if it is a base inode. */  in ntfs_may_write_mft_record()
818 		ntfs_debug("Mft record 0x%llx is a base record, write it.",  in ntfs_may_write_mft_record()
823 	 * This is an extent mft record.  Check if the inode corresponding to  in ntfs_may_write_mft_record()
824 	 * its base mft record is in icache and obtain a reference to it if it  in ntfs_may_write_mft_record()
829 	ntfs_debug("Mft record 0x%llx is an extent record.  Looking for base inode 0x%llx in icache.",  in ntfs_may_write_mft_record()
846 	 * corresponding to this extent mft record attached.  in ntfs_may_write_mft_record()
853 		 * extent mft record.  in ntfs_may_write_mft_record()
857 		ntfs_debug("Base inode 0x%llx has no attached extent inodes, write the extent record.",  in ntfs_may_write_mft_record()
867 			 * mft record.  in ntfs_may_write_mft_record()
875 	 * extent mft record.  in ntfs_may_write_mft_record()
880 …ntfs_debug("Extent inode 0x%llx is not attached to its base inode 0x%llx, write the extent record.…  in ntfs_may_write_mft_record()
898 	 * Found the extent inode coresponding to this extent mft record.  in ntfs_may_write_mft_record()
899 	 * Try to take the mft record lock.  in ntfs_may_write_mft_record()
904 		ntfs_debug("Extent mft record 0x%llx is already locked, do not write it.",  in ntfs_may_write_mft_record()
908 	ntfs_debug("Managed to lock extent mft record 0x%llx, write it.",  in ntfs_may_write_mft_record()
911 	 * The write has to occur while we hold the mft record lock so return  in ntfs_may_write_mft_record()
924  * @vol:	volume on which to search for a free mft record
925  * @base_ni:	open base inode if allocating an extent mft record or NULL
927  * Search for a free mft record in the mft bitmap attribute on the ntfs volume
932  * If @base_ni is not NULL start the search at the mft record after the base
933  * mft record @base_ni.
935  * Return the free mft record on success and -errno on error.  An error code of
952 	ntfs_debug("Searching for free mft record in the currently initialized mft bitmap.");  in ntfs_mft_bitmap_find_and_alloc_free_rec_nolock()
990 	/* Loop until a free mft record is found. */  in ntfs_mft_bitmap_find_and_alloc_free_rec_nolock()
1021 				 * mft record (base record) then give up searching since  in ntfs_mft_bitmap_find_and_alloc_free_rec_nolock()
1022 				 * no guarantee that the found record will be accessible.  in ntfs_mft_bitmap_find_and_alloc_free_rec_nolock()
1048 					ntfs_debug("Done.  (Found and allocated mft record 0x%llx.)",  in ntfs_mft_bitmap_find_and_alloc_free_rec_nolock()
1245 	 * Update the attribute record as well.  Note: @rl is the last  in ntfs_mft_bitmap_extend_allocation_nolock()
1250 		ntfs_error(vol->sb, "Failed to map mft record.");  in ntfs_mft_bitmap_extend_allocation_nolock()
1289 	/* Expand the attribute record if necessary. */  in ntfs_mft_bitmap_extend_allocation_nolock()
1302 	/* Generate the mapping pairs array directly into the attr record. */  in ntfs_mft_bitmap_extend_allocation_nolock()
1315 	 * Reflect this in the struct ntfs_inode structure and the attribute record.  in ntfs_mft_bitmap_extend_allocation_nolock()
1401 			ntfs_error(vol->sb, "Failed to restore attribute record.%s", es);  in ntfs_mft_bitmap_extend_allocation_nolock()
1446 	/* Get the attribute record. */  in ntfs_mft_bitmap_extend_initialized_nolock()
1449 		ntfs_error(vol->sb, "Failed to map mft record.");  in ntfs_mft_bitmap_extend_initialized_nolock()
1500 		ntfs_error(vol->sb, "Failed to map mft record.%s", es);  in ntfs_mft_bitmap_extend_initialized_nolock()
1550  * worth of clusters or if not enough space for this by one mft record worth
1605 	/* Minimum allocation is one mft record worth of clusters. */  in ntfs_mft_data_extend_allocation_nolock()
1623 …"Cannot allocate mft record because the maximum number of inodes (2^32) has already been reached."…  in ntfs_mft_data_extend_allocation_nolock()
1681 	/* Update the attribute record as well. */  in ntfs_mft_data_extend_allocation_nolock()
1684 		ntfs_error(vol->sb, "Failed to map mft record.");  in ntfs_mft_data_extend_allocation_nolock()
1727 	/* Expand the attribute record if necessary. */  in ntfs_mft_data_extend_allocation_nolock()
1740 	/* Generate the mapping pairs array directly into the attr record. */  in ntfs_mft_data_extend_allocation_nolock()
1752 	 * Reflect this in the struct ntfs_inode structure and the attribute record.  in ntfs_mft_data_extend_allocation_nolock()
1834 				ntfs_error(vol->sb, "Failed to restore attribute record.%s", es);  in ntfs_mft_data_extend_allocation_nolock()
1850  * ntfs_mft_record_layout - layout an mft record into a memory buffer
1851  * @vol:	volume to which the mft record will belong
1852  * @mft_no:	mft reference specifying the mft record number
1855  * Layout an empty, unused mft record with the mft record number @mft_no into
1856  * the buffer @m.  The volume @vol is needed because the mft record structure
1858  * record will be used on.
1867 	ntfs_debug("Entering for mft record 0x%llx.", (long long)mft_no);  in ntfs_mft_record_layout()
1869 		ntfs_error(vol->sb, "Mft record number 0x%llx exceeds maximum of 2^32.",  in ntfs_mft_record_layout()
1873 	/* Start by clearing the whole mft record to gives us a clean slate. */  in ntfs_mft_record_layout()
1894 …"Sector size is bigger than mft record size.  Setting usa_count to 1.  If chkdsk reports this as c…  in ntfs_mft_record_layout()
1926  * ntfs_mft_record_format - format an mft record on an ntfs volume
1927  * @vol:	volume on which to format the mft record
1928  * @mft_no:	mft record number to format
1930  * Format the mft record @mft_no in $MFT/$DATA, i.e. lay out an empty, unused
1931  * mft record into the appropriate place of the mft data attribute.  This is
1946 	ntfs_debug("Entering for mft record 0x%llx.", (long long)mft_no);  in ntfs_mft_record_format()
1949 	 * page of the wanted mft record.  in ntfs_mft_record_format()
1959 			ntfs_error(vol->sb, "Tried to format non-existing mft record 0x%llx.",  in ntfs_mft_record_format()
1965 	/* Read, map, and pin the folio containing the mft record. */  in ntfs_mft_record_format()
1968 		ntfs_error(vol->sb, "Failed to map page containing mft record to format 0x%llx.",  in ntfs_mft_record_format()
1977 		ntfs_error(vol->sb, "Failed to layout mft record 0x%llx.",  in ntfs_mft_record_format()
1988 	 * Make sure the mft record is written out to disk.  We could use  in ntfs_mft_record_format()
1990 	 * unnecessary as ntfs_writepage() will write the dirty record anyway.  in ntfs_mft_record_format()
2001  * ntfs_mft_record_alloc - allocate an mft record on an ntfs volume
2002  * @vol:	[IN]  volume on which to allocate the mft record
2005  * @base_ni:	[IN]  open base inode if allocating an extent mft record or NULL
2006  * @ni_mrec:	[OUT] on successful return this is the mapped mft record
2008  * Allocate an mft record in $MFT/$DATA of an open ntfs volume @vol.
2010  * If @base_ni is NULL make the mft record a base mft record, i.e. a file or
2016  * If @base_ni is not NULL make the allocated mft record an extent record,
2017  * allocate it starting at the mft record after the base mft record and attach
2023  * allocated mft record.  *@mrec is then set to the allocated, mapped, pinned,
2024  * and locked mft record.  If IS_ERR() is true, the function failed and the
2030  * To find a free mft record, we scan the mft bitmap for a zero bit.  To
2045  * record.  If there are no free records left in the range 64 to number of
2049  * is not sufficient space to do this, we try to extend by a single mft record
2050  * or one cluster, if cluster size is above the mft record size.
2053  * allocated mft record, incrementing mft data size and initialized size
2056  * mft records until we reach record 64 which we consider as the first free mft
2057  * record for use by normal files.
2068  *	- set the bit corresponding to the mft record being allocated in the
2070  *	- opened an struct ntfs_inode for the allocated mft record, and we will have
2072  *	  locked mft record.
2074  * On error, the volume will be left in a consistent state and no record will
2081  * the base mft record and an attribute list is used, it is very important that
2117 		ntfs_debug("Entering (allocating an extent mft record for base mft record 0x%llx).",  in ntfs_mft_record_alloc()
2120 		ntfs_debug("Entering (allocating a base mft record).");  in ntfs_mft_record_alloc()
2133 		ntfs_debug("Found and allocated free record (#1), bit 0x%llx.",  in ntfs_mft_record_alloc()
2154 	 * so we can simply allocate the first unused mft record.  in ntfs_mft_record_alloc()
2172 		ntfs_debug("Found free record (#2), bit 0x%llx.",  in ntfs_mft_record_alloc()
2178 	 * mft record that we can allocate.  in ntfs_mft_record_alloc()
2179 	 * Note: The smallest mft record we allocate is mft record 24.  in ntfs_mft_record_alloc()
2230 	ntfs_debug("Found free record (#3), bit 0x%llx.", (long long)bit);  in ntfs_mft_record_alloc()
2232 	/* @bit is the found free mft record, allocate it in the mft bitmap. */  in ntfs_mft_record_alloc()
2247 	 * will impact decisions for mft bitmap and mft record allocation done  in ntfs_mft_record_alloc()
2249 	 * parallel allocation could allocate the same mft record as this one.  in ntfs_mft_record_alloc()
2256 		ntfs_debug("Allocated mft record already initialized.");  in ntfs_mft_record_alloc()
2259 	ntfs_debug("Initializing allocated mft record.");  in ntfs_mft_record_alloc()
2261 	 * The mft record is outside the initialized data.  Extend the mft data  in ntfs_mft_record_alloc()
2262 	 * attribute until it covers the allocated record.  The loop is only  in ntfs_mft_record_alloc()
2293 	 * the allocated mft record, formatting the mft records allong the way.  in ntfs_mft_record_alloc()
2296 	 * record itself in one fell swoop later on.  in ntfs_mft_record_alloc()
2310 		ntfs_debug("Initializing mft record 0x%llx.",  in ntfs_mft_record_alloc()
2314 			ntfs_error(vol->sb, "Failed to format mft record.");  in ntfs_mft_record_alloc()
2322 	/* Update the mft data attribute record to reflect the new sizes. */  in ntfs_mft_record_alloc()
2325 		ntfs_error(vol->sb, "Failed to map mft record.");  in ntfs_mft_record_alloc()
2356 …ntfs_debug("Status of mft data after mft record initialization: allocated_size 0x%llx, data_size 0…  in ntfs_mft_record_alloc()
2366 	 * allocated mft record still needs to be marked as in use to match the  in ntfs_mft_record_alloc()
2368 	 * this mft record is not referenced from anywhere yet and the fact  in ntfs_mft_record_alloc()
2375 	 * We now have allocated and initialized the mft record.  Calculate the  in ntfs_mft_record_alloc()
2376 	 * index of and the offset within the page cache page the record is in.  in ntfs_mft_record_alloc()
2380 	/* Read, map, and pin the folio containing the mft record. */  in ntfs_mft_record_alloc()
2383 		ntfs_error(vol->sb, "Failed to map page containing allocated mft record 0x%llx.",  in ntfs_mft_record_alloc()
2391 	/* If we just formatted the mft record no need to do it again. */  in ntfs_mft_record_alloc()
2393 		/* Sanity check that the mft record is really not in use. */  in ntfs_mft_record_alloc()
2397 …"Mft record 0x%llx was marked free in mft bitmap but is marked used itself. Unmount and run chkdsk…  in ntfs_mft_record_alloc()
2407 		 * We need to (re-)format the mft record, preserving the  in ntfs_mft_record_alloc()
2411 		 * wrong with the previous mft record.  in ntfs_mft_record_alloc()
2417 			ntfs_error(vol->sb, "Failed to layout allocated mft record 0x%llx.",  in ntfs_mft_record_alloc()
2431 	/* Set the mft record itself in use. */  in ntfs_mft_record_alloc()
2440 		 * Setup the base mft record in the extent mft record.  This  in ntfs_mft_record_alloc()
2441 		 * completes initialization of the allocated extent mft record  in ntfs_mft_record_alloc()
2447 		 * Allocate an extent inode structure for the new mft record,  in ntfs_mft_record_alloc()
2449 		 * its, i.e. the allocated, mft record.  in ntfs_mft_record_alloc()
2455 			ntfs_error(vol->sb, "Failed to map allocated extent mft record 0x%llx.",  in ntfs_mft_record_alloc()
2458 			/* Set the mft record itself not in use. */  in ntfs_mft_record_alloc()
2461 			/* Make sure the mft record is written out to disk. */  in ntfs_mft_record_alloc()
2470 		 * Make sure the allocated mft record is written out to disk.  in ntfs_mft_record_alloc()
2473 		 * record (e.g. at a minimum a new attribute will be added to  in ntfs_mft_record_alloc()
2474 		 * the mft record.  in ntfs_mft_record_alloc()
2486 		 * Manually map, pin, and lock the mft record as we already  in ntfs_mft_record_alloc()
2491 		 * Make sure the allocated mft record is written out to disk.  in ntfs_mft_record_alloc()
2496 		 * that anyway after finishing with the new mft record (e.g. at  in ntfs_mft_record_alloc()
2498 		 * record.  in ntfs_mft_record_alloc()
2525 	 * Return the opened, allocated inode of the allocated mft record as  in ntfs_mft_record_alloc()
2526 	 * well as the mapped, pinned, and locked mft record.  in ntfs_mft_record_alloc()
2558 …"Cannot allocate mft record because the maximum number of inodes (2^32) has already been reached."…  in ntfs_mft_record_alloc()
2568  * ntfs_mft_record_free - free an mft record on an ntfs volume
2569  * @vol:	volume on which to free the mft record
2570  * @ni:		open ntfs inode of the mft record to free
2572  * Free the mft record of the open inode @ni on the mounted ntfs volume @vol.
2600 	/* Mark the mft record as not in use. */  in ntfs_mft_record_free()
2623 	 * record to be freed is guaranteed to do it already.  in ntfs_mft_record_free()
2635 	/* Clear the bit in the $MFT/$BITMAP corresponding to this record. */  in ntfs_mft_record_free()
2749 		/* Get the mft record number. */  in ntfs_write_mft_block()
2753 		/* Check whether to write this mft record. */  in ntfs_write_mft_block()
2775 			 * The record should be written.  If a locked ntfs  in ntfs_write_mft_block()