We've only got 9 GFS2 patches for this merge window:

1. Abhi Das contributed a patch to report journal recovery times more
    accurately during journal replay.
 2. Andreas Gruenbacher contributed a patch to fix fallocate chunk size.
 3. Andreas added a patch to correctly dirty inodes during rename.
 4. Andreas added a patch to improve the comment for function gfs2_block_map.
 5. Andreas added a patch to improve kernel trace point iomap end:
    The physical block address was added.
 6. Andreas added a patch to fix a nasty file system corruption bug that
    surfaced in xfstests 476 in punch-hole/truncate.
 7. Andreas fixed a problem Christoph Helwig pointed out, namely, that GFS2
    was misusing the IOMAP_ZERO flag. The zeroing of new blocks was moved
    to the proper fallocate code.
 8. I contributed a patch to declare function gfs2_remove_from_ail as static.
 9. I added a patch to only set PageChecked for jdata page writes.
 -----BEGIN PGP SIGNATURE-----
 
 iQEcBAABAgAGBQJaw7mxAAoJENeLYdPf93o7kbUH/3MVpJKJe+jdtb5zMGE+byPZ
 T4x+kjZcGwIjWtqIMnga2X1mMIpJ/MCQzZriT+zno2BRRxPS8MQsKp9rcmEGiwoP
 mKZnVGm85xvBDg/JaAnCOeH2S90BlXX3roaB4U4031tAmcZOz/tVbBiTCBXbZ7FP
 F9tnKt/CkRwBBE8CBvstpgYYvbM+kkfiQ8x38FJFx79lOBytIRhPCHpbR3mgko+p
 HrJNTadGJie65nsplKZcQpDQTPrslwP/ynyGh313ReztThHfCB6teL6sNOgJ5u+T
 ThCA95QesH3P45Dd+xl/SuCDd0pZLLslNgOl/q7qiFWaoxFFBKldN3Bnc7ItnDw=
 =iw4e
 -----END PGP SIGNATURE-----

Merge tag 'gfs2-4.17.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2

Pull gfs2 updates from Bob Peterson:
 "We've only got nine GFS2 patches for this merge window:

   - report journal recovery times more accurately during journal replay
     (Abhi Das)

   - fix fallocate chunk size (Andreas Gruenbacher)

   - correctly dirty inodes during rename (Andreas Gruenbacher)

   - improve the comment for function gfs2_block_map (Andreas
     Gruenbacher)

   - improve kernel trace point iomap end: The physical block address
     was added (Andreas Gruenbacher)

   - fix a nasty file system corruption bug that surfaced in xfstests
     476 in punch-hole/truncate (Andreas Gruenbacher)

   - fix a problem Christoph Helwig pointed out, namely, that GFS2 was
     misusing the IOMAP_ZERO flag. The zeroing of new blocks was moved
     to the proper fallocate code (Andreas Gruenbacher)

   - declare function gfs2_remove_from_ail as static (Bob Peterson)

   - only set PageChecked for jdata page writes (Bob Peterson)"

* tag 'gfs2-4.17.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
  gfs2: time journal recovery steps accurately
  gfs2: Zero out fallocated blocks in fallocate_chunk
  gfs2: Check for the end of metadata in punch_hole
  gfs2: gfs2_iomap_end tracepoint: log block address
  gfs2: Improve gfs2_block_map comment
  GFS2: Only set PageChecked for jdata pages
  GFS2: Make function gfs2_remove_from_ail static
  gfs2: Dirty source inode during rename
  gfs2: Fix fallocate chunk size

fs/gfs2/aops.c

@@ -940,13 +940,13 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping,
 }
 
 /**
- * gfs2_set_page_dirty - Page dirtying function
+ * jdata_set_page_dirty - Page dirtying function
  * @page: The page to dirty
  *
  * Returns: 1 if it dirtyed the page, or 0 otherwise
  */
  
-static int gfs2_set_page_dirty(struct page *page)
+static int jdata_set_page_dirty(struct page *page)
 {
 	SetPageChecked(page);
 	return __set_page_dirty_buffers(page);
@@ -1214,7 +1214,7 @@ static const struct address_space_operations gfs2_ordered_aops = {
 	.readpages = gfs2_readpages,
 	.write_begin = gfs2_write_begin,
 	.write_end = gfs2_write_end,
-	.set_page_dirty = gfs2_set_page_dirty,
+	.set_page_dirty = __set_page_dirty_buffers,
 	.bmap = gfs2_bmap,
 	.invalidatepage = gfs2_invalidatepage,
 	.releasepage = gfs2_releasepage,
@@ -1231,7 +1231,7 @@ static const struct address_space_operations gfs2_jdata_aops = {
 	.readpages = gfs2_readpages,
 	.write_begin = gfs2_write_begin,
 	.write_end = gfs2_write_end,
-	.set_page_dirty = gfs2_set_page_dirty,
+	.set_page_dirty = jdata_set_page_dirty,
 	.bmap = gfs2_bmap,
 	.invalidatepage = gfs2_invalidatepage,
 	.releasepage = gfs2_releasepage,

fs/gfs2/bmap.c

@@ -491,14 +491,12 @@ static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap,
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
-	struct super_block *sb = sdp->sd_vfs;
 	struct buffer_head *dibh = mp->mp_bh[0];
 	u64 bn;
 	unsigned n, i, blks, alloced = 0, iblks = 0, branch_start = 0;
 	unsigned dblks = 0;
 	unsigned ptrs_per_blk;
 	const unsigned end_of_metadata = mp->mp_fheight - 1;
-	int ret;
 	enum alloc_state state;
 	__be64 *ptr;
 	__be64 zero_bn = 0;
@@ -607,15 +605,6 @@ static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap,
 			iomap->flags |= IOMAP_F_NEW;
 			while (n-- > 0)
 				*ptr++ = cpu_to_be64(bn++);
-			if (flags & IOMAP_ZERO) {
-				ret = sb_issue_zeroout(sb, iomap->addr >> inode->i_blkbits,
-						       dblks, GFP_NOFS);
-				if (ret) {
-					fs_err(sdp,
-					       "Failed to zero data buffers\n");
-					flags &= ~IOMAP_ZERO;
-				}
-			}
 			break;
 		}
 	} while (iomap->addr == IOMAP_NULL_ADDR);
@@ -812,15 +801,22 @@ int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length,
 }
 
 /**
- * gfs2_block_map - Map a block from an inode to a disk block
+ * gfs2_block_map - Map one or more blocks of an inode to a disk block
  * @inode: The inode
  * @lblock: The logical block number
  * @bh_map: The bh to be mapped
  * @create: True if its ok to alloc blocks to satify the request
  *
- * Sets buffer_mapped() if successful, sets buffer_boundary() if a
+ * The size of the requested mapping is defined in bh_map->b_size.
- * read of metadata will be required before the next block can be
+ *
- * mapped. Sets buffer_new() if new blocks were allocated.
+ * Clears buffer_mapped(bh_map) and leaves bh_map->b_size unchanged
+ * when @lblock is not mapped.  Sets buffer_mapped(bh_map) and
+ * bh_map->b_size to indicate the size of the mapping when @lblock and
+ * successive blocks are mapped, up to the requested size.
+ *
+ * Sets buffer_boundary() if a read of metadata will be required
+ * before the next block can be mapped. Sets buffer_new() if new
+ * blocks were allocated.
  *
  * Returns: errno
  */
@@ -839,8 +835,6 @@ int gfs2_block_map(struct inode *inode, sector_t lblock,
 
 	if (create)
 		flags |= IOMAP_WRITE;
-	if (buffer_zeronew(bh_map))
-		flags |= IOMAP_ZERO;
 	ret = gfs2_iomap_begin(inode, (loff_t)lblock << inode->i_blkbits,
 			       bh_map->b_size, flags, &iomap);
 	if (ret) {
@@ -1344,6 +1338,7 @@ static inline bool walk_done(struct gfs2_sbd *sdp,
 static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+	u64 maxsize = sdp->sd_heightsize[ip->i_height];
 	struct metapath mp = {};
 	struct buffer_head *dibh, *bh;
 	struct gfs2_holder rd_gh;
@@ -1359,6 +1354,14 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length)
 	u64 prev_bnr = 0;
 	__be64 *start, *end;
 
+	if (offset >= maxsize) {
+		/*
+		 * The starting point lies beyond the allocated meta-data;
+		 * there are no blocks do deallocate.
+		 */
+		return 0;
+	}
+
 	/*
 	 * The start position of the hole is defined by lblock, start_list, and
 	 * start_aligned.  The end position of the hole is defined by lend,
@@ -1372,7 +1375,6 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length)
 	 */
 
 	if (length) {
-		u64 maxsize = sdp->sd_heightsize[ip->i_height];
 		u64 end_offset = offset + length;
 		u64 lend;
 

fs/gfs2/dir.c

@@ -1940,7 +1940,6 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
 {
 	struct buffer_head *bh;
 	struct gfs2_dirent *dent;
-	int error;
 
 	dent = gfs2_dirent_search(&dip->i_inode, filename, gfs2_dirent_find, &bh);
 	if (!dent) {
@@ -1953,18 +1952,10 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
 	gfs2_trans_add_meta(dip->i_gl, bh);
 	gfs2_inum_out(nip, dent);
 	dent->de_type = cpu_to_be16(new_type);
-
+	brelse(bh);
-	if (dip->i_diskflags & GFS2_DIF_EXHASH) {
-		brelse(bh);
-		error = gfs2_meta_inode_buffer(dip, &bh);
-		if (error)
-			return error;
-		gfs2_trans_add_meta(dip->i_gl, bh);
-	}
 
 	dip->i_inode.i_mtime = dip->i_inode.i_ctime = current_time(&dip->i_inode);
-	gfs2_dinode_out(dip, bh->b_data);
+	mark_inode_dirty_sync(&dip->i_inode);
-	brelse(bh);
 	return 0;
 }
 

fs/gfs2/file.c

@@ -729,11 +729,12 @@ static ssize_t gfs2_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
 			   int mode)
 {
+	struct super_block *sb = inode->i_sb;
 	struct gfs2_inode *ip = GFS2_I(inode);
+	loff_t end = offset + len;
 	struct buffer_head *dibh;
+	struct iomap iomap;
 	int error;
-	unsigned int nr_blks;
-	sector_t lblock = offset >> inode->i_blkbits;
 
 	error = gfs2_meta_inode_buffer(ip, &dibh);
 	if (unlikely(error))
@@ -747,21 +748,19 @@ static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
 			goto out;
 	}
 
-	while (len) {
+	while (offset < end) {
-		struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 };
+		error = gfs2_iomap_begin(inode, offset, end - offset,
-		bh_map.b_size = len;
+					 IOMAP_WRITE, &iomap);
-		set_buffer_zeronew(&bh_map);
+		if (error)
-
-		error = gfs2_block_map(inode, lblock, &bh_map, 1);
-		if (unlikely(error))
 			goto out;
-		len -= bh_map.b_size;
+		offset = iomap.offset + iomap.length;
-		nr_blks = bh_map.b_size >> inode->i_blkbits;
+		if (iomap.type != IOMAP_HOLE)
-		lblock += nr_blks;
-		if (!buffer_new(&bh_map))
 			continue;
-		if (unlikely(!buffer_zeronew(&bh_map))) {
+		error = sb_issue_zeroout(sb, iomap.addr >> inode->i_blkbits,
-			error = -EIO;
+					 iomap.length >> inode->i_blkbits,
+					 GFP_NOFS);
+		if (error) {
+			fs_err(GFS2_SB(inode), "Failed to zero data buffers\n");
 			goto out;
 		}
 	}
@@ -809,7 +808,7 @@ static long __gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_alloc_parms ap = { .aflags = 0, };
 	unsigned int data_blocks = 0, ind_blocks = 0, rblocks;
-	loff_t bytes, max_bytes, max_blks = UINT_MAX;
+	loff_t bytes, max_bytes, max_blks;
 	int error;
 	const loff_t pos = offset;
 	const loff_t count = len;
@@ -861,7 +860,8 @@ static long __gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t
 			return error;
 		/* ap.allowed tells us how many blocks quota will allow
 		 * us to write. Check if this reduces max_blks */
-		if (ap.allowed && ap.allowed < max_blks)
+		max_blks = UINT_MAX;
+		if (ap.allowed)
 			max_blks = ap.allowed;
 
 		error = gfs2_inplace_reserve(ip, &ap);

fs/gfs2/incore.h

@@ -130,15 +130,12 @@ static inline bool gfs2_rbm_eq(const struct gfs2_rbm *rbm1,
 enum gfs2_state_bits {
 	BH_Pinned = BH_PrivateStart,
 	BH_Escaped = BH_PrivateStart + 1,
-	BH_Zeronew = BH_PrivateStart + 2,
 };
 
 BUFFER_FNS(Pinned, pinned)
 TAS_BUFFER_FNS(Pinned, pinned)
 BUFFER_FNS(Escaped, escaped)
 TAS_BUFFER_FNS(Escaped, escaped)
-BUFFER_FNS(Zeronew, zeronew)
-TAS_BUFFER_FNS(Zeronew, zeronew)
 
 struct gfs2_bufdata {
 	struct buffer_head *bd_bh;

fs/gfs2/inode.c

@@ -1326,19 +1326,11 @@ static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to)
 static int update_moved_ino(struct gfs2_inode *ip, struct gfs2_inode *ndip,
 			    int dir_rename)
 {
-	int error;
-	struct buffer_head *dibh;
-
 	if (dir_rename)
 		return gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR);
 
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error)
-		return error;
 	ip->i_inode.i_ctime = current_time(&ip->i_inode);
-	gfs2_trans_add_meta(ip->i_gl, dibh);
+	mark_inode_dirty_sync(&ip->i_inode);
-	gfs2_dinode_out(ip, dibh->b_data);
-	brelse(dibh);
 	return 0;
 }
 

fs/gfs2/log.c

@@ -73,7 +73,7 @@ unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
  *
  */
 
-void gfs2_remove_from_ail(struct gfs2_bufdata *bd)
+static void gfs2_remove_from_ail(struct gfs2_bufdata *bd)
 {
 	bd->bd_tr = NULL;
 	list_del_init(&bd->bd_ail_st_list);

fs/gfs2/log.h

@@ -70,7 +70,6 @@ extern void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd,
 extern void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl,
 			   u32 type);
 extern void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans);
-extern void gfs2_remove_from_ail(struct gfs2_bufdata *bd);
 extern void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc);
 
 extern void gfs2_log_shutdown(struct gfs2_sbd *sdp);

fs/gfs2/quota.h

@@ -45,6 +45,8 @@ static inline int gfs2_quota_lock_check(struct gfs2_inode *ip,
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	int ret;
+
+	ap->allowed = UINT_MAX; /* Assume we are permitted a whole lot */
 	if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF)
 		return 0;
 	ret = gfs2_quota_lock(ip, NO_UID_QUOTA_CHANGE, NO_GID_QUOTA_CHANGE);

fs/gfs2/recovery.c

@@ -15,6 +15,7 @@
 #include <linux/gfs2_ondisk.h>
 #include <linux/crc32.h>
 #include <linux/crc32c.h>
+#include <linux/ktime.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -409,12 +410,13 @@ void gfs2_recover_func(struct work_struct *work)
 	struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
 	struct gfs2_log_header_host head;
 	struct gfs2_holder j_gh, ji_gh, thaw_gh;
-	unsigned long t;
+	ktime_t t_start, t_jlck, t_jhd, t_tlck, t_rep;
 	int ro = 0;
 	unsigned int pass;
 	int error;
 	int jlocked = 0;
 
+	t_start = ktime_get();
 	if (sdp->sd_args.ar_spectator ||
 	    (jd->jd_jid != sdp->sd_lockstruct.ls_jid)) {
 		fs_info(sdp, "jid=%u: Trying to acquire journal lock...\n",
@@ -446,6 +448,7 @@ void gfs2_recover_func(struct work_struct *work)
 		fs_info(sdp, "jid=%u, already locked for use\n", jd->jd_jid);
 	}
 
+	t_jlck = ktime_get();
 	fs_info(sdp, "jid=%u: Looking at journal...\n", jd->jd_jid);
 
 	error = gfs2_jdesc_check(jd);
@@ -455,13 +458,12 @@ void gfs2_recover_func(struct work_struct *work)
 	error = gfs2_find_jhead(jd, &head);
 	if (error)
 		goto fail_gunlock_ji;
+	t_jhd = ktime_get();
 
 	if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) {
 		fs_info(sdp, "jid=%u: Acquiring the transaction lock...\n",
 			jd->jd_jid);
 
-		t = jiffies;
-
 		/* Acquire a shared hold on the freeze lock */
 
 		error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED,
@@ -495,6 +497,7 @@ void gfs2_recover_func(struct work_struct *work)
 			goto fail_gunlock_thaw;
 		}
 
+		t_tlck = ktime_get();
 		fs_info(sdp, "jid=%u: Replaying journal...\n", jd->jd_jid);
 
 		for (pass = 0; pass < 2; pass++) {
@@ -509,9 +512,14 @@ void gfs2_recover_func(struct work_struct *work)
 		clean_journal(jd, &head);
 
 		gfs2_glock_dq_uninit(&thaw_gh);
-		t = DIV_ROUND_UP(jiffies - t, HZ);
+		t_rep = ktime_get();
-		fs_info(sdp, "jid=%u: Journal replayed in %lus\n",
+		fs_info(sdp, "jid=%u: Journal replayed in %lldms [jlck:%lldms, "
-			jd->jd_jid, t);
+			"jhead:%lldms, tlck:%lldms, replay:%lldms]\n",
+			jd->jd_jid, ktime_ms_delta(t_rep, t_start),
+			ktime_ms_delta(t_jlck, t_start),
+			ktime_ms_delta(t_jhd, t_jlck),
+			ktime_ms_delta(t_tlck, t_jhd),
+			ktime_ms_delta(t_rep, t_tlck));
 	}
 
 	gfs2_recovery_done(sdp, jd->jd_jid, LM_RD_SUCCESS);

fs/gfs2/trace_gfs2.h

@@ -515,6 +515,7 @@ TRACE_EVENT(gfs2_iomap_end,
 		__field(	u64,	inum			)
 		__field(	loff_t, offset			)
 		__field(	ssize_t, length			)
+		__field(	sector_t, pblock		)
 		__field(	u16,	flags			)
 		__field(	u16,	type			)
 		__field(	int,	ret			)
@@ -525,16 +526,20 @@ TRACE_EVENT(gfs2_iomap_end,
 		__entry->inum		= ip->i_no_addr;
 		__entry->offset		= iomap->offset;
 		__entry->length		= iomap->length;
+		__entry->pblock		= iomap->addr == IOMAP_NULL_ADDR ? 0 :
+					 (iomap->addr >> ip->i_inode.i_blkbits);
 		__entry->flags		= iomap->flags;
 		__entry->type		= iomap->type;
 		__entry->ret		= ret;
 	),
 
-	TP_printk("%u,%u bmap %llu iomap end %llu/%lu ty:%d flags:%08x rc:%d",
+	TP_printk("%u,%u bmap %llu iomap end %llu/%lu to %llu ty:%d flags:%08x rc:%d",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  (unsigned long long)__entry->inum,
 		  (unsigned long long)__entry->offset,
-		  (unsigned long)__entry->length, (u16)__entry->type,
+		  (unsigned long)__entry->length,
+		  (long long)__entry->pblock,
+		  (u16)__entry->type,
 		  (u16)__entry->flags, __entry->ret)
 );