Merge tag 'gfs2-v5.7-rc1.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2

Pull gfs2 fixes from Andreas Gruenbacher:
 "Various gfs2 fixes.

  Fixes for bugs prior to v5.7:
   - Fix random block reads when reading fragmented journals (v5.2)
   - Fix a possible random memory access in gfs2_walk_metadata (v5.3)

  Fixes for v5.7:
   - Fix several overlooked gfs2_qa_get / gfs2_qa_put imbalances
   - Fix several bugs in the new filesystem withdraw logic"

* tag 'gfs2-v5.7-rc1.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
  Revert "gfs2: Don't demote a glock until its revokes are written"
  gfs2: If go_sync returns error, withdraw but skip invalidate
  gfs2: Grab glock reference sooner in gfs2_add_revoke
  gfs2: don't call quota_unhold if quotas are not locked
  gfs2: move privileged user check to gfs2_quota_lock_check
  gfs2: remove check for quotas on in gfs2_quota_check
  gfs2: Change BUG_ON to an assert_withdraw in gfs2_quota_change
  gfs2: Fix problems regarding gfs2_qa_get and _put
  gfs2: More gfs2_find_jhead fixes
  gfs2: Another gfs2_walk_metadata fix
  gfs2: Fix use-after-free in gfs2_logd after withdraw
  gfs2: Fix BUG during unmount after file system withdraw
  gfs2: Fix error exit in do_xmote
  gfs2: fix withdraw sequence deadlock

fs/gfs2/bmap.c

@@ -528,10 +528,12 @@ static int gfs2_walk_metadata(struct inode *inode, struct metapath *mp,
 
 		/* Advance in metadata tree. */
 		(mp->mp_list[hgt])++;
-		if (mp->mp_list[hgt] >= sdp->sd_inptrs) {
-			if (!hgt)
+		if (hgt) {
+			if (mp->mp_list[hgt] >= sdp->sd_inptrs)
+				goto lower_metapath;
+		} else {
+			if (mp->mp_list[hgt] >= sdp->sd_diptrs)
 				break;
-			goto lower_metapath;
 		}
 
 fill_up_metapath:
@@ -876,10 +878,9 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
 					ret = -ENOENT;
 					goto unlock;
 				} else {
-					/* report a hole */
 					iomap->offset = pos;
 					iomap->length = length;
-					goto do_alloc;
+					goto hole_found;
 				}
 			}
 			iomap->length = size;
@@ -933,8 +934,6 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
 	return ret;
 
 do_alloc:
-	iomap->addr = IOMAP_NULL_ADDR;
-	iomap->type = IOMAP_HOLE;
 	if (flags & IOMAP_REPORT) {
 		if (pos >= size)
 			ret = -ENOENT;
@@ -956,6 +955,9 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
 		if (pos < size && height == ip->i_height)
 			ret = gfs2_hole_size(inode, lblock, len, mp, iomap);
 	}
+hole_found:
+	iomap->addr = IOMAP_NULL_ADDR;
+	iomap->type = IOMAP_HOLE;
 	goto out;
 }
 

fs/gfs2/glock.c

@@ -613,7 +613,7 @@ __acquires(&gl->gl_lockref.lock)
 				fs_err(sdp, "Error %d syncing glock \n", ret);
 				gfs2_dump_glock(NULL, gl, true);
 			}
-			return;
+			goto skip_inval;
 		}
 	}
 	if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags)) {
@@ -633,6 +633,7 @@ __acquires(&gl->gl_lockref.lock)
 		clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);
 	}
 
+skip_inval:
 	gfs2_glock_hold(gl);
 	/*
 	 * Check for an error encountered since we called go_sync and go_inval.
@@ -722,9 +723,6 @@ __acquires(&gl->gl_lockref.lock)
 			goto out_unlock;
 		if (nonblock)
 			goto out_sched;
-		smp_mb();
-		if (atomic_read(&gl->gl_revokes) != 0)
-			goto out_sched;
 		set_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags);
 		GLOCK_BUG_ON(gl, gl->gl_demote_state == LM_ST_EXCLUSIVE);
 		gl->gl_target = gl->gl_demote_state;

fs/gfs2/inode.c

@@ -622,7 +622,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 				error = finish_no_open(file, NULL);
 		}
 		gfs2_glock_dq_uninit(ghs);
-		return error;
+		goto fail;
 	} else if (error != -ENOENT) {
 		goto fail_gunlock;
 	}
@@ -764,9 +764,11 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 		error = finish_open(file, dentry, gfs2_open_common);
 	}
 	gfs2_glock_dq_uninit(ghs);
+	gfs2_qa_put(ip);
 	gfs2_glock_dq_uninit(ghs + 1);
 	clear_bit(GLF_INODE_CREATING, &io_gl->gl_flags);
 	gfs2_glock_put(io_gl);
+	gfs2_qa_put(dip);
 	return error;
 
 fail_gunlock3:
@@ -776,7 +778,6 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
 	clear_bit(GLF_INODE_CREATING, &io_gl->gl_flags);
 	gfs2_glock_put(io_gl);
 fail_free_inode:
-	gfs2_qa_put(ip);
 	if (ip->i_gl) {
 		glock_clear_object(ip->i_gl, ip);
 		gfs2_glock_put(ip->i_gl);
@@ -1005,7 +1006,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
 out_child:
 	gfs2_glock_dq(ghs);
 out_parent:
-	gfs2_qa_put(ip);
+	gfs2_qa_put(dip);
 	gfs2_holder_uninit(ghs);
 	gfs2_holder_uninit(ghs + 1);
 	return error;

fs/gfs2/log.c

@@ -669,13 +669,13 @@ void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd)
 	struct buffer_head *bh = bd->bd_bh;
 	struct gfs2_glock *gl = bd->bd_gl;
 
+	sdp->sd_log_num_revoke++;
+	if (atomic_inc_return(&gl->gl_revokes) == 1)
+		gfs2_glock_hold(gl);
 	bh->b_private = NULL;
 	bd->bd_blkno = bh->b_blocknr;
 	gfs2_remove_from_ail(bd); /* drops ref on bh */
 	bd->bd_bh = NULL;
-	sdp->sd_log_num_revoke++;
-	if (atomic_inc_return(&gl->gl_revokes) == 1)
-		gfs2_glock_hold(gl);
 	set_bit(GLF_LFLUSH, &gl->gl_flags);
 	list_add(&bd->bd_list, &sdp->sd_log_revokes);
 }
@@ -1131,6 +1131,10 @@ int gfs2_logd(void *data)
 
 	while (!kthread_should_stop()) {
 
+		if (gfs2_withdrawn(sdp)) {
+			msleep_interruptible(HZ);
+			continue;
+		}
 		/* Check for errors writing to the journal */
 		if (sdp->sd_log_error) {
 			gfs2_lm(sdp,
@@ -1139,6 +1143,7 @@ int gfs2_logd(void *data)
 				"prevent further damage.\n",
 				sdp->sd_fsname, sdp->sd_log_error);
 			gfs2_withdraw(sdp);
+			continue;
 		}
 
 		did_flush = false;

fs/gfs2/lops.c

@@ -263,7 +263,7 @@ static struct bio *gfs2_log_alloc_bio(struct gfs2_sbd *sdp, u64 blkno,
 	struct super_block *sb = sdp->sd_vfs;
 	struct bio *bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES);
 
-	bio->bi_iter.bi_sector = blkno << (sb->s_blocksize_bits - 9);
+	bio->bi_iter.bi_sector = blkno << sdp->sd_fsb2bb_shift;
 	bio_set_dev(bio, sb->s_bdev);
 	bio->bi_end_io = end_io;
 	bio->bi_private = sdp;
@@ -509,7 +509,7 @@ int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head,
 	unsigned int bsize = sdp->sd_sb.sb_bsize, off;
 	unsigned int bsize_shift = sdp->sd_sb.sb_bsize_shift;
 	unsigned int shift = PAGE_SHIFT - bsize_shift;
-	unsigned int readahead_blocks = BIO_MAX_PAGES << shift;
+	unsigned int max_bio_size = 2 * 1024 * 1024;
 	struct gfs2_journal_extent *je;
 	int sz, ret = 0;
 	struct bio *bio = NULL;
@@ -537,12 +537,17 @@ int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head,
 				off = 0;
 			}
 
-			if (!bio || (bio_chained && !off)) {
+			if (!bio || (bio_chained && !off) ||
+			    bio->bi_iter.bi_size >= max_bio_size) {
 				/* start new bio */
 			} else {
-				sz = bio_add_page(bio, page, bsize, off);
-				if (sz == bsize)
-					goto block_added;
+				sector_t sector = dblock << sdp->sd_fsb2bb_shift;
+
+				if (bio_end_sector(bio) == sector) {
+					sz = bio_add_page(bio, page, bsize, off);
+					if (sz == bsize)
+						goto block_added;
+				}
 				if (off) {
 					unsigned int blocks =
 						(PAGE_SIZE - off) >> bsize_shift;
@@ -568,7 +573,7 @@ int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head,
 			off += bsize;
 			if (off == PAGE_SIZE)
 				page = NULL;
-			if (blocks_submitted < blocks_read + readahead_blocks) {
+			if (blocks_submitted < 2 * max_bio_size >> bsize_shift) {
 				/* Keep at least one bio in flight */
 				continue;
 			}

fs/gfs2/meta_io.c

@@ -252,7 +252,7 @@ int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags,
 	int num = 0;
 
 	if (unlikely(gfs2_withdrawn(sdp)) &&
-	    (!sdp->sd_jdesc || (blkno != sdp->sd_jdesc->jd_no_addr))) {
+	    (!sdp->sd_jdesc || gl != sdp->sd_jinode_gl)) {
 		*bhp = NULL;
 		return -EIO;
 	}

fs/gfs2/quota.c

@@ -1051,8 +1051,7 @@ int gfs2_quota_lock(struct gfs2_inode *ip, kuid_t uid, kgid_t gid)
 	u32 x;
 	int error = 0;
 
-	if (capable(CAP_SYS_RESOURCE) ||
-	    sdp->sd_args.ar_quota != GFS2_QUOTA_ON)
+	if (sdp->sd_args.ar_quota != GFS2_QUOTA_ON)
 		return 0;
 
 	error = gfs2_quota_hold(ip, uid, gid);
@@ -1125,7 +1124,7 @@ void gfs2_quota_unlock(struct gfs2_inode *ip)
 	int found;
 
 	if (!test_and_clear_bit(GIF_QD_LOCKED, &ip->i_flags))
-		goto out;
+		return;
 
 	for (x = 0; x < ip->i_qadata->qa_qd_num; x++) {
 		struct gfs2_quota_data *qd;
@@ -1162,7 +1161,6 @@ void gfs2_quota_unlock(struct gfs2_inode *ip)
 			qd_unlock(qda[x]);
 	}
 
-out:
 	gfs2_quota_unhold(ip);
 }
 
@@ -1210,9 +1208,6 @@ int gfs2_quota_check(struct gfs2_inode *ip, kuid_t uid, kgid_t gid,
 	if (!test_bit(GIF_QD_LOCKED, &ip->i_flags))
 		return 0;
 
-        if (sdp->sd_args.ar_quota != GFS2_QUOTA_ON)
-                return 0;
-
 	for (x = 0; x < ip->i_qadata->qa_qd_num; x++) {
 		qd = ip->i_qadata->qa_qd[x];
 
@@ -1270,7 +1265,9 @@ void gfs2_quota_change(struct gfs2_inode *ip, s64 change,
 	if (ip->i_diskflags & GFS2_DIF_SYSTEM)
 		return;
 
-	BUG_ON(ip->i_qadata->qa_ref <= 0);
+	if (gfs2_assert_withdraw(sdp, ip->i_qadata &&
+				 ip->i_qadata->qa_ref > 0))
+		return;
 	for (x = 0; x < ip->i_qadata->qa_qd_num; x++) {
 		qd = ip->i_qadata->qa_qd[x];
 

fs/gfs2/quota.h

@@ -44,7 +44,8 @@ static inline int gfs2_quota_lock_check(struct gfs2_inode *ip,
 	int ret;
 
 	ap->allowed = UINT_MAX; /* Assume we are permitted a whole lot */
-	if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF)
+	if (capable(CAP_SYS_RESOURCE) ||
+	    sdp->sd_args.ar_quota == GFS2_QUOTA_OFF)
 		return 0;
 	ret = gfs2_quota_lock(ip, NO_UID_QUOTA_CHANGE, NO_GID_QUOTA_CHANGE);
 	if (ret)

fs/gfs2/super.c

@@ -1404,7 +1404,6 @@ static void gfs2_evict_inode(struct inode *inode)
 	if (ip->i_qadata)
 		gfs2_assert_warn(sdp, ip->i_qadata->qa_ref == 0);
 	gfs2_rs_delete(ip, NULL);
-	gfs2_qa_put(ip);
 	gfs2_ordered_del_inode(ip);
 	clear_inode(inode);
 	gfs2_dir_hash_inval(ip);

fs/gfs2/util.c

@@ -119,6 +119,12 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp)
 	if (!sb_rdonly(sdp->sd_vfs))
 		ret = gfs2_make_fs_ro(sdp);
 
+	if (sdp->sd_lockstruct.ls_ops->lm_lock == NULL) { /* lock_nolock */
+		if (!ret)
+			ret = -EIO;
+		clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
+		goto skip_recovery;
+	}
 	/*
 	 * Drop the glock for our journal so another node can recover it.
 	 */
@@ -159,10 +165,6 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp)
 		wait_on_bit(&gl->gl_flags, GLF_FREEING, TASK_UNINTERRUPTIBLE);
 	}
 
-	if (sdp->sd_lockstruct.ls_ops->lm_lock == NULL) { /* lock_nolock */
-		clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
-		goto skip_recovery;
-	}
 	/*
 	 * Dequeue the "live" glock, but keep a reference so it's never freed.
 	 */