Merge tag 'gfs2-for-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2

Pull gfs2 updates from Andreas Gruenbacher:

 - Prevent cluster nodes from trying to recover their own filesystems
   during a withdraw

 - Add two missing migrate_folio aops and an additional exhash directory
   consistency check (both triggered by syzbot bug reports)

 - Sanitize how dlm results are processed and clean up a few quirks in
   the glock code

 - Minor stuff: Get rid of the GIF_ALLOC_FAILED flag; use SECTOR_SIZE
   and SECTOR_SHIFT

* tag 'gfs2-for-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
  gfs2: No more self recovery
  gfs2: Validate i_depth for exhash directories
  gfs2: Set .migrate_folio in gfs2_{rgrp,meta}_aops
  gfs2: a minor finish_xmote cleanup
  gfs2: simplify finish_xmote
  gfs2: sanitize the gdlm_ast -> finish_xmote interface
  gfs2: Minor do_xmote cancelation fix
  gfs2: Remove GIF_ALLOC_FAILED flag
  gfs2: Use SECTOR_SIZE and SECTOR_SHIFT

12 files changed, 74 insertions, 73 deletions

diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index dbf1aede744c..509e2f0d97e7 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -60,6 +60,7 @@
 #include <linux/crc32.h>
 #include <linux/vmalloc.h>
 #include <linux/bio.h>
+#include <linux/log2.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -912,7 +913,6 @@ static int dir_make_exhash(struct inode *inode)
 	struct qstr args;
 	struct buffer_head *bh, *dibh;
 	struct gfs2_leaf *leaf;
-	int y;
 	u32 x;
 	__be64 *lp;
 	u64 bn;
@@ -979,9 +979,7 @@ static int dir_make_exhash(struct inode *inode)
 	i_size_write(inode, sdp->sd_sb.sb_bsize / 2);
 	gfs2_add_inode_blocks(&dip->i_inode, 1);
 	dip->i_diskflags |= GFS2_DIF_EXHASH;
-
-	for (x = sdp->sd_hash_ptrs, y = -1; x; x >>= 1, y++) ;
-	dip->i_depth = y;
+	dip->i_depth = ilog2(sdp->sd_hash_ptrs);
 
 	gfs2_dinode_out(dip, dibh->b_data);
 
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index ba25b884169e..b6fd1cb17de7 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -590,35 +590,31 @@ static void gfs2_demote_wake(struct gfs2_glock *gl)
 static void finish_xmote(struct gfs2_glock *gl, unsigned int ret)
 {
 	const struct gfs2_glock_operations *glops = gl->gl_ops;
-	struct gfs2_holder *gh;
-	unsigned state = ret & LM_OUT_ST_MASK;
 
-	trace_gfs2_glock_state_change(gl, state);
-	state_change(gl, state);
-	gh = find_first_waiter(gl);
+	if (!(ret & ~LM_OUT_ST_MASK)) {
+		unsigned state = ret & LM_OUT_ST_MASK;
+
+		trace_gfs2_glock_state_change(gl, state);
+		state_change(gl, state);
+	}
+
 
 	/* Demote to UN request arrived during demote to SH or DF */
 	if (test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags) &&
-	    state != LM_ST_UNLOCKED && gl->gl_demote_state == LM_ST_UNLOCKED)
+	    gl->gl_state != LM_ST_UNLOCKED &&
+	    gl->gl_demote_state == LM_ST_UNLOCKED)
 		gl->gl_target = LM_ST_UNLOCKED;
 
 	/* Check for state != intended state */
-	if (unlikely(state != gl->gl_target)) {
-		if (gh && (ret & LM_OUT_CANCELED))
-			gfs2_holder_wake(gh);
+	if (unlikely(gl->gl_state != gl->gl_target)) {
+		struct gfs2_holder *gh = find_first_waiter(gl);
+
 		if (gh && !test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags)) {
 			if (ret & LM_OUT_CANCELED) {
 				list_del_init(&gh->gh_list);
 				trace_gfs2_glock_queue(gh, 0);
+				gfs2_holder_wake(gh);
 				gl->gl_target = gl->gl_state;
-				gh = find_first_waiter(gl);
-				if (gh) {
-					gl->gl_target = gh->gh_state;
-					if (do_promote(gl))
-						goto out;
-					do_xmote(gl, gh, gl->gl_target);
-					return;
-				}
 				goto out;
 			}
 			/* Some error or failed "try lock" - report it */
@@ -629,7 +625,7 @@ static void finish_xmote(struct gfs2_glock *gl, unsigned int ret)
 				goto out;
 			}
 		}
-		switch(state) {
+		switch(gl->gl_state) {
 		/* Unlocked due to conversion deadlock, try again */
 		case LM_ST_UNLOCKED:
 			do_xmote(gl, gh, gl->gl_target);
@@ -640,8 +636,10 @@ static void finish_xmote(struct gfs2_glock *gl, unsigned int ret)
 			do_xmote(gl, gh, LM_ST_UNLOCKED);
 			break;
 		default: /* Everything else */
-			fs_err(gl->gl_name.ln_sbd, "wanted %u got %u\n",
-			       gl->gl_target, state);
+			fs_err(gl->gl_name.ln_sbd,
+			       "glock %u:%llu requested=%u ret=%u\n",
+			       gl->gl_name.ln_type, gl->gl_name.ln_number,
+			       gl->gl_req, ret);
 			GLOCK_BUG_ON(gl, 1);
 		}
 		return;
@@ -650,7 +648,7 @@ static void finish_xmote(struct gfs2_glock *gl, unsigned int ret)
 	/* Fast path - we got what we asked for */
 	if (test_and_clear_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags))
 		gfs2_demote_wake(gl);
-	if (state != LM_ST_UNLOCKED) {
+	if (gl->gl_state != LM_ST_UNLOCKED) {
 		if (glops->go_xmote_bh) {
 			int rv;
 
@@ -802,7 +800,8 @@ skip_inval:
 			 * We skip telling dlm to do the locking, so we won't get a
 			 * reply that would otherwise clear GLF_LOCK. So we clear it here.
 			 */
-			clear_bit(GLF_LOCK, &gl->gl_flags);
+			if (!test_bit(GLF_CANCELING, &gl->gl_flags))
+				clear_bit(GLF_LOCK, &gl->gl_flags);
 			clear_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags);
 			gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD);
 			return;
diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h
index c171f745650f..9339a3bff6ee 100644
--- a/fs/gfs2/glock.h
+++ b/fs/gfs2/glock.h
@@ -92,12 +92,22 @@ enum {
  * LM_OUT_ST_MASK
  * Masks the lower two bits of lock state in the returned value.
  *
+ * LM_OUT_TRY_AGAIN
+ * The trylock request failed.
+ *
+ * LM_OUT_DEADLOCK
+ * The lock request failed because it would deadlock.
+ *
  * LM_OUT_CANCELED
  * The lock request was canceled.
  *
+ * LM_OUT_ERROR
+ * The lock request timed out or failed.
  */
 
 #define LM_OUT_ST_MASK		0x00000003
+#define LM_OUT_TRY_AGAIN	0x00000020
+#define LM_OUT_DEADLOCK		0x00000010
 #define LM_OUT_CANCELED		0x00000008
 #define LM_OUT_ERROR		0x00000004
 
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index cebd66b22694..fe0faad4892f 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -11,6 +11,7 @@
 #include <linux/bio.h>
 #include <linux/posix_acl.h>
 #include <linux/security.h>
+#include <linux/log2.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -450,6 +451,11 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf)
 		gfs2_consist_inode(ip);
 		return -EIO;
 	}
+	if ((ip->i_diskflags & GFS2_DIF_EXHASH) &&
+	    depth < ilog2(sdp->sd_hash_ptrs)) {
+		gfs2_consist_inode(ip);
+		return -EIO;
+	}
 	ip->i_depth = (u8)depth;
 	ip->i_entries = be32_to_cpu(str->di_entries);
 
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 0a41c4e76b32..d4ad82f47eee 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -375,7 +375,6 @@ struct gfs2_glock {
 
 enum {
 	GIF_QD_LOCKED		= 1,
-	GIF_ALLOC_FAILED	= 2,
 	GIF_SW_PAGED		= 3,
 	GIF_FREE_VFS_INODE      = 5,
 	GIF_GLOP_PENDING	= 6,
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 187d789a8f1e..8760e7e20c9d 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -444,11 +444,9 @@ static void gfs2_final_release_pages(struct gfs2_inode *ip)
 	struct inode *inode = &ip->i_inode;
 	struct gfs2_glock *gl = ip->i_gl;
 
-	if (unlikely(!gl)) {
-		/* This can only happen during incomplete inode creation. */
-		BUG_ON(!test_bit(GIF_ALLOC_FAILED, &ip->i_flags));
+	/* This can only happen during incomplete inode creation. */
+	if (unlikely(!gl))
 		return;
-	}
 
 	truncate_inode_pages(gfs2_glock2aspace(gl), 0);
 	truncate_inode_pages(&inode->i_data, 0);
@@ -902,7 +900,6 @@ fail_gunlock3:
 fail_gunlock2:
 	gfs2_glock_put(io_gl);
 fail_dealloc_inode:
-	set_bit(GIF_ALLOC_FAILED, &ip->i_flags);
 	dealloc_error = 0;
 	if (ip->i_eattr)
 		dealloc_error = gfs2_ea_dealloc(ip, xattr_initialized);
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index eafe123617e6..811a0bd3792c 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -44,17 +44,17 @@ static inline int gfs2_is_dir(const struct gfs2_inode *ip)
 
 static inline void gfs2_set_inode_blocks(struct inode *inode, u64 blocks)
 {
-	inode->i_blocks = blocks << (inode->i_blkbits - 9);
+	inode->i_blocks = blocks << (inode->i_blkbits - SECTOR_SHIFT);
 }
 
 static inline u64 gfs2_get_inode_blocks(const struct inode *inode)
 {
-	return inode->i_blocks >> (inode->i_blkbits - 9);
+	return inode->i_blocks >> (inode->i_blkbits - SECTOR_SHIFT);
 }
 
 static inline void gfs2_add_inode_blocks(struct inode *inode, s64 change)
 {
-	change <<= inode->i_blkbits - 9;
+	change <<= inode->i_blkbits - SECTOR_SHIFT;
 	gfs2_assert(GFS2_SB(inode), (change >= 0 || inode->i_blocks >= -change));
 	inode->i_blocks += change;
 }
diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c
index 7cb9d216d8bb..cee5d199d2d8 100644
--- a/fs/gfs2/lock_dlm.c
+++ b/fs/gfs2/lock_dlm.c
@@ -119,7 +119,7 @@ static inline void gfs2_update_request_times(struct gfs2_glock *gl)
 static void gdlm_ast(void *arg)
 {
 	struct gfs2_glock *gl = arg;
-	unsigned ret = gl->gl_state;
+	unsigned ret;
 
 	/* If the glock is dead, we only react to a dlm_unlock() reply. */
 	if (__lockref_is_dead(&gl->gl_lockref) &&
@@ -139,13 +139,16 @@ static void gdlm_ast(void *arg)
 		gfs2_glock_free(gl);
 		return;
 	case -DLM_ECANCEL: /* Cancel while getting lock */
-		ret |= LM_OUT_CANCELED;
+		ret = LM_OUT_CANCELED;
 		goto out;
 	case -EAGAIN: /* Try lock fails */
+		ret = LM_OUT_TRY_AGAIN;
+		goto out;
 	case -EDEADLK: /* Deadlock detected */
+		ret = LM_OUT_DEADLOCK;
 		goto out;
 	case -ETIMEDOUT: /* Canceled due to timeout */
-		ret |= LM_OUT_ERROR;
+		ret = LM_OUT_ERROR;
 		goto out;
 	case 0: /* Success */
 		break;
diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c
index 9dc8885c95d0..7fb11ff71b5a 100644
--- a/fs/gfs2/meta_io.c
+++ b/fs/gfs2/meta_io.c
@@ -103,6 +103,7 @@ const struct address_space_operations gfs2_meta_aops = {
 	.invalidate_folio = block_invalidate_folio,
 	.writepages = gfs2_aspace_writepages,
 	.release_folio = gfs2_release_folio,
+	.migrate_folio = buffer_migrate_folio_norefs,
 };
 
 const struct address_space_operations gfs2_rgrp_aops = {
@@ -110,6 +111,7 @@ const struct address_space_operations gfs2_rgrp_aops = {
 	.invalidate_folio = block_invalidate_folio,
 	.writepages = gfs2_aspace_writepages,
 	.release_folio = gfs2_release_folio,
+	.migrate_folio = buffer_migrate_folio_norefs,
 };
 
 /**
@@ -228,7 +230,7 @@ static void gfs2_submit_bhs(blk_opf_t opf, struct buffer_head *bhs[], int num)
 		struct bio *bio;
 
 		bio = bio_alloc(bh->b_bdev, num, opf, GFP_NOIO);
-		bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
+		bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> SECTOR_SHIFT);
 		while (num > 0) {
 			bh = *bhs;
 			if (!bio_add_folio(bio, bh->b_folio, bh->b_size, bh_offset(bh))) {
@@ -443,11 +445,9 @@ void gfs2_journal_wipe(struct gfs2_inode *ip, u64 bstart, u32 blen)
 	struct buffer_head *bh;
 	int ty;
 
-	if (!ip->i_gl) {
-		/* This can only happen during incomplete inode creation. */
-		BUG_ON(!test_bit(GIF_ALLOC_FAILED, &ip->i_flags));
+	/* This can only happen during incomplete inode creation. */
+	if (!ip->i_gl)
 		return;
-	}
 
 	gfs2_ail1_wipe(sdp, bstart, blen);
 	while (blen) {
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 85c491fcf1a3..a19d7e431c8e 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -163,7 +163,7 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent)
 		return -EINVAL;
 	}
 
-	if (sb->sb_bsize < 512 || sb->sb_bsize > PAGE_SIZE ||
+	if (sb->sb_bsize < SECTOR_SIZE || sb->sb_bsize > PAGE_SIZE ||
 	    (sb->sb_bsize & (sb->sb_bsize - 1))) {
 		pr_warn("Invalid block size\n");
 		return -EINVAL;
@@ -224,8 +224,8 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent)
 	if (unlikely(!sb))
 		return -ENOMEM;
 	err = bdev_rw_virt(sdp->sd_vfs->s_bdev,
-			sector * (sdp->sd_vfs->s_blocksize >> 9), sb, PAGE_SIZE,
-			REQ_OP_READ | REQ_META);
+			   sector << (sdp->sd_vfs->s_blocksize_bits - SECTOR_SHIFT),
+			   sb, PAGE_SIZE, REQ_OP_READ | REQ_META);
 	if (err) {
 		pr_warn("error %d reading superblock\n", err);
 		kfree(sb);
@@ -257,7 +257,7 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent)
 		return error;
 	}
 
-	sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - 9;
+	sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - SECTOR_SHIFT;
 	sdp->sd_fsb2bb = BIT(sdp->sd_fsb2bb_shift);
 	sdp->sd_diptrs = (sdp->sd_sb.sb_bsize -
 			  sizeof(struct gfs2_dinode)) / sizeof(u64);
@@ -1155,12 +1155,12 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc)
 
 	/* Set up the buffer cache and fill in some fake block size values
 	   to allow us to read-in the on-disk superblock. */
-	sdp->sd_sb.sb_bsize = sb_min_blocksize(sb, 512);
+	sdp->sd_sb.sb_bsize = sb_min_blocksize(sb, SECTOR_SIZE);
 	error = -EINVAL;
 	if (!sdp->sd_sb.sb_bsize)
 		goto fail_free;
 	sdp->sd_sb.sb_bsize_shift = sb->s_blocksize_bits;
-	sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - 9;
+	sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - SECTOR_SHIFT;
 	sdp->sd_fsb2bb = BIT(sdp->sd_fsb2bb_shift);
 
 	sdp->sd_tune.gt_logd_secs = sdp->sd_args.ar_commit;
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 7c518c4ff638..b42e2110084b 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -487,11 +487,9 @@ static void gfs2_dirty_inode(struct inode *inode, int flags)
 	int need_endtrans = 0;
 	int ret;
 
-	if (unlikely(!ip->i_gl)) {
-		/* This can only happen during incomplete inode creation. */
-		BUG_ON(!test_bit(GIF_ALLOC_FAILED, &ip->i_flags));
+	/* This can only happen during incomplete inode creation. */
+	if (unlikely(!ip->i_gl))
 		return;
-	}
 
 	if (gfs2_withdrawing_or_withdrawn(sdp))
 		return;
diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c
index d5a1e63fa257..24864a66074b 100644
--- a/fs/gfs2/util.c
+++ b/fs/gfs2/util.c
@@ -232,32 +232,23 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp)
 	 */
 	ret = gfs2_glock_nq(&sdp->sd_live_gh);
 
+	gfs2_glock_put(live_gl); /* drop extra reference we acquired */
+	clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
+
 	/*
 	 * If we actually got the "live" lock in EX mode, there are no other
-	 * nodes available to replay our journal. So we try to replay it
-	 * ourselves. We hold the "live" glock to prevent other mounters
-	 * during recovery, then just dequeue it and reacquire it in our
-	 * normal SH mode. Just in case the problem that caused us to
-	 * withdraw prevents us from recovering our journal (e.g. io errors
-	 * and such) we still check if the journal is clean before proceeding
-	 * but we may wait forever until another mounter does the recovery.
+	 * nodes available to replay our journal.
 	 */
 	if (ret == 0) {
-		fs_warn(sdp, "No other mounters found. Trying to recover our "
-			"own journal jid %d.\n", sdp->sd_lockstruct.ls_jid);
-		if (gfs2_recover_journal(sdp->sd_jdesc, 1))
-			fs_warn(sdp, "Unable to recover our journal jid %d.\n",
-				sdp->sd_lockstruct.ls_jid);
-		gfs2_glock_dq_wait(&sdp->sd_live_gh);
-		gfs2_holder_reinit(LM_ST_SHARED,
-				   LM_FLAG_NOEXP | GL_EXACT | GL_NOPID,
-				   &sdp->sd_live_gh);
-		gfs2_glock_nq(&sdp->sd_live_gh);
+		fs_warn(sdp, "No other mounters found.\n");
+		/*
+		 * We are about to release the lockspace.  By keeping live_gl
+		 * locked here, we ensure that the next mounter coming along
+		 * will be a "first" mounter which will perform recovery.
+		 */
+		goto skip_recovery;
 	}
 
-	gfs2_glock_put(live_gl); /* drop extra reference we acquired */
-	clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags);
-
 	/*
 	 * At this point our journal is evicted, so we need to get a new inode
 	 * for it. Once done, we need to call gfs2_find_jhead which