Skip to contentSkip to conversation

xfs: improve RT geometry validation

PATCH In review Christoph Hellwig 3 participants 5 messages May 7 – May 9 XFS filesystem

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 xfs: improve RT geometry validation

Make sure the rtgcount is plausible for the given RT group size.

---
 fs/xfs/libxfs/xfs_sb.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)
fs/xfs/libxfs/xfs_sb.c +15 -7
index 47322adb7690..c0354f43704f 100644
@@ -302,17 +302,25 @@ xfs_validate_rt_geometry(
 		return false;
 
 	if (xfs_sb_is_v5(sbp) &&
-	    (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED)) {
+	    (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_METADIR)) {
+		uint64_t		nr_rgs;
 		uint32_t		mod;
 
 		/*
-		 * Zoned RT devices must be aligned to the RT group size,
-		 * because garbage collection assumes that all zones have the
-		 * same size to avoid insane complexity if that weren't the
-		 * case.
+		 * Check that the number of rgcounts is plausible for the RG
+		 * size.
+		 */
+		nr_rgs = div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
+		if (nr_rgs != sbp->sb_rgcount + !!mod)
+			return false;
+
+		/*
+		 * Zoned RT devices must be aligned to the RT group size because
+		 * garbage collection assumes that all zones have the same size
+		 * to avoid insane complexity if that weren't the case.
 		 */
-		div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
-		if (mod)
+		if ((sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED) &&
+		    mod > 0)
 			return false;
 	}
 
Signature · 1 line
2.53.0
maintainer Reviewed-by on v1 1/1
#1 lore raw
Quoted 33 lines · Christoph Hellwig

Make sure the rtgcount is plausible for the given RT group size.

---
 fs/xfs/libxfs/xfs_sb.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)
fs/xfs/libxfs/xfs_sb.c +10 -5
index 47322adb7690..c0354f43704f 100644
@@ -302,17 +302,25 @@ xfs_validate_rt_geometry(
 		return false;
 
 	if (xfs_sb_is_v5(sbp) &&
-	    (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED)) {
+	    (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_METADIR)) {
+		uint64_t		nr_rgs;
 		uint32_t		mod;
 
 		/*
-		 * Zoned RT devices must be aligned to the RT group size,
-		 * because garbage collection assumes that all zones have the
-		 * same size to avoid insane complexity if that weren't the
-		 * case.
+		 * Check that the number of rgcounts is plausible for the RG
+		 * size.
+		 */
+		nr_rgs = div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
+		if (nr_rgs != sbp->sb_rgcount + !!mod)
+			return false;
+
+		/*

fs/xfs/libxfs/xfs_sb.c xfs_validate_rt_geometry( · line 325

Quoted 10 lines
fs/xfs/libxfs/xfs_sb.c +5 -2
+		 * Zoned RT devices must be aligned to the RT group size because
+		 * garbage collection assumes that all zones have the same size
+		 * to avoid insane complexity if that weren't the case.
 		 */
-		div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
-		if (mod)
+		if ((sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED) &&
+		    mod > 0)
 			return false;
 	}

Looks good

Quoted 4 lines
fs/xfs/libxfs/xfs_sb.c +0 -1
--

2.53.0

#2 lore raw
In reply to Christoph Hellwig
Quoted 21 lines · Christoph Hellwig

Make sure the rtgcount is plausible for the given RT group size.

---
 fs/xfs/libxfs/xfs_sb.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)
fs/xfs/libxfs/xfs_sb.c +2 -1
index 47322adb7690..c0354f43704f 100644
@@ -302,17 +302,25 @@ xfs_validate_rt_geometry(
 		return false;
 
 	if (xfs_sb_is_v5(sbp) &&
-	    (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED)) {
+	    (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_METADIR)) {
+		uint64_t		nr_rgs;
 		uint32_t		mod;
 
 		/*

fs/xfs/libxfs/xfs_sb.c xfs_validate_rt_geometry( · line 315

Quoted 10 lines
fs/xfs/libxfs/xfs_sb.c +6 -4
-		 * Zoned RT devices must be aligned to the RT group size,
-		 * because garbage collection assumes that all zones have the
-		 * same size to avoid insane complexity if that weren't the
-		 * case.
+		 * Check that the number of rgcounts is plausible for the RG
+		 * size.
+		 */
+		nr_rgs = div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
+		if (nr_rgs != sbp->sb_rgcount + !!mod)
+			return false;

Isn't this already covered by xfs_validate_sb_rtgroups?

	groups = howmany_64(sbp->sb_rextents, sbp->sb_rgextents);
	if (groups != sbp->sb_rgcount) {
		xfs_warn(mp,
"Realtime groups (%u) do not cover the entire rt section; need (%llu) groups.",
				sbp->sb_rgcount, groups);
		return -EINVAL;
	}

fs/xfs/libxfs/xfs_sb.c xfs_validate_rt_geometry( · line 324

Quoted 11 lines
fs/xfs/libxfs/xfs_sb.c +7 -2
+
+		/*
+		 * Zoned RT devices must be aligned to the RT group size because
+		 * garbage collection assumes that all zones have the same size
+		 * to avoid insane complexity if that weren't the case.
 		 */
-		div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
-		if (mod)
+		if ((sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED) &&
+		    mod > 0)
 			return false;

And shouldn't this be in xfs_validate_sb_zoned?

Or do you want this in xfs_validate_rt_geometry because that's what xfs_repair calls to check the superblock?

--D

#3 lore raw
Quoted 1 line · Darrick J. Wong

Isn't this already covered by xfs_validate_sb_rtgroups?

From a quick look it should...

Quoted 7 lines

fs/xfs/libxfs/xfs_sb.c xfs_validate_rt_geometry( · line 324

Quoted 5 lines
fs/xfs/libxfs/xfs_sb.c +2 -2
-		div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
-		if (mod)
+		if ((sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED) &&
+		    mod > 0)
 			return false;

And shouldn't this be in xfs_validate_sb_zoned?

This just slightly moves code.

Quoted 2 lines

Or do you want this in xfs_validate_rt_geometry because that's what xfs_repair calls to check the superblock?

I guess. At least it was repair that went off on my badly messed up file system without this fix.

Maybe it's time to go back to the drawing board for the sb validation.

#4 lore raw
Quoted 15 lines · Christoph Hellwig
Quoted 1 line · Darrick J. Wong

Isn't this already covered by xfs_validate_sb_rtgroups?

From a quick look it should...

Quoted 7 lines

fs/xfs/libxfs/xfs_sb.c xfs_validate_rt_geometry( · line 324

Quoted 5 lines
fs/xfs/libxfs/xfs_sb.c +2 -2
-		div_u64_rem(sbp->sb_rextents, sbp->sb_rgextents, &mod);
-		if (mod)
+		if ((sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_ZONED) &&
+		    mod > 0)
 			return false;

And shouldn't this be in xfs_validate_sb_zoned?

This just slightly moves code.

Quoted 7 lines
Quoted 2 lines

Or do you want this in xfs_validate_rt_geometry because that's what xfs_repair calls to check the superblock?

I guess. At least it was repair that went off on my badly messed up file system without this fix.

Maybe it's time to go back to the drawing board for the sb validation.

Yeah, it's unfortunate that xfs_repair duplicates most of that sb validation code just so it can return an enum instead of xfs_err()ing all over the place. :/

--D