Skip to contentSkip to conversation

nfsd: only call inode_query_iversion in the I_VERSION case

PATCH In review J. Bruce Fields 2 participants 8 messages Nov 30, 2020

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/5 nfsd: only call inode_query_iversion in the I_VERSION case

From: "J. Bruce Fields" <[email protected]>

inode_query_iversion() can modify i_version. Depending on the exported filesystem, that may not be safe. For example, if you're re-exporting NFS, NFS stores the server's change attribute in i_version and does not expect it to be modified locally. This has been observed causing unnecessary cache invalidations.

The way a filesystem indicates that it's OK to call inode_query_iverson() is by setting SB_I_VERSION.

So, move the I_VERSION check out of encode_change(), where it's used only in FATTR responses, to nfsd4_changeattr(), which is also called for pre- and post- operation attributes.

(Note we could also pull the NFSEXP_V4ROOT case into nfsd4_change_attribute as well. That would actually be a no-op, since pre/post attrs are only used for metadata-modifying operations, and V4ROOT exports are read-only. But we might make the change in the future just for simplicity.)

---
 fs/nfsd/nfs3xdr.c |  5 ++---
 fs/nfsd/nfs4xdr.c |  6 +-----
 fs/nfsd/nfsfh.h   | 14 ++++++++++----
 3 files changed, 13 insertions(+), 12 deletions(-)
fs/nfsd/nfs3xdr.c +2 -3
index 2277f83da250..dfbf390ff40c 100644
@@ -291,14 +291,13 @@ void fill_post_wcc(struct svc_fh *fhp)
 		printk("nfsd: inode locked twice during operation.\n");
 
 	err = fh_getattr(fhp, &fhp->fh_post_attr);
-	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
-						     d_inode(fhp->fh_dentry));
 	if (err) {
 		fhp->fh_post_saved = false;
-		/* Grab the ctime anyway - set_change_info might use it */
 		fhp->fh_post_attr.ctime = d_inode(fhp->fh_dentry)->i_ctime;
 	} else
 		fhp->fh_post_saved = true;
+	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
+						     d_inode(fhp->fh_dentry));
 }
 
 /*
fs/nfsd/nfs4xdr.c +1 -5
index 833a2c64dfe8..56fd5f6d5c44 100644
@@ -2298,12 +2298,8 @@ static __be32 *encode_change(__be32 *p, struct kstat *stat, struct inode *inode,
 	if (exp->ex_flags & NFSEXP_V4ROOT) {
 		*p++ = cpu_to_be32(convert_to_wallclock(exp->cd->flush_time));
 		*p++ = 0;
-	} else if (IS_I_VERSION(inode)) {
+	} else
 		p = xdr_encode_hyper(p, nfsd4_change_attribute(stat, inode));
-	} else {
-		*p++ = cpu_to_be32(stat->ctime.tv_sec);
-		*p++ = cpu_to_be32(stat->ctime.tv_nsec);
-	}
 	return p;
 }
 
fs/nfsd/nfsfh.h +10 -4
index 56cfbc361561..3faf5974fa4e 100644
@@ -261,10 +261,16 @@ static inline u64 nfsd4_change_attribute(struct kstat *stat,
 {
 	u64 chattr;
 
-	chattr =  stat->ctime.tv_sec;
-	chattr <<= 30;
-	chattr += stat->ctime.tv_nsec;
-	chattr += inode_query_iversion(inode);
+	if (IS_I_VERSION(inode)) {
+		chattr =  stat->ctime.tv_sec;
+		chattr <<= 30;
+		chattr += stat->ctime.tv_nsec;
+		chattr += inode_query_iversion(inode);
+	} else {
+		chattr = cpu_to_be32(stat->ctime.tv_sec);
+		chattr <<= 32;
+		chattr += cpu_to_be32(stat->ctime.tv_nsec);
+	}
 	return chattr;
 }
 
Signature · 1 line
2.28.0
#1 lore raw

PATCH 2/5 nfsd: simplify nfsd4_change_info

From: "J. Bruce Fields" <[email protected]>

It doesn't make sense to carry all these extra fields around. Just make everything into change attribute from the start.

This is just cleanup, there should be no change in behavior.

---
 fs/nfsd/nfs4xdr.c | 11 ++---------
 fs/nfsd/xdr4.h    | 11 -----------
 2 files changed, 2 insertions(+), 20 deletions(-)
fs/nfsd/nfs4xdr.c +2 -9
index 56fd5f6d5c44..18c912930947 100644
@@ -2331,15 +2331,8 @@ static __be32 *encode_time_delta(__be32 *p, struct inode *inode)
 static __be32 *encode_cinfo(__be32 *p, struct nfsd4_change_info *c)
 {
 	*p++ = cpu_to_be32(c->atomic);
-	if (c->change_supported) {
-		p = xdr_encode_hyper(p, c->before_change);
-		p = xdr_encode_hyper(p, c->after_change);
-	} else {
-		*p++ = cpu_to_be32(c->before_ctime_sec);
-		*p++ = cpu_to_be32(c->before_ctime_nsec);
-		*p++ = cpu_to_be32(c->after_ctime_sec);
-		*p++ = cpu_to_be32(c->after_ctime_nsec);
-	}
+	p = xdr_encode_hyper(p, c->before_change);
+	p = xdr_encode_hyper(p, c->after_change);
 	return p;
 }
 
fs/nfsd/xdr4.h +0 -11
index 679d40af1bbb..3a59053084e6 100644
@@ -76,12 +76,7 @@ static inline bool nfsd4_has_session(struct nfsd4_compound_state *cs)
 
 struct nfsd4_change_info {
 	u32		atomic;
-	bool		change_supported;
-	u32		before_ctime_sec;
-	u32		before_ctime_nsec;
 	u64		before_change;
-	u32		after_ctime_sec;
-	u32		after_ctime_nsec;
 	u64		after_change;
 };
 
@@ -768,15 +763,9 @@ set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp)
 {
 	BUG_ON(!fhp->fh_pre_saved);
 	cinfo->atomic = (u32)fhp->fh_post_saved;
-	cinfo->change_supported = IS_I_VERSION(d_inode(fhp->fh_dentry));
 
 	cinfo->before_change = fhp->fh_pre_change;
 	cinfo->after_change = fhp->fh_post_change;
-	cinfo->before_ctime_sec = fhp->fh_pre_ctime.tv_sec;
-	cinfo->before_ctime_nsec = fhp->fh_pre_ctime.tv_nsec;
-	cinfo->after_ctime_sec = fhp->fh_post_attr.ctime.tv_sec;
-	cinfo->after_ctime_nsec = fhp->fh_post_attr.ctime.tv_nsec;
-
 }
 
 
Signature · 1 line
2.28.0
#2 lore raw

PATCH 3/5 nfsd: minor nfsd4_change_attribute cleanup

In reply to J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

Minor cleanup, no change in behavior.

Also pull out a common helper that'll be useful elsewhere.

---
 fs/nfsd/nfsfh.h          | 13 +++++--------
 include/linux/iversion.h | 13 +++++++++++++
 2 files changed, 18 insertions(+), 8 deletions(-)
fs/nfsd/nfsfh.h +5 -8
index 3faf5974fa4e..45bd776290d5 100644
@@ -259,19 +259,16 @@ fh_clear_wcc(struct svc_fh *fhp)
 static inline u64 nfsd4_change_attribute(struct kstat *stat,
 					 struct inode *inode)
 {
-	u64 chattr;
-
 	if (IS_I_VERSION(inode)) {
+		u64 chattr;
+
 		chattr =  stat->ctime.tv_sec;
 		chattr <<= 30;
 		chattr += stat->ctime.tv_nsec;
 		chattr += inode_query_iversion(inode);
-	} else {
-		chattr = cpu_to_be32(stat->ctime.tv_sec);
-		chattr <<= 32;
-		chattr += cpu_to_be32(stat->ctime.tv_nsec);
-	}
-	return chattr;
+		return chattr;
+	} else
+		return time_to_chattr(&stat->ctime);
 }
 
 extern void fill_pre_wcc(struct svc_fh *fhp);
include/linux/iversion.h +13 -0
index 2917ef990d43..3bfebde5a1a6 100644
@@ -328,6 +328,19 @@ inode_query_iversion(struct inode *inode)
 	return cur >> I_VERSION_QUERIED_SHIFT;
 }
 
+/*
+ * For filesystems without any sort of change attribute, the best we can
+ * do is fake one up from the ctime:
+ */
+static inline u64 time_to_chattr(struct timespec64 *t)
+{
+	u64 chattr = t->tv_sec;
+
+	chattr <<= 32;
+	chattr += t->tv_nsec;
+	return chattr;
+}
+
 /**
  * inode_eq_iversion_raw - check whether the raw i_version counter has changed
  * @inode: inode to check
Signature · 1 line
2.28.0
#3 lore raw

PATCH 4/5 nfsd4: don't query change attribute in v2/v3 case

In reply to J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

inode_query_iversion() has side effects, and there's no point calling it when we're not even going to use it.

We check whether we're currently processing a v4 request by checking fh_maxsize, which is arguably a little hacky; we could add a flag to svc_fh instead.

---
 fs/nfsd/nfs3xdr.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)
fs/nfsd/nfs3xdr.c +9 -5
index dfbf390ff40c..f4fd54dc6965 100644
@@ -259,11 +259,11 @@ void fill_pre_wcc(struct svc_fh *fhp)
 {
 	struct inode    *inode;
 	struct kstat	stat;
+	bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE);
 	__be32 err;
 
 	if (fhp->fh_pre_saved)
 		return;
-
 	inode = d_inode(fhp->fh_dentry);
 	err = fh_getattr(fhp, &stat);
 	if (err) {
@@ -272,11 +272,12 @@ void fill_pre_wcc(struct svc_fh *fhp)
 		stat.ctime = inode->i_ctime;
 		stat.size  = inode->i_size;
 	}
+	if (v4)
+		fhp->fh_pre_change = nfsd4_change_attribute(&stat, inode);
 
 	fhp->fh_pre_mtime = stat.mtime;
 	fhp->fh_pre_ctime = stat.ctime;
 	fhp->fh_pre_size  = stat.size;
-	fhp->fh_pre_change = nfsd4_change_attribute(&stat, inode);
 	fhp->fh_pre_saved = true;
 }
 
@@ -285,6 +286,8 @@ void fill_pre_wcc(struct svc_fh *fhp)
  */
 void fill_post_wcc(struct svc_fh *fhp)
 {
+	bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE);
+	struct inode *inode = d_inode(fhp->fh_dentry);
 	__be32 err;
 
 	if (fhp->fh_post_saved)
@@ -293,11 +296,12 @@ void fill_post_wcc(struct svc_fh *fhp)
 	err = fh_getattr(fhp, &fhp->fh_post_attr);
 	if (err) {
 		fhp->fh_post_saved = false;
-		fhp->fh_post_attr.ctime = d_inode(fhp->fh_dentry)->i_ctime;
+		fhp->fh_post_attr.ctime = inode->i_ctime;
 	} else
 		fhp->fh_post_saved = true;
-	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
-						     d_inode(fhp->fh_dentry));
+	if (v4)
+		fhp->fh_post_change =
+			nfsd4_change_attribute(&fhp->fh_post_attr, inode);
 }
 
 /*
Signature · 1 line
2.28.0
#4 lore raw

PATCH 5/5 Revert "nfsd4: support change_attr_type attribute"

In reply to J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

This reverts commit a85857633b04d57f4524cca0a2bfaf87b2543f9f.

We're still factoring ctime into our change attribute even in the IS_I_VERSION case. If someone sets the system time backwards, a client could see the change attribute go backwards. Maybe we can just say "well, don't do that", but there's some question whether that's good enough, or whether we need a better guarantee.

Also, the client still isn't actually using the attribute.

While we're still figuring this out, let's just stop returning this attribute.

---
 fs/nfsd/nfs4xdr.c    | 10 ----------
 fs/nfsd/nfsd.h       |  1 -
 include/linux/nfs4.h |  8 --------
 3 files changed, 19 deletions(-)
fs/nfsd/nfs4xdr.c +0 -10
index 18c912930947..c33838caf8c6 100644
@@ -3183,16 +3183,6 @@ nfsd4_encode_fattr(struct xdr_stream *xdr, struct svc_fh *fhp,
 			goto out;
 	}
 
-	if (bmval2 & FATTR4_WORD2_CHANGE_ATTR_TYPE) {
-		p = xdr_reserve_space(xdr, 4);
-		if (!p)
-			goto out_resource;
-		if (IS_I_VERSION(d_inode(dentry)))
-			*p++ = cpu_to_be32(NFS4_CHANGE_TYPE_IS_MONOTONIC_INCR);
-		else
-			*p++ = cpu_to_be32(NFS4_CHANGE_TYPE_IS_TIME_METADATA);
-	}
-
 #ifdef CONFIG_NFSD_V4_SECURITY_LABEL
 	if (bmval2 & FATTR4_WORD2_SECURITY_LABEL) {
 		status = nfsd4_encode_security_label(xdr, rqstp, context,
fs/nfsd/nfsd.h +0 -1
index cb742e17e04a..40cb40ac0a65 100644
@@ -387,7 +387,6 @@ void		nfsd_lockd_shutdown(void);
 
 #define NFSD4_2_SUPPORTED_ATTRS_WORD2 \
 	(NFSD4_1_SUPPORTED_ATTRS_WORD2 | \
-	FATTR4_WORD2_CHANGE_ATTR_TYPE | \
 	FATTR4_WORD2_MODE_UMASK | \
 	NFSD4_2_SECURITY_ATTRS | \
 	FATTR4_WORD2_XATTR_SUPPORT)
include/linux/nfs4.h +0 -8
index 9dc7eeac924f..5b4c67c91f56 100644
@@ -385,13 +385,6 @@ enum lock_type4 {
 	NFS4_WRITEW_LT = 4
 };
 
-enum change_attr_type4 {
-	NFS4_CHANGE_TYPE_IS_MONOTONIC_INCR = 0,
-	NFS4_CHANGE_TYPE_IS_VERSION_COUNTER = 1,
-	NFS4_CHANGE_TYPE_IS_VERSION_COUNTER_NOPNFS = 2,
-	NFS4_CHANGE_TYPE_IS_TIME_METADATA = 3,
-	NFS4_CHANGE_TYPE_IS_UNDEFINED = 4
-};
 
 /* Mandatory Attributes */
 #define FATTR4_WORD0_SUPPORTED_ATTRS    (1UL << 0)
@@ -459,7 +452,6 @@ enum change_attr_type4 {
 #define FATTR4_WORD2_LAYOUT_BLKSIZE     (1UL << 1)
 #define FATTR4_WORD2_MDSTHRESHOLD       (1UL << 4)
 #define FATTR4_WORD2_CLONE_BLKSIZE	(1UL << 13)
-#define FATTR4_WORD2_CHANGE_ATTR_TYPE	(1UL << 15)
 #define FATTR4_WORD2_SECURITY_LABEL     (1UL << 16)
 #define FATTR4_WORD2_MODE_UMASK		(1UL << 17)
 #define FATTR4_WORD2_XATTR_SUPPORT	(1UL << 18)
Signature · 1 line
2.28.0
#5 lore raw
In reply to J. Bruce Fields

This patch fixes the bug Daire saw. The rest is a replacement for the series I sent before, except that I haven't included the last few patches, which will need review from other filesystem people.

--b.

Quoted 95 lines · J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

inode_query_iversion() can modify i_version. Depending on the exported filesystem, that may not be safe. For example, if you're re-exporting NFS, NFS stores the server's change attribute in i_version and does not expect it to be modified locally. This has been observed causing unnecessary cache invalidations.

The way a filesystem indicates that it's OK to call inode_query_iverson() is by setting SB_I_VERSION.

So, move the I_VERSION check out of encode_change(), where it's used only in FATTR responses, to nfsd4_changeattr(), which is also called for pre- and post- operation attributes.

(Note we could also pull the NFSEXP_V4ROOT case into nfsd4_change_attribute as well. That would actually be a no-op, since pre/post attrs are only used for metadata-modifying operations, and V4ROOT exports are read-only. But we might make the change in the future just for simplicity.)

---
 fs/nfsd/nfs3xdr.c |  5 ++---
 fs/nfsd/nfs4xdr.c |  6 +-----
 fs/nfsd/nfsfh.h   | 14 ++++++++++----
 3 files changed, 13 insertions(+), 12 deletions(-)
fs/nfsd/nfs3xdr.c +2 -3
index 2277f83da250..dfbf390ff40c 100644
@@ -291,14 +291,13 @@ void fill_post_wcc(struct svc_fh *fhp)
 		printk("nfsd: inode locked twice during operation.\n");
 
 	err = fh_getattr(fhp, &fhp->fh_post_attr);
-	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
-						     d_inode(fhp->fh_dentry));
 	if (err) {
 		fhp->fh_post_saved = false;
-		/* Grab the ctime anyway - set_change_info might use it */
 		fhp->fh_post_attr.ctime = d_inode(fhp->fh_dentry)->i_ctime;
 	} else
 		fhp->fh_post_saved = true;
+	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
+						     d_inode(fhp->fh_dentry));
 }
 
 /*
fs/nfsd/nfs4xdr.c +1 -5
index 833a2c64dfe8..56fd5f6d5c44 100644
@@ -2298,12 +2298,8 @@ static __be32 *encode_change(__be32 *p, struct kstat *stat, struct inode *inode,
 	if (exp->ex_flags & NFSEXP_V4ROOT) {
 		*p++ = cpu_to_be32(convert_to_wallclock(exp->cd->flush_time));
 		*p++ = 0;
-	} else if (IS_I_VERSION(inode)) {
+	} else
 		p = xdr_encode_hyper(p, nfsd4_change_attribute(stat, inode));
-	} else {
-		*p++ = cpu_to_be32(stat->ctime.tv_sec);
-		*p++ = cpu_to_be32(stat->ctime.tv_nsec);
-	}
 	return p;
 }
 
fs/nfsd/nfsfh.h +10 -4
index 56cfbc361561..3faf5974fa4e 100644
@@ -261,10 +261,16 @@ static inline u64 nfsd4_change_attribute(struct kstat *stat,
 {
 	u64 chattr;
 
-	chattr =  stat->ctime.tv_sec;
-	chattr <<= 30;
-	chattr += stat->ctime.tv_nsec;
-	chattr += inode_query_iversion(inode);
+	if (IS_I_VERSION(inode)) {
+		chattr =  stat->ctime.tv_sec;
+		chattr <<= 30;
+		chattr += stat->ctime.tv_nsec;
+		chattr += inode_query_iversion(inode);
+	} else {
+		chattr = cpu_to_be32(stat->ctime.tv_sec);
+		chattr <<= 32;
+		chattr += cpu_to_be32(stat->ctime.tv_nsec);
+	}
 	return chattr;
 }
 
Signature · 1 line
2.28.0
#6 lore raw
In reply to J. Bruce Fields
Quoted 19 lines · J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

inode_query_iversion() can modify i_version. Depending on the exported filesystem, that may not be safe. For example, if you're re-exporting NFS, NFS stores the server's change attribute in i_version and does not expect it to be modified locally. This has been observed causing unnecessary cache invalidations.

The way a filesystem indicates that it's OK to call inode_query_iverson() is by setting SB_I_VERSION.

So, move the I_VERSION check out of encode_change(), where it's used only in FATTR responses, to nfsd4_changeattr(), which is also called for pre- and post- operation attributes.

Quoted 8 lines

(Note we could also pull the NFSEXP_V4ROOT case into nfsd4_change_attribute as well. That would actually be a no-op, since pre/post attrs are only used for metadata-modifying operations, and V4ROOT exports are read-only. But we might make the change in the future just for simplicity.)

New sparse warnings after this one is applied:

/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: warning: incorrect type in assignment (different base types)
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: expected unsigned long long [assigned] [usertype] chattr
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: got restricted __be32 [usertype]
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: warning: invalid assignment: +=
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: left side has type unsigned long long
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: right side has type restricted __be32
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: warning: incorrect type in assignment (different base types)
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: expected unsigned long long [assigned] [usertype] chattr
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: got restricted __be32 [usertype]
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: warning: invalid assignment: +=
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: left side has type unsigned long long
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: right side has type restricted __be32

Quoted 73 lines

---

fs/nfsd/nfs3xdr.c |  5 ++---
fs/nfsd/nfs4xdr.c |  6 +-----
fs/nfsd/nfsfh.h   | 14 ++++++++++----
3 files changed, 13 insertions(+), 12 deletions(-)
fs/nfsd/nfs3xdr.c +2 -3
index 2277f83da250..dfbf390ff40c 100644
@@ -291,14 +291,13 @@ void fill_post_wcc(struct svc_fh *fhp)
 		printk("nfsd: inode locked twice during operation.\n");
 
 	err = fh_getattr(fhp, &fhp->fh_post_attr);
-	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
-						     d_inode(fhp->fh_dentry));
 	if (err) {
 		fhp->fh_post_saved = false;
-		/* Grab the ctime anyway - set_change_info might use it */
 		fhp->fh_post_attr.ctime = d_inode(fhp->fh_dentry)->i_ctime;
 	} else
 		fhp->fh_post_saved = true;
+	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
+						     d_inode(fhp->fh_dentry));
 }
 
 /*
fs/nfsd/nfs4xdr.c +1 -5
index 833a2c64dfe8..56fd5f6d5c44 100644
@@ -2298,12 +2298,8 @@ static __be32 *encode_change(__be32 *p, struct kstat *stat, struct inode *inode,
 	if (exp->ex_flags & NFSEXP_V4ROOT) {
 		*p++ = cpu_to_be32(convert_to_wallclock(exp->cd->flush_time));
 		*p++ = 0;
-	} else if (IS_I_VERSION(inode)) {
+	} else
 		p = xdr_encode_hyper(p, nfsd4_change_attribute(stat, inode));
-	} else {
-		*p++ = cpu_to_be32(stat->ctime.tv_sec);
-		*p++ = cpu_to_be32(stat->ctime.tv_nsec);
-	}
 	return p;
 }
 
fs/nfsd/nfsfh.h +10 -4
index 56cfbc361561..3faf5974fa4e 100644
@@ -261,10 +261,16 @@ static inline u64 nfsd4_change_attribute(struct kstat *stat,
 {
 	u64 chattr;
 
-	chattr =  stat->ctime.tv_sec;
-	chattr <<= 30;
-	chattr += stat->ctime.tv_nsec;
-	chattr += inode_query_iversion(inode);
+	if (IS_I_VERSION(inode)) {
+		chattr =  stat->ctime.tv_sec;
+		chattr <<= 30;
+		chattr += stat->ctime.tv_nsec;
+		chattr += inode_query_iversion(inode);
+	} else {
+		chattr = cpu_to_be32(stat->ctime.tv_sec);
+		chattr <<= 32;
+		chattr += cpu_to_be32(stat->ctime.tv_nsec);
+	}
 	return chattr;
 }
 
Signature · 1 line
2.28.0

--
Chuck Lever

#7 lore raw
Quoted 25 lines · Chuck Lever
Quoted 25 lines · J. Bruce Fields

From: "J. Bruce Fields" <[email protected]>

inode_query_iversion() can modify i_version. Depending on the exported filesystem, that may not be safe. For example, if you're re-exporting NFS, NFS stores the server's change attribute in i_version and does not expect it to be modified locally. This has been observed causing unnecessary cache invalidations.

The way a filesystem indicates that it's OK to call inode_query_iverson() is by setting SB_I_VERSION.

So, move the I_VERSION check out of encode_change(), where it's used only in FATTR responses, to nfsd4_changeattr(), which is also called for pre- and post- operation attributes.

(Note we could also pull the NFSEXP_V4ROOT case into nfsd4_change_attribute as well. That would actually be a no-op, since pre/post attrs are only used for metadata-modifying operations, and V4ROOT exports are read-only. But we might make the change in the future just for simplicity.)

Quoted 4 lines
Quoted 2 lines

New sparse warnings after this one is applied:

D'oh. I'll send again, thanks!

--b.

Quoted 93 lines

/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: warning: incorrect type in assignment (different base types)
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: expected unsigned long long [assigned] [usertype] chattr
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: got restricted __be32 [usertype]
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: warning: invalid assignment: +=
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: left side has type unsigned long long
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: right side has type restricted __be32
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: warning: incorrect type in assignment (different base types)
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: expected unsigned long long [assigned] [usertype] chattr
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:270:24: got restricted __be32 [usertype]
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: warning: invalid assignment: +=
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: left side has type unsigned long long
/home/cel/src/linux/linux/fs/nfsd/nfsfh.h:272:24: right side has type restricted __be32

Quoted 73 lines

---

fs/nfsd/nfs3xdr.c |  5 ++---
fs/nfsd/nfs4xdr.c |  6 +-----
fs/nfsd/nfsfh.h   | 14 ++++++++++----
3 files changed, 13 insertions(+), 12 deletions(-)
fs/nfsd/nfs3xdr.c +2 -3
index 2277f83da250..dfbf390ff40c 100644
@@ -291,14 +291,13 @@ void fill_post_wcc(struct svc_fh *fhp)
 		printk("nfsd: inode locked twice during operation.\n");
 
 	err = fh_getattr(fhp, &fhp->fh_post_attr);
-	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
-						     d_inode(fhp->fh_dentry));
 	if (err) {
 		fhp->fh_post_saved = false;
-		/* Grab the ctime anyway - set_change_info might use it */
 		fhp->fh_post_attr.ctime = d_inode(fhp->fh_dentry)->i_ctime;
 	} else
 		fhp->fh_post_saved = true;
+	fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr,
+						     d_inode(fhp->fh_dentry));
 }
 
 /*
fs/nfsd/nfs4xdr.c +1 -5
index 833a2c64dfe8..56fd5f6d5c44 100644
@@ -2298,12 +2298,8 @@ static __be32 *encode_change(__be32 *p, struct kstat *stat, struct inode *inode,
 	if (exp->ex_flags & NFSEXP_V4ROOT) {
 		*p++ = cpu_to_be32(convert_to_wallclock(exp->cd->flush_time));
 		*p++ = 0;
-	} else if (IS_I_VERSION(inode)) {
+	} else
 		p = xdr_encode_hyper(p, nfsd4_change_attribute(stat, inode));
-	} else {
-		*p++ = cpu_to_be32(stat->ctime.tv_sec);
-		*p++ = cpu_to_be32(stat->ctime.tv_nsec);
-	}
 	return p;
 }
 
fs/nfsd/nfsfh.h +10 -4
index 56cfbc361561..3faf5974fa4e 100644
@@ -261,10 +261,16 @@ static inline u64 nfsd4_change_attribute(struct kstat *stat,
 {
 	u64 chattr;
 
-	chattr =  stat->ctime.tv_sec;
-	chattr <<= 30;
-	chattr += stat->ctime.tv_nsec;
-	chattr += inode_query_iversion(inode);
+	if (IS_I_VERSION(inode)) {
+		chattr =  stat->ctime.tv_sec;
+		chattr <<= 30;
+		chattr += stat->ctime.tv_nsec;
+		chattr += inode_query_iversion(inode);
+	} else {
+		chattr = cpu_to_be32(stat->ctime.tv_sec);
+		chattr <<= 32;
+		chattr += cpu_to_be32(stat->ctime.tv_nsec);
+	}
 	return chattr;
 }
 
Signature · 1 line
2.28.0

--
Chuck Lever