Skip to contentSkip to conversation

nfsd: don't get write access twіce in nfsd_setattr

PATCH In review Christoph Hellwig 2 participants 7 messages Feb 7, 2017 – Feb 9, 2017

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

PATCH 1/1 nfsd: don't get write access twіce in nfsd_setattr

Turns out doing mnt_want_write twice for the same process makes lockdep unhappy, so move the fh_want_write down to after calling vfs_truncate in nfsd_setattr. No changes to error handling required as the want write state is automatically cleaned up by the caller based on a flag in the svc_fh.

Fixes 41f53350 ("nfsd: special case truncates some more")
fs/nfsd/vfs.c +6 -5
index ca13236dbb1f..a974368026a1 100644
@@ -359,11 +359,6 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 	err = fh_verify(rqstp, fhp, ftype, accmode);
 	if (err)
 		return err;
-	if (get_write_count) {
-		host_err = fh_want_write(fhp);
-		if (host_err)
-			goto out_host_err;
-	}
 
 	dentry = fhp->fh_dentry;
 	inode = d_inode(dentry);
@@ -416,6 +411,12 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 
 	iap->ia_valid |= ATTR_CTIME;
 
+	if (get_write_count) {
+		host_err = fh_want_write(fhp);
+		if (host_err)
+			goto out_host_err;
+	}
+
 	fh_lock(fhp);
 	host_err = notify_change(dentry, iap, NULL);
 	fh_unlock(fhp);
#1 lore raw
Quoted 9 lines · Christoph Hellwig

Turns out doing mnt_want_write twice for the same process makes lockdep unhappy, so move the fh_want_write down to after calling vfs_truncate in nfsd_setattr. No changes to error handling required as the want write state is automatically cleaned up by the caller based on a flag in the svc_fh.

Fixes 41f53350 ("nfsd: special case truncates some more")

Thanks, I'll see if I can squeeze this into 4.10.--b.

Quoted 30 lines
fs/nfsd/vfs.c +6 -5
index ca13236dbb1f..a974368026a1 100644
@@ -359,11 +359,6 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 	err = fh_verify(rqstp, fhp, ftype, accmode);
 	if (err)
 		return err;
-	if (get_write_count) {
-		host_err = fh_want_write(fhp);
-		if (host_err)
-			goto out_host_err;
-	}
 
 	dentry = fhp->fh_dentry;
 	inode = d_inode(dentry);
@@ -416,6 +411,12 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 
 	iap->ia_valid |= ATTR_CTIME;
 
+	if (get_write_count) {
+		host_err = fh_want_write(fhp);
+		if (host_err)
+			goto out_host_err;
+	}
+
 	fh_lock(fhp);
 	host_err = notify_change(dentry, iap, NULL);
 	fh_unlock(fhp);
#2 lore raw
Quoted 7 lines · J. Bruce Fields
Quoted 6 lines · Christoph Hellwig

Turns out doing mnt_want_write twice for the same process makes lockdep unhappy, so move the fh_want_write down to after calling vfs_truncate in nfsd_setattr. No changes to error handling required as the want write state is automatically cleaned up by the caller based on a flag in the svc_fh.

Quoted 5 lines
Quoted 3 lines
Fixes 41f53350 ("nfsd: special case truncates some more")

Thanks, I'll see if I can squeeze this into 4.10.--b.

FYI, Chuck just reported another issue, let's wait for that for now.

#3 lore raw
Quoted 8 lines · Christoph Hellwig
Quoted 7 lines · J. Bruce Fields
Quoted 6 lines · Christoph Hellwig

Turns out doing mnt_want_write twice for the same process makes lockdep unhappy, so move the fh_want_write down to after calling vfs_truncate in nfsd_setattr. No changes to error handling required as the want write state is automatically cleaned up by the caller based on a flag in the svc_fh.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
Fixes 41f53350 ("nfsd: special case truncates some more")

Thanks, I'll see if I can squeeze this into 4.10.--b.

FYI, Chuck just reported another issue, let's wait for that for now.

OK.--b.

#4 lore raw
Quoted 1 line · J. Bruce Fields
Quoted 1 line

FYI, Chuck just reported another issue, let's wait for that for now.

I can't really reproduce the issue Chuck reported. So let's get this fixup in for now, especially as Linus replied to the thread it was reported in and asked for it.

#5 lore raw
Quoted 6 lines · Christoph Hellwig
Quoted 1 line · J. Bruce Fields
Quoted 1 line

FYI, Chuck just reported another issue, let's wait for that for now.

I can't really reproduce the issue Chuck reported. So let's get this fixup in for now, especially as Linus replied to the thread it was reported in and asked for it.

I can't find that reply--am I missing a thread?

OK, so for 4.10 I have:

nfsd: don't get write access twice in nfsd_setattr
nfsd: restore owner override for truncate

--b.

#6 lore raw
Quoted 9 lines · J. Bruce Fields
Quoted 6 lines · Christoph Hellwig
Quoted 1 line · J. Bruce Fields
Quoted 1 line

FYI, Chuck just reported another issue, let's wait for that for now.

I can't really reproduce the issue Chuck reported. So let's get this fixup in for now, especially as Linus replied to the thread it was reported in and asked for it.

I can't find that reply--am I missing a thread?

Dave reported it in reply to Linus' 4.10-rc7 announcement. Looks like neither you nor linux-nfs were Cc'ed on that.

Quoted 4 lines

OK, so for 4.10 I have:

nfsd: don't get write access twice in nfsd_setattr
nfsd: restore owner override for truncate

Yepp. Sorry for the mess.