[patch 07/29] knfsd: fix spurious EINVAL errors on first access of new filesystem

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Greg Kroah-Hartman
Date: Tuesday, November 20, 2007 - 11:23 am

2.6.23-stable review patch.  If anyone has any objections, please let us
know.

------------------
From: J. Bruce Fields <bfields@citi.umich.edu>

patch ac8587dcb58e40dd336d99d60f852041e06cc3dd in mainline.

The v2/v3 acl code in nfsd is translating any return from fh_verify() to
nfserr_inval.  This is particularly unfortunate in the case of an
nfserr_dropit return, which is an internal error meant to indicate to
callers that this request has been deferred and should just be dropped
pending the results of an upcall to mountd.

Thanks to Roland <devzero@web.de> for bug report and data collection.

Cc: Roland <devzero@web.de>
Acked-by: Andreas Gruenbacher <agruen@suse.de>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Reviewed-By: NeilBrown <neilb@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/nfsd/nfs2acl.c |    2 +-
 fs/nfsd/nfs3acl.c |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

--- a/fs/nfsd/nfs2acl.c
+++ b/fs/nfsd/nfs2acl.c
@@ -41,7 +41,7 @@ static __be32 nfsacld_proc_getacl(struct
 
 	fh = fh_copy(&resp->fh, &argp->fh);
 	if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP)))
-		RETURN_STATUS(nfserr_inval);
+		RETURN_STATUS(nfserr);
 
 	if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT))
 		RETURN_STATUS(nfserr_inval);
--- a/fs/nfsd/nfs3acl.c
+++ b/fs/nfsd/nfs3acl.c
@@ -37,7 +37,7 @@ static __be32 nfsd3_proc_getacl(struct s
 
 	fh = fh_copy(&resp->fh, &argp->fh);
 	if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP)))
-		RETURN_STATUS(nfserr_inval);
+		RETURN_STATUS(nfserr);
 
 	if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT))
 		RETURN_STATUS(nfserr_inval);

-- 
-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[patch 00/29] 2.6.23-stable review, Greg Kroah-Hartman, (Tue Nov 20, 11:22 am)
[patch 01/29] i2c-pasemi: Fix NACK detection, Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 02/29] i2c/eeprom: Recognize VGN as a valid Sony Va ..., Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 03/29] i2c/eeprom: Hide Sony Vaio serial numbers, Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 04/29] drivers/video/ps3fb: fix memset size error, Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 06/29] raid5: fix unending write sequence, Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 07/29] knfsd: fix spurious EINVAL errors on first a ..., Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 08/29] nfsd4: recheck for secure ports in fh_verify, Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 09/29] dmaengine: fix broken device refcounting, Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 10/29] x86: disable preemption in delay_tsc(), Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 11/29] reiserfs: dont drop PG_dirty when releasing ..., Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 13/29] libata: sata_sis: use correct S/G table size, Greg Kroah-Hartman, (Tue Nov 20, 11:23 am)
[patch 12/29] sata_sis: fix SCR read breakage, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 15/29] Fix divide-by-zero in the 2.6.23 scheduler code, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 16/29] geode: Fix not inplace encryption, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 17/29] libcrc32c: keep intermediate crc state in cp ..., Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 18/29] i386: avoid temporarily inconsistent pte-s, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 19/29] x86: fix off-by-one in find_next_zero_string, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 20/29] x86: mark read_crX() asm code as volatile, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 22/29] x86: return correct error code from child_ri ..., Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 23/29] ntp: fix typo that makes sync_cmos_clock erratic, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 24/29] x86: fix freeze in x86_64 RTC update code in ..., Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 25/29] softlockup watchdog fixes and cleanups, Greg Kroah-Hartman, (Tue Nov 20, 11:24 am)
[patch 26/29] softlockup: use cpu_clock() instead of sched ..., Greg Kroah-Hartman, (Tue Nov 20, 11:25 am)
[patch 27/29] USB: unusual_devs modification for Nikon D200, Greg Kroah-Hartman, (Tue Nov 20, 11:25 am)
[patch 28/29] USB: Nikon D40X unusual_devs entry, Greg Kroah-Hartman, (Tue Nov 20, 11:25 am)
Re: [patch 00/29] 2.6.23-stable review, Greg Kroah-Hartman, (Tue Nov 20, 11:29 am)