ocfs2: Add readonly check in ocfs2_quota_on().

PATCH New Tristan Ye 3 participants 6 messages Jun 1, 2011 – Jun 3, 2011 OCFS2

Share & tools

Tristan Ye

author

2011-06-01 08:19 UTC #0 lore raw

PATCH 1/1 ocfs2: Add readonly check in ocfs2_quota_on().

Enabling quota on a readonly ocfs2 fs caused a 'NULL pointer dereference' in dquot_enable(), a very straightforward fix is to check readonly at the very beginning of ocfs2_quota_on(), which calls dquot_enable().

Signed-off-by Tristan Ye <[email protected]>

---
 fs/ocfs2/super.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

fs/ocfs2/super.c +4 -0

index 029c4cd..3cade2a 100644
@@ -995,6 +995,10 @@ static int ocfs2_quota_on(struct super_block *sb, int type, int format_id)
 	if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
 		return -EINVAL;
 
+	if (ocfs2_is_hard_readonly(OCFS2_SB(sb))) {
+		return -EROFS;
+	}
+
 	return dquot_enable(sb_dqopt(sb)->files[type], type,
 			    format_id, DQUOT_LIMITS_ENABLED);
 }

Signature · 1 line

1.5.5

Sunil Mushran

2011-06-01 18:25 UTC #1 lore raw

PATCH 1/1 ocfs2: Add readonly check in ocfs2_quota_on().

looks good.

Quoted 24 lines · Tristan Ye

Enabling quota on a readonly ocfs2 fs caused a 'NULL pointer dereference' in dquot_enable(), a very straightforward fix is to check readonly at the very beginning of ocfs2_quota_on(), which calls dquot_enable().
Signed-off-by Tristan Ye<[email protected]>
---
  fs/ocfs2/super.c |    4 ++++
  1 files changed, 4 insertions(+), 0 deletions(-)
fs/ocfs2/super.c +4 -0
index 029c4cd..3cade2a 100644
@@ -995,6 +995,10 @@ static int ocfs2_quota_on(struct super_block *sb, int type, int format_id)
  	if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
  		return -EINVAL;
 
+	if (ocfs2_is_hard_readonly(OCFS2_SB(sb))) {
+		return -EROFS;
+	}
+
  	return dquot_enable(sb_dqopt(sb)->files[type], type,
  			    format_id, DQUOT_LIMITS_ENABLED);
  }

Jan Kara

2011-06-02 12:44 UTC #2 lore raw

PATCH 1/1 ocfs2: Add readonly check in ocfs2_quota_on().

Quoted 1 line · Sunil Mushran

looks good.

Yes, looks OK to me as well. Only I'd be interested what NULL pointer
dereference happened because I don't see how it could. Tristan, do you have the trace?

Honza

Quoted 26 lines · Tristan Ye

Quoted 24 lines
Enabling quota on a readonly ocfs2 fs caused a 'NULL pointer dereference' in dquot_enable(), a very straightforward fix is to check readonly at the very beginning of ocfs2_quota_on(), which calls dquot_enable().
Signed-off-by Tristan Ye<[email protected]>
---
 fs/ocfs2/super.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)
fs/ocfs2/super.c +4 -0
index 029c4cd..3cade2a 100644
@@ -995,6 +995,10 @@ static int ocfs2_quota_on(struct super_block *sb, int type, int format_id)
 	if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
 		return -EINVAL;
 
+	if (ocfs2_is_hard_readonly(OCFS2_SB(sb))) {
+		return -EROFS;
+	}
+
 	return dquot_enable(sb_dqopt(sb)->files[type], type,
 			    format_id, DQUOT_LIMITS_ENABLED);
 }

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Tristan Ye

author

2011-06-02 16:21 UTC #3 lore raw

PATCH 1/1 ocfs2: Add readonly check in ocfs2_quota_on().

Quoted 5 lines · Jan Kara

Quoted 1 line · Sunil Mushran
looks good.
Yes, looks OK to me as well. Only I'd be interested what NULL pointer
dereference happened because I don't see how it could. Tristan, do you have the trace?

Sure,
StackTrace attached.
Tristan

Quoted 29 lines

Honza
Quoted 26 lines · Tristan Ye
Quoted 24 lines
Enabling quota on a readonly ocfs2 fs caused a 'NULL pointer dereference' in dquot_enable(), a very straightforward fix is to check readonly at the very beginning of ocfs2_quota_on(), which calls dquot_enable().
Signed-off-by Tristan Ye<[email protected]>
---
 fs/ocfs2/super.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)
fs/ocfs2/super.c +4 -0
index 029c4cd..3cade2a 100644
@@ -995,6 +995,10 @@ static int ocfs2_quota_on(struct super_block *sb, int type, int format_id)
 	if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
 		return -EINVAL;
 
+	if (ocfs2_is_hard_readonly(OCFS2_SB(sb))) {
+		return -EROFS;
+	}
+
 	return dquot_enable(sb_dqopt(sb)->files[type], type,
 			    format_id, DQUOT_LIMITS_ENABLED);
 }

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: dquot_enable_oops
Url: http://oss.oracle.com/pipermail/ocfs2-devel/attachments/20110603/c97455ef/attachment.pl

Jan Kara

2011-06-02 16:50 UTC #4 lore raw

PATCH 1/1 ocfs2: Add readonly check in ocfs2_quota_on().

Quoted 9 lines · Tristan Ye

Quoted 5 lines · Jan Kara
Quoted 1 line · Sunil Mushran
looks good.
Yes, looks OK to me as well. Only I'd be interested what NULL pointer
dereference happened because I don't see how it could. Tristan, do you have the trace?
Sure,
StackTrace attached.

Ah, right, I see now... Thanks for info.

Honza

Quoted 66 lines

Quoted 18 lines

Quoted 18 lines · Tristan Ye
Quoted 17 lines
Enabling quota on a readonly ocfs2 fs caused a 'NULL pointer dereference' in dquot_enable(), a very straightforward fix is to check readonly at the very beginning of ocfs2_quota_on(), which calls dquot_enable().
Signed-off-by Tristan Ye<[email protected]>
---
 fs/ocfs2/super.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)
fs/ocfs2/super.c +0 -0
index 029c4cd..3cade2a 100644
@@ -995,6 +995,10 @@ static int ocfs2_quota_on(struct super_block *sb, int type, int format_id)
 	if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
 		return -EINVAL;
 

Quoted 8 lines

Quoted 8 lines
fs/ocfs2/super.c static int ocfs2_quota_on(struct super_block *sb, int type, int format_id) · line 1004
Quoted 7 lines
fs/ocfs2/super.c +4 -0
+	if (ocfs2_is_hard_readonly(OCFS2_SB(sb))) {
+		return -EROFS;
+	}
+
 	return dquot_enable(sb_dqopt(sb)->files[type], type,
 			    format_id, DQUOT_LIMITS_ENABLED);
 }

Jun  1 12:38:08 ocfs2-box4 kernel: BUG: unable to handle kernel NULL pointer dereference at 0000000000000018
Jun  1 12:38:08 ocfs2-box4 kernel: IP: [<ffffffff8110dc88>] dquot_enable+0x19/0xf4
Jun  1 12:38:08 ocfs2-box4 kernel: PGD 0
Jun  1 12:38:08 ocfs2-box4 kernel: Oops: 0000 [#1] SMP
Jun  1 12:38:08 ocfs2-box4 kernel: CPU 1
Jun  1 12:38:08 ocfs2-box4 kernel: Modules linked in: ocfs2 ocfs2_dlmfs ocfs2_stack_o2cb ocfs2_dlm ocfs2_nodemanager ocfs2_stackglue configfs i915 drm_kms_helper drm i2c_algo_bit video deflate zlib_deflate ctr camellia cast5 rmd160 crypto_null netconsole ccm serpent blowfish twofish_generic twofish_x86_64 twofish_common ecb xcbc cbc md5 sha256_generic sha512_generic des_generic aes_x86_64 aes_generic ah6 ah4 esp6 esp4 xfrm4_tunnel tunnel4 xfrm4_mode_tunnel xfrm4_mode_transport xfrm6_mode_transport xfrm6_mode_beet xfrm6_mode_tunnel ipcomp ipcomp6 xfrm_ipcomp xfrm6_tunnel tunnel6 af_key autofs4 sunrpc p4_clockmod speedstep_lib ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr ipv6 iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi dm_mirror dm_region_hash dm_log dm_mod uinput snd_intel8x0 snd_ac97_codec ac97_bus snd_seq snd_seq_device snd_pcm ppdev snd_timer i2c_i801 iTCO_wdt ehci_hcd tg3 snd parport_pc parport libphy sg i2c_core soundcore snd_page_alloc uhci_hcd pcspkr serio_raw dcdbas ext4 jbd2 crc1
6 sd_mod
Jun  1 12:38:08 ocfs2-box4 kernel: pata_acpi ata_generic ata_piix libata scsi_mod ide_cd_mod cdrom button [last unloaded: scsi_wait_scan]
Jun  1 12:38:08 ocfs2-box4 kernel:
Jun  1 12:38:08 ocfs2-box4 kernel: Pid: 3911, comm: quotaon Not tainted 2.6.39+ #3 Dell Inc.                 OptiPlex GX620               /0FH884
Jun  1 12:38:08 ocfs2-box4 kernel: RIP: 0010:[<ffffffff8110dc88>]  [<ffffffff8110dc88>] dquot_enable+0x19/0xf4
Jun  1 12:38:08 ocfs2-box4 kernel: RSP: 0018:ffff88007b971e18  EFLAGS: 00010246
Jun  1 12:38:08 ocfs2-box4 kernel: RAX: 0000000000000001 RBX: 0000000000000000 RCX: 0000000000000002
Jun  1 12:38:08 ocfs2-box4 kernel: RDX: 0000000000000002 RSI: 0000000000000001 RDI: 0000000000000000
Jun  1 12:38:08 ocfs2-box4 kernel: RBP: 0000000000000001 R08: 0000000000000004 R09: ffffffff817254e0
Jun  1 12:38:08 ocfs2-box4 kernel: R10: 0000000001c044d0 R11: ffff88007b971d34 R12: 0000000000000002
Jun  1 12:38:08 ocfs2-box4 kernel: R13: 0000000000000002 R14: 00007f28de3bc24e R15: ffff88007b971f28
Jun  1 12:38:08 ocfs2-box4 kernel: FS:  00007f28de38b700(0000) GS:ffff88007f480000(0000) knlGS:0000000000000000
Jun  1 12:38:08 ocfs2-box4 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jun  1 12:38:08 ocfs2-box4 kernel: CR2: 0000000000000018 CR3: 0000000078f5f000 CR4: 00000000000006e0
Jun  1 12:38:08 ocfs2-box4 kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Jun  1 12:38:08 ocfs2-box4 kernel: DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Jun  1 12:38:08 ocfs2-box4 kernel: Process quotaon (pid: 3911, threadinfo ffff88007b970000, task ffff8800794a6100)
Jun  1 12:38:08 ocfs2-box4 kernel: Stack:
Jun  1 12:38:08 ocfs2-box4 kernel: ffff88007b971f28 0000000000000001 00007f28de3bc24e ffff880078c39400
Jun  1 12:38:08 ocfs2-box4 kernel: 0000000000000001 0000000000800002 0000000000000002 00007f28de3bc24e
Jun  1 12:38:08 ocfs2-box4 kernel: ffff88007b971f28 ffffffff811103ac ffff880078f7c0e0 0000000400000001
Jun  1 12:38:08 ocfs2-box4 kernel: Call Trace:
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff811103ac>] ? do_quotactl+0x1ae/0x42a
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff810de0eb>] ? dput+0x27/0x152
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff81110753>] ? sys_quotactl+0x12b/0x160
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff8132097b>] ? system_call_fastpath+0x16/0x1b
Jun  1 12:38:08 ocfs2-box4 kernel: Code: 04 fd ff 8b 44 24 08 48 83 c4 18 5b 5d 41 5c 41 5d c3 41 57 41 56 41 55 41 54 41 89 cc 55 89 f5 53 48 89 fb 48 83 ec 18 f6 c1 04 <4c> 8b 6f 18 74 04 0f 0b eb fe 45 31 f6 85 c9 0f 84 b4 00 00 00
Jun  1 12:38:08 ocfs2-box4 kernel: RIP  [<ffffffff8110dc88>] dquot_enable+0x19/0xf4
Jun  1 12:38:08 ocfs2-box4 kernel: RSP <ffff88007b971e18>
Jun  1 12:38:08 ocfs2-box4 kernel: CR2: 0000000000000018
Jun  1 12:38:08 ocfs2-box4 kernel: ---[ end trace 317402b4fb3b74e7 ]---

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Tristan Ye

author

2011-06-03 01:05 UTC #5 lore raw

PATCH 1/1 ocfs2: Add readonly check in ocfs2_quota_on().

Quoted 11 lines · Jan Kara

Quoted 9 lines · Tristan Ye
Quoted 5 lines · Jan Kara
Quoted 1 line · Sunil Mushran
looks good.
Yes, looks OK to me as well. Only I'd be interested what NULL pointer
dereference happened because I don't see how it could. Tristan, do you have the trace?
Sure,
StackTrace attached.
Ah, right, I see now... Thanks for info.

And I believe another patch is being on its way now;-)

Quoted 37 lines

Honza

Quoted 34 lines

Quoted 18 lines

Quoted 18 lines · Tristan Ye
Quoted 17 lines
Enabling quota on a readonly ocfs2 fs caused a 'NULL pointer dereference' in dquot_enable(), a very straightforward fix is to check readonly at the very beginning of ocfs2_quota_on(), which calls dquot_enable().
Signed-off-by Tristan Ye<[email protected]>
---
 fs/ocfs2/super.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)
fs/ocfs2/super.c +0 -0
index 029c4cd..3cade2a 100644
@@ -995,6 +995,10 @@ static int ocfs2_quota_on(struct super_block *sb, int type, int format_id)
 	if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
 		return -EINVAL;
 

Quoted 8 lines

Quoted 8 lines
fs/ocfs2/super.c static int ocfs2_quota_on(struct super_block *sb, int type, int format_id) · line 1004
Quoted 7 lines
fs/ocfs2/super.c +4 -0
+	if (ocfs2_is_hard_readonly(OCFS2_SB(sb))) {
+		return -EROFS;
+	}
+
 	return dquot_enable(sb_dqopt(sb)->files[type], type,
 			    format_id, DQUOT_LIMITS_ENABLED);
 }

Jun  1 12:38:08 ocfs2-box4 kernel: BUG: unable to handle kernel NULL pointer dereference at 0000000000000018
Jun  1 12:38:08 ocfs2-box4 kernel: IP: [<ffffffff8110dc88>] dquot_enable+0x19/0xf4
Jun  1 12:38:08 ocfs2-box4 kernel: PGD 0
Jun  1 12:38:08 ocfs2-box4 kernel: Oops: 0000 [#1] SMP
Jun  1 12:38:08 ocfs2-box4 kernel: CPU 1
Jun  1 12:38:08 ocfs2-box4 kernel: Modules linked in: ocfs2 ocfs2_dlmfs ocfs2_stack_o2cb ocfs2_dlm ocfs2_nodemanager ocfs2_stackglue configfs i915 drm_kms_helper drm i2c_algo_bit video deflate zlib_deflate ctr camellia cast5 rmd160 crypto_null netconsole ccm serpent blowfish twofish_generic twofish_x86_64 twofish_common ecb xcbc cbc md5 sha256_generic sha512_generic des_generic aes_x86_64 aes_generic ah6 ah4 esp6 esp4 xfrm4_tunnel tunnel4 xfrm4_mode_tunnel xfrm4_mode_transport xfrm6_mode_transport xfrm6_mode_beet xfrm6_mode_tunnel ipcomp ipcomp6 xfrm_ipcomp xfrm6_tunnel tunnel6 af_key autofs4 sunrpc p4_clockmod speedstep_lib ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr ipv6 iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi dm_mirror dm_region_hash dm_log dm_mod uinput snd_intel8x0 snd_ac97_codec ac97_bus snd_seq snd_seq_device snd_pcm ppdev snd_timer i2c_i801 iTCO_wdt ehci_hcd tg3 snd parport_pc parport libphy sg i2c_core soundcore snd_page_alloc uhci_hcd pc

spkr serio_raw dcdbas ext4 jbd2 crc1

Quoted 33 lines

Quoted 32 lines

6 sd_mod

Jun  1 12:38:08 ocfs2-box4 kernel: pata_acpi ata_generic ata_piix libata scsi_mod ide_cd_mod cdrom button [last unloaded: scsi_wait_scan]
Jun  1 12:38:08 ocfs2-box4 kernel:
Jun  1 12:38:08 ocfs2-box4 kernel: Pid: 3911, comm: quotaon Not tainted 2.6.39+ #3 Dell Inc.                 OptiPlex GX620               /0FH884
Jun  1 12:38:08 ocfs2-box4 kernel: RIP: 0010:[<ffffffff8110dc88>]  [<ffffffff8110dc88>] dquot_enable+0x19/0xf4
Jun  1 12:38:08 ocfs2-box4 kernel: RSP: 0018:ffff88007b971e18  EFLAGS: 00010246
Jun  1 12:38:08 ocfs2-box4 kernel: RAX: 0000000000000001 RBX: 0000000000000000 RCX: 0000000000000002
Jun  1 12:38:08 ocfs2-box4 kernel: RDX: 0000000000000002 RSI: 0000000000000001 RDI: 0000000000000000
Jun  1 12:38:08 ocfs2-box4 kernel: RBP: 0000000000000001 R08: 0000000000000004 R09: ffffffff817254e0
Jun  1 12:38:08 ocfs2-box4 kernel: R10: 0000000001c044d0 R11: ffff88007b971d34 R12: 0000000000000002
Jun  1 12:38:08 ocfs2-box4 kernel: R13: 0000000000000002 R14: 00007f28de3bc24e R15: ffff88007b971f28
Jun  1 12:38:08 ocfs2-box4 kernel: FS:  00007f28de38b700(0000) GS:ffff88007f480000(0000) knlGS:0000000000000000
Jun  1 12:38:08 ocfs2-box4 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jun  1 12:38:08 ocfs2-box4 kernel: CR2: 0000000000000018 CR3: 0000000078f5f000 CR4: 00000000000006e0
Jun  1 12:38:08 ocfs2-box4 kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Jun  1 12:38:08 ocfs2-box4 kernel: DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Jun  1 12:38:08 ocfs2-box4 kernel: Process quotaon (pid: 3911, threadinfo ffff88007b970000, task ffff8800794a6100)
Jun  1 12:38:08 ocfs2-box4 kernel: Stack:
Jun  1 12:38:08 ocfs2-box4 kernel: ffff88007b971f28 0000000000000001 00007f28de3bc24e ffff880078c39400
Jun  1 12:38:08 ocfs2-box4 kernel: 0000000000000001 0000000000800002 0000000000000002 00007f28de3bc24e
Jun  1 12:38:08 ocfs2-box4 kernel: ffff88007b971f28 ffffffff811103ac ffff880078f7c0e0 0000000400000001
Jun  1 12:38:08 ocfs2-box4 kernel: Call Trace:
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff811103ac>] ? do_quotactl+0x1ae/0x42a
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff810de0eb>] ? dput+0x27/0x152
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff81110753>] ? sys_quotactl+0x12b/0x160
Jun  1 12:38:08 ocfs2-box4 kernel: [<ffffffff8132097b>] ? system_call_fastpath+0x16/0x1b
Jun  1 12:38:08 ocfs2-box4 kernel: Code: 04 fd ff 8b 44 24 08 48 83 c4 18 5b 5d 41 5c 41 5d c3 41 57 41 56 41 55 41 54 41 89 cc 55 89 f5 53 48 89 fb 48 83 ec 18 f6 c1 04 <4c> 8b 6f 18 74 04 0f 0b eb fe 45 31 f6 85 c9 0f 84 b4 00 00 00
Jun  1 12:38:08 ocfs2-box4 kernel: RIP  [<ffffffff8110dc88>] dquot_enable+0x19/0xf4
Jun  1 12:38:08 ocfs2-box4 kernel: RSP <ffff88007b971e18>
Jun  1 12:38:08 ocfs2-box4 kernel: CR2: 0000000000000018
Jun  1 12:38:08 ocfs2-box4 kernel: ---[ end trace 317402b4fb3b74e7 ]---