[PATCH AUTOSEL 5.3 12/71] ceph: fix directories inode i_blkbits initialization

[PATCH AUTOSEL 5.3 12/71] ceph: fix directories inode i_blkbits initialization

[Sasha Levin]

From: Luis Henriques <lhenriques@suse.com>

[ Upstream commit 750670341a24cb714e624e0fd7da30900ad93752 ]

When filling an inode with info from the MDS, i_blkbits is being
initialized using fl_stripe_unit, which contains the stripe unit in
bytes.  Unfortunately, this doesn't make sense for directories as they
have fl_stripe_unit set to '0'.  This means that i_blkbits will be set
to 0xff, causing an UBSAN undefined behaviour in i_blocksize():

  UBSAN: Undefined behaviour in ./include/linux/fs.h:731:12
  shift exponent 255 is too large for 32-bit type 'int'

Fix this by initializing i_blkbits to CEPH_BLOCK_SHIFT if fl_stripe_unit
is zero.

Signed-off-by: Luis Henriques <lhenriques@suse.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/ceph/inode.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 18500edefc56f..3b537e7038c7a 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -801,7 +801,12 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 
 	/* update inode */
 	inode->i_rdev = le32_to_cpu(info->rdev);
-	inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1;
+	/* directories have fl_stripe_unit set to zero */
+	if (le32_to_cpu(info->layout.fl_stripe_unit))
+		inode->i_blkbits =
+			fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1;
+	else
+		inode->i_blkbits = CEPH_BLOCK_SHIFT;
 
 	__ceph_update_quota(ci, iinfo->max_bytes, iinfo->max_files);
 
-- 
2.20.1

[PATCH AUTOSEL 5.3 13/71] ceph: fetch cap_gen under spinlock in ceph_add_cap

[Sasha Levin]

From: Jeff Layton <jlayton@kernel.org>

[ Upstream commit 606d102327a45a49d293557527802ee7fbfd7af1 ]

It's protected by the s_gen_ttl_lock, so we should fetch under it
and ensure that we're using the same generation in both places.

Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: "Yan, Zheng" <zyan@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/ceph/caps.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index ce0f5658720ab..8fd5301128106 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -645,6 +645,7 @@ void ceph_add_cap(struct inode *inode,
 	struct ceph_cap *cap;
 	int mds = session->s_mds;
 	int actual_wanted;
+	u32 gen;
 
 	dout("add_cap %p mds%d cap %llx %s seq %d\n", inode,
 	     session->s_mds, cap_id, ceph_cap_string(issued), seq);
@@ -656,6 +657,10 @@ void ceph_add_cap(struct inode *inode,
 	if (fmode >= 0)
 		wanted |= ceph_caps_for_mode(fmode);
 
+	spin_lock(&session->s_gen_ttl_lock);
+	gen = session->s_cap_gen;
+	spin_unlock(&session->s_gen_ttl_lock);
+
 	cap = __get_cap_for_mds(ci, mds);
 	if (!cap) {
 		cap = *new_cap;
@@ -681,7 +686,7 @@ void ceph_add_cap(struct inode *inode,
 		list_move_tail(&cap->session_caps, &session->s_caps);
 		spin_unlock(&session->s_cap_lock);
 
-		if (cap->cap_gen < session->s_cap_gen)
+		if (cap->cap_gen < gen)
 			cap->issued = cap->implemented = CEPH_CAP_PIN;
 
 		/*
@@ -775,7 +780,7 @@ void ceph_add_cap(struct inode *inode,
 	cap->seq = seq;
 	cap->issue_seq = seq;
 	cap->mseq = mseq;
-	cap->cap_gen = session->s_cap_gen;
+	cap->cap_gen = gen;
 
 	if (fmode >= 0)
 		__ceph_get_fmode(ci, fmode);
-- 
2.20.1

[PATCH AUTOSEL 5.3 14/71] ceph: reconnect connection if session hang in opening state

[Sasha Levin]

From: Erqi Chen <chenerqi@gmail.com>

[ Upstream commit 71a228bc8d65900179e37ac309e678f8c523f133 ]

If client mds session is evicted in CEPH_MDS_SESSION_OPENING state,
mds won't send session msg to client, and delayed_work skip
CEPH_MDS_SESSION_OPENING state session, the session hang forever.

Allow ceph_con_keepalive to reconnect a session in OPENING to avoid
session hang. Also, ensure that we skip sessions in RESTARTING and
REJECTED states since those states can't be resurrected by issuing
a keepalive.

Link: https://tracker.ceph.com/issues/41551
Signed-off-by: Erqi Chen chenerqi@gmail.com
Reviewed-by: "Yan, Zheng" <zyan@redhat.com>
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/ceph/mds_client.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 920e9f048bd8f..b11af7d8e8e93 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -4044,7 +4044,9 @@ static void delayed_work(struct work_struct *work)
 				pr_info("mds%d hung\n", s->s_mds);
 			}
 		}
-		if (s->s_state < CEPH_MDS_SESSION_OPEN) {
+		if (s->s_state == CEPH_MDS_SESSION_NEW ||
+		    s->s_state == CEPH_MDS_SESSION_RESTARTING ||
+		    s->s_state == CEPH_MDS_SESSION_REJECTED) {
 			/* this mds is failed or recovering, just wait */
 			ceph_put_mds_session(s);
 			continue;
-- 
2.20.1

[PATCH AUTOSEL 5.3 15/71] rbd: fix response length parameter for encoded strings

[Sasha Levin]

From: Dongsheng Yang <dongsheng.yang@easystack.cn>

[ Upstream commit 5435d2069503e2aa89c34a94154f4f2fa4a0c9c4 ]

rbd_dev_image_id() allocates space for length but passes a smaller
value to rbd_obj_method_sync().  rbd_dev_v2_object_prefix() doesn't
allocate space for length.  Fix both to be consistent.

Signed-off-by: Dongsheng Yang <dongsheng.yang@easystack.cn>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/block/rbd.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index c8fb886aebd4e..69db7385c8df5 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -5669,17 +5669,20 @@ static int rbd_dev_v2_image_size(struct rbd_device *rbd_dev)
 
 static int rbd_dev_v2_object_prefix(struct rbd_device *rbd_dev)
 {
+	size_t size;
 	void *reply_buf;
 	int ret;
 	void *p;
 
-	reply_buf = kzalloc(RBD_OBJ_PREFIX_LEN_MAX, GFP_KERNEL);
+	/* Response will be an encoded string, which includes a length */
+	size = sizeof(__le32) + RBD_OBJ_PREFIX_LEN_MAX;
+	reply_buf = kzalloc(size, GFP_KERNEL);
 	if (!reply_buf)
 		return -ENOMEM;
 
 	ret = rbd_obj_method_sync(rbd_dev, &rbd_dev->header_oid,
 				  &rbd_dev->header_oloc, "get_object_prefix",
-				  NULL, 0, reply_buf, RBD_OBJ_PREFIX_LEN_MAX);
+				  NULL, 0, reply_buf, size);
 	dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
 	if (ret < 0)
 		goto out;
@@ -6696,7 +6699,6 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
 	dout("rbd id object name is %s\n", oid.name);
 
 	/* Response will be an encoded string, which includes a length */
-
 	size = sizeof (__le32) + RBD_IMAGE_ID_LEN_MAX;
 	response = kzalloc(size, GFP_NOIO);
 	if (!response) {
@@ -6708,7 +6710,7 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
 
 	ret = rbd_obj_method_sync(rbd_dev, &oid, &rbd_dev->header_oloc,
 				  "get_id", NULL, 0,
-				  response, RBD_IMAGE_ID_LEN_MAX);
+				  response, size);
 	dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
 	if (ret == -ENOENT) {
 		image_id = kstrdup("", GFP_KERNEL);
-- 
2.20.1

Re: [PATCH AUTOSEL 5.3 15/71] rbd: fix response length parameter for encoded strings

[Ilya Dryomov]

On Tue, Oct 1, 2019 at 6:39 PM Sasha Levin <sashal@kernel.org> wrote:
>
> From: Dongsheng Yang <dongsheng.yang@easystack.cn>
>
> [ Upstream commit 5435d2069503e2aa89c34a94154f4f2fa4a0c9c4 ]
>
> rbd_dev_image_id() allocates space for length but passes a smaller
> value to rbd_obj_method_sync().  rbd_dev_v2_object_prefix() doesn't
> allocate space for length.  Fix both to be consistent.
>
> Signed-off-by: Dongsheng Yang <dongsheng.yang@easystack.cn>
> Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
> Signed-off-by: Sasha Levin <sashal@kernel.org>
> ---
>  drivers/block/rbd.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
> index c8fb886aebd4e..69db7385c8df5 100644
> --- a/drivers/block/rbd.c
> +++ b/drivers/block/rbd.c
> @@ -5669,17 +5669,20 @@ static int rbd_dev_v2_image_size(struct rbd_device *rbd_dev)
>
>  static int rbd_dev_v2_object_prefix(struct rbd_device *rbd_dev)
>  {
> +       size_t size;
>         void *reply_buf;
>         int ret;
>         void *p;
>
> -       reply_buf = kzalloc(RBD_OBJ_PREFIX_LEN_MAX, GFP_KERNEL);
> +       /* Response will be an encoded string, which includes a length */
> +       size = sizeof(__le32) + RBD_OBJ_PREFIX_LEN_MAX;
> +       reply_buf = kzalloc(size, GFP_KERNEL);
>         if (!reply_buf)
>                 return -ENOMEM;
>
>         ret = rbd_obj_method_sync(rbd_dev, &rbd_dev->header_oid,
>                                   &rbd_dev->header_oloc, "get_object_prefix",
> -                                 NULL, 0, reply_buf, RBD_OBJ_PREFIX_LEN_MAX);
> +                                 NULL, 0, reply_buf, size);
>         dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
>         if (ret < 0)
>                 goto out;
> @@ -6696,7 +6699,6 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
>         dout("rbd id object name is %s\n", oid.name);
>
>         /* Response will be an encoded string, which includes a length */
> -
>         size = sizeof (__le32) + RBD_IMAGE_ID_LEN_MAX;
>         response = kzalloc(size, GFP_NOIO);
>         if (!response) {
> @@ -6708,7 +6710,7 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
>
>         ret = rbd_obj_method_sync(rbd_dev, &oid, &rbd_dev->header_oloc,
>                                   "get_id", NULL, 0,
> -                                 response, RBD_IMAGE_ID_LEN_MAX);
> +                                 response, size);
>         dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
>         if (ret == -ENOENT) {
>                 image_id = kstrdup("", GFP_KERNEL);

Hi Sasha,

This patch just made things consistent, there was no bug here.  I don't
think it should be backported.

Thanks,

                Ilya

Re: [PATCH AUTOSEL 5.3 15/71] rbd: fix response length parameter for encoded strings

[Sasha Levin]

On Tue, Oct 01, 2019 at 07:15:49PM +0200, Ilya Dryomov wrote:
>On Tue, Oct 1, 2019 at 6:39 PM Sasha Levin <sashal@kernel.org> wrote:
>>
>> From: Dongsheng Yang <dongsheng.yang@easystack.cn>
>>
>> [ Upstream commit 5435d2069503e2aa89c34a94154f4f2fa4a0c9c4 ]
>>
>> rbd_dev_image_id() allocates space for length but passes a smaller
>> value to rbd_obj_method_sync().  rbd_dev_v2_object_prefix() doesn't
>> allocate space for length.  Fix both to be consistent.
>>
>> Signed-off-by: Dongsheng Yang <dongsheng.yang@easystack.cn>
>> Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
>> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
>> Signed-off-by: Sasha Levin <sashal@kernel.org>
>> ---
>>  drivers/block/rbd.c | 10 ++++++----
>>  1 file changed, 6 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
>> index c8fb886aebd4e..69db7385c8df5 100644
>> --- a/drivers/block/rbd.c
>> +++ b/drivers/block/rbd.c
>> @@ -5669,17 +5669,20 @@ static int rbd_dev_v2_image_size(struct rbd_device *rbd_dev)
>>
>>  static int rbd_dev_v2_object_prefix(struct rbd_device *rbd_dev)
>>  {
>> +       size_t size;
>>         void *reply_buf;
>>         int ret;
>>         void *p;
>>
>> -       reply_buf = kzalloc(RBD_OBJ_PREFIX_LEN_MAX, GFP_KERNEL);
>> +       /* Response will be an encoded string, which includes a length */
>> +       size = sizeof(__le32) + RBD_OBJ_PREFIX_LEN_MAX;
>> +       reply_buf = kzalloc(size, GFP_KERNEL);
>>         if (!reply_buf)
>>                 return -ENOMEM;
>>
>>         ret = rbd_obj_method_sync(rbd_dev, &rbd_dev->header_oid,
>>                                   &rbd_dev->header_oloc, "get_object_prefix",
>> -                                 NULL, 0, reply_buf, RBD_OBJ_PREFIX_LEN_MAX);
>> +                                 NULL, 0, reply_buf, size);
>>         dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
>>         if (ret < 0)
>>                 goto out;
>> @@ -6696,7 +6699,6 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
>>         dout("rbd id object name is %s\n", oid.name);
>>
>>         /* Response will be an encoded string, which includes a length */
>> -
>>         size = sizeof (__le32) + RBD_IMAGE_ID_LEN_MAX;
>>         response = kzalloc(size, GFP_NOIO);
>>         if (!response) {
>> @@ -6708,7 +6710,7 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
>>
>>         ret = rbd_obj_method_sync(rbd_dev, &oid, &rbd_dev->header_oloc,
>>                                   "get_id", NULL, 0,
>> -                                 response, RBD_IMAGE_ID_LEN_MAX);
>> +                                 response, size);
>>         dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
>>         if (ret == -ENOENT) {
>>                 image_id = kstrdup("", GFP_KERNEL);
>
>Hi Sasha,
>
>This patch just made things consistent, there was no bug here.  I don't
>think it should be backported.

I'll drop it, thanks!

-- 
Thanks,
Sasha

Re: [PATCH AUTOSEL 5.3 15/71] rbd: fix response length parameter for encoded strings

[Jens Axboe]

On 10/8/19 3:29 PM, Sasha Levin wrote:
> On Tue, Oct 01, 2019 at 07:15:49PM +0200, Ilya Dryomov wrote:
>> On Tue, Oct 1, 2019 at 6:39 PM Sasha Levin <sashal@kernel.org> wrote:
>>>
>>> From: Dongsheng Yang <dongsheng.yang@easystack.cn>
>>>
>>> [ Upstream commit 5435d2069503e2aa89c34a94154f4f2fa4a0c9c4 ]
>>>
>>> rbd_dev_image_id() allocates space for length but passes a smaller
>>> value to rbd_obj_method_sync().  rbd_dev_v2_object_prefix() doesn't
>>> allocate space for length.  Fix both to be consistent.
>>>
>>> Signed-off-by: Dongsheng Yang <dongsheng.yang@easystack.cn>
>>> Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
>>> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
>>> Signed-off-by: Sasha Levin <sashal@kernel.org>
>>> ---
>>>   drivers/block/rbd.c | 10 ++++++----
>>>   1 file changed, 6 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
>>> index c8fb886aebd4e..69db7385c8df5 100644
>>> --- a/drivers/block/rbd.c
>>> +++ b/drivers/block/rbd.c
>>> @@ -5669,17 +5669,20 @@ static int rbd_dev_v2_image_size(struct rbd_device *rbd_dev)
>>>
>>>   static int rbd_dev_v2_object_prefix(struct rbd_device *rbd_dev)
>>>   {
>>> +       size_t size;
>>>          void *reply_buf;
>>>          int ret;
>>>          void *p;
>>>
>>> -       reply_buf = kzalloc(RBD_OBJ_PREFIX_LEN_MAX, GFP_KERNEL);
>>> +       /* Response will be an encoded string, which includes a length */
>>> +       size = sizeof(__le32) + RBD_OBJ_PREFIX_LEN_MAX;
>>> +       reply_buf = kzalloc(size, GFP_KERNEL);
>>>          if (!reply_buf)
>>>                  return -ENOMEM;
>>>
>>>          ret = rbd_obj_method_sync(rbd_dev, &rbd_dev->header_oid,
>>>                                    &rbd_dev->header_oloc, "get_object_prefix",
>>> -                                 NULL, 0, reply_buf, RBD_OBJ_PREFIX_LEN_MAX);
>>> +                                 NULL, 0, reply_buf, size);
>>>          dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
>>>          if (ret < 0)
>>>                  goto out;
>>> @@ -6696,7 +6699,6 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
>>>          dout("rbd id object name is %s\n", oid.name);
>>>
>>>          /* Response will be an encoded string, which includes a length */
>>> -
>>>          size = sizeof (__le32) + RBD_IMAGE_ID_LEN_MAX;
>>>          response = kzalloc(size, GFP_NOIO);
>>>          if (!response) {
>>> @@ -6708,7 +6710,7 @@ static int rbd_dev_image_id(struct rbd_device *rbd_dev)
>>>
>>>          ret = rbd_obj_method_sync(rbd_dev, &oid, &rbd_dev->header_oloc,
>>>                                    "get_id", NULL, 0,
>>> -                                 response, RBD_IMAGE_ID_LEN_MAX);
>>> +                                 response, size);
>>>          dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret);
>>>          if (ret == -ENOENT) {
>>>                  image_id = kstrdup("", GFP_KERNEL);
>>
>> Hi Sasha,
>>
>> This patch just made things consistent, there was no bug here.  I don't
>> think it should be backported.
> 
> I'll drop it, thanks!

How did it even get picked up, it's not marked for stable?

-- 
Jens Axboe