- 13 11月, 2014 1 次提交
-
-
由 Johan Hedberg 提交于
There's no reason why all users of L2CAP would need to worry about initializing chan->nesting to L2CAP_NESTING_NORMAL (which is important since 0 is the same as NESTING_SMP). This patch moves the initialization to the common place that's used to create all new channels, i.e. the l2cap_chan_create() function. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 29 10月, 2014 1 次提交
-
-
由 Stephen Hemminger 提交于
Fix spelling errors in comments. Signed-off-by: NStephen Hemminger <stephen@networkplumber.org> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 25 10月, 2014 2 次提交
-
-
由 Martin Townsend 提交于
In the error case where credits is greater than max_credits there is a missing l2cap_chan_unlock before returning. Signed-off-by: NMartin Townsend <mtownsend1973@gmail.com> Tested-by: NJukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Alfonso Acosta 提交于
The values of a lot of the mgmt_device_connected() parameters come straight from a hci_conn object. We can simplify the function by passing the full hci_conn pointer to it. Signed-off-by: NAlfonso Acosta <fons@spotify.com> Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com>
-
- 02 10月, 2014 1 次提交
-
-
由 Johan Hedberg 提交于
The L2CAP connection's channel list lock (conn->chan_lock) must never be taken while already holding a channel lock (chan->lock) in order to avoid lock-inversion and lockdep warnings. So far the l2cap_chan_connect function has acquired the chan->lock early in the function and then later called l2cap_chan_add(conn, chan) which will try to take the conn->chan_lock. This violates the correct order of taking the locks and may lead to the following type of lockdep warnings: -> #1 (&conn->chan_lock){+.+...}: [<c109324d>] lock_acquire+0x9d/0x140 [<c188459c>] mutex_lock_nested+0x6c/0x420 [<d0aab48e>] l2cap_chan_add+0x1e/0x40 [bluetooth] [<d0aac618>] l2cap_chan_connect+0x348/0x8f0 [bluetooth] [<d0cc9a91>] lowpan_control_write+0x221/0x2d0 [bluetooth_6lowpan] -> #0 (&chan->lock){+.+.+.}: [<c10928d8>] __lock_acquire+0x1a18/0x1d20 [<c109324d>] lock_acquire+0x9d/0x140 [<c188459c>] mutex_lock_nested+0x6c/0x420 [<d0ab05fd>] l2cap_connect_cfm+0x1dd/0x3f0 [bluetooth] [<d0a909c4>] hci_le_meta_evt+0x11a4/0x1260 [bluetooth] [<d0a910eb>] hci_event_packet+0x3ab/0x3120 [bluetooth] [<d0a7cb08>] hci_rx_work+0x208/0x4a0 [bluetooth] CPU0 CPU1 ---- ---- lock(&conn->chan_lock); lock(&chan->lock); lock(&conn->chan_lock); lock(&chan->lock); Before calling l2cap_chan_add() the channel is not part of the conn->chan_l list, and can therefore only be accessed by the L2CAP user (such as l2cap_sock.c). We can therefore assume that it is the responsibility of the user to handle mutual exclusion until this point (which we can see is already true in l2cap_sock.c by it in many places touching chan members without holding chan->lock). Since the hci_conn and by exctension l2cap_conn creation in the l2cap_chan_connect() function depend on chan details we cannot simply add a mutex_lock(&conn->chan_lock) in the beginning of the function (since the conn object doesn't yet exist there). What we can do however is move the chan->lock taking later into the function where we already have the conn object and can that way take conn->chan_lock first. This patch implements the above strategy and does some other necessary changes such as using __l2cap_chan_add() which assumes conn->chan_lock is held, as well as adding a second needed label so the unlocking happens as it should. Reported-by: NJukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Tested-by: NJukka Rissanen <jukka.rissanen@linux.intel.com> Acked-by: NJukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 11 9月, 2014 1 次提交
-
-
由 Johan Hedberg 提交于
Even if we have no connection-oriented channels we should perform the L2CAP Information Request procedures before notifying L2CAP channels of the connection. This is so that the L2CAP channel implementations can perform checks on what the remote side supports (e.g. does it support the fixed channel in question). So far the code has relied on the l2cap_do_start() function to initiate the Information Request, however l2cap_do_start() is used on a per-channel basis and only for connection-oriented channels. This means that if there are no connection-oriented channels on the system we would never start the Information Request procedure. This patch creates a new l2cap_request_info() helper function to initiate the Information Request procedure, and ensures that it is called whenever a BR/EDR connection has been established. The patch also updates fixed channels to be notified of connection readiness only once the Information Request procedure has completed. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 09 9月, 2014 8 次提交
-
-
由 Johan Hedberg 提交于
The identity address update of all channels for an l2cap_conn needs to take the lock for each channel, i.e. it's safest to do this by a separate workqueue callback. Previously this was partially solved by moving the entire SMP key distribution behind a workqueue. However, if we want SMP context locking to be correct and safe we should always use the l2cap_chan lock when accessing it, meaning even smp_distribute_keys needs to take that lock which would once again create a dead lock when updating the identity address. The simplest way to solve this is to have l2cap_conn manage the deferred work which is what this patch does. A subsequent patch will remove the now unnecessary SMP key distribution work struct. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
The "pending" L2CAP response value is not defined for LE CoC. This patch adds a clarifying comment to the code so that the reader will not think there is a bug in trying to use this value for LE CoC. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
Now that there are no more users of the l2cap_conn_shutdown API (since smp.c switched to using hci_disconnect) we can simply remove it along with all of it's l2cap_conn variables. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
When the l2cap_conn_del() function is used we do not want to wait around "in case something happens" before disconnecting. This patch sets the disconnection timeout to 0 so that the disconnection routines get immediately scheduled. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
It's natural to have *_get() functions that increment the reference count of an object to return the object type itself. This way it's simple to make a copy of the object pointer and increase the reference count in a single step. This patch updates two such get() functions, namely hci_conn_get() and l2cap_conn_get(), and updates the users to take advantage of the new API. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
Now that SMP has been converted to use fixed channels we've got a bit of a problem with the hci_conn reference counting. So far the L2CAP code has kept a reference for each L2CAP channel that was notified of the connection. With SMP however this would mean that the connection is never dropped even though there are no other users of it. Furthermore, SMP already does its own hci_conn reference counting internally, starting from a security or pairing request and ending with the key distribution. This patch makes L2CAP fixed channels default to the L2CAP core not keeping a hci_conn reference for them. A new FLAG_HOLD_HCI_CONN flag is added so that L2CAP users can declare an exception to this rule and hold a reference even for their fixed channels. One such exception is the L2CAP socket layer which does want a reference for each socket (e.g. an ATT socket which uses a fixed channel). Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
The l2cap_chan_add() function doesn't require the channel to be unlocked. It only requires the l2cap_conn to be unlocked. Therefore, it's unnecessary to unlock a channel before calling l2cap_chan_add(). This patch removes such unnecessary unlocking from the l2cap_chan_connect() function. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
The l2cap_create_le_flowctl_pdu() function that l2cap_segment_le_sdu() calls is perfectly capable of doing packet fragmentation if given bigger PDUs than the HCI buffers allow. Forcing the PDU length based on the HCI MTU (conn->mtu) would therefore needlessly strict operation on hardware with limited LE buffers (e.g. both Intel and Broadcom seem to have this set to just 27 bytes). This patch removes the restriction and makes it possible to send PDUs of the full length that the remote MPS value allows. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org> Cc: stable@vger.kernel.org
-
- 14 8月, 2014 16 次提交
-
-
由 Lukasz Rymanowski 提交于
I-Frame which is going to be resend already has FCS field added and set (if it was required). Adding additional FCS field calculated from data + old FCS in resend function is incorrect. This patch fix that. Issue has been found during PTS testing. Signed-off-by: NLukasz Rymanowski <lukasz.rymanowski@tieto.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Lukasz Rymanowski 提交于
There is no need to decrease pdu size with L2CAP SDU lenght in Start L2CAP SDU frame. Start packtet is just 2 bytes longer as specified and we can keep payload as long as possible. When testing SAR L2CAP against PTS, L2CAP channel is usually configured in that way, that SDU = MPS * 3. PTS expets then 3 I-Frames from IUT: Start, Continuation and End frame. Without this fix, we sent 4 I-Frames. We could pass a test by using -b option in l2test and send just two bytes less than SDU length. With this patch no need to use -b option. Signed-off-by: NLukasz Rymanowski <lukasz.rymanowski@tieto.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
Now that there are no-longer any users for l2cap_conn->security_timer we can go ahead and simply remove it. The patch makes initialization of the conn->info_timer unconditional since it's better not to leave any l2cap_conn data structures uninitialized no matter what the underlying transport. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
Since we no-longer do special handling of SMP within l2cap_core.c we don't have any code for calling l2cap_conn_del() when smp.c doesn't like the data it gets. At the same time we cannot simply export l2cap_conn_del() since it will try to lock the channels it calls into whereas we already hold the lock in the smp.c l2cap_chan callbacks (i.e. it'd lead to a deadlock). This patch adds a new l2cap_conn_shutdown() API which is very similar to l2cap_conn_del() except that it defers the call to l2cap_conn_del() through a workqueue, thereby making it safe to use it from an L2CAP channel callback. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
There's no need to export the smp_distribute_keys() function since the resume callback is called in the same scenario. This patch makes the smp_notify_keys function private (at the same time moving it higher up in smp.c to avoid forward declarations) and adds a resume callback for SMP to call it from there instead. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
Now that we have all the necessary pieces in place we can fully convert SMP to use the L2CAP channel infrastructure. This patch adds the necessary callbacks and removes the now unneeded conn->smp_chan pointer. One notable behavioral change in this patch comes from the following code snippet: - case L2CAP_CID_SMP: - if (smp_sig_channel(conn, skb)) - l2cap_conn_del(conn->hcon, EACCES); This piece of code was essentially forcing a disconnection if garbage SMP data was received. The l2cap_conn_del() function is private to l2cap_conn.c so we don't have access to it anymore when using the L2CAP channel callbacks. Therefore, the behavior of the new code is simply to return errors in the recv() callback (which is simply the old smp_sig_channel()), but no disconnection will occur. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
For most cases it makes no difference whether l2cap_le_conn_ready() is called before or after calling the channel ready() callbacks, however for upcoming SMP code we need this as the ready() callback initializes certain structures that a call to smp_conn_security() from l2cap_le_conn_ready() depends on. Therefore, move the call to l2cap_le_conn_ready() after iterating through and notifying channels. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
L2CAP channel implementations may want to still access the chan->conn pointer. This will particularly be the case for SMP that will want to clear a reference to the SMP channel in the l2cap_conn structure. The only user of the teardown callback so far is l2cap_sock.c and for the code there it makes no difference whether the callback is called before or after clearing the chan->conn pointer. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
The l2cap_add_scid function is used for registering a fixed L2CAP channel. Instead of having separate initialization of the channel type and outgoing MTU in l2cap_sock.c it's more intuitive to do these things in the l2cap_add_scid function itself (and thereby make the functionality available to other users besides l2cap_sock.c). Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
Now that we've got the fixed channel infrastructure cleaned up in a generic way there's no longer a need to have a dedicated function for handling data on the ATT channel. Instead the generic l2cap_data_channel() handler will be able to do the exact same thing. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
When notifying global fixed channels of new connections it doesn't make sense to consider channels meant for a different link type than the one available. This patch adds an extra parameter to the l2cap_global_fixed_chan() lookup function and ensures that only channels matching the current hci_conn type are looked up. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
In order to remove special handling of fixed L2CAP channels we need to start creating them in a single place instead of having per-channel exceptions. The most natural place is the l2cap_conn_cfm() function which is called whenever there is a new baseband link. The only really special case so far has been the ATT socket, so in order not to break the code in between this patch removes the ATT special handling at the same time as it adds the generic fixed channel handling from l2cap_le_conn_ready() into the hci_conn_cfm() function. As a related change the channel locking in l2cap_conn_ready() becomes simpler and we can thereby move the smp_conn_security() call into the l2cap_le_conn_ready() function. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
This patch is a simple refactoring of l2cap_connect_cfm to allow easier extension of the function. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
With the update to sk->resume() and __l2cap_no_conn_pending() we no-longer need to have special handling of ATT channels in the l2cap_security_cfm() function. The chan->sec_level update when encryption has been enabled is safe to do for any kind of channel, and the loop takes later care of calling chan->ready() or chan->resume() if necessary. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
The __l2cap_no_conn_pending() function would previously only return a meaningful value for connection oriented channels and was therefore not useful for anything else. As preparation of making the L2CAP code more generic allow the function to be called for other channel types as well by returning a meaningful value for them. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
When looking up entries from the global L2CAP channel list there needs to be a guarantee that other code doesn't go and remove the entry after a channel has been returned by the lookup function. This patch makes sure that the channel reference is incremented before the read lock is released in the global channel lookup functions. The patch also adds the corresponding l2cap_chan_put() calls once the channels pointers are no-longer needed. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 21 7月, 2014 1 次提交
-
-
由 Johan Hedberg 提交于
It's safer practice to use sizeof(*ptr) instead of sizeof(ptr_type) when allocating memory in case the type changes. This also fixes the following style of warnings from static analyzers: CHECK: Prefer kzalloc(sizeof(*ie)...) over kzalloc(sizeof(struct inquiry_entry)...) + ie = kzalloc(sizeof(struct inquiry_entry), GFP_KERNEL); Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 18 7月, 2014 1 次提交
-
-
由 Johan Hedberg 提交于
The EOPNOTSUPP and ENOTSUPP errors are very similar in meaning, but ENOTSUPP is a fairly new addition to POSIX. Not all libc versions know about the value the kernel uses for ENOTSUPP so it's better to use EOPNOTSUPP to ensure understandable error messages. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 17 7月, 2014 1 次提交
-
-
由 Johan Hedberg 提交于
We're interested in whether an authentication request is because of a remote or local action. So far hci_conn_security() has been used both for incoming and outgoing actions (e.g. RFCOMM or L2CAP connect requests) so without some modifications it cannot know which peer is responsible for requesting authentication. This patch adds a new "bool initiator" parameter to hci_conn_security() to indicate which side is responsible for the request and updates the current users to pass this information correspondingly. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 16 7月, 2014 2 次提交
-
-
由 Johan Hedberg 提交于
To make the code more understandable it makes sense to use the new HCI defines for connection role instead of a "bool master" parameter. This makes it immediately clear when looking at the function calls what the last parameter is describing. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
由 Johan Hedberg 提交于
Having a dedicated u8 role variable in the hci_conn struct greatly simplifies tracking of the role, since this is the native way that it's represented on the HCI level. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 14 7月, 2014 1 次提交
-
-
由 Marcel Holtmann 提交于
The spinlock protecting the L2CAP ident number can be converted into a mutex since the whole processing is run in a workqueue. So instead of using a spinlock, just use a mutex here. Signed-off-by: NMarcel Holtmann <marcel@holtmann.org> Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com>
-
- 11 7月, 2014 1 次提交
-
-
由 Marcel Holtmann 提交于
The list of L2CAP fixed channels increased with newer versions of the specification. This just updates the constants for it. Signed-off-by: NMarcel Holtmann <marcel@holtmann.org> Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com>
-
- 09 7月, 2014 2 次提交
-
-
由 Marcel Holtmann 提交于
When ERTM support is disabled, then do not even send ERTM configuration option even if the remote side supports it. Signed-off-by: NMarcel Holtmann <marcel@holtmann.org> Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com>
-
由 Johan Hedberg 提交于
We already have several lists with struct bdaddr_list entries, and there will be more in the future. Since the operations for adding, removing, looking up and clearing entries in these lists are exactly the same it doesn't make sense to define new functions for every single list. This patch unifies the functions by passing the list_head to them instead of a hci_dev pointer. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-
- 07 7月, 2014 1 次提交
-
-
由 Johan Hedberg 提交于
If we have both LE scanning and advertising simultaneously enabled we need a way to tell hci_connect_le() in which role to initiate a connection. This patch adds a new parameter to the function to give it the necessary information. For auto-connect and mgmt_pair_device we always use master role, whereas for L2CAP users (in practice sockets) we use slave role whenever HCI_ADVERTISING is set and master role otherwise. Signed-off-by: NJohan Hedberg <johan.hedberg@intel.com> Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
-