- 13 Feb, 2012 37 commits
-
-
David Herrmann authored
Since we remove the owner field of hci_dev hci_dev_put and __hci_dev_put do the same so we can merge them into one function. Same for hci_dev_hold and __hci_dev_hold. Signed-off-by:
David Herrmann <dh.herrmann@googlemail.com> Acked-by:
Marcel Holtmann <marcel@holtmann.org> Signed-off-by:
Johan Hedberg <johan.hedberg@intel.com>
-
David Herrmann authored
The hci_dev->dev device structure has an internal refcount. This refcount is used to protect the whole hci_dev object. However, we currently do not use it. Therefore, if someone calls hci_free_dev() we currently immediately destroy the hci_dev object because we never took the device refcount. This even happens if the hci_dev->refcnt is not 0. In fact, the hci_dev->refcnt is totally useless in its current state. Therefore, we simply remove hci_dev->refcnt and instead use hci_dev->dev refcnt. This fixes all the symptoms and also correctly integrates the device structure into our bluetooth bus system. Signed-off-by:
-
David Herrmann authored
After unregistering an hci_dev object a bluetooth driver does not have any callbacks in the hci_dev structure left over. Therefore, there is no need to keep a reference to the module. Previously, we needed this to protect the hci-destruct callback. However, this callback is no longer available so we do not need this owner field, anymore. Drivers now call hci_unregister_dev() and they are done with the object. Signed-off-by:
-
David Herrmann authored
We provide a device-object to other subsystems and we provide our own release-function. Therefore, the device-object must own a reference to our module, otherwise the release-function may get deleted before the device-object does. Signed-off-by:
-
David Herrmann authored
The hci-destruct callback is not used by any driver so we can remove it. There is no reason to keep it alive, anymore. Drivers can free their internal data on driver-release and we do not need to provide a public destruct callback. Internally, we still use a destruct callback inside of hci_sysfs.c. This one is used to correctly free our hci_dev data structure if no more users have a reference to it. Signed-off-by:
-
David Herrmann authored
We currently leak the hci_uart object if HCI_UART_PROTO_SET is never set because the hci-destruct callback will then never be called. This fix removes the hci-destruct callback and frees the driver internal private hci_uart object directly on tty-close. We call hci_unregister_dev() here so the hci-core will never call our callbacks again (except destruct). Therefore, we can safely free the driver internal data right away and set the destruct callback to NULL. Signed-off-by:
-
David Herrmann authored
Instead of waiting for the hdev object to get freed we now free the private driver-internal data on SDIO shutdown. This allows us to remove the obsolete hci-destruct callback and free our data object right away after calling hci_unregister_dev(). The HCI-core does not call any callbacks after this so we are never called again and can safely exit the module. Signed-off-by:
-
David Herrmann authored
Instead of waiting for the hci-device to be destroyed we now free the private driver data on driver shutdown right away. We call hci_unregister_dev() on driver shutdown, that means, the hci-core will never ever call our callbacks again except the destruct callback. It also does not access hdev->driver_data so there is no reason to keep that alive. We simply set the destruct cb to NULL to avoid getting called again. Signed-off-by:
-
David Herrmann authored
Instead of using the hci-destruct callback we free our private driver data on USB shutdown. We already called hci_unregister_dev() here so the hci core will never ever call our callbacks again except the destruct callback. However, there is no reason to keep our *private* driver data alive if we get never called again and the hci-core does never touch it the data. So we simply free it right away and set the destruct callback to NULL. Signed-off-by:
-
David Herrmann authored
This frees the private driver data on USB shutdown instead of using the hci-destruct callback. We already call usb_set_intfdata(intf, NULL) but we do not do the same with the hci object. This would be totally safe, though. After calling hci_unregister_dev()/hci_free_dev() the hdev object will never call any callback of us again except the destruct callback. Therefore, we can safely set the destruct callback to NULL and free the driver data right away. This allows to unload the module without waiting for the hdev device to be released. Signed-off-by:
-
David Herrmann authored
This removes the hci-destruct callback and instead frees the private driver data in the vhci_release file release function. There is no reason to keep private driver data available if the driver has already shut down. After vhci_release is called our module can be unloaded. The only reason it is kept alive is the hci-core having a module-ref on us because of our destruct callback. However, this callback only frees hdev->driver_data. That is, we wait for the hdev-device to get destroyed to free our internal driver-data. In fact, the hci-core does never touch hdev->driver_data so it doesn't care if it is NULL. Therefore, we simply free it when unloading the driver. Another important fact is that the hdev core does not call any callbacks other than the destruct-cb after hci_unregister_dev() has been called. So there is no function of our module that will be called nor does the hci-core touch hdev->driver_data. Hence, no other code can touch hdev->driver_data after our cleanup so the destruct callback is definitely unnecessary here. Signed-off-by:
-
David Herrmann authored
The destruct cb is optional so we can safely remove our dummy cb. Signed-off-by:
-
David Herrmann authored
The destruct cb is optional so remove our empty dummy cb. Signed-off-by:
-
David Herrmann authored
The destruct callback is optional and we provide an empty callback so remove it. Signed-off-by:
-
David Herrmann authored
The callback is optional and we provide an empty callback so remove it entirely. Signed-off-by:
-
David Herrmann authored
The callback is optional and we provide an empty callback so remove it. Signed-off-by:
-
David Herrmann authored
The destruct callback is optional and we provide an empty callback so remove it entirely to avoid unnecessary code. Signed-off-by:
-
David Herrmann authored
Several drivers already provide an empty callback so we can actually make this optional and then remove all those empty callbacks in the drivers. This callback isn't needed at all by most drivers as they can remove their allocated structures on device disconnect and not on hci destruction. Signed-off-by:
-
David Herrmann authored
We currently leak the driver info structure if dtl1_config fails. If we add a kfree() to dtl1_release to fix this, then dtl1_release and dtl1_detach are identical so this merges both functions. Signed-off-by:
-
David Herrmann authored
The uart_proto open() callback is not called in atomic context so we can safely sleep here. The caller hci_uart_set_proto() in hci_ldisc.c is an ioctl-handler and therefore can sleep. Signed-off-by:
-
David Herrmann authored
The uart_proto open() callback is not called in atomic context so we can safely sleep here. The caller hci_uart_set_proto() in hci_ldisc.c is an ioctl-handler and therefore can sleep. Signed-off-by:
-
David Herrmann authored
The uart_proto open() callback is not called in atomic context so we can safely sleep here. The caller hci_uart_set_proto() in hci_ldisc.c is an ioctl-handler and therefore can sleep. Signed-off-by:
-
David Herrmann authored
The uart_proto open() callback is not called in atomic context so we can safely sleep here. The caller hci_uart_set_proto() in hci_ldisc.c is an ioctl() handler and therefore can sleep. Signed-off-by:
-
Vinicius Costa Gomes authored
queue_delayed_work() expects a relative time for when that work should be scheduled. Signed-off-by:
Vinicius Costa Gomes <vinicius.gomes@openbossa.org> Acked-by:
-
Andrei Emeltchenko authored
Adds HCI_ACL_TX_TIMEOUT and clear conversion from msec to jiffies Signed-off-by:
Andrei Emeltchenko <andrei.emeltchenko@intel.com> Acked-by:
-
Andrzej Kaczmarek authored
After moving L2CAP timers to workqueues l2cap_set_timer expects timeout value to be specified in jiffies but constants defined in miliseconds are used. This makes timeouts unreliable when CONFIG_HZ is not set to 1000. __set_chan_timer macro still uses jiffies as input to avoid multiple conversions from/to jiffies for sk_sndtimeo value which is already specified in jiffies. Signed-off-by:
Andrzej Kaczmarek <andrzej.kaczmarek@tieto.com> Ackec-by:
-
Andrzej Kaczmarek authored
sk_sndtime value should be specified in jiffies thus initial value needs to be converted from miliseconds. Otherwise this timeout is unreliable when CONFIG_HZ is not set to 1000. Signed-off-by:
-
Andrei Emeltchenko authored
Adds support for Number Of Completed Data Blocks Event. Signed-off-by:
-
Daniel Wagner authored
[ 2096.384084] btusb_send_frame:684: hci0 [ 2096.384087] usb 3-1: BOGUS urb flags, 2 --> 0 [ 2096.384091] Bluetooth: hci0 urb ffff8801b61d3a80 submission failed (22) According the documentation in usb_submit_urb() URB_ISO_ASAP flag is only allowed for endpoints of type USB_ENDPOINT_XFER_ISOC. This reverts commit b8aabfc9. Signed-off-by:
Daniel Wagner <daniel.wagner@bmw-carit.de> Acked-by:
Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Acked-by:
-
Johan Hedberg authored
This patch adds the necessary logic to perform name lookups after inquiry completes. This is done by checking for entries in the resolve list after each inquiry complete and remote name complete HCI event. Signed-off-by:
-
Johan Hedberg authored
This patch adds proper state tracking to the device discovery process. This makes it possible to return appropriate errors when trying to stop a non-active discovery or start discovery when it is already ongoing. Once name resolving is implemented this also makes it possible to know what the right action to do is when a remote name lookup is cancelled. Signed-off-by:
-
Johan Hedberg authored
This struct is used for not just inquiry caching but also for general device discovery state tracking so it's better to rename it to something more appropriate. Signed-off-by:
-
Johan Hedberg authored
For the remote name state tracking for the management interface to work the cache needs to be flushed whenever inquiry is started. The hci_do_inquiry function is only used by the management interface so by having the flushing done from it ensures that old ioctl based functionality isn't affected. Signed-off-by:
-
Johan Hedberg authored
If user-space has already confirmed the name for a remote device we shouldn't request confirmation again. The simplest way to do this is to return the name state from hci_inquiry_cache_update (if it is anything else than unknown then we do not need confirmation from user-space). Signed-off-by:
-
Johan Hedberg authored
This patch adds initial support for mgmt_confirm_name. It adds the necessary tracking of the name state by extending the inquiry cache. The actual name resolving operation (to be done once inquiry is finished) is not yet part of this patch. Signed-off-by:
-
Johan Hedberg authored
The EIR defines are needed also outside of mgmt.c (e.g. in hci_event.c to check if EIR data has the complete name) so it's better to have them in a single public place, i.e. hci.h. Signed-off-by:
-
Johan Hedberg authored
This makes it possible to use the convenience functions provided for standard kernel list types and it also makes it easier to extend the use of the cache for the management interface where e.g. name resolving control will be needed. Signed-off-by:
-
- 09 Feb, 2012 1 commit
-
-
John W. Linville authored
Fixes problem where caller would think routine succeeded when it failed leading to divide by zero panic. (This also reverts an earlier attempt, commit 42bc0c97 "rtlwifi: Return correct failure code on error". -- JWL) Signed-off-by:
Simon Graham <simon.graham@virtualcomputer.com> Acked-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
- 08 Feb, 2012 2 commits
-
-
Felix Fietkau authored
Most rate control implementations assume .get_rate and .tx_status are only called once the per-station data has been fully initialized. minstrel_ht crashes if this assumption is violated. Signed-off-by:
Felix Fietkau <nbd@openwrt.org> Tested-by:
Arend van Spriel <arend@broadcom.com> Signed-off-by:
-
Simon Graham authored
Callers of rtl_pci_init expect zero to be returned on error. Returning the error code leads to, amongst other things, divide by zero panics attempting to use the ring size that is set to zero. Signed-off-by:
-
