- Event thread wakes up and passes the event to
event_handler() to process.
- It processes and clears the USBIP_EH_SHUTDOWN
flag then returns.
- The outer event loop (event_handler_loop()) calls
wait_event_interruptible().
The processing of the second flag which is part of
VDEV_EVENT_DOWN (USBIP_EH_RESET) did not happen yet.
It is delayed until the next event.
This means the ->reset callback may not happen for
a long time (if ever), leaving the usbip port in a
weird state which prevents its reuse.
This patch changes the handler to process all flags
before waiting for another wakeup.
I have verified this change to fix a problem which
prevented reattach of a usbip device. It also helps
for socket errors which missed the RESET as well.
The delayed event processing also affects the stub
side of usbip and the error handling there.
Signed-off-by: Max Vozeler <mvz@vozeler.com> Reported-by: Marco Lancione <marco@optikam.com> Tested-by: Luc Jalbert <ljalbert@optikam.com> Cc: stable <stable@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 10:04:21 +0000 (13:04 +0300)]
staging: w35und: Add locking problems to TODO list
The w35und uses atomics such as "ThreadCount" and "FireCount" to emulate
locking in the TX paths, for example. Document this bug in the TODO list.
Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 10:01:03 +0000 (13:01 +0300)]
staging: w35und: Inline MLMESendFrame() to wbsoft_tx()
The wbsoft_tx() function is a simple wrapper on top of MLMESendFrame() so
inline the latter to the former.
Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 10:01:02 +0000 (13:01 +0300)]
staging: w35und: Use NETDEV_TX_BUSY if MLMESendFrame fails
This patch changes MLMESendFrame to return NETDEV_TX_BUSY if MLME frame is in
use so that wbsoft_tx() doesn't blindly return NETDEV_TX_OK in that case.
Cc: Sandro Bonazzola <sandro.bonazzola@gmail.com> Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 09:28:43 +0000 (12:28 +0300)]
staging: w35und: Remove unused fields from struct wbsoft_priv
This patch removes unused fields from "struct wbsoft_priv".
Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 09:28:42 +0000 (12:28 +0300)]
staging: w35und: Remove unused spinlocks
This patch removes unused spinlocks from "struct mlme_frame" and "struct
wbsoft_priv".
Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 09:28:41 +0000 (12:28 +0300)]
staging: w35und: Inline mlme_s.h to core.h
The mlme_s.h header is included in one place. As the header is very small, just
inline it to core.h.
Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 09:28:40 +0000 (12:28 +0300)]
staging: w35und: Remove dead code from mlme_s.h
There's bunch of macros in mlme_s.h that aren't used for anything. Kill them
off.
Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 09:28:39 +0000 (12:28 +0300)]
staging: w35und: Kill unused scan_s.h header
The scan_s.h header is not actually used for anything so just kill it off.
Cc: Lars Lindley <lindley@coyote.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch removes remaining typedef declarations from the w35und driver. Most
of them were unused so I just killed them off completely.
Cc: Lars Lindley <lindley@coyote.org> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pekka Enberg [Sun, 19 Sep 2010 09:28:37 +0000 (12:28 +0300)]
staging: w35und: Kill dead HAL macros in wbhal_f.h
Fixes an uninitialized variable access in _rx_iq_calibration_loop_winbond():
CC [M] drivers/staging/winbond/phy_calibration.o
drivers/staging/winbond/phy_calibration.c: In function ‘_rx_iq_calibration_loop_winbond’:
drivers/staging/winbond/phy_calibration.c:1138: warning: ‘val’ is used uninitialized in this function
Cc: Lars Lindley <lindley@coyote.org> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Ruslan Pisarev <ruslan@rpisarev.org.ua> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Jonathan Cameron [Tue, 21 Sep 2010 13:40:57 +0000 (14:40 +0100)]
staging: iio: amend macros for SCAN_EL creation to add separate index attr
As Manuel Stahl observed, putting data into the naming of an attribute
(beyond what it is for) breaks the one value per attribute rule (in spirit
at least). Hence we introduce a separate _index attribute for each scan
element to tell userspace the ordering in output from the buffer.
This will generate a lot of sparse warnings as all drivers will have unused
iio_const_attrs created. The rest of the set will clean these up.
The final patch will remove the old indexing method. It is not here so
as to maintain one or other interface for all devices as this set
progresses.
Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Manuel Stahl <manuel.stahl@iis.fraunhofer.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Jonathan Cameron [Tue, 21 Sep 2010 13:40:53 +0000 (14:40 +0100)]
staging: iio: adis16300 add _type attributes for all scan elements
Note some of the in driver type specifications do not match the
data sheet. I am assuming the datasheet is correct. This bug
would not have been apparent in the driver as these values were
unusued.
Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk> Signed-off-by: Manuel Stahl <manuel.stahl@iis.fraunhofer.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Henry Ptasinski [Tue, 21 Sep 2010 05:33:12 +0000 (22:33 -0700)]
staging: brcm80211: add fullmac driver
This patch to the existing bcm80211 directory in the staging tree adds fullmac
driver support for the BCM4329 SDIO chip from Broadcom. Configuration of the
mac80211 driver or the fullmac driver can be done through menuconfig.
Signed-off-by: Henry Ptasinski <henryp@broadcom.com> Signed-off-by: Nohee Ko <noheek@broadcom.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
move all the header information to ti_wilink_st.h.
This header would then not only serve the local modules like the st_core.c,
st_kim.c and st_ll.c but also the external modules/protocol drivers that are
dependent on the shared transport driver.
Modify the source files to include the new header ti_wilink_st.h,
replace uint and family by a much more proper u8/u16 types.
This is required specifically for data members of structures coming
in from the chip and data members parsed from the binary firmware file.
staging: ath6kl: Fixing a race condition during initialization
The commit fixes a race condition in the initialization portion of the
driver. The problem was observed while testing with the kernel in the
staging-next tree. The wait on the ready event is signalled prematurely
before setting the arVersion.wlan_ver and arVersion.abi_ver. The code
waiting on this semaphore checks for the validity of these values and
hence may proceed with an un-updated values of these fields. The wakeup
signal is now sent after these values are set.
The commit fixes a compilation error that was encountered while using
a specific kernel configuration file. The problem was the use of some
functions defined in <linux/semaphore.h> without including the header
file explicitly. It was probably working before because of the
dependency getting implicitly satisfied via some other header file.
Marek Lindner [Sat, 18 Sep 2010 19:01:22 +0000 (21:01 +0200)]
Staging: batman-adv: Introduce update_primary_addr to update mac address
set_primary_if is currently misused to update the mac address in vis
packets. This unneeded and introduces overhead due to other operations
which must be done when updating the primary interface.
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:21 +0000 (21:01 +0200)]
Staging: batman-adv: Track references of batman_if in set_primary_if
set_primary_if exchanges the current primary interfaces with a new one.
This is a new reference and thus we have to count it and decrease the
count of the old primary interface.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:20 +0000 (21:01 +0200)]
Staging: batman-adv: count batman_if list queries as reference
The return of get_batman_if_by_netdev and get_active_batman_if leaks a
pointer from the rcu protected list of interfaces. We must protect it to
prevent a too early release of the memory. Those functions must increase
the reference counter before rcu_read_unlock or it may be to late to
prevent a free.
hardif_add_interface must also increase the reference count for the
returned batman_if to make the behaviour consistent.
Reported-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:19 +0000 (21:01 +0200)]
Staging: batman-adv: Use refcnt to track usage count of batman_if
get_batman_if_by_netdev and get_active_batman_if may leak data from the
rcu protected list of interfaces. The rcu protected list of all gateway
nodes leaks the actual data outside the read-side critical area. This is
not valid as we may free the data using a call_rcu created callback
after we unlock using rcu_read_unlock. A workaround is to provide a
reference count to be sure that the memory isn't freed to early.
It is currently only to implement the already existing functionality and
doesn't provide the full tracking of all usage cases.
Additionally, we must hardif_hold inside the
rcu_read_lock()..rcu_read_unlock() before we attach to the structure
which "leaks" it. When another function now removed it from its usage
context (primary_if, usage on stack, ...) then we must hardif_put it. If
it is decremented to zero then we can issue the call_rcu to the freeing
function. So "put" is not allowed inside an rcu_read_lock.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:18 +0000 (21:01 +0200)]
Staging: batman-adv: Use synchronize_rcu instead of call_rcu
It is recommended [1] to use synchronize_rcu to simplify the code -
especially when otherwise extra locking is needed to protect other code
from picking stale elements. It also protects us for emitting to many
callbacks which may results in OOM conditions.
The only reason not to use it, would be in performance critical sections
or when we are not allowed to block.
[1] Documentation/RCU/checklist.txt
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Regions which do not use rcu functions don't need to protected by
rcu_read_lock. If we want to protect data from being freed than it must
be covered by the same read-side critical section or otherwise the grace
period may already ended and freed the memory before we called
rcu_read_lock again.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:15 +0000 (21:01 +0200)]
Staging: batman-adv: Introduce if_list_lock to protect if_list
The update critical sections of if_list must be protected by a locking
primitive other than RCU. The iterator must also be protected by the
chosen locking mechanism.
The rtnl_lock in hardif_remove_interfaces must also be moved outside the
iterator primitive to ensure that we don't deadlock the kernel due to
differently nested locks in hardif_remove_interfaces and hard_if_event.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:14 +0000 (21:01 +0200)]
Staging: batman-adv: wait for call_rcu before unloading module
synchronize_rcu respective synchronize_net only waits for the rcu grace
period to elapse and we may fail to finish the calls which were made to
call_rcu in that time. In result the module could be unloaded during the
execution of the RCU callbacks.
rcu_barrier[1] will now wait for all outstanding RCU callbacks to finish
before continuing.
[1] Documentation/RCU/rcubarrier.txt
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:12 +0000 (21:01 +0200)]
Staging: batman-adv: Move mailing list address to .org
The official mailing list is run on lists.open-mesh.org and it should be
avoided to sent them to lists.open-mesh.net to reduce the number of
receipents and double posts.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sven Eckelmann [Sat, 18 Sep 2010 19:01:11 +0000 (21:01 +0200)]
Staging: batman-adv: checkpatch cleanup of comments
checkpatch now detects the start of a comment and warns about usage of
multiple spaces at the beginning of a line. We have to replace the ' '
in multiple lines comments by ' * ' to fix it.
Checkpatch also wants a comment after a definition of a spinlock_t which
describes what it protects. It is currently not possible to add it
before the actual struct which includes the spinlock.
Signed-off-by: Sven Eckelmann <sven.eckelmann@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Platform devices tend to have id as -1 when only 1 device exist and
a value >=0 when multiple devices exist, since we plan to store all
these platform devices if multiple exist, there was a requirement to
have id to be >=0.
The patch fixes this problem.
projects / karo-tx-linux.git / log