With the previous changes, a disassociated global_cwq now can run as
an unbound one on its own - it can create workers as necessary to
drain remaining works after the CPU has been brought down and manage
the number of workers using the usual idle timer mechanism making
trustee completely redundant except for the actual unbinding
operation.
This patch removes the trustee and let a disassociated global_cwq
manage itself. Unbinding is moved to a work item (for CPU affinity)
which is scheduled and flushed from CPU_DONW_PREPARE.
This patch moves nr_running clearing outside gcwq and manager locks to
simplify the code. As nr_running is unused at the point, this is
safe.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Currently, during CPU offlining, after all pending work items are
drained, the trustee butchers all workers. Also, on CPU onlining
failure, workqueue_cpu_callback() ensures that the first idle worker
is destroyed. Combined, these guarantee that an offline CPU doesn't
have any worker for it once all the lingering work items are finished.
This guarantee isn't really necessary and makes CPU on/offlining more
expensive than needs to be, especially for platforms which use CPU
hotplug for powersaving.
This patch lets offline CPUs removes idle worker butchering from the
trustee and let a CPU which failed onlining keep the created first
worker. The first worker is created if the CPU doesn't have any
during CPU_DOWN_PREPARE and started right away. If onlining succeeds,
the rebind_workers() call in CPU_ONLINE will rebind it like any other
workers. If onlining fails, the worker is left alone till the next
try.
This makes CPU hotplugs cheaper by allowing global_cwqs to keep
workers across them and simplifies code.
Note that trustee doesn't re-arm idle timer when it's done and thus
the disassociated global_cwq will keep all workers until it comes back
online. This will be improved by further patches.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Currently, if there are left workers when a CPU is being brough back
online, the trustee kills all idle workers and scheduled rebind_work
so that they re-bind to the CPU after the currently executing work is
finished. This works for busy workers because concurrency management
doesn't try to wake up them from scheduler callbacks, which require
the target task to be on the local run queue. The busy worker bumps
concurrency counter appropriately as it clears WORKER_UNBOUND from the
rebind work item and it's bound to the CPU before returning to the
idle state.
To reduce CPU on/offlining overhead (as many embedded systems use it
for powersaving) and simplify the code path, workqueue is planned to
be modified to retain idle workers across CPU on/offlining. This
patch reimplements CPU online rebinding such that it can also handle
idle workers.
As noted earlier, due to the local wakeup requirement, rebinding idle
workers is tricky. All idle workers must be re-bound before scheduler
callbacks are enabled. This is achieved by interlocking idle
re-binding. Idle workers are requested to re-bind and then hold until
all idle re-binding is complete so that no bound worker starts
executing work item. Only after all idle workers are re-bound and
parked, CPU_ONLINE proceeds to release them and queue rebind work item
to busy workers thus guaranteeing scheduler callbacks aren't invoked
until all idle workers are ready.
worker_rebind_fn() is renamed to busy_worker_rebind_fn() and
idle_worker_rebind() for idle workers is added. Rebinding logic is
moved to rebind_workers() and now called from CPU_ONLINE after
flushing trustee. While at it, add CPU sanity check in
worker_thread().
Note that now a worker may become idle or the manager between trustee
release and rebinding during CPU_ONLINE. As the previous patch
updated create_worker() so that it can be used by regular manager
while unbound and this patch implements idle re-binding, this is safe.
This prepares for removal of trustee and keeping idle workers across
CPU hotplugs.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Currently, create_worker()'s callers are responsible for deciding
whether the newly created worker should be bound to the associated CPU
and create_worker() sets WORKER_UNBOUND only for the workers for the
unbound global_cwq. Creation during normal operation is always via
maybe_create_worker() and @bind is true. For workers created during
hotplug, @bind is false.
Normal operation path is planned to be used even while the CPU is
going through hotplug operations or offline and this static decision
won't work.
Drop @bind from create_worker() and decide whether to bind by looking
at GCWQ_DISASSOCIATED. create_worker() will also set WORKER_UNBOUND
autmatically if disassociated. To avoid flipping GCWQ_DISASSOCIATED
while create_worker() is in progress, the flag is now allowed to be
changed only while holding all manager_mutexes on the global_cwq.
This requires that GCWQ_DISASSOCIATED is not cleared behind trustee's
back. CPU_ONLINE no longer clears DISASSOCIATED before flushing
trustee, which clears DISASSOCIATED before rebinding remaining workers
if asked to release. For cases where trustee isn't around, CPU_ONLINE
clears DISASSOCIATED after flushing trustee. Also, now, first_idle
has UNBOUND set on creation which is explicitly cleared by CPU_ONLINE
while binding it. These convolutions will soon be removed by further
simplification of CPU hotplug path.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
POOL_MANAGING_WORKERS is used to ensure that at most one worker takes
the manager role at any given time on a given global_cwq. Trustee
later hitched on it to assume manager adding blocking wait for the
bit. As trustee already needed a custom wait mechanism, waiting for
MANAGING_WORKERS was rolled into the same mechanism.
Trustee is scheduled to be removed. This patch separates out
MANAGING_WORKERS wait into per-pool mutex. Workers use
mutex_trylock() to test for manager role and trustee uses mutex_lock()
to claim manager roles.
gcwq_claim/release_management() helpers are added to grab and release
manager roles of all pools on a global_cwq. gcwq_claim_management()
always grabs pool manager mutexes in ascending pool index order and
uses pool index as lockdep subclass.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Currently, WORKER_UNBOUND is used to mark workers for the unbound
global_cwq and WORKER_ROGUE is used to mark workers for disassociated
per-cpu global_cwqs. Both are used to make the marked worker skip
concurrency management and the only place they make any difference is
in worker_enter_idle() where WORKER_ROGUE is used to skip scheduling
idle timer, which can easily be replaced with trustee state testing.
This patch replaces WORKER_ROGUE with WORKER_UNBOUND and drops
WORKER_ROGUE. This is to prepare for removing trustee and handling
disassociated global_cwqs as unbound.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Workqueue used CPU_DYING notification to mark GCWQ_DISASSOCIATED.
This was necessary because workqueue's CPU_DOWN_PREPARE happened
before other DOWN_PREPARE notifiers and workqueue needed to stay
associated across the rest of DOWN_PREPARE.
After the previous patch, workqueue's DOWN_PREPARE happens after
others and can set GCWQ_DISASSOCIATED directly. Drop CPU_DYING and
let the trustee set GCWQ_DISASSOCIATED after disabling concurrency
management.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Currently, all workqueue cpu hotplug operations run off
CPU_PRI_WORKQUEUE which is higher than normal notifiers. This is to
ensure that workqueue is up and running while bringing up a CPU before
other notifiers try to use workqueue on the CPU.
Per-cpu workqueues are supposed to remain working and bound to the CPU
for normal CPU_DOWN_PREPARE notifiers. This holds mostly true even
with workqueue offlining running with higher priority because
workqueue CPU_DOWN_PREPARE only creates a bound trustee thread which
runs the per-cpu workqueue without concurrency management without
explicitly detaching the existing workers.
However, if the trustee needs to create new workers, it creates
unbound workers which may wander off to other CPUs while
CPU_DOWN_PREPARE notifiers are in progress. Furthermore, if the CPU
down is cancelled, the per-CPU workqueue may end up with workers which
aren't bound to the CPU.
While reliably reproducible with a convoluted artificial test-case
involving scheduling and flushing CPU burning work items from CPU down
notifiers, this isn't very likely to happen in the wild, and, even
when it happens, the effects are likely to be hidden by the following
successful CPU down.
Fix it by using different priorities for up and down notifiers - high
priority for up operations and low priority for down operations.
Workqueue cpu hotplug operations will soon go through further cleanup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: stable@vger.kernel.org
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
As discussed in
http://thread.gmane.org/gmane.linux.kernel/1249726/focus=1288990,
the capability introduced in 4d7e30d989
to govern EPOLLWAKEUP seems misnamed: this capability is about governing
the ability to suspend the system, not using a particular API flag
(EPOLLWAKEUP). We should make the name of the capability more general
to encourage reuse in related cases. (Whether or not this capability
should also be used to govern the use of /sys/power/wake_lock is a
question that needs to be separately resolved.)
This patch renames the capability to CAP_BLOCK_SUSPEND. In order to ensure
that the old capability name doesn't make it out into the wild, could you
please apply and push up the tree to ensure that it is incorporated
for the 3.5 release.
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
b43 with open firmware crashes mac80211 because
it changes the number of queues at runtime which,
while it was never really supported, now crashes
mac80211 due to the new hardware queue logic.
Fix this by detecting open vs. proprietary fw
earlier and registering with mac80211 with the
right number of queues.
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Cc: stable@vger.kernel.org (depends on commit a6f38ac3)
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Headers should really include all the needed prototypes, types, defines
etc. to be self-contained. This is a long-standing issue, but apparently
the new tracing code unearthed it (SMP=n is also a prerequisite):
In file included from fs/pstore/internal.h:4:0,
from fs/pstore/ftrace.c:21:
include/linux/pstore.h:43:15: error: field ‘read_mutex’ has incomplete type
While at it, I also added the following:
linux/types.h -> size_t, phys_addr_t, uXX and friends
linux/spinlock.h -> spinlock_t
linux/errno.h -> Exxxx
linux/time.h -> struct timespec (struct passed by value)
struct module and rs_control forward declaration (passed via pointers).
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using work_pending() to defer certain operations when
a HW-reset work has been queued is racy since the check
would return false when the work item is actually in
execution. Use SC_OP_HW_RESET instead to fix this race.
Also, unify the reset debug statistics maintenance.
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
When an interface in AP or P2P-GO mode is removed,
check whether a station interface is already present and
reconfigure the beacon timers etc. properly if it's
associated.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Currently, there are problems with how ANI is handled in
multi-VIF scenarios. This patch addresses them by unifying
the start/stop logic.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Remove unused variables, use a helper function to choose
the slot and reset beaconing status at one place.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Setup the beacon queue parameters after disabling
interrupts. Also, remove the redundant call in conf_tx()
for IBSS mode since the queue would be configured
with the appropriate cwmin/cwmax values when beaconing
is enabled.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In the tx_last_beacon() callback, mac80211's beaconing
status can be used instead. The beacon tasklet doesn't require
it because it is disabled when removing a slot.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
* The beaconing status routine is not required, since in
multi-VIF cases the HW beacon parameters should not be
re-configured.
* Remove SC_OP_TSF_RESET - when a beaconing interface comes
up the first time, the TSF has to be reset.
* Simplify ath9k_allow_beacon_config().
* Handle setting/clearing the SWBA interrupt properly.
* Remove the TSF mangling in IBSS mode, it is not required.
* General code cleanup.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Cleanup the messy logic dealing with station association
and disassociation.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
* Do not set/clear TSF when adding/deleting an interface.
This should be done when the BSS is set up and should also
take into account the existence of other interfaces.
* Set opmode explicitly.
* ANI setup needs to be decided based on multiple interfaces.
This can be done via the bss_info_changed() callback.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This patch revamps interface addition and deletion and simplifies
slot allocation. There is no need to setup the beacon buffer
in add/remove interface, remove this and use simple APIs for
assigning/deleting slots.
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
We found a deadlock in the handling of command failures/reset conditions.
For example:
1. Two commands are in the queue.
2. The first command is sent, but causes a timeout, which kicks off an
asynchronous device reset
3. The second command is queued (but not yet sent to the hardware)
4. The device reset kicks in, causing the if_usb disconnect handler to
set the "surprise removed" flag to be set as the device disappears
from the bus. This causes lbs_thread to stop processing things
("adapter removed; waiting to die"), not processing any further
commands, leaving the second queued command "in the air", causing a
deadlock.
Fix this by removing the surpriseremoved flag setting in if_usb. I can't
see any reason why this needs to be done so early. lbs_remove_card will set
this flag at an appropriate time - i.e. after all pending commands have
been completed or cancelled, avoiding this deadlock.
Signed-off-by: Daniel Drake <dsd@laptop.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Fail commands immediately when the request cannot be sent to the hardware.
This solves the following deadlock:
1. Two commands are in the queue.
2. The first command is sent, but causes a timeout, which kicks off an
asynchronous device reset
3. The second command is submitted to the device, and fails. The failure
is noted but the existing code waits for the timeout handler to take
care of the failure.
4. The device reset kicks in, causing the device "surprise removed" flag
to be set as the device disappears from the bus.
5. lbs_thread notes this and enters "adapter removed; waiting to die"
mode, without processing any further command timeouts.
While adjusting lbs thread logic to handle this situation may be one way
to fix this, it seems more practical to simplify handling of host_to_card
failure so that the commands are failed immediately without waiting for
more compliated timeout logic to kick in.
Signed-off-by: Daniel Drake <dsd@laptop.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
reg_notifier can be called before the interface is up.
Handle this correctly by storing the requested country code, then
apply the relevant configuration when the interface is brought up.
Signed-off-by: Daniel Drake <dsd@laptop.org>
Acked-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
AR9485, AR9330 and AR9340 are the chips that this is *NOT* supposed to be
applied on.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
- add an inline function for getting the correct modal EEPROM struct
- remove unnecessary indirection through ath9k_hw_ar9300_get_eeprom
access the relevant fields directly
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
If the aggregate size exceeds the TXOP limit, it leads to lots of unnecessary
hardware and software retries.
The previous 4ms frame limit table was completely undocumented, the commit
that updated it only vaguely referenced and equation from the standard,
but I've been unable to replicate its results.
Fix this by using a formula based on the code in ath_pkt_duration, which is
more likely to be correct for this case.
Reported-by: Dave Täht <dave.taht@gmail.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Prepare for using different queue size defaults for each AC.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In all those years apparently nobody noticed that the txop limit programmed
into the chip was off by a factor of 32 (!), probably because the VI and VO
queues aren't used that much aside from mgmt frames on VO.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The row/column sizes can be derived from the array argument within the macro
itself, which is less error prone. In a few cases the supplied column size
was actually wrong.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Use the EEPROM information to choose the right tx gain table
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Many chips are not able to deal with non-consecutive rx antenna selections
and respond with calibration errors, reset errors, etc.
When an antenna is selected as a tx antenna, also flag it for rx to avoid
chip issues.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This patch configures data rates to firmware using bitrate mask
provided by cfg80211.
Earlier we used to only update band information in this handler
which will be used later for ibss network. Due to recent
modifications in ibss join code we don't need to do that.
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Avinash Patil <patila@marvell.com>
Signed-off-by: Kiran Divekar <dkiran@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In mwifiex_set_rf_channel() ibss specific flags were unnecessarily
getting modified for infra and AP mode. This patch removes
mwifiex_set_rf_channel() function and adds equivalant code in infra,
ibss and AP path.
For ibss, now we are chosing band based on channel type and basic
rates provided in ibss join request. We can start ibss network in
A only, B only, G only, BG, BGN, AN mode.
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
1) Remove unnecessary wrapper functions.
2) Currently we don't have command to set Tx data rate, so
mwifiex_rate_ioctl_set_rate_value() function and related code
can be removed.
3) "ds_rate" filled by mwifiex_ret_tx_rate_cfg() is never used.
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
1) Recently we removed set_channel cfg80211 handler. Also, cfg80211
blocks ibss connection requests if ibss network is already started
/joined. Hence the code to restart ibss network in new channel
(mwifiex_drv_change_adhoc_chan() function) becomes redundant.
2) mwifiex_bss_set_channel() function is redundant. It does some
error checking and calculate adhoc start band and adhoc channel.
Cfg80211 already takes care of error checking and provides correct
channel information to the driver. Adhoc start band is already
calculated in mwifiex_set_rf_channel() function.
Other associated code is also removed in this patch.
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The old code was an accidental copy&paste of the 2.4 GHz version,
which doesn't work.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Commit d83579e2a5 incorporated some
changes from the vendor driver that made it newly important that the
calculated hardware version correctly include the CHIP_92D bit, as all
of the IS_92D_* macros were changed to depend on it. However, this bit
was being unset for dual-mac, dual-phy devices. The vendor driver
behavior was modified to not do this, but unfortunately this change was
not picked up along with the others. This caused scanning in the 2.4GHz
band to be broken, and possibly other bugs as well.
This patch brings the version calculation logic in parity with the
vendor driver in this regard, and in doing so fixes the regression.
However, the version calculation code in general continues to be largely
incoherent and messy, and needs to be cleaned up.
Signed-off-by: Forest Bond <forest.bond@rapidrollout.com>
Cc: Stable <stable@vger.kernel.org> [v3.2+]
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The mac80211 rate_index changed to be a u8, so
can't hold the negative error value properly.
Use a temporary variable for error checking.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Turns out every most standard Linux distributions enable
CONFIG_EXPERT, so use the shiny new CFG80211_CERTIFICATION_ONUS
which is meant by design to not be enabled by all Linux
distributions.
Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
