When an array is being reshaped to change the number of devices,
the two halves can be differently degraded. e.g. one could be
missing a device and the other not.
So we need to be more careful about calculating the 'degraded'
attribute.
Instead of just inc/dec at appropriate times, perform a full
re-calculation examining both possible cases. This doesn't happen
often so it not a big cost, and we already have most of the code to
do it.
Signed-off-by: NeilBrown <neilb@suse.de>
We have a variable 'mddev' in this function, but repeatedly get the
same value by dereferencing bitmap->mddev.
There is room for simplification here...
Signed-off-by: NeilBrown <neilb@suse.de>
The info is already available in /proc/mdstat and /sys/block in
an accessible form so there is no point in putting a road-block in
the ioctl for information gathering.
Signed-off-by: NeilBrown <neilb@suse.de>
commit d0a4bb4927 introduced a
regression which is annoying but fairly harmless.
When writing to an array that is undergoing recovery (a spare
in being integrated into the array), writing to the array will
set bits in the bitmap, but they will not be cleared when the
write completes.
For bits covering areas that have not been recovered yet this is not a
problem as the recovery will clear the bits. However bits set in
already-recovered region will stay set and never be cleared.
This doesn't risk data integrity. The only negatives are:
- next time there is a crash, more resyncing than necessary will
be done.
- the bitmap doesn't look clean, which is confusing.
While an array is recovering we don't want to update the
'events_cleared' setting in the bitmap but we do still want to clear
bits that have very recently been set - providing they were written to
the recovering device.
So split those two needs - which previously both depended on 'success'
and always clear the bit of the write went to all devices.
Signed-off-by: NeilBrown <neilb@suse.de>
Before performing a recovery we try to remove any spares that
might not be working, then add any that might have become relevant.
Currently we abort on the first spare that cannot be added.
This is a false optimisation.
It is conceivable that - depending on rules in the personality - a
subsequent spare might be accepted.
Also the loop does other things like count the available spares and
reset the 'recovery_offset' value.
If we abort early these might not happen properly.
So remove the early abort.
In particular if you have an array what is undergoing recovery and
which has extra spares, then the recovery may not restart after as
reboot as the could of 'spares' might end up as zero.
Reported-by: Anssi Hannula <anssi.hannula@iki.fi>
Signed-off-by: NeilBrown <neilb@suse.de>
While reshaping a degraded array (as when reshaping a RAID0 by first
converting it to a degraded RAID4) we currently get confused about
which devices are in_sync. In most cases we get it right, but in the
region that is being reshaped we need to treat non-failed devices as
in-sync when we have the data but haven't actually written it out yet.
Reported-by: Adam Kwolek <adam.kwolek@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>
commit d70ed2e4fa
broke hot-add to a linear array.
After that commit, metadata if not written to devices until they
have been fully integrated into the array as determined by
saved_raid_disk. That patch arranged to clear that field after
a recovery completed.
However for linear arrays, there is no recovery - the integration is
instantaneous. So we need to explicitly clear the saved_raid_disk
field.
Signed-off-by: NeilBrown <neilb@suse.de>
Convert rtllib from registering the crypt drivers against rtllib_crypt
and instead register the against lib80211. The crypto functions have
R- prepended (R-CCMP, R-TKIP, R-WEP) so they will not clash with the
lib80211 versions.
We cannot use the lib80211 crypt drivers since the rtl8192e has some
hardware support that is not handled by the lib80211 crypt drivers.
Signed-off-by: Sean MacLennan <seanm@seanm.ca>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Convert rtllib_crypt_data to lib80211_crypt_data and
rtllib_crypt_ops to lib80211_crypt_ops.
This is almost a 1:1 replacement, only extra_prefix_len and
extra_postfix_len changed.
Signed-off-by: Sean MacLennan <seanm@seanm.ca>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Add lib80211.h header file to rtllib.h and get it compiling.
Signed-off-by: Sean MacLennan <seanm@seanm.ca>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This silently was working for many years and stopped working on
Niagara-T3 machines.
We need to set the MSIQ to VALID before we can set it's state to IDLE.
On Niagara-T3, setting the state to IDLE first was causing HV_EINVAL
errors. The hypervisor documentation says, rather ambiguously, that
the MSIQ must be "initialized" before one can set the state.
I previously understood this to mean merely that a successful setconf()
operation has been performed on the MSIQ, which we have done at this
point. But it seems to also mean that it has been set VALID too.
Signed-off-by: David S. Miller <davem@davemloft.net>
Uncached/writecombine buffers need to be clean before creating uc/wc
userspace mappings, lest dirty cache lines fall down and corrupt the
contents of the buffer.
Signed-off-by: Rob Clark <rob@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch removes the int return of function INTnsProcessData, because
nobody uses this return.
Signed-off-by: Marcos Paulo de Souza <marcos.mage@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Removed remnants of old function definitions which have since been
removed in earlier cleanup work.
Signed-off-by: Bart Westgeest <bart@elbrys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The usbip_xmit function supported sending and receiving data, however
the sending part of the function was never used/executed. Renamed the
function to usbip_recv, and removed the unused code.
Signed-off-by: Bart Westgeest <bart@elbrys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
These callbacks are currently used by the individual buffer implementations to
ensure that the request_update callback is not issued while the buffer is in use.
But the core already provides sufficient measures to prevent this from happening
in the first place. So it is safe to remove them.
There is one functional change due to this patch. Since the buffer is no longer
marked as in use when the chrdev is opened, it is now possible to enable the
buffer while it is opened. This did not work before, because mark_param_change
did fail if the buffer was marked as in use.
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Right now we have a mark_param_change callback in the buffer access
functions struct, which should be called whenever the parameters (length,
bytes per datum) of the buffer change. But it is only called when the user
changes the buffer size, not when the bytes per datum change. Additionally each
buffer implementation already keeps track internally whether its parameters
have changed, making the call to mark_param_change after changing the buffer
length redundant. Since each buffer implementation knows best when one of its
parameters has changed just make tracking of this internal and drop the
mark_param_change callback.
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Currently none of the buffer implementations implements the enable() or
is_enable() nor does core code try to call these. So it is safe to remove them.
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The flag is only cleared, never set or tested, so it is safe to remove it.
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Our buffer implementation does not support multiple concurrent readers. So we
have to ensure that a device is only opened once at a time. So do the same thing
we do for the event fd and introduce a per device busy flag. The flag gets set
when opening the device and gets cleared when closing the device. If a open is
attempted while the busy flag is set we return -EBUSY.
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The buffer buffer storage is only update when enabling the buffer. Changing the
buffer size while the buffer is enabled will confuse the buffer in regard to
its actual buffer size and can cause potential memory corruption. Thus it is
only safe to modify the buffer size when the buffer is disabled.
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Currently we only disallow changing the scan elements, while the buffer is
enabled, in triggered buffer mode. This patch changes it to disallow it for all
buffered modes. Disabling or enabling scan elements while the buffer is enabled
will cause undefined behavior since the reader will not be able to tell samples
with the new and old scan element set apart and thus wont be able to extract
any meaningful data from the buffer.
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch removes the return of the function CARDbSetMediaChannel, that
always return TRUE value.
Signed-off-by: Marcos Paulo de Souza <marcos.mage@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This function always return TRUE, and it is not used by the funtions
who calls it.
Signed-off-by: Marcos Paulo de Souza <marcos.mage@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch fixes some 80 chatacters limit warnings in the lowmemorykiller.c file
Signed-off-by: Marco Navarra <fromenglish@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch fixes a simple tab-space warning in binder.h found by checkpatch tool
Signed-off-by: Marco Navarra <fromenglish@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
gma_resume_display() wants 'struct pci_dev *' as it's parameter, so lets pass dev->pdev instead of dev. Fixes the following warning.
drivers/staging/gma500/power.c: In function ‘gma_power_begin’:
drivers/staging/gma500/power.c:269:3: warning: passing argument 1 of ‘gma_resume_display’ from incompatible pointer type
drivers/staging/gma500/power.c:99:13: note: expected ‘struct pci_dev *’ but argument is of type ‘struct drm_device *’
Signed-off-by: Rakib Mullick <rakib.mullick@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Variable IoBuffer.InputLength is chosen from userspace,
and can therefore be less than the intended size. In this
case,the memory from the kmalloc call is eventually cast
to a PBULKWRM_BUFFER. If the IoBuffer.InputLength does not
meet the minimum size of PBULKWRM_BUFFER, then we will get
a kernel Oops. To resolve this issue, this patch verifies
IoBuffer.InputLength meets the minimum size before invoking
the kmalloc call.
Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This ioctl, IOCTL_BCM_GET_DRIVER_VERSION, is
responsible for sending the driver version
to userspace. However, the requested size stored
in IoBuffer.OutputLength may be incorrect.
Therefore, we altered the code to send the
exact length of the version, plus one for the
null character.
Signed-off-by: Kevin McKinney <klmckinney1@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* 'usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
USB: Fix usb/isp1760 build on sparc
usb: gadget: epautoconf: do not change number of streams
usb: dwc3: core: fix cached revision on our structure
usb: musb: fix reset issue with full speed device
They don't need to disable interrupts anymore, we only run in process
context now.
Acked-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
i810 uses a userspace provided mmio map using the drm core map
infrastructure. By the time we reach lastclose, this is all gone
and our mmio_map pointer points at freed memory. Depending upon
luck that still works, most often it just oopses.
Aside: drm maps aren't refcounted, so userspace can essentially oops
the kernel any time it wants to. Who cares.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
My dear old i815 always hits the deadlocked on reclaim_buffers
warning. Switch over to the idlelock duct-tape on hope that
works better. I've fired up my i815 and now closing glxgears doesn't
take 5 seconds anymore. \o/
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We simply say that regular this_cpu use must be safe regardless of
preemption and interrupt state. That has no material change for x86
and s390 implementations of this_cpu operations. However, arches that
do not provide their own implementation for this_cpu operations will
now get code generated that disables interrupts instead of preemption.
-tj: This is part of on-going percpu API cleanup. For detailed
discussion of the subject, please refer to the following thread.
http://thread.gmane.org/gmane.linux.kernel/1222078
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
LKML-Reference: <alpine.DEB.2.00.1112221154380.11787@router.home>
