Daniel Vetter [Wed, 10 Jul 2013 12:11:45 +0000 (14:11 +0200)]
drm/imx: kill firstopen callback
This thing seems to do some kind of delayed setup. Really, real kms
drivers shouldn't do that at all. Either stuff needs to be dynamically
hotplugged or the driver setup sequence needs to be fixed.
This patch here just moves the setup at the very end of the driver
load callback, with the locking adjusted accordingly.
v2: Also move the corresponding put from ->lastclose to ->unload.
imx-drm: imx-tve: Remove unneeded check for platform_get_resource()
As devm_ioremap_resource() is used on probe, there is no need to explicitly
check the return value from platform_get_resource(), as this is something that
devm_ioremap_resource() takes care by itself.
drivers/staging/comedi/drivers/addi_apci_2032.c: In function 'apci2032_auto_attach':
drivers/staging/comedi/drivers/addi_apci_2032.c:328:3: error: implicit declaration of function 'kzalloc' [-Werror=implicit-function-declaration]
drivers/staging/comedi/drivers/addi_apci_2032.c:328:11: warning: assignment makes pointer from integer without a cast [enabled by default]
drivers/staging/comedi/drivers/addi_apci_2032.c: In function 'apci2032_detach':
drivers/staging/comedi/drivers/addi_apci_2032.c:350:3: error: implicit declaration of function 'kfree' [-Werror=implicit-function-declaration]
drivers/staging/comedi/drivers/amplc_pci224.c: In function 'pci224_attach_common':
drivers/staging/comedi/drivers/amplc_pci224.c:1289:2: error: implicit declaration of function 'kmalloc' [-Werror=implicit-function-declaration]
drivers/staging/comedi/drivers/amplc_pci224.c:1289:23: warning: assignment makes pointer from integer without a cast [enabled by default]
drivers/staging/comedi/drivers/amplc_pci224.c:1296:24: warning: assignment makes pointer from integer without a cast [enabled by default]
drivers/staging/comedi/drivers/amplc_pci224.c:1303:25: warning: assignment makes pointer from integer without a cast [enabled by default]
drivers/staging/comedi/drivers/amplc_pci224.c:1348:42: warning: assignment makes pointer from integer without a cast [enabled by default]
drivers/staging/comedi/drivers/amplc_pci224.c: In function 'pci224_detach':
drivers/staging/comedi/drivers/amplc_pci224.c:1474:3: error: implicit declaration of function 'kfree' [-Werror=implicit-function-declaration]
staging: comedi: ii_pci20kc: cleanup the ai subdevice
The PCI-200341M-1 module can both provide an analog input subdevice
in this driver. The module provides four differential input channels
with four programmable gains.
Currently the gain is configured as an option passed when the board
is attached. Rewrite the ai subdevice support functions to properly
handle the programmable gain.
For aesthetics, redefine the memory map for the module.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: ii_pci20kc: cleanup the ao subdevice
The PCI-20006M-1 and PCI-20006M-2 modules can both provide an
analog output subdevice in this driver. Fix the module init code
to detect both of them.
The analog output can support +/-5, 0-10, or +/-10 volt ranges
depending on jumpers on the board. Report all the ranges to the
user instead of relying on the configuration options to set the
range. The user can then select the appropriate range depending
on how they have configured the hardware.
Tidy up the (*insn_{read,write})() functions to work like the
comedi core expects.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: ii_pci20kc.c: cleanup the dio subdevice
For asethetics, redefine the register map for the carrier board.
The 'PCI*' names lead one to think this is a PCI board not a
legacy ISA board.
Remove the #if 0'd out pci20xxx_{do,di}() functions. They are not
needed.
Rename, and fix, the pci20xxx_dio_config() function. The control
and direction registers are not readable. Use the new register map
defines instead of the crazy bit shifts.
Rename, and fix, the pci20xxx_dio_insn_config() function. This
function should treat data[0] as the 'instruction' to handle not
just use it to determine if the port is an input or output.
Rename, and tix, the pci20xxx_dio_insn_bits() function. The 'state'
only needs to be updated if the 'mask' indicates.
For aesthetics, abosrb the pci20xxx_dio_init() into the (*attach)
function.
Fix the carrier board id detect in the (*attach). There are two carrier
board types, one with 32 dio channels and one without. Use this info
when setting up the dio subdevice.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: ii_pci20kc: use comedi_alloc_spriv()
For aesthetic reasons, use the helper function to allocate the
subdevice private data instead of hanging it on the device
private data. The core will free the memory during the detach.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: range: tidy up comedi_check_chanlist()
The only difference in the if() and else if() check of the chanlist
is the source of the range table length. Consolidate the checks to
make the function a bit more concise.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Reviewed-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For aesthetic reasons, pass the comedi_device pointer to this function
instead of the private data pointer. Rename the local variable used
for the private data pointer to the comedi "norm".
Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For aesthetic reasons, pass the comedi_device pointer to this function
instead of the private data pointer. Rename the local variable used
for the private data pointer to the comedi "norm".
Remove the unnecessary sanity check of the private data pointer. This
function can only be called is the private data was allocated during
the attach.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: tidy up receive_dux_commands()
For aesthetic reasons, pass the comedi_device pointer to this function
instead of the private data pointer. Rename the local variable used
for the private data pointer to the comedi "norm".
Add a local variable for the usb_device pointer to tidy up the
usb_bulk_msg() call.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: tidy up send_dux_commands()
For aesthetic reasons, pass the comedi_device pointer to this function
instead of the private data pointer. Rename the local variable used
for the private data pointer to the comedi "norm".
Add a local variable for the usb_device pointer to tidy up the
usb_bulk_msg() call.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: tidy up usbduxsub_ao_isoc_irq()
Rename the local variables to the comedi "norm".
The comedi_subdevice in this function is actually the dev->write_subdev
that was initialized in the attach. Use that instead of accessing the
dev->subdevices array directly.
Use dev->class_dev as the device for all dev_printk() messages.
Instead of using the 'comedidev' back pointer in the private data, use
the comedi_device 'dev' that we already have.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: tidy up usbduxsub_ai_isoc_irq()
Rename the local variables to the comedi "norm".
The comedi_subdevice in this function is actually the dev->read_subdev
that was initialized in the attach. Use that instead of accessing the
dev->subdevices array directly.
Use dev->class_dev as the device for all dev_printk() messages.
Instead of using the 'comedidev' back pointer in the private data, use
the comedi_device 'dev' that we already have.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: absorb usbdux_attach_common into caller
This function is only called by usbdux_auto_attach(), absorb it.
Also, there is no reason to down/up the semaphore during the attach.
None of the subdevices are functioning yet so there are no commands
being sent to the usb device.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This define enables printing of the 'dux_commands' that is sent to
the usb device in send_dux_commands(). This type of development
debug should not be left in the final driver.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The tidy_up() function is only called by the (*detach). That function
unlinks any running urbs and frees all the allocated urbs and buffers
used by the driver.
Rename tidy_up() to usbdux_free_usb_buffers() and move all the parts
that don't deal with the freeing of the buffers directly into the
(*detach).
Also, remove all the unnecessary clearing of the pointers. The comedi
core will kfree() the private data after calling the (*detach).
Foe aesthetic reasons, do the kfree()'ing of the buffers and urbs in
the reverse order that they were allocated.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the comedi_driver (*auto_attach) fails, the comedi core will call
the (*detach) function to do any cleanup. It's not necessary to do
the cleanup in the (*auto_attach).
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: push usb (*probe) into comedi (*auto_attach)
The usb_driver (*probe) calls comedi_usb_auto_config() after finding a
free slot for the static private data and doind some initial allocation
and setup. The comedi_usb_auto_config() will then call the comedi_driver
(*auto_attach).
Move all the probe/auto_attach into the comedi_driver and just have the
usb_driver call comedi_usb_auto_config(). This allows the comedi_driver
to allocate the private data and removes the need for the static array
and the semaphore that protects it.
Since all the probe/auto_attach is now handled by the comedi_driver, we
can also remove the 'attached' and 'probed' information from the private
data. The subdevice functions will only be hooked up if the usb device
has been probed and the comedi device is attached.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: push usb (*disconnect) into comedi (*detach)
The usb_driver (*disconnect) calls comedi_usb_auto_unconfig() which will
call the comedi_driver (*detach). Just move all the disconnect/detach
into the comedi_driver and use comedi_usb_auto_unconfig() directly for
the (*disconnect).
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: move usb buffer allocation into new function
Move all the usb buffer allocation code in the usb_driver (*probe)
into a new function, usbdux_alloc_usb_buffers(). This allows tidying
up the error path in the (*probe).
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: usbdux: tidy up usbdux_usb_probe()
Use a pointer to the struct usbdux_private data being configured
in the probe. Also use a pointer to the struct urb being setup in
the buffer allocations.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add BLOCK depends in Kconfig for LUSTRE to fix this:
drivers/staging/lustre/lustre/fid/../include/linux/lustre_compat25.h:117:2:
error: implicit declaration of function ‘unregister_blkdev’
This isn't needed, userspace can trigger off of the device id if they
need to do something "special" for different devices in special modes
depending on the device's endian issues.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Cc: Won Kang <wonkang@gctsemi.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Joe Perches [Wed, 24 Jul 2013 21:13:03 +0000 (14:13 -0700)]
staging: gdm724x: Update logging
Make the logging prefixes match the module names
by adding #define pr_fmt(fmt) KBUILD_MODNAME and
converting printks to netdev_<level> when a
struct netdevice * is available and and pr_<level>
when not.
Remove embedded prefixes.
Add missing terminating newlines.
Remove an unnecessary rx-dropped message.
Remove an unnecessary temporary variable.
Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: gdm724x: gdm_tty: fix tty api build warnings
This fixes the build warnings due to changes in the tty api. Note, I
don't know if the reference counting is correct, but at least there are
no more build warnings.
Also the tty code needs to be reworked, no static arrays of tty devices
and the like...
Cc: Won Kang <wonkang@gctsemi.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Liu Xuezhao [Wed, 24 Jul 2013 18:08:17 +0000 (02:08 +0800)]
staging/lustre/obdclass: add obd_target.h
Move some server-side data structures form obd.h to obd_target.h. To
remove some code dependencies use HAVE_SERVER_SUPPORT to protect some
llog/fsfilt_* functions.
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-1330
Lustre-change: http://review.whamcloud.com/2678 Signed-off-by: Liu Xuezhao <xuezhao.liu@emc.com> Signed-off-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Peng Tao <bergwolf@gmail.com> Reviewed-by: Mike Pershin <mike.pershin@intel.com> Signed-off-by: Peng Tao <tao.peng@emc.com> Signed-off-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
John L. Hammond [Wed, 24 Jul 2013 17:17:30 +0000 (01:17 +0800)]
staging/lustre/obdclass: use a dummy structure for lu_ref_link
Move the definition of struct lu_ref_link to lu_ref.h. If USE_LU_REF
is not defined then define it to be the empty struct. Change the
return type of lu_ref_add() and lu_ref_add_atomic() to void. Add
lu_ref_add_at() taking same arguments as lu_ref_add() togerther with a
pointer to a struct lu_ref_link and returning void. Adjust all
structures containing a lu_ref_link pointer to contain a struct
lu_ref_link instead. Use lu_ref_add_at() and lu_ref_del_at() to
handle embedded lu_ref_links.
[Original patch mainly changes lu_ref.[ch] but in upstream client we don't
have lu_ref code in the hope of moving to kobject refcounting. So the patch
only picks the other pieces of original patch, in order to keep code in
sync and make it easier to port future patches.]
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3059
Lustre-change: http://review.whamcloud.com/5920 Signed-off-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com> Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by: Mike Pershin <mike.pershin@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Peng Tao <tao.peng@emc.com> Signed-off-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging/lustre/llite: fix 'data race condition' issues of lli_flags
Fix 'data race condition' defects found by Coverity version
6.5.0:
Data race condition (MISSING_LOCK)
Accessing variable without holding lock. Elsewhere,
this variable is accessed with lock held.
lli->lli_flags need to be protected by lli->lli_lock.
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2744
Lustre-change: http://review.whamcloud.com/6571 Signed-off-by: Sebastien Buisson <sebastien.buisson@bull.net> Reviewed-by: Fan Yong <fan.yong@intel.com> Reviewed-by: Keith Mannthey <keith.mannthey@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Peng Tao <tao.peng@emc.com> Signed-off-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>