staging: mt29f_spinand: Export OF module alias information
The SPI core always reports the MODALIAS uevent as "spi:<modalias>"
regardless of the mechanism that was used to register the device
(i.e: OF or board code) and the table that is used later to match
the driver with the device (i.e: SPI id table or OF match table).
So drivers needs to export the SPI id table and this be built into
the module or udev won't have the necessary information to autoload
the needed driver module when the device is added.
But this means that OF-only drivers needs to have both OF and SPI id
tables that have to be kept in sync and also the dev node compatible
manufacturer prefix is stripped when reporting the MODALIAS. Which can
lead to issues if two vendors use the same SPI device name for example.
To avoid the above, the SPI core behavior may be changed in the future
to not require an SPI device table for OF-only drivers and report the
OF module alias. So, it's better to also export the OF table even when
is unused now to prevent breaking module loading when the core changes.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fbtft_par_dbg(DEBUG_SET_ADDR_WIN.. ) is repeated in every
set_addr_win() handler, this could be replicated by using the kernel
function tracer instead.
Mike Rapoport [Thu, 3 Sep 2015 05:13:10 +0000 (08:13 +0300)]
staging: fbtft: remove unused bl_ops from fbtft_unregister_backlight
The only usage of bl_ops variable in fbtft_unregister_backlight function
was assigment of a value to that variable, therefore the assignment and
the variable itself can be safely removed
Signed-off-by: Mike Rapoport <mike.rapoport@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Quoted strings should not be split to help text grep in the source.
All quoted strings that were split have thus been merged to one unique
quoted string each to follow the code style.
Raphaël Beamonte [Tue, 18 Aug 2015 16:58:12 +0000 (12:58 -0400)]
staging: rtl8192u: r8192U_core: remove forward declarations in .c file
Checkpatch was giving a "externs should be avoided in .c files" because
of these forward declarations. As these were not useful in this case,
they have been removed.
Clean-up the file by using a cleaner spacing around symbols and
words. Mostly use the automatic checkpatch whitespacing fixes.
This takes care of the consistent spacing errors reported by
checkpatch.
Raphaël Beamonte [Tue, 18 Aug 2015 16:58:06 +0000 (12:58 -0400)]
staging: rtl8192u: r8192U_core: fix missing struct leading to consistent spacing code style error
A missing struct keyword in variable declaration triggered a
need consistent spacing around '*' code style error. The struct
keyword thus has been added everywhere for the rtl8192_rx_info
struct, and therefore its typedef removed as not needed anymore.
Sudip Mukherjee [Tue, 18 Aug 2015 15:18:28 +0000 (20:48 +0530)]
staging: most: fix Makefile
The Makefile is including "drivers/media/video". But there is no such
directory in kernel tree. Since it is aim-v4l2 this might have been
"drivers/media/v4l2-core", but the Kconfig already mentions that it
depends on VIDEO_V4L2. So no need to mention that again in the Makefile.
Johnny Kim [Thu, 20 Aug 2015 07:32:53 +0000 (16:32 +0900)]
staging: wilc1000: change address to fixed value
The linux_wlan_init_test_config() is called once when net driver is loaded.
And because the pointer type of the pstrWFIDrv is changed with the interger
type, this patch replaces it with designated value instead of pointer.
Signed-off-by: Johnny Kim <johnny.kim@atmel.com> Signed-off-by: Tony Cho <tony.cho@atmel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Johnny Kim [Thu, 20 Aug 2015 07:32:51 +0000 (16:32 +0900)]
staging: wilc1000: add an argument for Handle_SetWfiDrvHandler
Similar to functions of same layer, this patch add an argument for
Handle_SetWfiDrvHandler function. As a result, the redundant typecasting is
removed.
Signed-off-by: Johnny Kim <johnny.kim@atmel.com> Signed-off-by: Tony Cho <tony.cho@atmel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Johnny Kim [Thu, 20 Aug 2015 07:32:50 +0000 (16:32 +0900)]
staging: wilc1000: use the real data type
This patch changes the type of gu8FlushedJoinReqDrvHandler with his real
data type becasue typecasting is not necessary. In result, typecasting
which is not necessary and some building warnings is removed.
Signed-off-by: Johnny Kim <johnny.kim@atmel.com> Signed-off-by: Tony Cho <tony.cho@atmel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Raphaël Beamonte [Wed, 19 Aug 2015 03:14:50 +0000 (23:14 -0400)]
staging: wilc1000: replace MALLOC_WILC_BUFFER() macro to avoid possible memory leak
The MALLOC_WILC_BUFFER() macro was using a return statement, and didn't
take care of possible memory leaks and subsequent bugs when it was failing
after succeeding some allocations. This patch corrects this behavior.
staging: comedi: usbduxfast: remove unnecessary clearing of private data
The clearing of the private data members in the (*detatch) is not
necessary. The comedi core is going to kfree the private data as soon
as the (*detach) finishes.
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>
The semaphore down/up in usbduxfast_attach_common() is not necessary.
This function is only called as part of the (*auto_attach) and does
not talk to the USB device.
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>
staging: comedi: usbduxfast: document strange ai maxdata
The subdevice 'maxdata' is typically a mask of the valid bits that can
be returned by a subdevice, (1 << bits) - 1. The analog inputs of this
device have a resolution of 12-bits so the 'maxdata' should be 0x0fff.
But, this hardware can produce a value of 0x1000 indicating an overflow
from the ADC. The comedilib library's comedi_to_phys() function will
then return NAN when this value is read from the hardware.
Add a comment to clarify the strage 'maxdata' value.
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>
staging: comedi: usbduxfast: TRIG_EXT is only valid for cmd->start_src
According to the code, and verified with the author Bernd Porr, the
TRIG_EXT trigger is only valid for the cmd->start_src. Remove this
trigger from the cmd->scan_begin_src and cmd->convert_src.
The only valid scan_begin_src is then TRIG_FOLLOW and the convert_src
is only TRIG_TIMER. Remove the redundant unique trigger checks in Step
2b of the (*do_cmdtest) as well as any other unnecessary checks of
those trigger sources.
Also remove the incorrect comment for a cmd->stop_src of TRIG_EXT.
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>
The private data 'devpriv' is allocated in the (*auto_attach) before
the subdevices are initialized and the (*auto_attach) has to complete
successfully before the core will call any of the subdevice operations.
Because of this the (!devpriv) checks in the subdevice code can never
fail. Remove them.
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>
staging: comedi: usbduxfast: only chanlist of 1, 2, 3, or 16 are supported
The analog input (*do_cmd) only supports channel lists of 1, 2, 3, or 16
channels. Add a check for this to usbduxfast_ai_check_chanlist() and remove
the check from usbduxfast_ai_cmd().
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>
staging: comedi: usbduxfast: chanlist check is Step 5 of (*do_cmdtest)
The channel list should be checked in Step 5 of the (*do_cmdtest) not
as part of the (*do_cmd). Factor the check out of usbduxfast_ai_cmd().
Tidy up the factored out code. The channel number 'i' will never be
greater than NUMCHANNELS due to the subdevice setup and the checks
done in the code. The up/down of the semaphore is also not needed
because the (*do_cmdtest) never actually tries to access the hardware.
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>
staging: comedi: s526: remove #if 0'ed out code block
This #if 0'ed out code is not necessary. It's actually just a copy
of the psuedo code in the S526 Hardware Manual that shows an example
of a counter application. Remove it.
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>
This (*insn_config) does not follow the comedi core API. It also
would not work as expected.
It appears to be trying to configure the analog input subdevice so
that the (*insn_read) would read multiple channels (data[0]) and
optionally enable the 15us delay (data[1]) needed for the multiplexor
to change channels between samples.
Unfortunately, the comedi core expects (*insn_read) operations to
return 1 or more samples for a single channel, which is what the
(*insn_read) in this driver does.
The (*insn_config) is also enabling the analog input end-of-conversion
interrupt. This isn't needed, and might be a problem since the driver
does not currently request and interrupt. The enable bit does not
need to be set for the end-of-conversion to occur in the interrupt
status register.
Remove the (*insn_config) and modify the (*insn_read) to automatically
handle the 15us delay when needed.
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>
staging: comedi: s526: wait for analog output conversions to complete
It's possible for the user to write more than one value to an analog
output channel with the (*insn_write) operation. Use comedi_timeout()
to check the interrupt status register to ensure that each data
conversion is complete before writing the next value.
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>
projects / linux-beck.git / log