- 01 Aug, 2013 11 commits
-
-
H Hartley Sweeten authored
All the boards supported by this driver that have analog outputs use the same function to handle the (*insn_write) for the subdevice. They all also have 12-bit resolution (maxdata = 0x0fff). Remove the 'ao' and 'ao_nbits' members from the boardinfo and replace them with a simple bit-field flag 'has_ao'. Tidy up the subdevice init a bit as well as the (*insn_write) function. 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>
-
H Hartley Sweeten authored
All the boards supported by this driver have digital outputs and all of then use the same function to handle the (*insn_bits) for the subdevice. Remove the 'do_' member from the boardinfo and always initialize the digital output subdevice during the board attach. Tidy up the subdevice init a bit as well as the (*insn_bits) function. Remove SDF_READABLE from the subdev_flags. This is a pure digital output subdevice and the outputs are not readable. Signed-off-by:
-
H Hartley Sweeten authored
All the boards supported by this driver have digital inputs and all of then use the same function to handle the (*insn_bits) for the subdevice. Remove the 'di' member from the boardinfo and always initialize the digital input subdevice during the board attach. Tidy up the subdevice init a bit as well as the (*insn_bits) function. The (*insn_bits) function does not need to clear data[0]. Signed-off-by:
-
H Hartley Sweeten authored
The subdevice functions that used this boardinfo can use the subdevice 's->maxdata' value instead. Tidy up those functions. Change the boardinfo so that the calculation is not needed during the attach. Signed-off-by:
-
H Hartley Sweeten authored
All the boards supported by this driver have analog inputs and all of then use the same function to handle the (*insn_read) for the subdevice. Remove the 'ai' member from the boardinfo and always initalize the analog input subdevice during the board attach. Tidy up the subdevice init a bit. Signed-off-by:
-
H Hartley Sweeten authored
DMA support is optional in this driver. To enable it, the user passes the desired DMA channel during the board attach. A timer is then used by the driver to periodically get the analog input data from the DMA. Since the DMA support is optional, modify the init code so that it only fails if the DMA channel is unavailable or the DMA buffers can not be allocated. Don't fail just because the user passed an invalid DMA channel (only DMA channels 1 and 3 are valid). Remove the printk() noise about the DMA. Change the printk() when the request_dma() fails into a dev_err(). Move the timer initialization so it's only setup if DMA is available. It's not needed otherwise. Also, only hook up the subdevice command support functions if DMA is available. This allows removing a couple sanity checks in the command support. Signed-off-by:
-
H Hartley Sweeten authored
As noted in the driver, a timer is always used by this driver to handle the DMA because samples could be dropped while waiting for the DMA done interrupt. Remove the irq setup code as well as the interrupt handler. Signed-off-by:
-
H Hartley Sweeten authored
As noted in the driver, a timer is always used by this driver to handle the DMA due to buggy DMA controllers and the lack of a hardware fifo in some of the supported boards. Remove all the non-timer mode specific code. Signed-off-by:
-
Jens Frederich authored
The FB_OLPC_DCON help section is to short according to checkpatch.pl. We want more information about the controller type, its task, its video pipeline position and so on. There are no style issues, remove checkpatch.pl TODO entry. Signed-off-by:
Jens Frederich <jfrederich@gmail.com> Acked-by:
Andres Salomon <dilinger@queued.net> Signed-off-by:
-
Li Zefan authored
The debug file is opened with single_open(), but there's no single_release(). Signed-off-by:
Li Zefan <lizefan@huawei.com> Signed-off-by:
-
Li Zefan authored
The debug file is opened with single_open(), but there's no single_release(). Signed-off-by:
-
- 29 Jul, 2013 29 commits
-
-
Greg Kroah-Hartman authored
The driver uses the tty layer, so explicitly say that, to prevent randomconfig build errors. Reported-by:
Randy Dunlap <rdunlap@infradead.org> Signed-off-by:
-
Aaro Koskinen authored
Reformat the high-level documentation to human readable plain text by removing HTML and other tags. Signed-off-by:
Aaro Koskinen <aaro.koskinen@iki.fi> Signed-off-by:
-
Aaro Koskinen authored
Delete CVS keyword markers. Signed-off-by:
-
Aaro Koskinen authored
Convert "@return" to "Returns: ". Signed-off-by:
-
Aaro Koskinen authored
Convert "@param name" inside comments to "@name:". Signed-off-by:
-
Aaro Koskinen authored
Delete duplicated function call documentation from the .h file. These functions are just internal anyway, and the .h file will be completely removed in the future when the driver will be made a single self-contained .c file. Signed-off-by:
-
Aaro Koskinen authored
Delete "@INTERNAL" from comments. Signed-off-by:
-
Aaro Koskinen authored
Reformat long comments according to multi-line comment style. Signed-off-by:
-
Aaro Koskinen authored
Fix most of the whitespace issues reported by checkpatch: - use tabs for indentation - brace placement vs. newlines - (foo_t*) -> (foo_t *) - i=0 -> i = 0 Signed-off-by:
-
Aaro Koskinen authored
Delete a redundant function. Signed-off-by:
-
Aaro Koskinen authored
Replace cvmx_le16_to_cpu() with le16_to_cpu(). Signed-off-by:
-
Aaro Koskinen authored
Just use udelay(). Signed-off-by:
-
Aaro Koskinen authored
Replace long udelays with mdelays. Signed-off-by:
-
Aaro Koskinen authored
Just use likely/unlikely. Signed-off-by:
-
Aaro Koskinen authored
Use generic Linux error codes to replace cvmx_usb_status_t. The conversion table: Before After cvmx_usb_status_t int CVMX_USB_SUCCESS 0 CVMX_USB_INVALID_PARAM -EINVAL CVMX_USB_NO_MEMORY -ENOMEM CVMX_USB_BUSY -EBUSY CVMX_USB_TIMEOUT -ETIMEDOUT CVMX_USB_INCORRECT_MODE Deleted (was never used) Signed-off-by:
-
Aaro Koskinen authored
Delete __cplusplus. Signed-off-by:
-
Aaro Koskinen authored
Delete redundant example code found in comments. It's already there in the driver proper. Signed-off-by:
-
Aaro Koskinen authored
Delete redudant debug flags and dead code behind them. Signed-off-by:
-
Aaro Koskinen authored
Delete call, parameter, and return value tracing. Signed-off-by:
-
Aaro Koskinen authored
Use CVMX_BOARD_TYPE_UBNT_E100 for EdgeRouter Lite. Signed-off-by:
-
Aaro Koskinen authored
The USB driver should depend on SOC instead of CPU. Signed-off-by:
-
Lilis Iskandar authored
Fixed a pointer asterisk placement issue Signed-off-by:
Lilis Iskandar <veeableful@gmail.com> Signed-off-by:
-
Lilis Iskandar authored
Fixed code indent issues Signed-off-by:
-
Lilis Iskandar authored
Fixed C99 comments Signed-off-by:
-
Lilis Iskandar authored
Fixed brace issues Signed-off-by:
-
Lilis Iskandar authored
This one fixes spacing/tabbing issues. Signed-off-by:
-
Marc Dietrich authored
On resume the slave controller is reinitialized. The tegra i2c master controller disables the clock at the end of the initialiation, propably to save some power, and enables it again on each transfer. We don't do this yet and also forgot to enable the clock on resume. Fix this copy-paste error by not disabling the clock after initialization. This didn't striked us yet because suspend/resume hasn't landed in mainline yet, but will soon. Signed-off-by:
Marc Dietrich <marvin24@gmx.de> Signed-off-by:
-
Won Kang authored
Conflict resolution for function name set_endian when building for powerpc The name changed to gdm_set_endian Reported-by:
kbuild test robot <fengguang.wu@intel.com> Signed-off-by:
Won Kang <wonkang@gctsemi.com> Signed-off-by:
-
Wei Yongjun authored
Remove duplicated include. Signed-off-by:
Wei Yongjun <yongjun_wei@trendmicro.com.cn> Signed-off-by:
-
