On 13/08/15 10:41, Ian Abbott wrote:
On 12/08/15 21:25, H Hartley Sweeten wrote:The watchdog is stopped in apci3501_write_insn_timer() by writing a 0 to the timer control register. There is no need to read the register first and mask it (as done when the timer is used as a timer). Reported-by: coverity (CID 1227052) Signed-off-by: H Hartley Sweeten <hsweeten@xxxxxxxxxxxxxxxxxxx> Cc: Ian Abbott <abbotti@xxxxxxxxx> Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/staging/comedi/drivers/addi-data/hwdrv_apci3501.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/staging/comedi/drivers/addi-data/hwdrv_apci3501.c b/drivers/staging/comedi/drivers/addi-data/hwdrv_apci3501.c index 1f2f781..e12b2bc 100644 --- a/drivers/staging/comedi/drivers/addi-data/hwdrv_apci3501.c +++ b/drivers/staging/comedi/drivers/addi-data/hwdrv_apci3501.c @@ -102,8 +102,6 @@ static int apci3501_write_insn_timer(struct comedi_device *dev, /* Enable the Watchdog */ outl(ul_Command1, dev->iobase + APCI3501_TIMER_CTRL_REG); } else if (data[1] == 0) { /* Stop The Watchdog */ - ul_Command1 = inl(dev->iobase + APCI3501_TIMER_CTRL_REG); - ul_Command1 = ul_Command1 & 0xFFFFF9FEUL; outl(0x0, dev->iobase + APCI3501_TIMER_CTRL_REG); } else if (data[1] == 2) { ul_Command1 = inl(dev->iobase + APCI3501_TIMER_CTRL_REG);I wonder if that was a bug in the original code. Shouldn't it have just cleared bit 0 to stop the watchdog?
Never mind. Configuring the timer as a watchdog clears the ACPI3501_TIMER_CTRL_REG anyway, so there's no need to preserve the other bits.
-- -=( Ian Abbott @ MEV Ltd. E-mail: <abbotti@xxxxxxxxx> )=- -=( Web: http://www.mev.co.uk/ )=- _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
