On 28/01/15 22:34, Rickard Strandqvist wrote:
Variable ar assigned a value that is never used.
I have also removed all the code that thereby serves no purpose.
This was found using a static code analysis program called cppcheck
Signed-off-by: Rickard Strandqvist <rickard_strandqvist@xxxxxxxxxxxxxxxxxx>
---
drivers/staging/comedi/drivers/dt2815.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/staging/comedi/drivers/dt2815.c b/drivers/staging/comedi/drivers/dt2815.c
index a98fb66..c5c0490 100644
--- a/drivers/staging/comedi/drivers/dt2815.c
+++ b/drivers/staging/comedi/drivers/dt2815.c
@@ -98,12 +98,11 @@ static int dt2815_ao_insn(struct comedi_device *dev, struct comedi_subdevice *s,
struct dt2815_private *devpriv = dev->private;
int i;
int chan = CR_CHAN(insn->chanspec);
- unsigned int lo, hi;
+ unsigned int lo;
int ret;
for (i = 0; i < insn->n; i++) {
lo = ((data[i] & 0x0f) << 4) | (chan << 1) | 0x01;
- hi = (data[i] & 0xff0) >> 4;
ret = comedi_timeout(dev, s, insn, dt2815_ao_status, 0x00);
if (ret)
I think this highlights a bug in the driver. It ought to write the hi
byte to the DT2815_DATA register, but I can't find any register
programming manuals for this board (only a Windows driver manual - and
that's for Windows 3.1 and Windows 95).
My guess (based on kernel log messages that have since been removed from
the driver) is that the sequence should be:
1. wait for the DT2815_STATUS register to read 0x00
2. write the lo byte to the DT2815_DATA register
3. wait for the DT2815_STATUS register to read 0x10
4. write the hi byte to the DT2815_DATA register
and that step 4 is missing from the driver.
--
-=( 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
