On 07/10/15 19:09, H Hartley Sweeten wrote:
According the the PCI-2006M user's manual, bipolar ranges use 2's
complement coding and unipolar ranges are straight binary. Fix
ii20k_ao_insn_write() to use the correct coding based on the range.
For aesthetics, use the comedi_offset_munge() helper to handle the
munging of the data.
Signed-off-by: H Hartley Sweeten <hswee...@visionengravers.com>
Cc: Ian Abbott <abbo...@mev.co.uk>
Cc: Greg Kroah-Hartman <gre...@linuxfoundation.org>
---
drivers/staging/comedi/drivers/ii_pci20kc.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/staging/comedi/drivers/ii_pci20kc.c
b/drivers/staging/comedi/drivers/ii_pci20kc.c
index 4d3f89c..6e2ce93 100644
--- a/drivers/staging/comedi/drivers/ii_pci20kc.c
+++ b/drivers/staging/comedi/drivers/ii_pci20kc.c
@@ -146,6 +146,7 @@ static int ii20k_ao_insn_write(struct comedi_device *dev,
{
void __iomem *iobase = ii20k_module_iobase(dev, s);
unsigned int chan = CR_CHAN(insn->chanspec);
+ unsigned int range = CR_RANGE(insn->chanspec);
int i;
for (i = 0; i < insn->n; i++) {
@@ -153,9 +154,9 @@ static int ii20k_ao_insn_write(struct comedi_device *dev,
s->readback[chan] = val;
- /* munge data */
- val += ((s->maxdata + 1) >> 1);
- val &= s->maxdata;
+ /* munge the offset binary to 2's complement if needed */
+ if (comedi_range_is_bipolar(s, range))
+ val = comedi_offset_munge(s, val);
writeb(val & 0xff, iobase + II20K_AO_LSB_REG(chan));
writeb((val >> 8) & 0xff, iobase + II20K_AO_MSB_REG(chan));
I'm not sure about that as there are discrepancies in the manual. The
one I found was at:
https://www.uni-due.de/~hp0259/agwiki/lib/exe/fetch.php?media=anleitungen:pci-20006m.pdf
In "5. Specifications" it has:
CODE
UNIPOLAR Binary
BIPOLAR Complementary Binary
But then under "D/A Codes" it has:
OUTDATA Volts Unipolar Volts Bipolar
------- -------------- -------------
32767 +Full Scale -1 LSB *
0 +½ Full Scale 0
-1 +½ F.S. - 1 LSB -1 LSB
-32768 0 -Full Scale
± 5V Range: OUTDATA = VOLTS * (65536 / 10)
± 10V Range: OUTDATA = VOLTS * (65536 / 20)
0 to 10V Range: OUTDATA = (VOLTS * (65536 / 10)) - 32768
* I think that's a typo and should be "+Full Scale - 1 LSB".
The table (with typo corrected) and formulae indicate that for both
unipolar and bipolar, -32768 is the lowest voltage and +32767 is the
highest voltage, so we should munge comedi sample values regardless of
range.
--
-=( Ian Abbott @ MEV Ltd. E-mail: <abbo...@mev.co.uk> )=-
-=( Web: http://www.mev.co.uk/ )=-
_______________________________________________
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
