The `insn_read` handler for the EEPROM subdevice on E-series boards
(`ni_eeprom_insn_read()`) currently ignores `insn->n` (the number of
samples to read) and assumes a single sample is to be read into
`data[0]`. Fortunately, the Comedi core ensures that `data[]` has a
length of at least 16 so there is no problem with array bounds.
The usual Comedi convention for `insn_read` handlers is to read the same
channel `insn->n` times into successive elements of `data[]` so let's do
that. (Each channel number corresponds to a single EEPROM address.)
Since we do not expect the EEPROM data at a particular address to change
between readings, let's just read it once and copy the value `insn->n`
times.
Also, follow the usual Comedi convention and return `insn->n` from the
handler to indicate success (although any non-negative value will do).
Signed-off-by: Ian Abbott <abbotti@xxxxxxxxx>
---
drivers/staging/comedi/drivers/ni_mio_common.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/staging/comedi/drivers/ni_mio_common.c b/drivers/staging/comedi/drivers/ni_mio_common.c
index e008095436d7..5b397ad2a604 100644
--- a/drivers/staging/comedi/drivers/ni_mio_common.c
+++ b/drivers/staging/comedi/drivers/ni_mio_common.c
@@ -4507,9 +4507,15 @@ static int ni_eeprom_insn_read(struct comedi_device *dev,
struct comedi_insn *insn,
unsigned int *data)
{
- data[0] = ni_read_eeprom(dev, CR_CHAN(insn->chanspec));
+ unsigned int val;
+ unsigned int i;
- return 1;
+ if (insn->n) {
+ val = ni_read_eeprom(dev, CR_CHAN(insn->chanspec));
+ for (i = 0; i < insn->n; i++)
+ data[i] = val;
+ }
+ return insn->n;
}
static int ni_m_series_eeprom_insn_read(struct comedi_device *dev,
--
2.20.1
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
