| Message ID | 1445796586-28683-2-git-send-email-bjorn.andersson@sonymobile.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Sun, 2015-10-25 at 11:09 -0700, Bjorn Andersson wrote: > Print an error message to indicate that invalid configuration data was > provided in the platform_data, rather than just aborting initialization. Perhaps it'd be nicer to show the 3 values. > diff --git a/drivers/iio/light/lm3533-als.c b/drivers/iio/light/lm3533-als.c [] > @@ -743,8 +743,10 @@ static int lm3533_als_set_resistor(struct lm3533_als *als, u8 val) > { > int ret; > > - if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) > + if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) { > + dev_err(&als->pdev->dev, "invalid resistor value\n"); > return -EINVAL; > + }; > > ret = lm3533_write(als->lm3533, LM3533_REG_ALS_RESISTOR_SELECT, val); > if (ret) {
On Tue 27 Oct 12:18 PDT 2015, Joe Perches wrote: > On Sun, 2015-10-25 at 11:09 -0700, Bjorn Andersson wrote: > > Print an error message to indicate that invalid configuration data was > > provided in the platform_data, rather than just aborting initialization. > > Perhaps it'd be nicer to show the 3 values. > I do agree that it would be helpful to hint the developer about the expected range. The problem I see here though is that, in the case of devicetree we've now moved from resistance to a register value. E.g. specifying 500kOhm in dt would make this piece complain about a resistance of 0 being invalid and I would have to read the code and/or datasheet anyways to find the valid range. The reason I did not put this error print in the dt parser was that I did not want to replicate the resistor limits and I figured it would be useful for platform_data users as well. But if this is going to be informative I think I need to move it. So would you be okay with not improving the message? Or would you like me to just move it into the code in patch 1 - and leave the platform_data case as it was (with a silent failure)? > > diff --git a/drivers/iio/light/lm3533-als.c b/drivers/iio/light/lm3533-als.c > [] > > @@ -743,8 +743,10 @@ static int lm3533_als_set_resistor(struct lm3533_als *als, u8 val) > > { > > int ret; > > > > - if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) > > + if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) { > > + dev_err(&als->pdev->dev, "invalid resistor value\n"); > > return -EINVAL; > > + }; > > Regards, Bjorn
On Wed, 2015-10-28 at 11:37 -0700, Bjorn Andersson wrote: > On Tue 27 Oct 12:18 PDT 2015, Joe Perches wrote: > > On Sun, 2015-10-25 at 11:09 -0700, Bjorn Andersson wrote: > > > Print an error message to indicate that invalid configuration data was > > > provided in the platform_data, rather than just aborting initialization. > > Perhaps it'd be nicer to show the 3 values. > I do agree that it would be helpful to hint the developer about the > expected range. The problem I see here though is that, in the case of > devicetree we've now moved from resistance to a register value. [] > So would you be okay with not improving the message? Or would you like > me to just move it into the code in patch 1 - and leave the > platform_data case as it was (with a silent failure)? [] > > > diff --git a/drivers/iio/light/lm3533-als.c b/drivers/iio/light/lm3533-als.c > > [] > > > @@ -743,8 +743,10 @@ static int lm3533_als_set_resistor(struct lm3533_als *als, u8 val) > > > { > > > int ret; > > > > > > - if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) > > > + if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) { > > > + dev_err(&als->pdev->dev, "invalid resistor value\n"); > > > return -EINVAL; > > > + }; > > > What I suggested was just a trivial proposal. Whatever you do is fine with me.
diff --git a/drivers/iio/light/lm3533-als.c b/drivers/iio/light/lm3533-als.c index 076bc46fad03..e56937c40a18 100644 --- a/drivers/iio/light/lm3533-als.c +++ b/drivers/iio/light/lm3533-als.c @@ -743,8 +743,10 @@ static int lm3533_als_set_resistor(struct lm3533_als *als, u8 val) { int ret; - if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) + if (val < LM3533_ALS_RESISTOR_MIN || val > LM3533_ALS_RESISTOR_MAX) { + dev_err(&als->pdev->dev, "invalid resistor value\n"); return -EINVAL; + }; ret = lm3533_write(als->lm3533, LM3533_REG_ALS_RESISTOR_SELECT, val); if (ret) {
Print an error message to indicate that invalid configuration data was provided in the platform_data, rather than just aborting initialization. Signed-off-by: Bjorn Andersson <bjorn.andersson@sonymobile.com> --- This patch can be picked independently of the other 2, but included in the series for "completeness". drivers/iio/light/lm3533-als.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)