| Message ID | 3fd11489356b1c73a3d7b4bd9dec7e12c9fe8788.1660934107.git.mazziesaccount@gmail.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Headers | show |
| Series | Use devm helpers for regulator get and enable | expand |
On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen <mazziesaccount@gmail.com> wrote: > > Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, > bulk-enable, add-action-to-disable-at-detach - pattern. ... > int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, > const char *name) > { > struct bmg160_data *data; > struct iio_dev *indio_dev; > int ret; > + static const char * const regulators[] = {"vdd", "vddio"}; Please, keep this following the "longest line first" rule. Note, in this case you even can move it out of the function, so we will see clearly that this is (not a hidden) global variable. P.S. Same applies for the rest of the similar places in your series.
Thanks for the review Andy On 8/20/22 02:30, Andy Shevchenko wrote: > On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen > <mazziesaccount@gmail.com> wrote: >> >> Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, >> bulk-enable, add-action-to-disable-at-detach - pattern. > > ... > >> int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, >> const char *name) >> { >> struct bmg160_data *data; >> struct iio_dev *indio_dev; >> int ret; >> + static const char * const regulators[] = {"vdd", "vddio"}; > > Please, keep this following the "longest line first" rule. Note, in This was not following the (IMO slightly silly) rule even prior my patch. I can for sure move my line up - but that won't give you the "reverse X-mas tree". I don't have any real objections on changing the styling though - I don't expect this to be merged before the dependency is in rc1 - so I guess I will anyways need to respin this for next cycle. I can do the styling then. > this case you even can move it out of the function, so we will see > clearly that this is (not a hidden) global variable. Here I do disagree with you. Moving the array out of the function makes it _much_ less obvious it is not used outside this function. Reason for making is "static const" is to allow the data be placed in read-only area (thanks to Guenter who originally gave me this tip). > P.S. Same applies for the rest of the similar places in your series. > Br, -- Matti
On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti <Matti.Vaittinen@fi.rohmeurope.com> wrote: > On 8/20/22 02:30, Andy Shevchenko wrote: > > On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen > > <mazziesaccount@gmail.com> wrote: ... > >> struct bmg160_data *data; > >> struct iio_dev *indio_dev; > >> int ret; > >> + static const char * const regulators[] = {"vdd", "vddio"}; > > > > Please, keep this following the "longest line first" rule. Note, in > > This was not following the (IMO slightly silly) rule even prior my > patch. I can for sure move my line up - but that won't give you the > "reverse X-mas tree". What do you mean by this? In the above case the rule does exactly give you "reversed xmas tree order". What did I miss? > I don't have any real objections on changing the styling though - I > don't expect this to be merged before the dependency is in rc1 - so I > guess I will anyways need to respin this for next cycle. I can do the > styling then. Fine with me. > > this case you even can move it out of the function, so we will see > > clearly that this is (not a hidden) global variable. > > Here I do disagree with you. Moving the array out of the function makes > it _much_ less obvious it is not used outside this function. Reason for > making is "static const" is to allow the data be placed in read-only > area (thanks to Guenter who originally gave me this tip). "static" in C language means two things (that's what come to my mind): - for functions this tells that a function is not used outside of the module; - for variables that it is a _global_ variable. Hiding static inside functions is not a good coding practice since it hides scope of the variable. And if you look into the kernel code, I believe the use you are proposing is in minority.
On 8/20/22 09:25, Andy Shevchenko wrote: > On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti > <Matti.Vaittinen@fi.rohmeurope.com> wrote: >> On 8/20/22 02:30, Andy Shevchenko wrote: >>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen >>> <mazziesaccount@gmail.com> wrote: > > What did I miss? >>>> struct bmg160_data *data; >>>> struct iio_dev *indio_dev; This does already violate the rule. > >>> this case you even can move it out of the function, so we will see >>> clearly that this is (not a hidden) global variable. >> >> Here I do disagree with you. Moving the array out of the function makes >> it _much_ less obvious it is not used outside this function. Reason for >> making is "static const" is to allow the data be placed in read-only >> area (thanks to Guenter who originally gave me this tip). > > "static" in C language means two things (that's what come to my mind): > - for functions this tells that a function is not used outside of the module; > - for variables that it is a _global_ variable. > > Hiding static inside functions is not a good coding practice since it > hides scope of the variable. For const arrays the static in function does make sense. Being able to place the data in read-only areas do help with the memory on limited systems. > And if you look into the kernel code, I > believe the use you are proposing is in minority. I don't know about the statistics. What I know is that we do have a technical benefits when we use static const arrays instead of non static ones in the functions. I do also believe placing the variables in blocks is a good practice. I tend to agree with you that using local, non const statics has pitfalls - but the pitfalls do not really apply with const ones. You know the value and have no races. Benefit is that just by seeing that no pointer is returned you can be sure that no "sane code" uses the data outside the function it resides. Best Regards -- Matti
On Sat, Aug 20, 2022 at 9:48 AM Vaittinen, Matti <Matti.Vaittinen@fi.rohmeurope.com> wrote: > On 8/20/22 09:25, Andy Shevchenko wrote: > > On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti > > <Matti.Vaittinen@fi.rohmeurope.com> wrote: > >> On 8/20/22 02:30, Andy Shevchenko wrote: > >>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen > >>> <mazziesaccount@gmail.com> wrote: > > > > What did I miss? > > >>>> struct bmg160_data *data; > >>>> struct iio_dev *indio_dev; > > This does already violate the rule. Indeed, I am reading this with an MTA that has True Type fonts, and I can't see it at the first glance. But this breaks that rule slightly while your added line breaks it significantly. > >>> this case you even can move it out of the function, so we will see > >>> clearly that this is (not a hidden) global variable. > >> > >> Here I do disagree with you. Moving the array out of the function makes > >> it _much_ less obvious it is not used outside this function. Reason for > >> making is "static const" is to allow the data be placed in read-only > >> area (thanks to Guenter who originally gave me this tip). > > > > "static" in C language means two things (that's what come to my mind): > > - for functions this tells that a function is not used outside of the module; > > - for variables that it is a _global_ variable. > > > > Hiding static inside functions is not a good coding practice since it > > hides scope of the variable. > > For const arrays the static in function does make sense. Being able to > place the data in read-only areas do help with the memory on limited > systems. I'm not sure we are on the same page. I do not object to the "const" part and we are _not_ talking about that. > > And if you look into the kernel code, I > > believe the use you are proposing is in minority. > > I don't know about the statistics. What I know is that we do have a > technical benefits when we use static const arrays instead of non static > ones in the functions. I do also believe placing the variables in blocks > is a good practice. Yes, and global variables are better to be seen as global variables. > I tend to agree with you that using local, non const statics has > pitfalls - but the pitfalls do not really apply with const ones. You > know the value and have no races. Benefit is that just by seeing that no > pointer is returned you can be sure that no "sane code" uses the data > outside the function it resides. Putting a global variable (const or non-const) to the function will hide its scope and it is prone to getting two variables with the same or very similar names with quite different semantics. That's why it's really not good practice. I would rather see it outside of the function _esp_ because it's static const.
On 8/20/22 10:18, Andy Shevchenko wrote: > On Sat, Aug 20, 2022 at 9:48 AM Vaittinen, Matti > <Matti.Vaittinen@fi.rohmeurope.com> wrote: >> On 8/20/22 09:25, Andy Shevchenko wrote: >>> On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti >>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: >>>> On 8/20/22 02:30, Andy Shevchenko wrote: >>>>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen >>>>> <mazziesaccount@gmail.com> wrote: >>> >>> What did I miss? >> >> >>>> struct bmg160_data *data; >> >>>> struct iio_dev *indio_dev; >> >> This does already violate the rule. > > Indeed, I am reading this with an MTA that has True Type fonts, and I > can't see it at the first glance. But this breaks that rule slightly > while your added line breaks it significantly. Yes. As I said, I think the reverse xmas tree rule is not too well justified. Bunch of the subsystems do not really follow it, nor did this function. Yet, as I said, I can move the array to the first line in the function when I respin the series.. >>>>> this case you even can move it out of the function, so we will see >>>>> clearly that this is (not a hidden) global variable. >>>> >>>> Here I do disagree with you. Moving the array out of the function makes >>>> it _much_ less obvious it is not used outside this function. Reason for >>>> making is "static const" is to allow the data be placed in read-only >>>> area (thanks to Guenter who originally gave me this tip). Just wanted to correct - it was Sebastian Reichel, not Guenter who explained me why doing local static const arrays is better than plain const. >>> >>> "static" in C language means two things (that's what come to my mind): >>> - for functions this tells that a function is not used outside of the module; >>> - for variables that it is a _global_ variable. >>> >>> Hiding static inside functions is not a good coding practice since it >>> hides scope of the variable. >> >> For const arrays the static in function does make sense. Being able to >> place the data in read-only areas do help with the memory on limited >> systems. > > I'm not sure we are on the same page. I do not object to the "const" > part and we are _not_ talking about that. > Maybe the explanation by Sebastian here can put us on the same page: https://lore.kernel.org/all/20190502073539.GB7864@localhost.localdomain/ https://lore.kernel.org/all/322fa765ddd72972aba931c706657661ca685afa.camel@fi.rohmeurope.com/ >>> And if you look into the kernel code, I >>> believe the use you are proposing is in minority. >> >> I don't know about the statistics. What I know is that we do have a >> technical benefits when we use static const arrays instead of non static >> ones in the functions. I do also believe placing the variables in blocks >> is a good practice. > > Yes, and global variables are better to be seen as global variables. > >> I tend to agree with you that using local, non const statics has >> pitfalls - but the pitfalls do not really apply with const ones. You >> know the value and have no races. Benefit is that just by seeing that no >> pointer is returned you can be sure that no "sane code" uses the data >> outside the function it resides. > > Putting a global variable (const or non-const) to the function will > hide its scope and it is prone to getting two variables with the same > or very similar names with quite different semantics. I don't see how moving something from a local block to a global scope does make conflicts less of an issue? On the contrary, it makes things worse as then the moved variable will collide with any other variable in any of the functions in the whole file. Having the array as function local static makes the naming collisions to be issue only if another global variable has the same name. And if that happened - the chances are code would still be correct as the function here is clearly intended to use the local one. If someone really later adds a global with the same name - and uses the global in this function - then he should have noted we have local variable with same name. Additionally - such user would be using terribly bad name for a global variable. Please note that scope of the function local static variable is limited to function even if the life-time is not just the life-time of a function. > That's why it's > really not good practice. I would rather see it outside of the > function _esp_ because it's static const. Sorry, I really don't agree with your reasoning here. :(
On Fri, 19 Aug 2022 22:19:31 +0300 Matti Vaittinen <mazziesaccount@gmail.com> wrote: > Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, > bulk-enable, add-action-to-disable-at-detach - pattern. > > Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com> > > --- > v2 => v3 > Split to own patch. > --- > drivers/iio/gyro/bmg160_core.c | 24 +++--------------------- > 1 file changed, 3 insertions(+), 21 deletions(-) > > diff --git a/drivers/iio/gyro/bmg160_core.c b/drivers/iio/gyro/bmg160_core.c > index cedd9f02ea21..baa80980c99f 100644 > --- a/drivers/iio/gyro/bmg160_core.c > +++ b/drivers/iio/gyro/bmg160_core.c > @@ -93,7 +93,6 @@ > > struct bmg160_data { > struct regmap *regmap; > - struct regulator_bulk_data regulators[2]; > struct iio_trigger *dready_trig; > struct iio_trigger *motion_trig; > struct iio_mount_matrix orientation; > @@ -1067,19 +1066,13 @@ static const char *bmg160_match_acpi_device(struct device *dev) > return dev_name(dev); > } > > -static void bmg160_disable_regulators(void *d) > -{ > - struct bmg160_data *data = d; > - > - regulator_bulk_disable(ARRAY_SIZE(data->regulators), data->regulators); > -} > - > int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, > const char *name) > { > struct bmg160_data *data; > struct iio_dev *indio_dev; > int ret; > + static const char * const regulators[] = {"vdd", "vddio"}; As in previous, small preference for spaces after { and before } > > indio_dev = devm_iio_device_alloc(dev, sizeof(*data)); > if (!indio_dev) > @@ -1090,22 +1083,11 @@ int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, > data->irq = irq; > data->regmap = regmap; > > - data->regulators[0].supply = "vdd"; > - data->regulators[1].supply = "vddio"; > - ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(data->regulators), > - data->regulators); > + ret = devm_regulator_bulk_get_enable(dev, ARRAY_SIZE(regulators), > + regulators); > if (ret) > return dev_err_probe(dev, ret, "Failed to get regulators\n"); > > - ret = regulator_bulk_enable(ARRAY_SIZE(data->regulators), > - data->regulators); > - if (ret) > - return ret; > - > - ret = devm_add_action_or_reset(dev, bmg160_disable_regulators, data); > - if (ret) > - return ret; > - > ret = iio_read_mount_matrix(dev, &data->orientation); > if (ret) > return ret;
On Sat, 20 Aug 2022 06:19:00 +0000 "Vaittinen, Matti" <Matti.Vaittinen@fi.rohmeurope.com> wrote: > Thanks for the review Andy > > On 8/20/22 02:30, Andy Shevchenko wrote: > > On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen > > <mazziesaccount@gmail.com> wrote: > >> > >> Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, > >> bulk-enable, add-action-to-disable-at-detach - pattern. > > > > ... > > > >> int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, > >> const char *name) > >> { > >> struct bmg160_data *data; > >> struct iio_dev *indio_dev; > >> int ret; > >> + static const char * const regulators[] = {"vdd", "vddio"}; > > > > Please, keep this following the "longest line first" rule. Note, in > > This was not following the (IMO slightly silly) rule even prior my > patch. I can for sure move my line up - but that won't give you the > "reverse X-mas tree". > > I don't have any real objections on changing the styling though - I > don't expect this to be merged before the dependency is in rc1 - so I > guess I will anyways need to respin this for next cycle. I can do the > styling then. I was a bit surprised Mark didn't do an immutable branch for this, but indeed looks like it's going to be a multiple cycle thing - so we'll probably have a bunch of new cases introduced in the meantime that we need to tidy up. Ah well. > > > this case you even can move it out of the function, so we will see > > clearly that this is (not a hidden) global variable. > > Here I do disagree with you. Moving the array out of the function makes > it _much_ less obvious it is not used outside this function. Reason for > making is "static const" is to allow the data be placed in read-only > area (thanks to Guenter who originally gave me this tip). > > > P.S. Same applies for the rest of the similar places in your series. > > > > Br, > -- Matti >
On 8/20/22 14:38, Jonathan Cameron wrote: > On Sat, 20 Aug 2022 06:19:00 +0000 > "Vaittinen, Matti" <Matti.Vaittinen@fi.rohmeurope.com> wrote: > >> Thanks for the review Andy >> >> On 8/20/22 02:30, Andy Shevchenko wrote: >>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen >>> <mazziesaccount@gmail.com> wrote: >> >>I >> don't expect this to be merged before the dependency is in rc1 - so I >> guess I will anyways need to respin this for next cycle. I can do the >> styling then. > I was a bit surprised Mark didn't do an immutable branch for this, but > indeed looks like it's going to be a multiple cycle thing - so we'll > probably have a bunch of new cases introduced in the meantime that > we need to tidy up. Ah well. > I guess we can ask Mark what he thinks of the immutable branch ;) I admit I am not too keen on rebasing this - especially if I find the time to go through more drivers.
On 8/20/22 14:22, Jonathan Cameron wrote: > On Fri, 19 Aug 2022 22:19:31 +0300 > Matti Vaittinen <mazziesaccount@gmail.com> wrote: > >> Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, >> bulk-enable, add-action-to-disable-at-detach - pattern. >> >> Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com> >> >> --- >> v2 => v3 >> Split to own patch. >> --- >> drivers/iio/gyro/bmg160_core.c | 24 +++--------------------- >> 1 file changed, 3 insertions(+), 21 deletions(-) >> >> diff --git a/drivers/iio/gyro/bmg160_core.c b/drivers/iio/gyro/bmg160_core.c >> index cedd9f02ea21..baa80980c99f 100644 >> --- a/drivers/iio/gyro/bmg160_core.c >> +++ b/drivers/iio/gyro/bmg160_core.c >> @@ -93,7 +93,6 @@ >> >> struct bmg160_data { >> struct regmap *regmap; >> - struct regulator_bulk_data regulators[2]; >> struct iio_trigger *dready_trig; >> struct iio_trigger *motion_trig; >> struct iio_mount_matrix orientation; >> @@ -1067,19 +1066,13 @@ static const char *bmg160_match_acpi_device(struct device *dev) >> return dev_name(dev); >> } >> >> -static void bmg160_disable_regulators(void *d) >> -{ >> - struct bmg160_data *data = d; >> - >> - regulator_bulk_disable(ARRAY_SIZE(data->regulators), data->regulators); >> -} >> - >> int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, >> const char *name) >> { >> struct bmg160_data *data; >> struct iio_dev *indio_dev; >> int ret; >> + static const char * const regulators[] = {"vdd", "vddio"}; > > As in previous, small preference for spaces after { and before } Thanks. I'll fix it when I respin.
On Sat, Aug 20, 2022 at 1:05 PM Matti Vaittinen <mazziesaccount@gmail.com> wrote: > On 8/20/22 10:18, Andy Shevchenko wrote: > > On Sat, Aug 20, 2022 at 9:48 AM Vaittinen, Matti > > <Matti.Vaittinen@fi.rohmeurope.com> wrote: > >> On 8/20/22 09:25, Andy Shevchenko wrote: > >>> On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti > >>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: > >>>> On 8/20/22 02:30, Andy Shevchenko wrote: > >>>>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen > >>>>> <mazziesaccount@gmail.com> wrote: > >>> > >>> What did I miss? > >> > >> >>>> struct bmg160_data *data; > >> >>>> struct iio_dev *indio_dev; > >> > >> This does already violate the rule. > > > > Indeed, I am reading this with an MTA that has True Type fonts, and I > > can't see it at the first glance. But this breaks that rule slightly > > while your added line breaks it significantly. > > Yes. As I said, I think the reverse xmas tree rule is not too well > justified. Bunch of the subsystems do not really follow it, nor did this > function. Yet, as I said, I can move the array to the first line in the > function when I respin the series.. You still can do better in _your_ series, right? > >>>>> this case you even can move it out of the function, so we will see > >>>>> clearly that this is (not a hidden) global variable. > >>>> > >>>> Here I do disagree with you. Moving the array out of the function makes > >>>> it _much_ less obvious it is not used outside this function. Reason for > >>>> making is "static const" is to allow the data be placed in read-only > >>>> area (thanks to Guenter who originally gave me this tip). > > Just wanted to correct - it was Sebastian Reichel, not Guenter who > explained me why doing local static const arrays is better than plain const. Did he suggest putting it inside the function? > >>> "static" in C language means two things (that's what come to my mind): > >>> - for functions this tells that a function is not used outside of the module; > >>> - for variables that it is a _global_ variable. > >>> > >>> Hiding static inside functions is not a good coding practice since it > >>> hides scope of the variable. > >> > >> For const arrays the static in function does make sense. Being able to > >> place the data in read-only areas do help with the memory on limited > >> systems. > > > > I'm not sure we are on the same page. I do not object to the "const" > > part and we are _not_ talking about that. > > Maybe the explanation by Sebastian here can put us on the same page: > https://lore.kernel.org/all/20190502073539.GB7864@localhost.localdomain/ > https://lore.kernel.org/all/322fa765ddd72972aba931c706657661ca685afa.camel@fi.rohmeurope.com/ Again, you are too focused on "const", I'm talking about "static". The above doesn't clear a bit regarding why you hide the global variable inside a function. I don't see either Sebastian's clear point on this. > >>> And if you look into the kernel code, I > >>> believe the use you are proposing is in minority. > >> > >> I don't know about the statistics. What I know is that we do have a > >> technical benefits when we use static const arrays instead of non static > >> ones in the functions. I do also believe placing the variables in blocks > >> is a good practice. > > > > Yes, and global variables are better to be seen as global variables. > > > >> I tend to agree with you that using local, non const statics has > >> pitfalls - but the pitfalls do not really apply with const ones. You > >> know the value and have no races. Benefit is that just by seeing that no > >> pointer is returned you can be sure that no "sane code" uses the data > >> outside the function it resides. > > > > Putting a global variable (const or non-const) to the function will > > hide its scope and it is prone to getting two variables with the same > > or very similar names with quite different semantics. > > I don't see how moving something from a local block to a global scope > does make conflicts less of an issue? You may add a static variable inside each function in the same module and name it "foo" and there will be no conflict, but when you read the code your brain will be spoiled. This is simply _bad code practice_. I don't know how else I can explain this to you. > On the contrary, it makes things > worse as then the moved variable will collide with any other variable in > any of the functions in the whole file. Having the array as function > local static makes the naming collisions to be issue only if another > global variable has the same name. Again, you missed my point. I'm talking about reading and analysing the code. Otherwise (good) compiler should spill a lot of warnings in case you have global vs. local naming collision. > And if that happened - the chances > are code would still be correct as the function here is clearly intended > to use the local one. If someone really later adds a global with the > same name - and uses the global in this function - then he should have > noted we have local variable with same name. Additionally - such user > would be using terribly bad name for a global variable. > > Please note that scope of the function local static variable is limited > to function even if the life-time is not just the life-time of a function. Nope. The RO section might be very well flashed into ROM, so... > > That's why it's > > really not good practice. I would rather see it outside of the > > function _esp_ because it's static const. > > Sorry, I really don't agree with your reasoning here. :( So, whom should we listen to here? Because bad code is bad code. And this is code above.
On 8/20/22 19:21, Andy Shevchenko wrote: > On Sat, Aug 20, 2022 at 1:05 PM Matti Vaittinen > <mazziesaccount@gmail.com> wrote: >> On 8/20/22 10:18, Andy Shevchenko wrote: >>> On Sat, Aug 20, 2022 at 9:48 AM Vaittinen, Matti >>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: >>>> On 8/20/22 09:25, Andy Shevchenko wrote: >>>>> On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti >>>>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: >>>>>> On 8/20/22 02:30, Andy Shevchenko wrote: >>>>>>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen >>>>>>> <mazziesaccount@gmail.com> wrote: >>>>> >>>>> What did I miss? >>>> >>>> >>>> struct bmg160_data *data; >>>> >>>> struct iio_dev *indio_dev; >>>> >>>> This does already violate the rule. >>> >>> Indeed, I am reading this with an MTA that has True Type fonts, and I >>> can't see it at the first glance. But this breaks that rule slightly >>> while your added line breaks it significantly. >> >> Yes. As I said, I think the reverse xmas tree rule is not too well >> justified. Bunch of the subsystems do not really follow it, nor did this >> function. Yet, as I said, I can move the array to the first line in the >> function when I respin the series.. > > You still can do better in _your_ series, right? I don't see the benefit of the reverse xmas tree. We have discussed this already in the past :) I definitely have no need to start using reverse xmas tree thingee somewhere it has not been previously used. It may be better in _your_ opinion. >>>>>>> this case you even can move it out of the function, so we will see >>>>>>> clearly that this is (not a hidden) global variable. >>>>>> >>>>>> Here I do disagree with you. Moving the array out of the function makes >>>>>> it _much_ less obvious it is not used outside this function. Reason for >>>>>> making is "static const" is to allow the data be placed in read-only >>>>>> area (thanks to Guenter who originally gave me this tip). >> >> Just wanted to correct - it was Sebastian Reichel, not Guenter who >> explained me why doing local static const arrays is better than plain const. > > Did he suggest putting it inside the function? He asked me to convert a local array to static const. I though like you do now that the local array should not be static but just const - and he corrected me in his reply. This can be seen in the discussion I linked below. >>>>> "static" in C language means two things (that's what come to my mind): >>>>> - for functions this tells that a function is not used outside of the module; >>>>> - for variables that it is a _global_ variable. >>>>> >>>>> Hiding static inside functions is not a good coding practice since it >>>>> hides scope of the variable. >>>> >>>> For const arrays the static in function does make sense. Being able to >>>> place the data in read-only areas do help with the memory on limited >>>> systems. >>> >>> I'm not sure we are on the same page. I do not object to the "const" >>> part and we are _not_ talking about that. >> >> Maybe the explanation by Sebastian here can put us on the same page: >> https://lore.kernel.org/all/20190502073539.GB7864@localhost.localdomain/ >> https://lore.kernel.org/all/322fa765ddd72972aba931c706657661ca685afa.camel@fi.rohmeurope.com/ > > Again, you are too focused on "const", I'm talking about "static". The > above doesn't clear a bit regarding why you hide the global variable > inside a function. I don't see either Sebastian's clear point on this. I don't really see why you talk about "hiding a global variable in a function"? A static variable which is declared in function is not global. It is local. It causes no more name collisions than a regular local variable does so I really don't understand your reasoning. >>>>> And if you look into the kernel code, I >>>>> believe the use you are proposing is in minority. >>>> >>>> I don't know about the statistics. What I know is that we do have a >>>> technical benefits when we use static const arrays instead of non static >>>> ones in the functions. I do also believe placing the variables in blocks >>>> is a good practice. >>> >>> Yes, and global variables are better to be seen as global variables. >>> >>>> I tend to agree with you that using local, non const statics has >>>> pitfalls - but the pitfalls do not really apply with const ones. You >>>> know the value and have no races. Benefit is that just by seeing that no >>>> pointer is returned you can be sure that no "sane code" uses the data >>>> outside the function it resides. >>> >>> Putting a global variable (const or non-const) to the function will >>> hide its scope and it is prone to getting two variables with the same >>> or very similar names with quite different semantics. >> >> I don't see how moving something from a local block to a global scope >> does make conflicts less of an issue? > > You may add a static variable inside each function in the same module > and name it "foo" and there will be no conflict, but when you read the > code your brain will be spoiled. And how is it different from reading functions where the regular variables have identical names? I _really_ can't follow your reasoning. > This is simply _bad code practice_. I > don't know how else I can explain this to you. > >> On the contrary, it makes things >> worse as then the moved variable will collide with any other variable in >> any of the functions in the whole file. Having the array as function >> local static makes the naming collisions to be issue only if another >> global variable has the same name. > > Again, you missed my point. I'm talking about reading and analysing > the code. I _definitely_ miss your point here. I have zero problems reading code where static const variables are used in a function. I think it is pretty much as hard as seeing a #defined value - difference being that one can point to the variable. I admit that static variables whose value is changed can be more of a problem especially when access to function is not serialized. > Otherwise (good) compiler should spill a lot of warnings in > case you have global vs. local naming collision. > >> And if that happened - the chances >> are code would still be correct as the function here is clearly intended >> to use the local one. If someone really later adds a global with the >> same name - and uses the global in this function - then he should have >> noted we have local variable with same name. Additionally - such user >> would be using terribly bad name for a global variable. >> >> Please note that scope of the function local static variable is limited >> to function even if the life-time is not just the life-time of a function. > > Nope. The RO section might be very well flashed into ROM, so... ...so? >>> That's why it's >>> really not good practice. I would rather see it outside of the >>> function _esp_ because it's static const. >> >> Sorry, I really don't agree with your reasoning here. :( > > So, whom should we listen to here? Because bad code is bad code. And > this is code above. Bad is a subjective concept. I'd say the code gets much worse if we make the local variable a global one.
On Sat, Aug 20, 2022 at 8:27 PM Matti Vaittinen <mazziesaccount@gmail.com> wrote: > On 8/20/22 19:21, Andy Shevchenko wrote: > > On Sat, Aug 20, 2022 at 1:05 PM Matti Vaittinen > > <mazziesaccount@gmail.com> wrote: > >> On 8/20/22 10:18, Andy Shevchenko wrote: > >>> On Sat, Aug 20, 2022 at 9:48 AM Vaittinen, Matti > >>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: > >>>> On 8/20/22 09:25, Andy Shevchenko wrote: > >>>>> On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti > >>>>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: > >>>>>> On 8/20/22 02:30, Andy Shevchenko wrote: > >>>>>>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen > >>>>>>> <mazziesaccount@gmail.com> wrote: > >>>>> > >>>>> What did I miss? > >>>> > >>>> >>>> struct bmg160_data *data; > >>>> >>>> struct iio_dev *indio_dev; > >>>> > >>>> This does already violate the rule. > >>> > >>> Indeed, I am reading this with an MTA that has True Type fonts, and I > >>> can't see it at the first glance. But this breaks that rule slightly > >>> while your added line breaks it significantly. > >> > >> Yes. As I said, I think the reverse xmas tree rule is not too well > >> justified. Bunch of the subsystems do not really follow it, nor did this > >> function. Yet, as I said, I can move the array to the first line in the > >> function when I respin the series.. > > > > You still can do better in _your_ series, right? > > I don't see the benefit of the reverse xmas tree. We have discussed this > already in the past :) I definitely have no need to start using reverse > xmas tree thingee somewhere it has not been previously used. It may be > better in _your_ opinion. Yes, this is a style issue and not a real coding problem. As a reviewer with some of a background I feel that reversed xmas tree ordering is better to read, maintain, and review. So, as a reviewer I have an opinion, as a maintainer I can speak for the IIO subsystem, since it's not on my watch. > >>>>>>> this case you even can move it out of the function, so we will see > >>>>>>> clearly that this is (not a hidden) global variable. > >>>>>> > >>>>>> Here I do disagree with you. Moving the array out of the function makes > >>>>>> it _much_ less obvious it is not used outside this function. Reason for > >>>>>> making is "static const" is to allow the data be placed in read-only > >>>>>> area (thanks to Guenter who originally gave me this tip). > >> > >> Just wanted to correct - it was Sebastian Reichel, not Guenter who > >> explained me why doing local static const arrays is better than plain const. > > > > Did he suggest putting it inside the function? > > He asked me to convert a local array to static const. I though like you > do now that the local array should not be static but just const - and he > corrected me in his reply. This can be seen in the discussion I linked > below. Yes, and it's irrelevant to what I'm trying to tell you. > >>>>> "static" in C language means two things (that's what come to my mind): > >>>>> - for functions this tells that a function is not used outside of the module; > >>>>> - for variables that it is a _global_ variable. > >>>>> > >>>>> Hiding static inside functions is not a good coding practice since it > >>>>> hides scope of the variable. > >>>> > >>>> For const arrays the static in function does make sense. Being able to > >>>> place the data in read-only areas do help with the memory on limited > >>>> systems. > >>> > >>> I'm not sure we are on the same page. I do not object to the "const" > >>> part and we are _not_ talking about that. > >> > >> Maybe the explanation by Sebastian here can put us on the same page: > >> https://lore.kernel.org/all/20190502073539.GB7864@localhost.localdomain/ > >> https://lore.kernel.org/all/322fa765ddd72972aba931c706657661ca685afa.camel@fi.rohmeurope.com/ > > > > Again, you are too focused on "const", I'm talking about "static". The > > above doesn't clear a bit regarding why you hide the global variable > > inside a function. I don't see either Sebastian's clear point on this. > > I don't really see why you talk about "hiding a global variable in a > function"? A static variable which is declared in function is not > global. It is local. SInce it's static it's global by nature, but local by namespace. > It causes no more name collisions than a regular > local variable does so I really don't understand your reasoning. And I have no other words to explain it to you. You are using a global variable in the scope of function. This is not good for the maintenance and development as it's prone to get an issue in the future. > >>>>> And if you look into the kernel code, I > >>>>> believe the use you are proposing is in minority. > >>>> > >>>> I don't know about the statistics. What I know is that we do have a > >>>> technical benefits when we use static const arrays instead of non static > >>>> ones in the functions. I do also believe placing the variables in blocks > >>>> is a good practice. > >>> > >>> Yes, and global variables are better to be seen as global variables. > >>> > >>>> I tend to agree with you that using local, non const statics has > >>>> pitfalls - but the pitfalls do not really apply with const ones. You > >>>> know the value and have no races. Benefit is that just by seeing that no > >>>> pointer is returned you can be sure that no "sane code" uses the data > >>>> outside the function it resides. > >>> > >>> Putting a global variable (const or non-const) to the function will > >>> hide its scope and it is prone to getting two variables with the same > >>> or very similar names with quite different semantics. > >> > >> I don't see how moving something from a local block to a global scope > >> does make conflicts less of an issue? > > > > You may add a static variable inside each function in the same module > > and name it "foo" and there will be no conflict, but when you read the > > code your brain will be spoiled. > > And how is it different from reading functions where the regular > variables have identical names? I _really_ can't follow your reasoning. Because they are on the stack and not one per module. When you read the code it's very easy to miss that the variable is static if you have a lot of other variables defined. > > This is simply _bad code practice_. I > > don't know how else I can explain this to you. > > > >> On the contrary, it makes things > >> worse as then the moved variable will collide with any other variable in > >> any of the functions in the whole file. Having the array as function > >> local static makes the naming collisions to be issue only if another > >> global variable has the same name. > > > > Again, you missed my point. I'm talking about reading and analysing > > the code. > > I _definitely_ miss your point here. I have zero problems reading code > where static const variables are used in a function. I think it is > pretty much as hard as seeing a #defined value - difference being that > one can point to the variable. Good for you. > I admit that static variables whose value is changed can be more of a > problem especially when access to function is not serialized. > > > Otherwise (good) compiler should spill a lot of warnings in > > case you have global vs. local naming collision. > > > >> And if that happened - the chances > >> are code would still be correct as the function here is clearly intended > >> to use the local one. If someone really later adds a global with the > >> same name - and uses the global in this function - then he should have > >> noted we have local variable with same name. Additionally - such user > >> would be using terribly bad name for a global variable. > >> > >> Please note that scope of the function local static variable is limited > >> to function even if the life-time is not just the life-time of a function. > > > > Nope. The RO section might be very well flashed into ROM, so... > > ...so? It won't be created by function, it's created by the compiler / linker. It won't gone if function gone. e.g. __init foo() { static bar ...; } is nonsense. And it takes a ROM space. > >>> That's why it's > >>> really not good practice. I would rather see it outside of the > >>> function _esp_ because it's static const. > >> > >> Sorry, I really don't agree with your reasoning here. :( > > > > So, whom should we listen to here? Because bad code is bad code. And > > this is code above. > > Bad is a subjective concept. I'd say the code gets much worse if we make > the local variable a global one. ... To summarize, we have a huge disagreement on the placement of the static variables. Not sure we ever get into compromize here, so I leave it up to maintainers, but my opinion that it is simply a bad code practice. -- With Best Regards, Andy Shevchenko
On 8/21/22 16:08, Andy Shevchenko wrote: > On Sat, Aug 20, 2022 at 8:27 PM Matti Vaittinen > <mazziesaccount@gmail.com> wrote: >> On 8/20/22 19:21, Andy Shevchenko wrote: >>> On Sat, Aug 20, 2022 at 1:05 PM Matti Vaittinen >>> <mazziesaccount@gmail.com> wrote: >>>> On 8/20/22 10:18, Andy Shevchenko wrote: >>>>> On Sat, Aug 20, 2022 at 9:48 AM Vaittinen, Matti >>>>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: >>>>>> On 8/20/22 09:25, Andy Shevchenko wrote: >>>>>>> On Sat, Aug 20, 2022 at 9:19 AM Vaittinen, Matti >>>>>>> <Matti.Vaittinen@fi.rohmeurope.com> wrote: >>>>>>>> On 8/20/22 02:30, Andy Shevchenko wrote: >>>>>>>>> On Fri, Aug 19, 2022 at 10:21 PM Matti Vaittinen >>>>>>>>> <mazziesaccount@gmail.com> wrote: //snip > SInce it's static it's global by nature, but local by namespace. Which is an _improvement_ over having it in global namespace? >> It causes no more name collisions than a regular >> local variable does so I really don't understand your reasoning. > > And I have no other words to explain it to you. You are using a global > variable in the scope of function. This is not good for the > maintenance and development as it's prone to get an issue in the > future. If you foresee some issues, please describe them as I don't see one single problem with a local static const data. I have seen you telling me that "static const" variables are _harder_ for you to review. Could you please explain what are the potential mistake(s) a reviewer can do, and what is the 'issue' that mistake can cause? >>> So, whom should we listen to here? Because bad code is bad code. And >>> this is code above. >> >> Bad is a subjective concept. I'd say the code gets much worse if we make >> the local variable a global one. > > ... > > > To summarize, we have a huge disagreement on the placement of the > static variables. Not sure we ever get into compromize here, so I > leave it up to maintainers, but my opinion that it is simply a bad > code practice. Bad and good are labels we can place on things. We however need to have the reason for those labels to be meaningful. I am sorry but I don't want to label the local _const_ static variables bad without reason. This discussion starts to remind me on statements like "using goto is always bad" or "one must never use macros in C". Yeah - ultimately it is a maintainer decision. Best Regards -- Matti
On Fri, 19 Aug 2022 22:19:31 +0300 Matti Vaittinen <mazziesaccount@gmail.com> wrote: > Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, > bulk-enable, add-action-to-disable-at-detach - pattern. > > Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com> Applied with tweaks: - patch title includes gyro: - ordering as Andy suggested - spaces after { and before } > > --- > v2 => v3 > Split to own patch. > --- > drivers/iio/gyro/bmg160_core.c | 24 +++--------------------- > 1 file changed, 3 insertions(+), 21 deletions(-) > > diff --git a/drivers/iio/gyro/bmg160_core.c b/drivers/iio/gyro/bmg160_core.c > index cedd9f02ea21..baa80980c99f 100644 > --- a/drivers/iio/gyro/bmg160_core.c > +++ b/drivers/iio/gyro/bmg160_core.c > @@ -93,7 +93,6 @@ > > struct bmg160_data { > struct regmap *regmap; > - struct regulator_bulk_data regulators[2]; > struct iio_trigger *dready_trig; > struct iio_trigger *motion_trig; > struct iio_mount_matrix orientation; > @@ -1067,19 +1066,13 @@ static const char *bmg160_match_acpi_device(struct device *dev) > return dev_name(dev); > } > > -static void bmg160_disable_regulators(void *d) > -{ > - struct bmg160_data *data = d; > - > - regulator_bulk_disable(ARRAY_SIZE(data->regulators), data->regulators); > -} > - > int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, > const char *name) > { > struct bmg160_data *data; > struct iio_dev *indio_dev; > int ret; > + static const char * const regulators[] = {"vdd", "vddio"}; > > indio_dev = devm_iio_device_alloc(dev, sizeof(*data)); > if (!indio_dev) > @@ -1090,22 +1083,11 @@ int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, > data->irq = irq; > data->regmap = regmap; > > - data->regulators[0].supply = "vdd"; > - data->regulators[1].supply = "vddio"; > - ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(data->regulators), > - data->regulators); > + ret = devm_regulator_bulk_get_enable(dev, ARRAY_SIZE(regulators), > + regulators); > if (ret) > return dev_err_probe(dev, ret, "Failed to get regulators\n"); > > - ret = regulator_bulk_enable(ARRAY_SIZE(data->regulators), > - data->regulators); > - if (ret) > - return ret; > - > - ret = devm_add_action_or_reset(dev, bmg160_disable_regulators, data); > - if (ret) > - return ret; > - > ret = iio_read_mount_matrix(dev, &data->orientation); > if (ret) > return ret;
On 10/16/22 19:08, Jonathan Cameron wrote: > On Fri, 19 Aug 2022 22:19:31 +0300 > Matti Vaittinen <mazziesaccount@gmail.com> wrote: > >> Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, >> bulk-enable, add-action-to-disable-at-detach - pattern. >> >> Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com> > > Applied with tweaks: > - patch title includes gyro: > - ordering as Andy suggested > - spaces after { and before } Thanks Jonathan for saving me from respin the series :) Much apprecited! Yours -- Matti
diff --git a/drivers/iio/gyro/bmg160_core.c b/drivers/iio/gyro/bmg160_core.c index cedd9f02ea21..baa80980c99f 100644 --- a/drivers/iio/gyro/bmg160_core.c +++ b/drivers/iio/gyro/bmg160_core.c @@ -93,7 +93,6 @@ struct bmg160_data { struct regmap *regmap; - struct regulator_bulk_data regulators[2]; struct iio_trigger *dready_trig; struct iio_trigger *motion_trig; struct iio_mount_matrix orientation; @@ -1067,19 +1066,13 @@ static const char *bmg160_match_acpi_device(struct device *dev) return dev_name(dev); } -static void bmg160_disable_regulators(void *d) -{ - struct bmg160_data *data = d; - - regulator_bulk_disable(ARRAY_SIZE(data->regulators), data->regulators); -} - int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, const char *name) { struct bmg160_data *data; struct iio_dev *indio_dev; int ret; + static const char * const regulators[] = {"vdd", "vddio"}; indio_dev = devm_iio_device_alloc(dev, sizeof(*data)); if (!indio_dev) @@ -1090,22 +1083,11 @@ int bmg160_core_probe(struct device *dev, struct regmap *regmap, int irq, data->irq = irq; data->regmap = regmap; - data->regulators[0].supply = "vdd"; - data->regulators[1].supply = "vddio"; - ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(data->regulators), - data->regulators); + ret = devm_regulator_bulk_get_enable(dev, ARRAY_SIZE(regulators), + regulators); if (ret) return dev_err_probe(dev, ret, "Failed to get regulators\n"); - ret = regulator_bulk_enable(ARRAY_SIZE(data->regulators), - data->regulators); - if (ret) - return ret; - - ret = devm_add_action_or_reset(dev, bmg160_disable_regulators, data); - if (ret) - return ret; - ret = iio_read_mount_matrix(dev, &data->orientation); if (ret) return ret;
Use devm_regulator_bulk_get_enable() instead of open coded bulk-get, bulk-enable, add-action-to-disable-at-detach - pattern. Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com> --- v2 => v3 Split to own patch. --- drivers/iio/gyro/bmg160_core.c | 24 +++--------------------- 1 file changed, 3 insertions(+), 21 deletions(-)