On 01/05/2012 03:19 PM, J.I. Cameron wrote: > On Jan 5 2012, Lars-Peter Clausen wrote: > >> On 01/04/2012 01:30 PM, Manuel Stahl wrote: >>> Hi, >>> >>> I have a short question about the arguments of iio_buffer_register. Are >>> they really needed? >>> When we look at this code from lis3l02dq_core.c: >>> >>> indio_dev->channels = lis3l02dq_channels; >>> indio_dev->num_channels = ARRAY_SIZE(lis3l02dq_channels); >>> >>> <zip> >>> >>> ret = iio_buffer_register(indio_dev, >>> lis3l02dq_channels, >>> ARRAY_SIZE(lis3l02dq_channels)); >>> >>> The needed info is already available in indio_dev... >>> >> >> It might even cause problems if we pass another set of channels to >> iio_buffer_register as we assign to the indio_dev. So it properly is a good >> idea to update the function to only take the indio_dev as a parameter. > There certainly have been driver that have different sets of channels passed > into > the buffer registration code from elsewhere. These mostly came about > to keep things working with minimal changes as we moved over the channel spec. > > Also, we may need a clean way to say, 'this one does not go in the buffer' and > also a cleaner option for the opposite case of 'this one is buffered only'. > > Jonathan Just wrote a small cocci script which finds these drivers and there are indeed two which do this: ./adc/ad7298_core.c:223:7-26: Channels passed to iio_buffer_register differ ./adc/ad7298_core.c:223:7-26: Number of channels passed to iio_buffer_register differ ./impedance-analyzer/ad5933.c:734:7-26: Channels passed to iio_buffer_register differ ./impedance-analyzer/ad5933.c:734:7-26: Number of channels passed to iio_buffer_register differ So I guess best would be to add a flag to chan spec which allows us to specify that a channel is not buffer-able or buffered only. - Lars @r1@ struct iio_dev *indio_dev; expression ch; expression num; @@ indio_dev->channels = ch; indio_dev->num_channels = num; @r2 depends on r1@ expression indiodev; expression ch; expression num; position p; @@ iio_buffer_register@p(indiodev, ch, num) @script:python depends on r2@ indio_dev << r1.indio_dev; ch1 << r1.ch; ch2 << r2.ch; num1 << r1.num; num2 << r2.num; p << r2.p; @@ if (str(ch1) != str(ch2) and str(ch2) != "%s -> channels" % str(indio_dev)): coccilib.report.print_report(p[0], "Channels passed to iio_buffer_register differ") if (str(num1) != str(num2) and str(num2) != "%s -> num_channels" % str(indio_dev)): coccilib.report.print_report(p[0], "Number of channels passed " + "to iio_buffer_register differ") -- To unsubscribe from this list: send the line "unsubscribe linux-iio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
