On Mon, 2018-01-29 at 21:46 +0100, Jacek Anaszewski wrote: > On 01/28/2018 03:30 PM, Andy Shevchenko wrote: > > On Fri, 2018-01-26 at 21:39 +0100, Jacek Anaszewski wrote: > > > On 01/26/2018 08:38 AM, Andy Shevchenko wrote: > > > > + Note, LM3555 is driven by AS3645A. > > > > > > Could you please provide more details on this relationship? > > > Looking at the datasheet it seems that both chips are standalone > > > and controlled via I2C and/or via GPIO in case of LM3555. > > > > The only difference I see looking in datasheets again that AS3645A > > supports dual-led mode, by using STROBE signal in case of parallel > > interface or thru additional bits in I2C mode. > > > > Can you be more specific what exactly makes you feel that they are > > not > > compatible? > > The expression "LM3555 is driven by AS3645A" brings on mind rather > some type of hardware dependency between chips than drivers > similarity. > > Now I understand that you meant that both devices can be driven by any > of leds-lm355x or leds-as3645a driver, am I right? No. The chip in question is LM3555, and it's handled only in as3645a driver. > I suspected that I got something entirely wrong here, especially > that both you and Pavel agreed. > > But, given that it turned out that leds-lm355x.c doesn't support > LM3555 and its documentation [3] indeed shows the affinity with > AS3645, I suppose that this patch is a result of a premature > conclusion drawn basing on the driver's description rather than its > capabilities. > > [3] http://www.ti.com/lit/ds/symlink/lm3555.pdf > > Effectively, the only sensible change seems to be: > > - tristate "LED support for LM355x Chips, LM3554 and LM3556" > + tristate "LED support for LM3554 and LM3556 chips" > depends on LEDS_CLASS && I2C > select REGMAP_I2C > help > - This option enables support for LEDs connected to LM355x. > - LM355x includes Torch, Flash and Indicator functions. > + This option enables support for LEDs connected to LM3554 > + and LM3556. It includes Torch, Flash and Indicator > functions. I like patch Pavel sent previously. It feels more comprehensive than this one. -- Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> Intel Finland Oy
