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?
If so, then I don't see the compatibility:
Excerpt from both drivers showing register addresses:
drivers/leds/leds-as3645a.c:
#define AS_DESIGN_INFO_REG 0x00
#define AS_VERSION_CONTROL_REG 0x01
#define AS_INDICATOR_AND_TIMER_REG 0x02
#define AS_CURRENT_SET_REG 0x03
#define AS_CONTROL_REG 0x04
#define AS_FAULT_INFO_REG 0x05
#define AS_BOOST_REG 0x0d
#define AS_PASSWORD_REG 0x0f
drivers/leds/leds-lm355x.c:
struct lm355x_reg_data {
u8 regno; // this is register address
u8 mask;
u8 shift;
};
static struct lm355x_reg_data lm3554_regs[REG_MAX] = {
[REG_FLAG] = {0xD0, 0xBF, 0},
[REG_TORCH_CFG] = {0xE0, 0x80, 7},
[REG_TORCH_CTRL] = {0xA0, 0x38, 3},
[REG_STROBE_CFG] = {0xE0, 0x04, 2},
[REG_FLASH_CTRL] = {0xB0, 0x78, 3},
[REG_INDI_CFG] = {0xE0, 0x08, 3},
[REG_INDI_CTRL] = {0xA0, 0xC0, 6},
[REG_OPMODE] = {0xA0, 0x03, 0},
};
static struct lm355x_reg_data lm3556_regs[REG_MAX] = {
[REG_FLAG] = {0x0B, 0xFF, 0},
[REG_TORCH_CFG] = {0x0A, 0x10, 4},
[REG_TORCH_CTRL] = {0x09, 0x70, 4},
[REG_STROBE_CFG] = {0x0A, 0x20, 5},
[REG_FLASH_CTRL] = {0x09, 0x0F, 0},
[REG_INDI_CFG] = {0xFF, 0xFF, 0},
[REG_INDI_CTRL] = {0x09, 0x70, 4},
[REG_OPMODE] = {0x0A, 0x03, 0},
};
This is consistent with documentation:
[0] http://www.ti.com/lit/ds/symlink/lm3554.pdf
[1] http://www.ti.com/lit/ds/symlink/lm3556.pdf
[2] https://www.mouser.com/ds/2/588/AS3645_Datasheet_EN_v3-1214758.pdf
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.
--
Best regards,
Jacek Anaszewski
