Re: [PATCH v2] sensors-detect: Add code to detect TMP435

[Guenter Roeck <linux@xxxxxxxxxxxx>]

On 12/08/2014 02:40 AM, Jean Delvare wrote:
> Hi Guenter,
>
> On Sat,  6 Dec 2014 16:25:19 -0800, Guenter Roeck wrote:
> > Also strengthen chip detection for other TMP4xx chips,
> > and update driver support status for TMP431 and TMP432.
> >
> > Signed-off-by: Guenter Roeck <linux@xxxxxxxxxxxx>
> > ---
> > v2: Detect TMP435 on its entire address range
> >      Strengthen detect function for other TMP4xx chips
> >
> >   CHANGES                    |  1 +
> >   prog/detect/sensors-detect | 42 +++++++++++++++++++++++++++++++++---------
> >   2 files changed, 34 insertions(+), 9 deletions(-)
> >
> > diff --git a/CHANGES b/CHANGES
> > index 638a8bf..534b810 100644
> > --- a/CHANGES
> > +++ b/CHANGES
> > @@ -24,6 +24,7 @@ SVN HEAD
> >                     Document support for EMC1402, EMC1404, and EMC1424
> >                     Detect new revisions of EMC14xx
> >                     Add detection of EMC1422
> > +                  Add detection of TMP435
> >
> >   3.3.5 "Happy Birthday Beddy" (2014-01-22)
> >     libsensors: Improve documentation of two functions
> > diff --git a/prog/detect/sensors-detect b/prog/detect/sensors-detect
> > index 448cf22..55b3537 100755
> > --- a/prog/detect/sensors-detect
> > +++ b/prog/detect/sensors-detect
> > @@ -1029,15 +1029,20 @@ use vars qw(@i2c_adapter_names);
> >   		i2c_detect => sub { tmp42x_detect(@_, 2); },
> >   	}, {
> >   		name => "Texas Instruments TMP431",
> > -		driver => "to-be-written",		# tmp401
> > +		driver => "tmp401",
> >   		i2c_addrs => [0x4c, 0x4d],
> >   		i2c_detect => sub { lm90_detect(@_, 16); },
> >   	}, {
> >   		name => "Texas Instruments TMP432",
> > -		driver => "to-be-written",		# tmp401
> > +		driver => "tmp401",
> >   		i2c_addrs => [0x4c, 0x4d],
> >   		i2c_detect => sub { lm90_detect(@_, 17); },
> >   	}, {
> > +		name => "Texas Instruments TMP435",
> > +		driver => "tmp401",
> > +		i2c_addrs => [0x37, 0x48..0x4f],
> > +		i2c_detect => sub { lm90_detect(@_, 19); },
> > +	}, {
> >   		name => "Texas Instruments TMP441",
> >   		driver => "tmp421",
> >   		i2c_addrs => [0x1c..0x1f, 0x2a, 0x4c..0x4f],
> > @@ -4674,11 +4679,13 @@ sub max6680_95_detect
> >   #		  8 = W83L771W/G, 9 = TMP401, 10 = TMP411,
> >   #		  11 = W83L771AWG/ASG, 12 = MAX6690,
> >   #		  13 = ADT7461A/NCT1008, 14 = SA56004,
> > -#		  15 = G781, 16 = TMP431, 17 = TMP432, 18 = TMP451
> > +#		  15 = G781, 16 = TMP431, 17 = TMP432, 18 = TMP451,
> > +#		  19 = TMP435
> >   # Registers used:
> >   #   0x03: Configuration
> >   #   0x04: Conversion rate
> >   #   0xbf: Configuration 2 (National Semiconductor, Winbond, and Philips only)
> > +#   0x1a: Configuration 2 (TI only)
> >   #   0xfe: Manufacturer ID
> >   #   0xff: Chip ID / die revision
> >   sub lm90_detect
> > @@ -4688,7 +4695,13 @@ sub lm90_detect
> >   	my $cid = i2c_smbus_read_byte_data($file, 0xff);
> >   	my $conf = i2c_smbus_read_byte_data($file, 0x03);
> >   	my $rate = i2c_smbus_read_byte_data($file, 0x04);
> > -	my $conf2 = i2c_smbus_read_byte_data($file, 0xbf);
> > +	my $conf2;
> > +
> > +	if ($chip == 9 || $chip == 10 || ($chip >= 16 && $chip <= 19)) {
> > +		$conf2 = i2c_smbus_read_byte_data($file, 0x1a);
> > +	} else {
> > +		$conf2 = i2c_smbus_read_byte_data($file, 0xbf);
> > +	}
>
> The TMP431/TMP432 datasheet I have claims that Configuration Register 2
> is at 0x3f for the TMP432 (surprisingly, I have to say.)
You are right; I overlooked the second register table in the datasheet.
Those chips have weird register addresses.

> At this point I am wondering if it wouldn't make more sense to split
> support for the TI chips to a separate function.
Agreed, makes sense. I'll do that.

> >   	if ($chip == 0) {
> >   		return if ($conf & 0x2a) != 0;
> > @@ -4748,17 +4761,19 @@ sub lm90_detect
> >   	}
> >   	if ($chip == 9) {
> >   		return if ($conf & 0x1B) != 0;
> > +		return if ($conf2 & 0xfc) != 0x1c;
> >   		return if $rate > 0x0F;
> >   		return if $mid != 0x55;		# Texas Instruments
> >   		return 8 if $cid == 0x11;	# TMP401
> >   	}
> >   	if ($chip == 10) {
> >   		return if ($conf & 0x1B) != 0;
> > +		return if ($conf2 & 0xfc) != 0x1c;
> >   		return if $rate > 0x0F;
> >   		return if $mid != 0x55;		# Texas Instruments
> > -		return 6 if ($addr == 0x4c && $cid == 0x12); # TMP411A
> > -		return 6 if ($addr == 0x4d && $cid == 0x13); # TMP411B
> > -		return 6 if ($addr == 0x4e && $cid == 0x10); # TMP411C
> > +		return 8 if ($addr == 0x4c && $cid == 0x12); # TMP411A
> > +		return 8 if ($addr == 0x4d && $cid == 0x13); # TMP411B
> > +		return 8 if ($addr == 0x4e && $cid == 0x10); # TMP411C
> >   	}
> >   	if ($chip == 11) {
> >   		return if ($conf & 0x2a) != 0;
> > @@ -4794,15 +4809,17 @@ sub lm90_detect
> >   	}
> >   	if ($chip == 16) {
> >   		return if ($conf & 0x1B) != 0;
> > +		return if ($conf2 & 0xe3) != 0;
> >   		return if $rate > 0x0F;
> >   		return if $mid != 0x55;		# Texas Instruments
> > -		return 6 if ($cid == 0x31);	# TMP431A/B/C/D
> > +		return 8 if ($cid == 0x31);	# TMP431A/B/C/D
> >   	}
> >   	if ($chip == 17) {
> >   		return if ($conf & 0x1B) != 0;
> > +		return if ($conf2 & 0xe3) != 0;
>
> With the datasheet I have, the mask should be 0xC3 here.
Yes. Again, looked at wrong table.

Thanks,
Guenter

> >   		return if $rate > 0x0F;
> >   		return if $mid != 0x55;		# Texas Instruments
> > -		return 6 if ($cid == 0x32);	# TMP432A/B
> > +		return 8 if ($cid == 0x32);	# TMP432A/B
> >   	}
> >   	if ($chip == 18) {
> >   		return if ($conf & 0x1B) != 0;
> > @@ -4810,6 +4827,13 @@ sub lm90_detect
> >   		return if $mid != 0x55;		# Texas Instruments
> >   		return 4 if ($cid == 0x00);	# TMP451
> >   	}
> > +	if ($chip == 19) {
> > +		return if ($conf & 0x1B) != 0;
> > +		return if ($conf2 & 0xe3) != 0;
> > +		return if $rate > 0x0F;
> > +		return if $mid != 0x55;		# Texas Instruments
> > +		return 8 if ($cid == 0x35);	# TMP435
> > +	}
> >   	return;
> >   }


_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors