ASB100 PWM

ASB100 PWM

[Mark M. Hoffman]

vitalyb wrote in ticket #1437:

> asb100??s chip address: 0x2d
> pwm data address: 0x59
> default value: 0x8f
> 
> lower nybble (with a mask 0x0f) regulates Power and Chassis Fans, but not a CPU
> fan
> 
> it looks like if higher (7) bit is set, manual control is enabled, ie if
> register contains 0x85 fan??s speed will be around 5/16
> 
> but if bits 5 and/or 6 are set and bit 7 is not, fan speed is constantly
> lowering (approx. every 4 sec.)

So, I tried the attached patch (vs. current CVS) with no result here:
modifying the pwm did not change any fan speeds at all.  But maybe
there are different ASB100 revs?  Who knows?  The patch didn't
seem to hurt anything, so if it works for you I'll commit it.

BTW: thanks for the info... anything else you can find out about
that chip, please email (sensors@stimpy.netroedge.com) us.

Regards,

-- 
Mark M. Hoffman
mhoffman@lightlink.com

-------------- next part --------------
Index: kernel/chips/asb100.c
=================================RCS file: /home/cvs/lm_sensors2/kernel/chips/asb100.c,v
retrieving revision 1.1
diff -u -r1.1 asb100.c
--- kernel/chips/asb100.c	29 Oct 2003 05:37:52 -0000	1.1
+++ kernel/chips/asb100.c	11 Nov 2003 06:47:24 -0000
@@ -29,7 +29,7 @@
     This driver supports the hardware sensor chip: Asus ASB100(A) BACH
 
     Chip	#vin	#fanin	#pwm	#temp	wchipid	vendid	i2c	ISA
-    asb100	7	3	0	4	0x31	0x0694	yes	no
+    asb100	7	3	1	4	0x31	0x0694	yes	no
 */
 
 //#define DEBUG 1
@@ -102,7 +102,9 @@
 #define ASB100_REG_BEEP_INTS1	0x56
 #define ASB100_REG_BEEP_INTS2	0x57
 #define ASB100_REG_WCHIPID	0x58
-#define ASB100_REG_DIODE	0x59
+
+/* bit 7 -> enable, bits 0-3 -> duty cycle */
+#define ASB100_REG_PWM1		0x59
 
 /* <TODO> Does this exist on ASB100? */
 /* The following are undocumented in the data sheets however we
@@ -186,6 +188,19 @@
 	return ((s16)reg / 128) * 5;
 }
 
+/* PWM: 0 - 255 per sensors documentation
+   REG: (6.25% duty cycle per bit) */
+static u8 ASB100_PWM_TO_REG(int pwm)
+{
+	pwm = SENSORS_LIMIT(pwm, 0, 255);
+	return (u8)(pwm / 16);
+}
+
+static int ASB100_PWM_FROM_REG(u8 reg)
+{
+	return reg * 16;
+}
+
 #define ALARMS_FROM_REG(val) (val)
 #define BEEPS_FROM_REG(val) (val)
 #define BEEPS_TO_REG(val) ((val) & 0xffffff)
@@ -228,6 +243,7 @@
 	u16 temp_over[4];	/* Register value (0 and 3 are u8 only) */
 	u16 temp_hyst[4];	/* Register value (0 and 3 are u8 only) */
 	u8 fan_div[3];		/* Register encoding, right justified */
+	u8 pwm;			/* Register encoding */
 	u8 vid;			/* Register encoding, combined */
 	u32 alarms;		/* Register encoding, combined */
 	u32 beeps;		/* Register encoding, combined */
@@ -263,6 +279,8 @@
 		int ctl_name, int *nrels_mag, long *results);
 static void asb100_fan_div(struct i2c_client *client, int operation,
 		int ctl_name, int *nrels_mag, long *results);
+static void asb100_pwm(struct i2c_client *client, int operation,
+		int ctl_name, int *nrels_mag, long *results);
 
 static struct i2c_driver asb100_driver = {
 	.owner		= THIS_MODULE,
@@ -296,6 +314,8 @@
 #define ASB100_SYSCTL_VID	1300	/* Volts * 1000 */
 #define ASB100_SYSCTL_VRM	1301
 
+#define ASB100_SYSCTL_PWM1	1401	/* 0-255 => 0-100% duty cycle */
+
 #define ASB100_SYSCTL_FAN_DIV	2000	/* 1, 2, 4 or 8 */
 #define ASB100_SYSCTL_ALARMS	2001	/* bitvector */
 #define ASB100_SYSCTL_BEEP	2002	/* bitvector */
@@ -362,7 +382,8 @@
 	 &i2c_sysctl_real, NULL, &asb100_alarms},
 	{ASB100_SYSCTL_BEEP, "beep", NULL, 0, 0644, NULL, &i2c_proc_real,
 	 &i2c_sysctl_real, NULL, &asb100_beep},
-
+	{ASB100_SYSCTL_PWM1, "pwm1", NULL, 0, 0644, NULL, &i2c_proc_real,
+	 &i2c_sysctl_real, NULL, &asb100_pwm},
 	{0}
 };
 
@@ -784,6 +805,9 @@
 		data->fan_div[2] = (asb100_read_value(client,
 				ASB100_REG_PIN) >> 6) & 0x03;
 
+		/* PWM */
+		data->pwm = asb100_read_value(client, ASB100_REG_PWM1);
+
 		/* alarms */
 		data->alarms = asb100_read_value(client, ASB100_REG_ALARM1) +
 			(asb100_read_value(client, ASB100_REG_ALARM2) << 8);
@@ -1040,6 +1064,33 @@
 			data->fan_div[0] = DIV_TO_REG(results[0]);
 			old = (old & 0xcf) | ((data->fan_div[0] & 0x03) << 4);
 			asb100_write_value(client, ASB100_REG_VID_FANDIV, old);
+		}
+	}
+}
+
+void asb100_pwm(struct i2c_client *client, int operation, int ctl_name,
+		int *nrels_mag, long *results)
+{
+	struct asb100_data *data = client->data;
+
+	if (operation = SENSORS_PROC_REAL_INFO)
+		*nrels_mag = 0;
+	else if (operation = SENSORS_PROC_REAL_READ) {
+		asb100_update_client(client);
+		results[0] = ASB100_PWM_FROM_REG(data->pwm & 0x0f);
+		results[1] = (data->pwm & 0x80) ? 1 : 0;
+		*nrels_mag = 2;
+	} else if (operation = SENSORS_PROC_REAL_WRITE) {
+		u8 val = data->pwm;
+		if (*nrels_mag >= 1) {
+			val = 0x0f & ASB100_PWM_TO_REG(results[0]);
+			if (*nrels_mag >= 2) {
+				if (results[1])
+					val |= 0x80;
+				else
+					val &= ~0x80;
+			}
+			asb100_write_value(client, ASB100_REG_PWM1, val);
 		}
 	}
 }

ASB100 PWM

[Jean Delvare]

[Mark Hoffman asks:]

> But maybe there are different ASB100 revs?  Who knows?

Let's ask Alex about that.  Alex, do you know of different ASB100
("Bach") revisions that would handle PWM differently?

-- 
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/

ASB100 PWM

[Vitaly V. Bursov]

On Tue, 11 Nov 2003 01:29:29 -0500
"Mark M. Hoffman" <mhoffman@lightlink.com> wrote:

> vitalyb wrote in ticket #1437:
> 
> > asb100??s chip address: 0x2d
> > pwm data address: 0x59
> > default value: 0x8f
> > 
> > lower nybble (with a mask 0x0f) regulates Power and Chassis Fans, but not a CPU
> > fan
> > 
> > it looks like if higher (7) bit is set, manual control is enabled, ie if
> > register contains 0x85 fan??s speed will be around 5/16
> > 
> > but if bits 5 and/or 6 are set and bit 7 is not, fan speed is constantly
> > lowering (approx. every 4 sec.)
> 
> So, I tried the attached patch (vs. current CVS) with no result here:
> modifying the pwm did not change any fan speeds at all.  But maybe
> there are different ASB100 revs?  Who knows?  The patch didn't
> seem to hurt anything, so if it works for you I'll commit it.

$ echo $[ 0x80 ] 1 > pwm1
It works for me! Thanks!


I have an Asus A7V8X mobo rev 1.04, ASB100 chip marking follows

ASB100-A    Bach
TPV268981   0240E

i2cdump`s output for banks 0-4 is attached.

And I checked a photo in the manual. It is a mobo rev 1.02
ASB100 marking:

ASB100-A    Bach
TKF100471AE 0227

Looks differently...

> BTW: thanks for the info... anything else you can find out about
> that chip, please email (sensors@stimpy.netroedge.com) us.

OK.

-- 
Regards,
Vitaly V. Bursov                    mailto:vitalyb@mail333.com
GPG Key ID: F95A23B9                http://www.gnupg.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: asb100-dump.gz
Type: application/octet-stream
Size: 683 bytes
Desc: not available
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20031111/0c1fbb89/asb100-dump.obj

ASB100 PWM

[MBM Support]

sorry for the delay, lot of real life work :)

as far as I know there are 2 bach's but I have never worked much with the
fans, only how to read them, so not only late answer but also a crappy
one... sorry

Regards, Alex

- Please always attach all previous mails !
----------------------------------------------------------------------
The Motherboard Monitor:
http://mbm.livewiredev.com/
----------------------------------------------------------------------
For long you live and high you fly,
And smiles you'll give and tears you'll cry,
And all you touch and all you see,
Is all your life will ever be.
----------------------------------------------------------------------



-----Original Message-----
From: Jean Delvare [mailto:khali@linux-fr.org]
Sent: dinsdag 11 november 2003 08:56
To: Alex
Cc: Sensors; vitalyb@mail333.com
Subject: Re: ASB100 PWM




[Mark Hoffman asks:]

> But maybe there are different ASB100 revs?  Who knows?

Let's ask Alex about that.  Alex, do you know of different ASB100
("Bach") revisions that would handle PWM differently?

--
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/

ASB100 PWM

[Jean Delvare]

> as far as I know there are 2 bach's but I have never worked much with
> the fans, only how to read them, so not only late answer but also a
> crappy one... sorry

Maybe you can at least tell us how one can tell the difference between
the two revisions? Some register we can read the information in?

Thanks.

-- 
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/

ASB100 PWM

[MBM Support]

this is my detection routine for it:

for Z:=$29 to $2F do
begin
  SMBChipAddress:=Z;
  UsedBus:=btSMBus;
  IOByte($4E,$00,IOWrite);
  X:=IOByte($58,$0,IORead);

  if X=$31 then
  begin
    if (IOByte($4F,$0,IORead)=$94) or (IOByte($4F,$0,IORead)=$06) then
    begin
      ChipType:=ASB100;
      // etc etc
    end
    else
    begin
      ChipType:=AS99127F;
      // etc etc
    end;
  end;
  if (X=$56) or (X=$10) then
  begin
    if ((IOByte($4E,$0,IORead)=$94) and (IOByte($4F,$0,IORead)=$06))
    or ((IOByte($4E,$0,IORead)=$94) and (IOByte($4F,$0,IORead)=$36))
    or ((IOByte($4E,$0,IORead)=$5C) and (IOByte($4F,$0,IORead)=$A3))then

    //Asus ASM58 Mozart-2            $58=$56 $4E=$94 $4F=$36
    //Asus AS2K129R Mozart-2         $58=$56 $4E=$94 $4F=$06
    //Asus XXXXXXXX Mozart-2         $58=$10 $4E=$5C $4F=$A3
    begin
      ChipType:=MOZART2;
      // etc etc
    end;
  end;

question for you, heard/know of the VIA VT6307 ?

Regards, Alex

- Please always attach all previous mails !
----------------------------------------------------------------------
The Motherboard Monitor:
http://mbm.livewiredev.com/
----------------------------------------------------------------------
For long you live and high you fly,
And smiles you'll give and tears you'll cry,
And all you touch and all you see,
Is all your life will ever be.
----------------------------------------------------------------------   

 

-----Original Message-----
From: Jean Delvare [mailto:khali@linux-fr.org]
Sent: woensdag 12 november 2003 22:58
To: MBM Support
Cc: sensors@Stimpy.netroedge.com; vitalyb@mail333.com
Subject: Re: ASB100 PWM




> as far as I know there are 2 bach's but I have never worked much with
> the fans, only how to read them, so not only late answer but also a
> crappy one... sorry

Maybe you can at least tell us how one can tell the difference between
the two revisions? Some register we can read the information in?

Thanks.

-- 
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/

ASB100 PWM

[Jean Delvare]

>     if (IOByte($4F,$0,IORead)=$94) or (IOByte($4F,$0,IORead)=$06) then
>     begin
>       ChipType:=ASB100;
>       // etc etc
>     end

OK, so you don't actually make any difference between both revisions.
Thanks anyway.

> question for you, heard/know of the VIA VT6307 ?

No, never heard of it, sorry.

-- 
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/

ASB100 PWM

[Mark M. Hoffman]

* Vitaly V. Bursov <vitalyb@mail333.com> [2003-11-11 14:48:52 +0200]:

> $ echo $[ 0x80 ] 1 > pwm1
> It works for me! Thanks!

OK great!  It's committed.

> I have an Asus A7V8X mobo rev 1.04, ASB100 chip marking follows
> 
> ASB100-A    Bach
> TPV268981   0240E
> 
> i2cdump`s output for banks 0-4 is attached.
> 
> And I checked a photo in the manual. It is a mobo rev 1.02
> ASB100 marking:
> 
> ASB100-A    Bach
> TKF100471AE 0227
> 
> Looks differently...

Indeed, mine says this:

ASB100      Bach
TSR393731AD 0148

I'm therefore assuming that ASB100-A supports pwm1, while
plain ASB100 does not.  Looking through your attached
chip dump (quickly, I'll admit) I don't see a way to
distinguish the two.

Regards,

-- 
Mark M. Hoffman
mhoffman@lightlink.com

ASB100 PWM

[Jean Delvare]

> I'm therefore assuming that ASB100-A supports pwm1, while
> plain ASB100 does not.  Looking through your attached
> chip dump (quickly, I'll admit) I don't see a way to
> distinguish the two.

Care to send me your dump? I already have Vitaly's one. I might see
something you missed. I've become quite good at reading dumps ;)

-- 
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/

ASB100 PWM

[Mark M. Hoffman]

* Jean Delvare <khali@linux-fr.org> [2003-11-15 08:11:46 +0100]:
> 
> > I'm therefore assuming that ASB100-A supports pwm1, while
> > plain ASB100 does not.  Looking through your attached
> > chip dump (quickly, I'll admit) I don't see a way to
> > distinguish the two.
> 
> Care to send me your dump? I already have Vitaly's one. I might see
> something you missed. I've become quite good at reading dumps ;)

Thanks Jean...

  WARNING! This program can confuse your I2C bus, cause data loss and worse!
  I will probe file /dev/i2c-0, address 0x2d, mode byte
  You have five seconds to reconsider and press CTRL-C!

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: b9 bd 9d 71 bf c8 00 cd a7 f8 a5 ac 8c 68 54 f8    ???q??.??????hT?
10: a5 f8 a5 ff 00 00 00 d4 46 41 24 50 4b 00 00 00    ???....?FA$PK...
20: 5e 5f d0 bb b8 b8 b8 1b 7e 92 94 62 59 62 59 d8    ^_??????~??bYbY?
30: c4 c3 b0 d9 b2 00 00 00 00 2d 28 a9 a9 a9 00 00    ?????....-(???..
40: 01 00 00 00 00 00 00 6e 00 22 01 40 00 00 80 06    ?......n."?@..??
50: ff ff 00 ff ff ff ff ff 31 8f 00 d5 ff 48 e2 1f    ........1?.?.H??
60: 5f 5f d0 bb b7 b7 b7 1b 7e 93 94 62 59 62 59 d8    __??????~??bYbY?
70: c4 c3 b0 d9 b2 00 00 00 00 2d 28 a9 a9 a9 00 00    ?????....-(???..
80: b9 bc 9d 71 bf c8 00 cd a7 f8 a5 ac 8c 68 54 f8    ???q??.??????hT?
90: a5 f8 a5 ff 00 00 00 d4 46 41 24 50 4b 00 00 00    ???....?FA$PK...
a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
  WARNING! This program can confuse your I2C bus, cause data loss and worse!
  I will probe file /dev/i2c-0, address 0x2d, mode byte
  Probing bank 1 using bank register 0x4e.
  You have five seconds to reconsider and press CTRL-C!

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 80 00 00 00 ff ff ff ff ff ff ff ff ff ff ff ff    ?...............
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
20: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff 81 ff    ..............?.
50: 1c 80 00 32 00 3c 00 ff ff ff ff ff ff ff ff ff    ??.2.<..........
60: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
80: 1c 80 32 3c ff ff ff ff ff ff ff ff ff ff ff ff    ??2<............
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
  WARNING! This program can confuse your I2C bus, cause data loss and worse!
  I will probe file /dev/i2c-0, address 0x2d, mode byte
  Probing bank 2 using bank register 0x4e.
  You have five seconds to reconsider and press CTRL-C!

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 80 00 00 00 ff ff ff ff ff ff ff ff ff ff ff ff    ?...............
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
20: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff 82 ff    ..............?.
50: ff 80 00 28 00 2d 00 ff ff ff ff ff ff ff ff ff    .?.(.-..........
60: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
80: ff 80 28 2d ff ff ff ff ff ff ff ff ff ff ff ff    .?(-............
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
  WARNING! This program can confuse your I2C bus, cause data loss and worse!
  I will probe file /dev/i2c-0, address 0x2d, mode byte
  Probing bank 3 using bank register 0x4e.
  You have five seconds to reconsider and press CTRL-C!

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 00 00 ff ff 00 00 08 02 ff ff 00 ff 55 50 ff 00    ......??....UP..
10: 00 00 00 ff ff ff ff ff 11 01 ff ff 1f 9f 00 00    ........??..??..
20: 00 00 00 00 00 00 00 08 00 00 ff ff ff 00 ff ff    .......?........
30: 00 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff 83 ff    ..............?.
50: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
60: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
80: ff 1c 1c cf 00 00 00 ff ff ff ff ff ff ff ff ff    .???............
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ................
e0: 15 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff    ?...............
f0: 10 00 ff 00 00 00 ff ff ff ff ff ff ff ff ff ff    ?...............

Regards,

-- 
Mark M. Hoffman
mhoffman@lightlink.com

ASB100 PWM

[Jean Delvare]

> > Care to send me your dump? I already have Vitaly's one. I might see
> > something you missed. I've become quite good at reading dumps ;)
> 
> Thanks Jean...
> (...)
>   Probing bank 1 using bank register 0x4e.

Hu-oh. I had forgotten that this beast has several banks (I still can't
understand why manufacturers to this since there is less that 256
registers total).

Mark, can you explain to Vitaly how to produce a complete dump, as you
just did?

I'd also need at least a third user's dump, together with the model
information (A or not). Without that, there will be too many registers
that could be the one we are after. Of course, more dumps would be
welcome. The more we have, the easier the work will be.

-- 
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/

ASB100 PWM

[Mark M. Hoffman]

* Jean Delvare <khali@linux-fr.org> [2003-11-18 21:07:01 +0100]:

> Hu-oh. I had forgotten that this beast has several banks (I still can't
> understand why manufacturers to this since there is less that 256
> registers total).
> 
> Mark, can you explain to Vitaly how to produce a complete dump, as you
> just did?

Look again, he already did that.

> I'd also need at least a third user's dump, together with the model
> information (A or not). Without that, there will be too many registers
> that could be the one we are after. Of course, more dumps would be
> welcome. The more we have, the easier the work will be.

OK, if the opportunity comes up.  Really I'm not sure the effort 
would be worth it... my ASB100 !A ignores writes to pwm1 with no
harm done AFAICS.

Regards,

-- 
Mark M. Hoffman
mhoffman@lightlink.com

ASB100 PWM

[Jean Delvare]

> > Mark, can you explain to Vitaly how to produce a complete dump, as
> > you just did?
> 
> Look again, he already did that.

Oh, you're right. Sorry.

> > I'd also need at least a third user's dump, together with the model
> > information (A or not). Without that, there will be too many
> > registers that could be the one we are after. Of course, more dumps
> > would be welcome. The more we have, the easier the work will be.
> 
> OK, if the opportunity comes up.  Really I'm not sure the effort 
> would be worth it... my ASB100 !A ignores writes to pwm1 with no
> harm done AFAICS.

You're right. I sure have better things to do, and actually it's simply
impossible to come to a conclusion with only two dumps. Let's keep them
in a corner until the day we have at least two more, then we'll see.
After all, we don't have a datasheet for the ASB100 so we shouldn't even
be supporting it.

That said, had I to bet on a register, I would choose 0x85 in bank 3.
But I admit I must have less than 1% chance to have it right ;)

-- 
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/