Ticket 1267

Mark D. Studebaker mds at paradyne.com
Tue Jun 10 03:53:18 CEST 2003


thanks for testing the 801 fix.
We've added the ID to CVS.

On the 627THF, if you are feeling brave you can try investigating what device
ID is returned by super IO probing either by hacking sensors-detect
or at line 683 or so in the w83627hf driver.
Then add code to recognize it and see what happens.
There's no THF datasheet on the Winbond web site.

Matthias Hentges wrote:
> Hi again,
> 
> 
>>try adding 0x24d3 to the supported[] table in i2c-i801.c (2.7.0)
>>or to the pci_device_id[] table (CVS, get most recent)
>>let us know if it works, email us, ref. your ticket number.
> 
> 
> Adding the ID to supported[] did the trick with loading the module.
> At least for version 2.7.0, in CVS the modules loads, but the kernel
> oopses when running sensors-detect:
> 
> Jun  8 00:56:58 mhcln01 kernel: i2c-i801 version 2.8.0-CVS (2003????)
> Jun  8 00:56:58 mhcln01 kernel: i2c-dev.o: Registered '' as minor 0
> Jun  8 00:57:05 mhcln01 kernel: Unable to handle kernel paging request at virtua
> l address 0004002c
> Jun  8 00:57:05 mhcln01 kernel:  printing eip:
> Jun  8 00:57:05 mhcln01 kernel: f8889b20
> Jun  8 00:57:05 mhcln01 kernel: *pde = 00000000
> Jun  8 00:57:05 mhcln01 kernel: Oops: 0000
> Jun  8 00:57:05 mhcln01 kernel: CPU:    0
> Jun  8 00:57:05 mhcln01 kernel: EIP:    0010:[<f8889b20>]    Tainted: P 
> Jun  8 00:57:05 mhcln01 kernel: EFLAGS: 00010286
> Jun  8 00:57:05 mhcln01 kernel: eax: 00040004   ebx: 00000006   ecx: fffffffb   
> edx: f888bb3f
> Jun  8 00:57:05 mhcln01 kernel: esi: f3282000   edi: 00000000   ebp: 00000000   
> esp: f2833f44
> Jun  8 00:57:05 mhcln01 kernel: ds: 0018   es: 0018   ss: 0018
> Jun  8 00:57:05 mhcln01 kernel: Process sensors-detect (pid: 7759, stackpage=f28
> 33000)
> Jun  8 00:57:05 mhcln01 kernel: Stack: 00000000 00000c00 f3282000 00001000 fffff
> fea 00001000 c040b9b4 c015d54a 
> Jun  8 00:57:05 mhcln01 kernel:        f3282000 f2833f98 00000000 00000c00 f2833
> f94 00000000 00000000 f58c3200 
> :Jun  8 00:57:05 mhcln01 kernel:        ffffffea 00001000 f7188d80 00000000 00000
> 000 00000000 c013b047 f58c3200 
> Jun  8 00:57:05 mhcln01 kernel: Call Trace:    [<c015d54a>] [<c013b047>] [<c0108
> 9d3>]
> Jun  8 00:57:05 mhcln01 kernel: 
> Jun  8 00:57:05 mhcln01 kernel: Code: 83 78 28 00 74 1a 83 78 24 00 74 0a 83 c4 
> f8 68 40 bb 88 f8 
> 
> 
> On 2.7.0 sensors-detect finds several chips: two lm75, one pcf?? and two eeproms.
> The eeprom driver seems to work, but the two lm75 and the pcf?? don't.
> 
> Output of sensors-detect:
> 
> Next adapter: SMBus I801 adapter at 0400 (Non-I2C SMBus adapter)
> Do you want to scan it? (YES/no/selectively): 
> Client found at address 0x08
> Client found at address 0x2f
> [...]
> Client found at address 0x30
> Client found at address 0x44
> Client found at address 0x48
> Probing for `National Semiconductor LM75'... Success!
>     (confidence 3, driver `lm75')
> Probing for `Dallas Semiconductor DS1621'... Failed!
> Probing for `Philips Semiconductors PCF8591'... Success!
>     (confidence 1, driver `pcf8591')
> Client found at address 0x49
> Probing for `National Semiconductor LM75'... Success!
>     (confidence 3, driver `lm75')
> Probing for `Dallas Semiconductor DS1621'... Failed!
> Probing for `Philips Semiconductors PCF8591'... Success!
>     (confidence 1, driver `pcf8591')
> Client found at address 0x50
> Probing for `Serial EEPROM'... Success!
>     (confidence 8, driver `eeprom')
> Probing for `DDC monitor'... Failed!
> Client found at address 0x52
> Probing for `Serial EEPROM'... Success!
>     (confidence 8, driver `eeprom')
> Client found at address 0x61
> Probing for `SMBus 2.0 ARP-Capable Device'... Success!
>     (confidence 1, driver `smbus-arp')
> Client found at address 0x69
> 
> I have tried forcing the w83781d module:
> modprobe w83781d force_w83627hf=0,0x2f
> 
> i2cdump 0 0x2f:
> 
> [...]
> 
>      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
> 00: 00 00 00 00 00 00 00 00 00 80 00 00 01 00 00 ff    .........?..?...
> 10: 00 00 7f 00 00 10 00 00 00 00 00 00 00 00 00 00    ..?..?..........
> 20: 08 02 0d bd 02 02 02 7e ff ff ff 00 00 00 00 e3    ???????~.......?
> 30: ba cc a8 d8 b2 31 15 40 2c 3c 7f e1 e1 e1 ff ff    ?????1?@,<????..
> 40: 01 10 00 00 00 00 00 5f 2f 11 01 44 41 95 00 a3    ??....._/??DA?.?
> 50: ff ff ff ff ff ff 00 00 72 70 ff 00 19 01 05 ff    ........rp..???.
> 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: 01 ff 01 ff 00 00 00 00 01 01 01 01 3c 3c 0a 0a    ?.?.....????<<??
> 90: 00 00 00 01 ff 00 00 00 01 01 3c 18 00 00 01 01    ...?....??<?..??
> a0: ff ff ff 00 00 00 00 00 19 e7 6f 18 00 ff ff ff    ........??o?....
> b0: bc f3 02 ff 30 08 00 03 00 08 ff ff 00 04 ff ff    ???.0?.?.?...?..
> c0: 7e 80 00 32 00 3c 00 ff 7e 80 00 32 00 3c 00 ff    ~?.2.<..~?.2.<..
> 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 00 ff 00 01 00 ff ff 02 ff 00 00 0d ff 1d    .....?...?...?.?
> 
> The driver loads and can be accessed via /proc/sys/dev/sensors/w83627hf-i2c-0-2f
> but the values are all wrong.
> 
> 
>>On the 627THF, I´ll see if it is different than the regular HF if
>>I can find a datasheet.
>>Try modprobing i2c-isa and w83627hf anyway and see what happens.
> 
> 
> w83627hf can not be loaded (no such device).
> 
> Looks like the ISA bus can not be accessed.
> 
> Looks like i'm out of luck huh?
> 
> TIA



More information about the lm-sensors mailing list