Hello list, hello Jean,
I've put ogether a patchset against the CVS from about five minutes ago,
including Documentation and all changes proposed in new_drivers (except for
libsensors support, which I don't think is really necessary on such a
chip). Following changes have been made to the code:
- Updated chip-notes
- i2c_check_functionality is only used once
- saa1064_write_value has been replaced by the appropriate
i2c_smbus_write_... functions in-place
- use of memset in saa1064_init_client()
- use of i2c_smbus_write_i2c_block_data to write the digits
- replaced pwrloss sysctl with refresh sysctl (see doc)
- added detection (four bytes read, first byte must be 0x80 or 0x00, others
0x00)
- removed dynamic and blank{13,24}, instead always using dynamic mode, all
digits
Hope everything went ok, I have tried the detection procedure on the pcf8574
(the only chip besides the SAA1064 I can play with) by force, and it has
not been mis-detected (except by forcing it, of course).
Jean, I will now install perl and see how your detect-script reacts to those
two chips, report will follow ;-)
Greetings,
Sascha
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lm_sensors2-cvs-saa1064.diff
Type: text/x-diff
Size: 19587 bytes
Desc: not available
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20030719/a2d923fb/attachment.bin
