Hi!

After getting a couple reports that the Artop driver doesn't work, I
decided to take a look at it ... well and I decided that this is
impossible to debug, so I rewrote it from scratch instead. I have no
documentation nor the actual hardware, so it may be wrong somewhere,
but I think definitely less wrong than the original one.

I must say I also thought about preserving the original one for it's
humoristic value, consider what the following statements do:

1) #define SPLIT_BYTE(B,H,L)       ((H)=(B>>4), (L)=(B-((B>>4)<<4)))

... computation of L is just wonderful - first shift B down, then up
again, this clears the lower four bits, and then subtract that from B.
What a nice way to write ((B) & 0xf) ...

2) ((art&0x06)==0x06)?"5":((art&0x05)==0x05)?"4":((art&0x04)==0x04)?"3":((art&0x03)==0x03)?"2":((art&0x02)==0x02)?"1":((art&0x01)==0x01)?"0":"?"

... this is a real pearl - what a cool way to compute (art - 1).

2a)

switch(speed) {
         case 7: return "6";
         case 6: return "5";
         case 5: return "4";
         case 4: return "3";
         case 3: return "2";
         case 2: return "1";
         case 1: return "0";
         case 0: return "?";
}

... another one, for the same purpose with a different chip version.

3)  ((0x00 << (2*drive->dn))

... no comment

-- 
Vojtech Pavlik
SuSE Labs