Re: [coreboot] superiotool sometimes skips further SuperIO chips in the system - suggested remedy included

[Peter Stuge]

Frantisek Rysanek wrote:
> I'd like to offer a bit of my own time to add such a global index of 
> "ports occupied" (likely not "devices detected" - see above why).

Great! Please send the patch with Signed-off-by according to the
coreboot development guidelines: (they are short)

http://www.coreboot.org/Development_Guidelines#How_to_contribute
> Would you be interested?

Yes. I suggest using a fixed size array for the ports:

#define MAX_SUPERIOS 10
static unsigned char found_superios = 0;
static unsigned short superio_port[MAX_SUPERIOS];

and when looping over devices, skip already-known ports:

for(each device) {
  for(i=0;i<found_superios;i++)
    if(cur_device_port == superio_port[i])
      goto nextdev;
  if(probe(cur_device_port))
    superio_port[found_superios++] = cur_device_port;
nextdev:
}

//Peter