Am 04.06.2011 00:40 schrieb Peter Stuge:
Carl-Daniel Hailfinger wrote:
I had asked Steven to use decivolt precision to avoid wasting lots of space.
Oh dear.
Oh well. I hope he's not disappointed by having had to change this twice.
Did some magic chips appear which require millivolt precision?
One chip specs voltage with .05 V precision. It's a good idea to use SI standard units. It's a bad idea to use anything else. uint_16t also fits well for the so far only user of this information; the USB flashing protocol Stefan and I have worked on.
We'll see how it works out. I don't think a conversion to decivolts of already checked in code makes sense now, and I'd say we close this issue until we have serious problems with the size of the flashchip support table. My constification patch provided a 99% reduction for the .data section, so the immediate effect of a flashchip array shrink would only hit .rodata which is less of an issue.
Regards, Carl-Daniel