> I tend to agree this is not ideal. But IMO the issue is that family is > treated as an int throughout the code rather than u64. Even u32 would > have been better than int because negative numbers are not allowed AFAICT > just skimming the code. > > Unfortunately, ripping through the code to change family to u32 is > probably not worth the churn. I'll think on this but I'm tempted to apply > this. > > Ira Hi, I believe this patch is better suited for the stable branches. Additionally, replacing int to u32 or u64 in all relevant parts of the code seems too risky, as it could potentially introduce new bugs. Given the discussion so far, would it be appropriate to resend the same patch, but with a more detailed commit message this time? Thank you
