14.01.2016 11:23, Harald Welte пишет: > we generally don't use 'unsigned' in the code so far, but 'unsigned > int'. So it's a bit odd to see 'unsigned' without int here. In > general, I probably wouldn't care too much. However, given that it is a > single function that has one 'unsigned int' and one 'unsigned' argument, > it seems a bit odd.
I prefer 'unsigned' because it's less to type and neither provides any guarantees with regards to size of the type. I think size_t and uintXX_t would be better technically. On a related note, what's the status of https://patchwork.ozlabs.org/patch/559571/ ? Cause merging it would result in conflict with this patch so I'd prefer to rebase mine earlier if possible. cheers, Max.
