On 13 March 2014 21:28, Andreas Färber <afaerber@xxxxxxx> wrote: > int32 is a type used in softfloat that has weird at-least-as-wide > semantics and bit us in the past. I'm coming round to the opinion that we should just change all those to int32_t &c. I know we had a discussion about efficiency at one point and I was on the other side at that time, but I don't really think it matters much either way and having the types still lurking around is just confusing. Maybe after 2.0 (and we should nuke the STATUS macros too). Either way, int32 shouldn't be used outside the softfloat code and its callers. > I'm not sure if we have a policy about __u64 etc. in KVM code. Since > it'll be Linux-only I don't see problems currently; for cross-platform > parts we prefer uint64_t. Suggest to leave as is unless told otherwise. For ARM I took the view that __u64 were the kernel's types, not ours. The kernel header structs define fields as __u64 but for in-QEMU functions and variables we should use the posix uint64_t. thanks -- PMM -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
