binary.Read/Write already uses reflect as far as I can tell, so it wouldn't matter in that case. There is an optimization for numeric values in there, which may be what you are thinking of? But they are specified in the spec and that's not what we're talking about anyway (and if an optimization really had to go, for whatever reason, it wouldn't be the end of the world).
Basically, binary.Read/binary.Write are capable of reading/writing struct values. The worry would be that, if the Go compiler reordered fields under the hood, the order may differ between writing the data and reading it back (especially across different versions of Go).
However, because these functions use reflection, they likely wouldn't be affected. While the in-memory layout of the struct fields might get reordered, presumably the reflection order would match the declaration order. Indeed, there is already an Offset field on the reflect.StructField type, and there is no statement anywhere I can find that such offsets must increase monotonically.
So, the fields would remain in declaration order when inspected with reflection, but their offsets could jump around within the struct, yet well behaved reflection-based code should be agnostic to this change.
binary.Read/Write already uses reflect as far as I can tell, so it wouldn't matter in that case. There is an optimization for numeric values in there, which may be what you are thinking of? But they are specified in the spec and that's not what we're talking about anyway (and if an optimization really had to go, for whatever reason, it wouldn't be the end of the world).
Did you mean something else?