3.3: haven't you heard of set(cc) instruction?

[IAmAThousandTrees]

in section 3.3, "branchless programming" the statement "There are no Boolean types in assembly, nor any instructions that yield either one or zero based on the result of the comparison" is patently false. Although there are frequently better ways to implement predication than using set, claiming it doesn't exist is at least 20 years out of date. every ISA I know of has some equivalent condition-code→value coversion.

I notice you don't seem to look at or fix issues raised here any more, but I raise the issue anyway since I don't know how much of the following text and code samples would have to be changed...