This makes it so that less memory is used for IR instructions, as well
as catching bugs when one expected one kind of instruction and received
the other.
Previously, there was hacky code to deal with result locations and how
they work with regards to comptime values and runtime values. In
addition, there was a hacky "mem_slot" mechanism that managed the memory
for local variables, and acted differently depending on comptime vs
runtime situations. All that is deleted in this commit, and as a result,
result locations code has one less complication.
Importantly, this means that a comptime result location is now passed to
a function when it is evaluated at comptime.
This test causes many regressions, and some of the behavior tests are
disabled (commented out) in this commit. Future commits will re-enable
the tests before merging the branch.
Let's use the usual declaration-searching mechanism that resolves the
`usingnamespace` declarations on the go instead of directly peeking into
the symbol table.
Fixes#4134
Use a struct as second parameter to be future proof (and also allows to
specify default values for the parameters)
Closes#2679 as it was just a matter of a few lines of code.
Previously, the compiler had special logic to determine whether to
include the startup code, which was in `std/special/start.zig`. Now,
the file is moved to `std/start.zig`, and there is no special logic
in the compiler. Instead, the standard library unconditionally imports
the `start.zig` file, which then has a `comptime` block that does the
logic of determining what, if any, start symbols to export. Instead of
`start.zig` being in its own special package, it is just another normal
file that is part of the standard library.
`std.builtin.TestFn` is now part of the standard library rather than
specially generated by the compiler.
