…ent::DerefMove.

…he MIR body.

…_mir.

…thods.

…FnAbi.

…rtual.

although, not sure why this works - it wasn't needed before

Configure sanitize option when compiling with a sanitizer to make
it possible to execute different code depending on whether given
sanitizer is enabled or not.

rustc: split FnAbi's into definitions/direct calls ("of_instance") and indirect calls ("of_fn_ptr").

After this PR:
* `InstanceDef::Virtual` is only used for "direct" virtual calls, and shims around those calls use `InstanceDef::ReifyShim` (i.e. for `<dyn Trait as Trait>::f as fn(_)`)
  * this could easily be done for intrinsics as well, to allow their reification, but I didn't do it
* `FnAbi::of_instance` is **always** used for declaring/defining an `fn`, and for direct calls to an `fn`
  * this is great for e.g. rust-lang#65881 (`#[track_caller]`), which can introduce the "caller location" argument into "codegen signatures" by only changing `FnAbi::of_instance`, after this PR
* `FnAbi::of_fn_ptr` is used primarily for indirect calls, i.e. to `fn` pointers
  * *not* virtual calls (which use `FnAbi::of_instance` with `InstanceDef::Virtual`)
  * there's also a couple uses where the `rustc_codegen_llvm` needs to declare (i.e. FFI-import) an LLVM function that has no Rust declaration available at all
    * at least one of them could probably be a "weak lang item" instead

As there are many steps, this PR is best reviewed commit by commit - some of which arguably should be in their own PRs, I may have gotten carried away a bit.

cc @nagisa @rkruppe @oli-obk @anp

Conditional compilation for sanitizers

Configure sanitize option when compiling with a sanitizer to make
it possible to execute different code depending on whether given
sanitizer is enabled or not.

…oli-obk

handle diverging functions forwarding their return place

Fixes rust-lang/miri#1075: the shim around diverging closures turned into function pointers actually "obtains" a return place inside a diverging function, but just uses it as the return place for a diverging callee. Handle this by using NULL places.

This is kind of a hack as it breaks our invariant that all places are dereferencable, but we'd eventually let raw pointers break that anyway I assume so that seems fine.

r? @oli-obk

rustbuild fixes

When upgrading Debian's rustc to 1.38 I needed these patches:

(1) In order to cross-compile rustc 1.38 and take it through the full rustbuild process including install, I needed the first patch.

(2) In order to build rustc 1.38 using rustc 1.38 itself I need to set --cap-lints warn, otherwise I get this error:

~~~~
error: unnecessary `unsafe` block
   --> src/bootstrap/builder.rs:148:19
    |
148 |             name: unsafe { ::std::intrinsics::type_name::<S>() },
    |                   ^^^^^^ unnecessary `unsafe` block
    |
note: lint level defined here
   --> src/bootstrap/lib.rs:107:9
    |
107 | #![deny(warnings, rust_2018_idioms, unused_lifetimes)]
    |         ^^^^^^^^
    = note: `#[deny(unused_unsafe)]` implied by `#[deny(warnings)]`

error: aborting due to previous error

error: could not compile `bootstrap`.

~~~~

In order to set --cap-lints warn however, I need bootstrap.py not to clobber RUSTFLAGS. (This worked previously, not sure if it was broken intentionally but we would like support for it.)