rustbuild: Build libstd with ThinLTO

[alexcrichton]

This commit moves the standard library to get compiled with multiple codegen
units and ThinLTO like the compiler itself. This I would hope is the last major
step towards closing out #45320

[rust-highfive]

r? @nikomatsakis

(rust_highfive has picked a reviewer for you, use r? to override)

[alexcrichton]

This is a resubmission of #45529 now that #45655 has landed to hopefully enable this landing

Carrying over this approval as...

@bors: r=Mark-Simulacrum

[bors]

📌 Commit 832c943 has been approved by Mark-Simulacrum

[bors]

⌛ Testing commit 832c943 with merge abcf001e8772c24449cdd1401e599b62a1fa086e...

[bors]

💔 Test failed - status-travis

[kennytm]

dist-x86_64-musl, some debuginfo error which cannot be read because we assumed the output is valid UTF-8, which is not 🤦‍. Might be caused by #45511, assuming legit.

[00:57:34] ---- [debuginfo-gdb] debuginfo/vec.rs stdout ----
[00:57:34] NOTE: compiletest thinks it is using GDB without native rust support
[00:57:34] NOTE: compiletest thinks it is using GDB version 7011001
[00:57:34] thread '[debuginfo-gdb] debuginfo/vec.rs' panicked at 'called `Result::unwrap()` on an `Err` value: FromUtf8Error { bytes: [47, 99, 104, 101, 99, 107, 111, 117, 116, 47, 111, 98, 106, 47, 98, 117, 105, 108, 100, 47, 120, 56, 54, 95, 54, 52, 45, 117, 110, 107, 110, 111, 119, 110, 45, 108, 105, 110, 117, 120, 45, 103, 110, 117, 47, 116, 101, 115, 116, 47, 100, 101, 98, 117, 103, 105, 110, 102, 111, 47, 118, 101, 99, 46, 100, 101, 98, 117, 103, 103, 101, 114, 46, 115, 99, 114, 105, 112, 116, 58, 55, 58, 32, 69, 114, 114, 111, 114, 32, 105, 110, 32, 115, 111, 117, 114, 99, 101, 100, 32, 99, 111, 109, 109, 97, 110, 100, 32, 102, 105, 108, 101, 58, 10, 248, 236, 46, 249, 67, 127, 10], error: Utf8Error { valid_up_to: 114, error_len: Some(1) } }', /checkout/src/libcore/result.rs:906:4
[00:57:34] note: Run with `RUST_BACKTRACE=1` for a backtrace.
[00:57:34] [00:57:34] [00:57:34] failures:
[00:57:34] [debuginfo-gdb] debuginfo/vec.rs
[00:57:34] [00:57:34] test result: �[31mFAILED�(B�[m. 104 passed; 1 failed; 3 ignored; 0 measured; 0 filtered out

Content of those bytes:

/checkout/obj/build/x86_64-unknown-linux-gnu/test/debuginfo/vec.debugger.script:7: Error in sourced command file:
\xf8\xec.\xf9C\x7f

[alexcrichton]

The binary in question indeed does suffer from the same error message as #45511, so I think it's the same issue.

[alexcrichton]

I'm going to just assume this won't make progres until #45511 is fixed so I'm going to close this and focus efforts on that.