Associate an allocator to boxes

[glandium]

This turns Box<T> into Box<T, A: Alloc = Global>. This is a
minimalist change to achieve this, not touching anything that could have
backwards incompatible consequences like requiring type annotations in
places where they currently aren't required,
per #50822 (comment)

This is a rebased version of #50882, meant primarily to get CI results and see if #52694 is still a problem.

Cc: @SimonSapin

[rust-highfive]

r? @joshtriplett

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

[rust-highfive]

The job x86_64-gnu-llvm-5.0 of your PR failed on Travis (raw log). Through arcane magic we have determined that the following fragments from the build log may contain information about the problem.

Click to expand the log.

[00:47:26] .................................................................................................... 2200/4605
[00:47:30] ...................i................................................................................ 2300/4605
[00:47:34] .................................................................................................... 2400/4605
[00:47:37] .................................................................................................... 2500/4605
[00:47:41] ................................iiiiiiiii........................................................... 2600/4605
[00:47:47] .................................................................................................... 2800/4605
[00:47:51] .................................................................................................... 2900/4605
[00:47:54] ......................................................i............................................. 3000/4605
[00:47:57] .................................................................................................... 3100/4605
---
travis_time:start:test_codegen
Check compiletest suite=codegen mode=codegen (x86_64-unknown-linux-gnu -> x86_64-unknown-linux-gnu)
[01:00:38] 
[01:00:38] running 111 tests
[01:00:41] i..ii...iii.......i...i.........i..iii...........i.....i.....ii...i.i.ii..............i...ii..ii.i.. 100/111
[01:00:41] ..iiii.....
[01:00:41] 
[01:00:41]  finished in 3.387
[01:00:41] travis_fold:end:test_codegen

---
[01:15:27]  right: `2`', libstd/sync/mutex.rs:661:13
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/once.rs:582:28
[01:15:27] thread '<unnamed>' panicked at 'Once instance has previously been poisoned', libstd/sync/once.rs:372:21
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/once.rs:610:28
[01:15:27] thread '<unnamed>' panicked at 'assertion failed: t1.join().is_ok()', libstd/sync/once.rs:638:9
[01:15:27] thread '<unnamed>' panicked at 'test panic in inner thread to poison RwLock', libstd/sync/rwlock.rs:774:13
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/rwlock.rs:711:13
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/rwlock.rs:641:13
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/rwlock.rs:652:13
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/rwlock.rs:652:13
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/rwlock.rs:617:13
[01:15:27] thread '<unnamed>' panicked at 'explicit panic', libstd/sync/rwlock.rs:629:13
[01:15:28] thread '<unnamed>' panicked at 'What the answer to my lifetimes dilemma is?', libstd/sys_common/remutex.rs:241:13
[01:15:28] .......................................F............................................................ 700/768
[01:15:28] thread '<unnamed>' panicked at 'index 2 and/or 4 in `"aé 💩"` do not lie on character boundary', libstd/sys_common/wtf8.rs:784:5
[01:15:28] thread '<unnamed>' panicked at 'index 5 and/or 8 in `"aé 💩"` do not lie on character boundary', libstd/sys_common/wtf8.rs:784:5
[01:15:28] thread '<unnamed>' panicked at 'assertion failed: is_code_point_boundary(self, new_len)', libstd/sys_common/wtf8.rs:329:9
[01:15:28] thread '<unnamed>' panicked at 'assertion failed: is_code_point_boundary(self, new_len)', libstd/sys_common/wtf8.rs:329:9
---
[01:15:37] 
[01:15:37] error: test failed, to rerun pass '--lib'
[01:15:37] 
[01:15:37] 
[01:15:37] command did not execute successfully: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo" "test" "--target" "x86_64-unknown-linux-gnu" "-j" "4" "--release" "--locked" "--color" "always" "--features" "panic-unwind jemalloc backtrace" "--manifest-path" "/checkout/src/libstd/Cargo.toml" "-p" "std" "--" "--quiet"
[01:15:37] 
[01:15:37] 
[01:15:37] failed to run: /checkout/obj/build/bootstrap/debug/bootstrap test
[01:15:37] Build completed unsuccessfully in 0:32:49
[01:15:37] Build completed unsuccessfully in 0:32:49
[01:15:37] Makefile:58: recipe for target 'check' failed
[01:15:37] make: *** [check] Error 1

The command "stamp sh -x -c "$RUN_SCRIPT"" exited with 2.
travis_time:start:2f0cee8e
$ date && (curl -fs --head https://google.com | grep ^Date: | sed 's/Date: //g' || true)
---
travis_time:end:1a7273f0:start=1539765920243640848,finish=1539765920364819527,duration=121178679
travis_fold:end:after_failure.4
travis_fold:start:after_failure.5
travis_time:start:1070e04b
$ cat ./obj/build/x86_64-unknown-linux-gnu/native/asan/build/lib/asan/clang_rt.asan-dynamic-i386.vers || true
cat: ./obj/build/x86_64-unknown-linux-gnu/native/asan/build/lib/asan/clang_rt.asan-dynamic-i386.vers: No such file or directory
travis_fold:end:after_failure.5
travis_fold:start:after_failure.6
travis_time:start:12ccf678
$ dmesg | grep -i kill

I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact @TimNN. (Feature Requests)

[glandium]

As expected #52694 is still a problem. There is no way around it, short of requiring LLVM 7 for system LLVM builds. One problem here being that there isn't even a llvm-7-tools package for any version of Ubuntu (and travis builds are using 16.04, which doesn't even have LLVM 6)

[Centril]

This PR introduces some new unsafe { .. }. It would be nice to have a comment on each usage.

[SimonSapin]

impl<T> From<Unique<T>> for NonNull<T> is safe and can be used here.

[TimNN]

Ping from triage! Is there a way to work around the LLVM issues, short of requiring at least LLVM 7.0? Raising the minimum LLVM version would probably involve some kind of team RFC, but I'm not sure which team would be most appropriate.

[bors]

☔ The latest upstream changes (presumably #55316) made this pull request unmergeable. Please resolve the merge conflicts.

[TimNN]

I'm closing this as blocked on #55842.

@glandium: If you can come up with a way that does not require LLVM 7.0, please feel free to re-open. Thanks for your PR!