Summary:
`constexpr` All The Things! Although only very recent versions of modern compilers support this, FunctionRef can and should be `constexpr` (demonstration here: https://godbolt.org/g/NJ72RG).

This change also changes the `FunctionRef` constructor such that it no longer participates in overload resolution if the argument is not callable with the specified arguments. In addition, it perfectly forwards the callable object such that if it was an rvalue when it was passed to the constructor, it will be invoked as an rvalue in `operator()`.

Reviewed By: yfeldblum

Differential Revision: D5366720

fbshipit-source-id: bc64053213478aab5a5bd5950c7b2d6f364d86bd

Summary:
Provides a `make_mock` interface to `folly::LeakySingleton` for unit test
mocking.

Reviewed By: andriigrynenko

Differential Revision: D5382003

fbshipit-source-id: 9ab81dd64703f4f02772bfb8c85e020680abcc85

Summary: Because MSVC is not happy about how it's written now.

Reviewed By: yfeldblum

Differential Revision: D5378740

fbshipit-source-id: 8b77d1a05da75b331e0cb72bd0ad859b89e54b36

Summary: OpenSSL 1.1.0 drops support for SSLv2, and in doing so also removes the define that user-code used to guard on whether SSLv2 was supported. This adds the define to the portability header to indicate SSLv2 is not supported.

Reviewed By: anirudhvr

Differential Revision: D5376869

fbshipit-source-id: 04708db138dc13e7747cfff582f4cfed291648b3

Summary: Backport C++17's `std::launder`.

Reviewed By: yfeldblum, ot

Differential Revision: D5370531

fbshipit-source-id: b7cc6701d082b6c4855fa77bb644e151ce882deb

Summary:
The TestDoubleWidthTooBig test was performing an exact equality check on
two double values. This equality check fails when run on 32-bit
platforms.

Replace the EXPECT_EQ() check with EXPECT_NEAR(), and simply check that
the two values do not differ by more than 1e-14.

Reviewed By: Orvid

Differential Revision: D652767

fbshipit-source-id: 7bf7ed560a4180a2e2d07a6050d1df18e18d1eea

Summary:
Update LogLevel's operator+() to consist only of a single return statement.
This is required for pre-C++14 compiler support.

Reviewed By: yfeldblum

Differential Revision: D5368256

fbshipit-source-id: 9ecbcde5edd1d0b3e7580d6263ad926e44908219

Summary:
[Folly] `Promise<T>::makeEmpty()` and `Future<T>::makeEmpty()`.

These can currently be done by creating, and then moving away from, regular promises and futures. But that allocates; this is semantically equivalent but more efficient.

Differential Revision: D5368339

fbshipit-source-id: de054cfc75c701c5d39aac64d9a2949cd34b1896

Summary:
Avoid this false positive from clang:

  folly/futures/detail/Core.h:369:15: error: variable 'priority' may be uninitialized when used here [-Werror,-Wconditional-uninitialized]

Reviewed By: yfeldblum

Differential Revision: D5369916

fbshipit-source-id: 49649417d1b799cc3945cba456b42ab438ad2ddc

Summary: [Folly] Apply clang-format to `folly/stats/`.

Reviewed By: Orvid

Differential Revision: D5366745

fbshipit-source-id: 3b7419d4ab4c6203693603722cd8e707741d3953

Summary: [Folly] Apply `clang-format` to `folly/ssl/`.

Reviewed By: Orvid

Differential Revision: D5366760

fbshipit-source-id: ad4ec0570050f80d69c78626906e18b5892a999f

Summary: [Folly] Apply `clang-format` to `folly/concurrency/`.

Reviewed By: Orvid

Differential Revision: D5366946

fbshipit-source-id: 2c9de7c4d8534e02c367fd235fb1ea943ba0539e

Summary:
[Folly] Extract the namespace choice in `FBString.h`.

The funny `namespace` bits confuse `clang-format`, so let us remove the confusion.

Reviewed By: ot, Orvid

Differential Revision: D5367299

fbshipit-source-id: efca65fecc70076829c14e31a835bc91f3efb4a4

Summary: [Folly] Apply `clang-format` to `folly/io/` (namespace).

Reviewed By: Orvid, terrelln

Differential Revision: D5366477

fbshipit-source-id: b4efcdc66c62885f5574b2940e5138590710deae

Summary: The fix for this didn't get made before the commit landed.

Reviewed By: yfeldblum

Differential Revision: D5367593

fbshipit-source-id: 4ed9c781121d1b23db11d75d0f8d017d1d8663d9

Summary: When I moved the change internally, I failed to add the `#include` that was required :(

Reviewed By: yfeldblum

Differential Revision: D5366037

fbshipit-source-id: 7ddf8fc0b7a0468d0831e886d8025c1893f7f8ae

Summary: `constexpr` All The Things!

Reviewed By: Orvid

Differential Revision: D5364910

fbshipit-source-id: 11abeb8adc38aef3ac4f2596ecf0f533b9e74203

Summary:
[Folly] For `futures::retrying`, detect policy type using `std::result_of`.

`FOLLY_CREATE_HAS_MEMBER_FN_TRAITS` does not work in all cases. For example, it does not work with inherited members.

And it is not really what we want, anyway, which is to dispatch based on the result type of invoking the policy with a particular set of arguments. We can do that more directly using `std::result_of`. And if we did not have `std::result_of`, then we could do that with SFINAE directly.

Reviewed By: WillerZ

Differential Revision: D5361808

fbshipit-source-id: 23f969dfd6dbaaa944dc2288e70c3ea11d3398f0

Summary:
Allows you to construct a `Future<A>` from a `Future<B>` if A can be constructed from `B&&`.

The constructor is `implicit` if and only if `B&&` is convertible to `A`.

This is the same philosophy as Optional, and has the same use-cases.

Reviewed By: yfeldblum

Differential Revision: D5264185

fbshipit-source-id: e2fae373ab549c186a69dc5f4e32481ef7b11bba

Summary: There is no difference, as the statements are already implicitly volatile.

Reviewed By: yfeldblum

Differential Revision: D5363491

fbshipit-source-id: ab5bb878f5ca72aa0de96f82e345213729c071e2

Summary:
[Folly] In-place construction for `Try`.

Using `in_place` as the first argumenbt. Avoid move operations.

Reviewed By: Orvid

Differential Revision: D5362568

fbshipit-source-id: 5c2aaf5dba1253c59e384940add411f0f2b570b2

Summary:
[Folly] In-place construction for `Optional`.

Using `in_place` as the first argument. Avoid move operations.

Reviewed By: Orvid, ot

Differential Revision: D5362563

fbshipit-source-id: 771db2556842d5dec35d1bf2ad6c23358a417d54

Summary:
[Folly] Move `in_place` and friends to `Utility.h`.

Document that they are backports from C++17.

Reviewed By: Orvid

Differential Revision: D5362364

fbshipit-source-id: 495cb7f339fc558f20d89100c141fc7a243239d5

Summary:
[Folly] Remove `construct_in_place`.

It is an alias for `in_place`, and unnecessary.

Reviewed By: Orvid

Differential Revision: D5362354

fbshipit-source-id: 39ce07bad87185e506fe7d780789b78060008182

Summary:
[Folly] Disambiguate the various `in_place` overloads, following C++17.

Forwarding `folly::in_place` is ambiguous because there are multiple overloads and we depend on deduction to choose the correct overload. For example:

```lang=c++
enum struct Err {};
Expected<Expected<int, Err>, Err> val(in_place, in_place, 3); // fails to compile
```

So we must disambiguate the three overloads: the default, the typed, and the indexed.

C++17 defines `std::in_place`, `std::in_place_type<typename>`, and `std::in_place_index<std::size_t>`. Let us mimic that exactly, so that it becomes trivial to swap out our implementations for the standard implementations once we jump to C++17.

Reviewed By: Orvid

Differential Revision: D5362339

fbshipit-source-id: d4012b01796390e74d8c14cdf68af70102608039

Summary:
There is no way to lock a socket on Windows, so these calls will always fail. Just don't do the lock on Windows.
To do it properly with sockets, this should probably be forwarding to `sendmsg`/`recvmsg`, but this is good enough for now.

Reviewed By: simpkins

Differential Revision: D5307901

fbshipit-source-id: 9855274e932a3e2ec3cacae10363200187e0c01b

Summary: This also silences a few warnings in the stl vector tests.

Reviewed By: yfeldblum

Differential Revision: D5350383

fbshipit-source-id: 84df2a7dbfbb23ebb3f15a09f0cb58080b3e06d6

Summary: Xcode 9 is throwing errors because `-Wshadow-local` and `-Wshadow-compatible-local` are undefined in Clang.

Reviewed By: Orvid

Differential Revision: D5356476

fbshipit-source-id: 4cc9f48f796f66decb4beaa66d3e66f486449c19

Summary: Folly no longer supports GCC 4.8, so this is no longer needed.

Reviewed By: yfeldblum

Differential Revision: D5354542

fbshipit-source-id: bd9a2732f22f2176f64f92f0af6f7d59dd59a8d9

Summary: This is so we can do some magic internally.

Reviewed By: yfeldblum

Differential Revision: D5354341

fbshipit-source-id: 96abf92f4765f5ae21aa3c0dc4bcadcc22560ab5

Summary: This was breaking Windows builds.

Reviewed By: Orvid

Differential Revision: D5351468

fbshipit-source-id: 3385931b82b11d71d98861fb45efc71a632d470b

Summary: Because `#warn` isn't actually a thing.

Reviewed By: ot

Differential Revision: D5356878

fbshipit-source-id: 57751ca57e7f6aaf9f4d8be70958b9cacc4f7335

Summary: The empty file was left to stop the bleeding without breaking things. Time to remove it.

Reviewed By: mzlee

Differential Revision: D5354310

fbshipit-source-id: cd3039811cf9d7371578e4078cd7d32e3bc6d8a1

Summary: reserved

Reviewed By: djwatson

Differential Revision: D5352232

fbshipit-source-id: e3c27e22320e12bd635acb8a1b394a9de02cecfc

Summary:
- Fixed leak in hazptr_priv destruction
- Updated tests to detect leak

Reviewed By: djwatson

Differential Revision: D5351280

fbshipit-source-id: 724810cbbe0565f0cbd41f9d2121abefd98487bd

Summary: It doesn't need to exist anymore

Reviewed By: yfeldblum

Differential Revision: D5318746

fbshipit-source-id: c70b184f4b3fc12ede4632d6b3d43de16ed758c7

Summary:
Versions of gcc prior to 5.x only support C++11 constexpr functions, and
require that they consist of only a single return statement.

This updates isLogLevelFatal() to meet those requirements so that it can still
be compiled with gcc 4.9.

Reviewed By: ikobzar

Differential Revision: D5350304

fbshipit-source-id: 2a9c256236c484f8c3d5f83690fde8b7333aa4fc

Summary:
Using RSA free and removing unnecessary Bignum allocations to prevent
test failures due to memory leaks

Reviewed By: anirudhvr

Differential Revision: D5348232

fbshipit-source-id: 7c76f3dced26a3080fc82b4bacc06bc0768d6fda

Summary:
- Add support for private lists of retired objects
- Add an option for one domain
- More scalable thread cache managemnt
- hazptr_rec alignment
- FOLLY_ALWAYS_INLINE
- Refactor management of retired objects in hazptr_domain
- Refactor benchmarks

Reviewed By: davidtgoldblatt

Differential Revision: D5322646

fbshipit-source-id: 9d31582b9a8216861e7e78e2e1cd08dc11cf7f88

Summary: So we can use writeFile in projects which compile with -fno-rtti

Reviewed By: mzlee

Differential Revision: D5341469

fbshipit-source-id: bb77baf9576604ff862d54d59bf382862bd0dd58