Summary:
* `VirtualGen` gains a move-only counterpart, `VirtualGenMoveOnly`
* `virtualize` added as an operator to type-erase generator names, particularly for making errors and stack traces more legible.
* `FunctionRef` used to avoid heap allocations where short-lived references are sufficient.

Differential Revision: D28167691

fbshipit-source-id: 5beaab3f613666ffafc7ac1951be1933c2863d3f

Summary: `ConcurrentHashMap` has a `DCHECK` for use-after-destruction. One of these has a comment explaining what it's for, but `DCHECK` supports taking a message to show the user. Make it a message.

Reviewed By: yfeldblum, magedm

Differential Revision: D28609176

fbshipit-source-id: ea6e57b7650ad597f45ea3efc2fa0fb62d5de40e

Summary: Add sendmsg/recmsg io_uring support

Reviewed By: yfeldblum

Differential Revision: D27148068

fbshipit-source-id: efb9174fb40f5f5bf0d0318db12cb61075866381

Summary:
Use the new ConstructorCallback class to allow exterior functions/classes
to get a callback for each new AsyncSocket()

Implications:
This adds an optional and an atomic int to the size of each AsyncSocket so
there is a very small increase in mem usage even if this is unused (which will
be most code, at least initially).

There is also a very small amount of additional CPU on constructing an
AsyncSocket: checking the std::option to see if it's non-nullopt.

Reviewed By: bschlinker

Differential Revision: D27516379

fbshipit-source-id: 56f4411afcd404dcf0092cce562230e5c0dcdf17

Summary: Original commit changeset: ce083cec4c31

Differential Revision: D28558687

fbshipit-source-id: 5cf4ac838c80428902280f3f9ccdd1174410fbe6

Summary:
It is safe to expose `size()` without requirement to go through
accessor (which adds reference tracking overhead).

Reviewed By: yfeldblum

Differential Revision: D28591686

fbshipit-source-id: ea0be2a34df01488402e53fdfbe232ee710f03f7

Reviewed By: igorsugak

Differential Revision: D28582083

fbshipit-source-id: 9beedfccfd1fc6b9dc4f7ff8ab0139282bd285ab

Summary:
## Description

Hi!
I wanted to fix a build of folly on Android.

Trying to use folly's `MPMCQueue` on Android triggers the following error:
```
{xxx}/rnfbjni/android/app/build/third-party-ndk/folly/folly/Subprocess.cpp:542:19: error: use of undeclared identifier 'getdtablesize'; did you mean 'getpagesize'?
```
where `{xxx}` is a local path on my machine.

[Here](https://github.com/karol-bisztyga/rnfbjni/tree/patch-folly-repro) is a branch of my project that reproduces the problem(I've already fixed it by patching `folly` so you can see what patch has to be removed in order to make it fail [here](https://github.com/karol-bisztyga/rnfbjni/commit/e6d8059e55efda5f88d5edbc1e932b94223b052c)).

From [the man page for `getdtablesize`](https://linux.die.net/man/2/getdtablesize):

> SVr4, 4.4BSD (the getdtablesize() function first appeared in 4.2BSD). It is not specified in POSIX.1-2001; portable applications should employ sysconf(_SC_OPEN_MAX) instead of this call.

### Related issues:
https://github.com/facebook/folly/issues/1279
https://github.com/facebook/folly/issues/918 - probably, I cannot see bug details.

Pull Request resolved: https://github.com/facebook/folly/pull/1584

Reviewed By: simpkins

Differential Revision: D28547979

Pulled By: yfeldblum

fbshipit-source-id: 3b8d882ad2a4b7cb817a2c7bbc5f9f750b35770e

Summary:
Same plan as D28395301 (https://github.com/facebook/folly/commit/99f856ae2009a80b157b5121e44b1f70f61bd7c9), but with a better implementation and tests
in the diff to make sure it actually works.

Reviewed By: bschlinker

Differential Revision: D28421062

fbshipit-source-id: d9b052b38c7b938a78bbaccd4df8d276af2b6d6c

Summary:
Pull Request resolved: https://github.com/facebook/folly/pull/1585

FreeBSD uses libc++ with libcxxrt and not with llvm's libcxxabi so handle it separately.

Fixes: https://github.com/facebook/folly/issues/1582.

Reviewed By: luciang

Differential Revision: D28528248

fbshipit-source-id: c8a1c8a5af20b40e7028e8c2d12589edd3850d32

Summary:
Original commit changeset: c9b68c6b9fac
Backout: D26916419 (https://github.com/facebook/folly/commit/b9f3d9be479d4dccdc8e5094933229e676308872)

Differential Revision: D28532140

fbshipit-source-id: 15e5eb1262234039387bfda36d280391f52cc3b2

Summary:
Switches a thread search from linear-time std::find to logarithmic-time std::binary_search.

Extracts the comparison into a separate function in ThreadList class.

Reviewed By: yfeldblum

Differential Revision: D28301992

fbshipit-source-id: 199728d82ca7d7fe8534f7163c76a6473b6f7c4b

Summary:
Pull Request resolved: https://github.com/facebook/folly/pull/1524

In particular, every file in `folly/experimental/symbolizer/` and `folly/experimental/exception_tracer/`.

Permits including `folly/experimental/exception_tracer/` in the cmake build.

Closes: https://github.com/facebook/folly/pull/1524.

Reviewed By: luciang

Differential Revision: D24797973

fbshipit-source-id: 664febf885432795e54827e580cfe9eac1e75fba

Summary:
Reimplement `folly::exception_wrapper` thrown variant not to need to cache any pointers and rather to access the exception object and runtime type, and to perform runtime upcasting, via the platform runtime.

This both simplifies and extends the capability of the thrown variant.

Reviewed By: phoad, luciang

Differential Revision: D28415741

fbshipit-source-id: ce083cec4c31a8cc98e956c247229ac765c7f983

Summary: Add a separate tag to the blocking guards, derived by default from the names available in the executors. This allows more fine-grained information when reporting blocking operations on the executors.

Reviewed By: yfeldblum

Differential Revision: D28265296

fbshipit-source-id: 98d877ab13dc73c0b0b06edbdf15ec0e06f4d74a

Summary: Use the global immutable executor by default for SerialExecutor.

Reviewed By: yfeldblum

Differential Revision: D26916419

fbshipit-source-id: c9b68c6b9facd11c8768199f8bae3d6c9276f4bd

Summary: The current signature results in an attempt to construct a `Try<semi_await_result_t<SemiAwaitable>>` from a `semi_await_try_result_t<SemiAwaitable>::element_type`. While the inner types match for Task, they don't for AsyncGenerator - the former is `AsyncGenerator<T>::NextResult`, which is not constructible from a T. (We could alternatively add that constructor, but that would result in timeout returning a `Task<AsyncGenerator<T>::NextResult>`, which is undesirably indirect).

Reviewed By: yfeldblum

Differential Revision: D28475605

fbshipit-source-id: c30a60e3ec942f801653a08b020796d47d4ef763

Summary:
The function `folly::exception_ptr_get_type` under 32-bit arm-ehabi libc++ >= v5.0.1 and < v10.0.0-rc2, is broken because the `__cxa_exception` layout is changed in an incompatible way, with an extra 4 bytes padding incorrectly preceding the unwind field. Shift the `__cxa_exception` pointer before accessing fields through it.

Major thanks to Crystal Jin <crystalj@fb.com> for help debugging the problem and testing the fix.

Differential Revision: D28423734

fbshipit-source-id: fdf67fa5c1056df921fad23f86ae965761603738

Summary: Observer initialization is blocking and is using thread-local state to detect cycles/track dependencies. When running on a fiber we have to make sure the fiber doesn't get suspended during observer initialization.

Reviewed By: yfeldblum

Differential Revision: D28306797

fbshipit-source-id: 807a4553e89f9fa807e6c745a623041dbb0e47ce

Summary: Adding `addLifecycleObserver` and `removeLifecycleObserver` to support tests.

Reviewed By: yfeldblum

Differential Revision: D27163395

fbshipit-source-id: de749b2570c71a3f98cbbc8569c2990ce2c5be0c

Summary: When multiple observers attached and one is removed, removal code can end up calling `observerDetach` on the wrong observer. Simplified and fixed removal logic and added new tests.

Reviewed By: yfeldblum

Differential Revision: D27033014

fbshipit-source-id: df84bf65207a00f3b76d8950ac2c29f477508da1

Summary:
The initial implementation of ConstructorCallback used a global (pre-main)
initialization which was deemed safe.  While likely still safe, it turns out
that when trying to land diffs that used ConstructorCallback, many code bases
compile with -Wglobal-constructor which explicitly denies this type of
initialization (even if it is safe).

Re-worked the code to prevent any non-zero initialization and still maintain
the property that we don't lock unless someone is registering a callback.

Reviewed By: bschlinker

Differential Revision: D28395301

fbshipit-source-id: 2cdd82189410b3ccea1c892560d0d0cb5b6abc34

Summary:
So it can be used in cases where data is already hashed, and
order-independent way to reduce the set of hashes is required.

Reviewed By: ot

Differential Revision: D28382187

fbshipit-source-id: 4327ddd67b49a61d0172c4676090b2f96e84c065

Summary:
To facilitate debugging.

Privacy note: now `what()` message may contain part of JSON.

parser_error `what()` message:

- before the change:
`folly::toJson: JSON object value was a NaN`

- after the change:
`folly::toJson: JSON object value was a NaN when serializing value at "outerKey"->"innerKey"`

Reviewed By: yfeldblum

Differential Revision: D27623818

fbshipit-source-id: 86e61a0fed500eddafbd3e3bb1b9ed363ba8fcae

Summary:
This introduces `__folly_memcpy`, which is an implementation of `memcpy` that uses prefetch to speed up cold copies (data absent from L1) and uses overlapping copies to avoid as much branching as possible to speed up hot copies (data present in L1).

A description of the core ideas for this memcpy is in the file comment at the top of folly_memcpy.S.

`__folly_memcpy` *does* act as a `memmove`, although it isn't optimized for that purpose for copies of 257 or more bytes. This masks some undefined behavior bugs when code calls `memcpy` on an overlapping region of data. `perf` samples will show when `memmove` is called by `__folly_memcpy`, which will help identify these undefined behavior bugs for copies of 257 bytes or more.

Reviewed By: yfeldblum

Differential Revision: D23629205

fbshipit-source-id: 61ed66122cc8edf33154ea6e8b87f4223c0ffcc0

Summary:
Add missing hazard pointer protection of the new node inserted by a successful try_emplace.

Add test that breaks without this fix.

Reviewed By: yfeldblum

Differential Revision: D28361211

fbshipit-source-id: 963527a9c6b24b0cb569d5b49da9ef9a0e8b58cd

Summary:
There are lint rules that should discourage code from doing non-trivial
things as part of static initialization, e.g., calling a complex
constructor.

That said, probably someone somewhere in fbcode does that so I got
wrapped around the axel trying to change this code to prevent such a
race with ConstructorCallback.

The concern was that if someone created a Foo()
pre-main() there might be an initialization race for
ConstructorCallback<Foo>::nConstructorCallbacks_ and if we lost that race,
we would issue callbacks on garbage data.

BUT - apparently the C++ standard initialization requires that all static
zero values are initialized before more complex initilization, so this
can't happen (assuming the compiler correctly implements the spec).
So this code change is just to add a commment with the information so
no one else goes chasing down this (false) rat hole...

Differential Revision: D27969959

fbshipit-source-id: c135d1bb235d652d11e0541cc09a10a1bad1ce87

Summary: title

Reviewed By: yfeldblum

Differential Revision: D28345979

fbshipit-source-id: ea55db34e19d3881c65559172d4fc9df4808fdad

Summary:
TSAN catches this as a data race, though it's fairly benign. In the
assert we read non-atomic `state` without holding the lock. As a fix, acquire
the lock in the assert. The assert is skipped anyway in opt builds.

Reviewed By: yfeldblum

Differential Revision: D28302491

fbshipit-source-id: b92ca9f3d321d34301d96e59fab363d999bddb30

Summary: It cannot yet be assumed in all places.

Reviewed By: MartinSherburn

Differential Revision: D28325332

fbshipit-source-id: b48a1f9d83c2171ef3912b33c55e9cf8fe294c32

Summary: D26331579 (https://github.com/facebook/folly/commit/aa7f74aaf68ab0a7eb834f49a864c277dbe5fe09) was the offending diff that caused an insta crash on Messenger Android and Instagram Android. Backing out for now.

Reviewed By: JunqiWang

Differential Revision: D28308563

fbshipit-source-id: 87a547d9ba7cb3b33a4ddee9e273943f8379d990

Summary: Also backed out D28276400 (https://github.com/facebook/folly/commit/b24643035fb8778b87feefdb3a365c06583ce53f) because it has a dependency to D26331579 (https://github.com/facebook/folly/commit/aa7f74aaf68ab0a7eb834f49a864c277dbe5fe09)

Reviewed By: lhuang04, JunqiWang

Differential Revision: D28309181

fbshipit-source-id: 0e04e00c59d497290aa0067177a51fe1cfc5e144

Summary: No such function exists.

Reviewed By: luciang

Differential Revision: D28276400

fbshipit-source-id: 57d9ede6df0570f2cc8b0a0b5bd77f2cb6be9fd7

Summary: Adds a policy around whether token-bucket is used only non-concurrently, in which case compare-exchanges can be optimized to stores.

Differential Revision: D28230482

fbshipit-source-id: 2c00797dc943e9ea8985737a8c7dda059a195cfe

Summary:
Reimplement `folly::exception_wrapper` thrown variant not to need to cache any pointers and rather to access the exception object and runtime type, and to perform runtime upcasting, via the platform runtime.

This both simplifies and extends the capability of the thrown variant.

Reviewed By: luciang

Differential Revision: D26331579

fbshipit-source-id: a6fe0f10434956efa5ba4b4bcd1bd3a1b48466a4

Summary:
Starting with C++11, implicit generation of the copy constructor is declared as deprecated.
> § D.2 Implicit declaration of copy functions [depr.impldec]
> The implicit definition of a copy constructor as defaulted is deprecated if the class has a user-declared copy assignment operator or a user-declared destructor. The implicit definition of a copy assignment operator as defaulted is deprecated if the class has a user-declared copy constructor or a user-declared destructor (15.4, 15.8). In a future revision of this International Standard, these implicit definitions could become deleted (11.4).

Clang-12 produces diagnostics for these cases, which is turned into errors for folly:
```
folly/test/AtomicHashMapTest.cpp:330:12: error: definition of implicit copy constructor for 'Integer' is deprecated because it has a user-declared copy assignment operator [-Werror,-Wdeprecated-copy]
  Integer& operator=(const Integer& a) {
           ^
```

Reviewed By: yfeldblum

Differential Revision: D28239770

fbshipit-source-id: a28d19ec774264944b750aa267e93d4fa6d1744e

Summary: The atomic location is not used to order memory.

Reviewed By: jbeshay

Differential Revision: D28229348

fbshipit-source-id: c6206d4b72025e1f7f68adc822e9ae4ed13bd8b4

Summary: Items which are subject to policy: align, atom, and clock.

Differential Revision: D28229049

fbshipit-source-id: e65512265abaa8b777f95cf3233d96a70df73c5f

Summary:
Adding new `parseTo` overload for `std::string_view`, and related tests.

Facilitates migration from `folly::StringPiece` to `std::string_view` by supporting:

```lang=c++
splitTo<string_view>(...)
```

Reviewed By: yfeldblum

Differential Revision: D27971000

fbshipit-source-id: 732f968eb6797447ae12bdc70dbb82d5fc6a8a33

Summary: If inlining and constant folding can determine that the buffer is larger than the `to_ascii_size_max` constant, then there is no need to call the `to_ascii_size` function an extra time.

Differential Revision: D27986105

fbshipit-source-id: d088001a7b3632dabb794c86dafdaa93dd0c1348