Summary:
The PAUSE instruction's latency varies widely across x64
architectures, resulting in too much spinning on Skylake in some
scenarios.  This diff switches to using cycle counts to limit spinning
on x64, so that the minimum and maximum spin durations are independent
of asm_volatile_pause's latency.

Reviewed By: yfeldblum

Differential Revision: D17961663

fbshipit-source-id: 6170bfa48d007ca21b73b1a5c7e68da0043cda2c

Summary:
This avoids invalidating the entire build in response
to just running `hg amend`, which is frustrating and slow.

Reviewed By: chadaustin

Differential Revision: D18005409

fbshipit-source-id: ef93313859919298be78204046eb08bcadc5398e

Summary: [Folly] Backport invocability variable templates: `is_invocable_v`, `is_invocable_r_v`, `is_nothrow_invocable_v`, `is_nothrow_invocable_r_v`. And for the member-invoke and free-invoke traits.

Reviewed By: ericniebler

Differential Revision: D17950331

fbshipit-source-id: 5409d7a72116b0976573e1f64760c5ef9274966f

Summary:
If CH parsing is enabled, also parse out the ServerName extension. OpenSSL 1.1.1
changes the behavior of `SSL_get_servername`: an SNI value is stored in
the underlying SESSION if and only if both parties negotiated that SNI.

There are some situations where one would wish to retrieve the original
ServerName that the client sent.

Reviewed By: knekritz

Differential Revision: D17893443

fbshipit-source-id: b29ee42e90629c869dd5e68c93c7cb2abc19745f

Summary:
No longer allow a child task's modifications to the RequestContext made before it's first suspension point to bleed into the next child task's context.

Now always restore the parent task's RequestContext before starting subsequent tasks.

Reviewed By: andriigrynenko

Differential Revision: D17846070

fbshipit-source-id: 44e206d4c833759fc36c7d1bd0f3545ca6bb9ab1

Summary: Relax DFATAL message

Reviewed By: yfeldblum

Differential Revision: D18001089

fbshipit-source-id: 2027d0257e1cba6905dc4eb1863b70a8fe6a7366

Summary: trim, ltrim and rtrim are similar to ltrimWhitespace, rtrimWhitespace and trimWhitespace but you can specify what you want to remove

Reviewed By: yfeldblum

Differential Revision: D17946658

fbshipit-source-id: 9bdc5e6eb810e628a8c16b5142c17a3313f884f2

Summary:
ThreadPoolExecutor::getPoolStats depends on derived classes implementation.
Calling `getPoolStats` is specially useful from `ThreadPoolExecutor::withAll`.

This change moves registration of thread pool executor instances to derived classes
to avoid `ThreadPoolExecutor::withAll` racing with thread pool destruction while calling `getPoolStats`.

Reviewed By: yfeldblum

Differential Revision: D17965135

fbshipit-source-id: 637de55ed78b085dc86fffa0e0f4a66cd4cfcede

Summary: keepAliveRelease code reorg

Reviewed By: kevin-vigor

Differential Revision: D17835172

fbshipit-source-id: 48c60b04324f0d9bd1ac628e2625d3d4d2654c99

Summary: Add missing switch case

Reviewed By: yfeldblum

Differential Revision: D17945170

fbshipit-source-id: 8128c40e1c86b2a5ff8d1bad3c0e9fe197adc174

Summary:
this should help avoid this particular error:

```
IOError: [Errno socket error] [Errno 11001] getaddrinfo failed
```

Reviewed By: chadaustin

Differential Revision: D17886598

fbshipit-source-id: bd9f5b84ebd7ca5c339be3afec3a975fa907d052

Summary:
This diff allows passing a watchman version number override
via the environment as well as via the cmake `WATCHMAN_VERSION_OVERRIDE`
option.

To help invalidate the build I've added a new section to the manifest
files that allows listing out additional env vars that the project
hashes should be sensitive to.  The effect of this is that we'll
re-run the cmake configure step if the listed env vars are changed.

Reviewed By: Ben0mega

Differential Revision: D17865896

fbshipit-source-id: 8ea5572b0b9b7af95ec5c310e494cb17a139ced4

Summary: testpilot's defaults assume a bigger machine than some of our laptops

Reviewed By: fanzeyi

Differential Revision: D17878120

fbshipit-source-id: e01f1f9c77a4f5f011051c9c642dbe934c66bc0b

Summary: During shutdown there's a race condition in IOThreadPoolExecutor::getEventBase where a thread is created in ensureActiveThreads, and then destroyed before pickThread() is called. This adds the same guard that exists in add().

Reviewed By: yfeldblum

Differential Revision: D17955227

fbshipit-source-id: 042cb5c42056a9da4578571fb10936bbc477c4ff

Summary: Add a portability macro for testing whether the compiler supports C++17's inline variables.

Reviewed By: yfeldblum

Differential Revision: D17957184

fbshipit-source-id: bff7d2bc15867e32f787860b4f8e5a0176ac1017

Summary: [Folly] Fix wrong definition of `FOLLY_HAVE_GROUP_VARINT` when group varint is unsupported.

Reviewed By: Orvid

Differential Revision: D17961015

fbshipit-source-id: 3c82aaaf4b257180ab0956e4b181bfd619d7adda

Summary:
/home/folly/folly/experimental/test/QuotientMultiSetTest.cpp:60:29: error:
missing template arguments before 'reader'
     folly::QuotientMultiSet reader(data);

Reviewed By: vitaut

Differential Revision: D17959566

fbshipit-source-id: 02fec44b3d644eea0213895aa4df70349ef7d062

Differential Revision:
D17908896

Original commit changeset: 85847fc17f4a

fbshipit-source-id: 9b2bd8c11881cad18cffea0e14349e25d743b150

Summary:
Calling `folly::ThreadPoolExecutor::withAll` is powerfull when combined with `folly::ThreadPoolExecutor::getPoolStats`.
However getPendingTaskCountImpl being pure virtual can trigger `__cxa_pure_virtual ... std::terminate()` while object is partially destructed.

This change signals implementors of derived clases to write a body for getPendingTaskCountImpl while allowing calling getPoolStats -> getPendingTaskCountImpl within a try/catch block.

Reviewed By: yfeldblum

Differential Revision: D17915523

fbshipit-source-id: 2ffad7076313407969d88c75c7e457fa4e7e2d6d

Summary:
- There is not an easy way to check if a given type is an instantiation
  of a class template.
- The common solution is to write a custom trait each time and
  specialize it for the given instantiation so that the trait returns
  `std:true_type`.
- Add `is_instantiation_of` to help with DRY in custom traits for
  checking if a type is an instance of a given class template. Note that
  this does not work when the class template has a mix of type and
  non-type template parameters. It only works with types.
Pull Request resolved: https://github.com/facebook/folly/pull/1184

Reviewed By: vitaut

Differential Revision: D16161679

Pulled By: yfeldblum

fbshipit-source-id: c933fd3db7c56f61b6784fdcd199ff7963d61d8f

Summary:
The deadlock repro is described https://github.com/facebook/folly/issues/1211.
Fix is to check the working epoch before retire() starts the sync work.

Reviewed By: yfeldblum

Differential Revision: D17908896

fbshipit-source-id: 85847fc17f4a75058d4a72c43dba0c570a2593d6

Summary: As title.

Reviewed By: yfeldblum

Differential Revision: D17937014

fbshipit-source-id: a0ff5f5f9cbbb182794dc564610d7adb9c7c8d08

Summary: [Folly] Fix `Json.SortKeys` test when `F14NodeMap` may randomize table iteration order.

Reviewed By: shixiao

Differential Revision: D17903465

fbshipit-source-id: 00f1a8c441af3e11815a4312f6dc810ec9455a5e

Summary:
Unconditionally sleeping for 100ms in the wait polling loop has high average latency. If a process exits just after the initial wait() call, it's better to poll again sooner.

Start with a 2ms sleep and double it every sleep, capping out at 100ms.

Reviewed By: yfeldblum, mzhaom

Differential Revision: D17769851

fbshipit-source-id: efec8fdb8c2199b8db29b46b8c43fdaac56f69a8

Summary: `waitOrTerminateOrKill` fills the gap between `wait()` and `terminateOrKill`. It allows giving the subprocess a chance to shut down cleanly (after closing its stdin pipe, for example), but avoids waiting forever or leaking a child process.

Reviewed By: yfeldblum, mzhaom

Differential Revision: D17769090

fbshipit-source-id: 8940fd63f7eb9c09ef293c89a5e97f69805735ff

Summary:
QuotientMultiSet is a space-efficient static data structure to store a non-decreasing sequence of b-bit integers.

If the integers are uniformly distributed lookup is O(1)-time and performs a single random memory lookup with high probability.

Reviewed By: ot

Differential Revision: D17506766

fbshipit-source-id: b3e7a22dd193672fadb07d4cccb8b01bedae7cf9

Summary:
- Document in the `README.md` how to build tests as part of the CMake build.

Closes https://github.com/facebook/folly/issues/1240
Pull Request resolved: https://github.com/facebook/folly/pull/1244

Differential Revision: D17927395

Pulled By: yfeldblum

fbshipit-source-id: 6b638ac807886b505b69063f2a47162f3f3fc8de

Summary: The standard only requires `To` to be trivially copyable, not trivial.

Reviewed By: yfeldblum

Differential Revision: D17916612

fbshipit-source-id: 80ae35672ea4664fe4204c8181ac6402065c4e03

Summary: Made a slight tweak to the collectAll benchmark for the coroutine implementation to avoid allocating additional coroutine frames.

Reviewed By: andriigrynenko

Differential Revision: D17917156

fbshipit-source-id: 890a2ab33eb53050930d4547119d30a9c161887e

Summary: std::chrono is less prone to error than specifying timeouts in `int` seconds.

Reviewed By: yfeldblum, mzhaom

Differential Revision: D17767787

fbshipit-source-id: 5bb23d3a78e6e798e7b10331dc416e24d5a8746d

Summary:
Synchronized::operator-> is dangerous because it's very easy to
implicitly acquire locks in ways that don't form a coherent locking
model. Replace uses of it in folly in preparation for removing or
marking the function deprecated later on.

Reviewed By: yfeldblum

Differential Revision: D17862465

fbshipit-source-id: 45d3b0d738941c3faa6d73418e79dcb8a1259e98

Reviewed By: yfeldblum

Differential Revision: D17892633

fbshipit-source-id: 1b4b2cffb040b1abf709ea2ccd4015426398dda0

Reviewed By: yfeldblum

Differential Revision: D17889050

fbshipit-source-id: 323a2a5796b4f06c2fe4c84a714a763aa3da4866

Summary: [Folly] Apply extra formatting rules beyond black.

Reviewed By: zertosh

Differential Revision: D17898403

fbshipit-source-id: 6d73c75a35513f138bcc0d734aada0c99cff54c7

Summary: [Folly] Apply `black` formatter to `.py` files.

Reviewed By: zertosh

Differential Revision: D17898404

fbshipit-source-id: b597b55b646e539a288d175e0b2db8111b5b103b

Reviewed By: yfeldblum

Differential Revision: D17889052

fbshipit-source-id: e9733e6ab31d01064660043b0f4edc31b0f5377e

Reviewed By: andriigrynenko

Differential Revision: D17749079

fbshipit-source-id: 44c2c6dd55fe5be033bc4a5139941b5e3a004a67

Summary: Like subscribeInline but does work on an executor and returns a subscription that can be waited on as a SemiFuture or cancelled

Reviewed By: andriigrynenko

Differential Revision: D17700281

fbshipit-source-id: 52b2c63af53e8f87d921cb0a0ab28da0a800a9fc

Summary:
This will add support to set and get thread name for Android.

`pthread_setname_np` was introduced back in NDK API level 9: https://stackoverflow.com/questions/15241869/pthread-error-in-ndk-build?fbclid=IwAR0yvyaC4KNA6D1GlgIofQSFIuZAgL7qDR5L6m5tkY6lv-R6uTvzreAiFp4

Backwards support added for setting current thread name using `prctl(PR_SET_NAME)`, when `pthread_setname_np` is not available (for `__ANDROID_API__ < 9`).

Currently, Android NDK does not support `pthread_getname_np` so `prctl(PR_GET_NAME)` is used for this.

Reviewed By: yfeldblum

Differential Revision: D17815151

fbshipit-source-id: 474caee070fad26f3e6b118307058d5506f0467b

Summary: [Folly] Move `ThreadName.cpp` macro defs outside C++ section, including outside `namespace ...` blocks.

Reviewed By: vitaut

Differential Revision: D17868541

fbshipit-source-id: a0964ce1f1c6322185f94b8162df4c0e6eaa84bf