Summary: Prevent a race condition where a session's reference count can be decremented to 0 (causing the session to be freed) by one thread before another thread can increment the reference count.

Reviewed By: mingtaoy

Differential Revision: D26614966

fbshipit-source-id: 6c55321becfc1a3467f57c692065680a32ac21f3

Summary:
BoundedAsyncPipe::write() accepts arguments by-reference, so even
these constants may be out-of-scope by the time the coroutine actually runs,
causing use-after-scope problems which are detected by ASAN.

Wrapping in co_invoke allows us to scope the parameter so that it
lives long enough.

Reviewed By: lxfind

Differential Revision: D26649913

fbshipit-source-id: 5bc6b7f64a5e75c9386b245fa3fd1484efaf1d13

Summary:
This is needed if we have an AsyncStackRoot pointer, and we
want to call both `getTopFrame()` and `getNextRoot()`
(e.g. walking async stacks) without const-casting the pointer.

Reviewed By: yfeldblum

Differential Revision: D26675131

fbshipit-source-id: 84de25c67d563e118483d67d0fedb4ea7aef2604

Summary: Currently there is an extension point `folly_co_invoke` for providing customizations for `co_invoke`. But `tag_invoke` is intended as a generic mechanism for providing namespaced customizations. So let us use that.

Reviewed By: ericniebler

Differential Revision: D26674580

fbshipit-source-id: b1baea04996f162699510c3aba8d2c330fcf34bb

Reviewed By: zertosh

Differential Revision: D26660179

fbshipit-source-id: bc0ae7deeb3a3097bfb77cbf0bc471157f586326

Summary: As per title, stub it out in a way that will just trigger exceptions for everything not yet implemented.

Reviewed By: yfeldblum

Differential Revision: D26579892

fbshipit-source-id: 5bfc20b1f6737bbda80616ea1407a5cf108be3c3

Summary: Choose a single strategy for guarding coro files. Namely, all includes are unconditional, but otherwise the bodies of all headers, sources, and tests are conditional.

Reviewed By: ericniebler

Differential Revision: D26647625

fbshipit-source-id: 33eeb12fb231bd7ae7fe5cb97ead16ab017b15f5

Summary:
There is a Most Vexing Parse problem in `folly/MapUtil.h`:

```
xplat/folly/MapUtil.h:44:75: error: expected expression
  return (pos != map.end()) ? (pos->second) : M(std::forward<Value>(dflt));
                                                                          ^
```

Fix the issue by replacing paren-init with `static_cast`.

Reviewed By: yfeldblum, ispeters, ot

Differential Revision: D26610359

fbshipit-source-id: 89842b9b68cebaaab43ea5f82abe7e500f98ff2c

Summary:
Newer libiberty has limits that over constrain
the size of symbols that can be demangled.
This applies the same adjustment in the default
platform009 demangler, done in D26380825

Differential Revision: D26617797

fbshipit-source-id: 8d3e72a0807aa9ecff18d26f38c1acb83f9c051b

Summary:
Right now we have a single option verifyPeer_(SSLVerifyPeerEnum) that
controls how verification is done regardless of how the ssl context is
inevitably used (client or server). This is confusing for our users since it
is 1. unclear of what the current values mean 2. It is unclear that setting an
option such as VERIFY_REQ_CLIENT_CERT also sets your verification mode when
running as a client.In order to make this api more clear for
our users we separate out this enum to two separate enums for client and
server. When this is passed to openSSL we still pass the same single mode int.

Reviewed By: mingtaoy

Differential Revision: D18308192

fbshipit-source-id: 5d49d3dc5117075a55681cf67b5f8e72b5e62c89

Summary:
Original commit changeset: b48fe4683165

Our application can't terminate since this change was committed. 100% reproducible.

Reviewed By: WillerZ

Differential Revision: D26621716

fbshipit-source-id: 571ecd9a6f5286469877bf510f1f55b036df9217

Summary: This is the second attempt, the first one was D22958650. Decided to do a different diff as some affected files were moved from experimental/afrind/coro/h2proxy to proxygen/facebook/lib/experimental/coro/ which created some confusion while arc pulling

Reviewed By: yfeldblum

Differential Revision: D25432869

fbshipit-source-id: a183898302a79084d890548b9b7ecc4409f501d2

Summary:
`/proc/self/exe` doesn't exist on apple platforms, so provide an
implementation that does.

Reviewed By: yfeldblum

Differential Revision: D26599826

fbshipit-source-id: e00dde59d639063e59cb97feb75caf0a49a01954

Summary: Move `order_preserving_reinsertion_view` to a new header `folly/container/View.h` and revise its customization facility in terms of `folly::tag_invoke`.

Reviewed By: iahs

Differential Revision: D25787117

fbshipit-source-id: b7ccd3b6a1413a2639ef1beb1f1ee23bbed045e3

Summary:
This'll move getdeps dependency for katran to 0.3 (which is already the case
for travis build)

Reviewed By: avasylev

Differential Revision: D26129525

fbshipit-source-id: ec9d0615a3e02d75454a3f9dd974ac5c7589e40a

Summary: IoUringBackend benchmarks

Reviewed By: kevin-vigor

Differential Revision: D26594012

fbshipit-source-id: c6a08e6ec13b736c32619958713706118e2f1a49

Summary: No need to overload coroutine `promise_type::return_value` to handle initializer list expressions - just template and default the parameter.

Reviewed By: Orvid

Differential Revision: D26508901

fbshipit-source-id: 61c4a2225355b0044638cd89b9153d599317a453

Summary: When EventBase is destructed, the underlying local storage should be destructed prior to event loop being destroyed.

Reviewed By: andriigrynenko

Differential Revision: D26583298

fbshipit-source-id: 9bfb8442b2a827aa2b3564b27ea765ec1e7b26a2

Summary: Owing to a set of bugs, it is not currently possible to actually declare ranges::enable_view without actually defining it, so we have to include the actual definition when we're in compilation units with it available. If the definition is not available, then the `const` form is sufficient to keep things compiling.

Reviewed By: yfeldblum

Differential Revision: D26337214

fbshipit-source-id: fb042b62967e3e6f536e8268e1b6dd6ccd20b810

Summary: Some compilers fail when `friend` declarations have mismatched `constexpr`-specifiers.

Reviewed By: Mizuchi

Differential Revision: D26601032

fbshipit-source-id: a49e327ff9353ddb21eac96776f97d4089c85374

Summary: Use `std::filesystem` or `std::experimental::filesystem`, via `folly/portability/Filesystem.h`, in `folly/test/FileTest.cpp`. Replaces the use of `boost::filesystem`.

Reviewed By: Orvid

Differential Revision: D26572916

fbshipit-source-id: f769e00b2772f3bfaed899f5694d89db87089e35

Summary:
Make the implementation of `folly::fs::lexically_normal` match C++17 `std::filesystem::path::lexically_normal`. The implementation from `boost::filesystem::path::lexically_normal` is easy to fall back on but has some differences.

The implementation is copied mutatis mutandis from https://github.com/gulrak/filesystem/ at tag v1.5.0, in which `ghc::filesystem` is a complete portable standalone implementation of the `std::filesystem` spec and released under the MIT license.

Reviewed By: Orvid

Differential Revision: D26572520

fbshipit-source-id: 5ce03f37c078ee4f62e3480d4abc259199867c8e

Summary: [Folly] Prefer to nest helper functions in coro tests within the specific tests which use them.

Reviewed By: ispeters, Orvid

Differential Revision: D20324066

fbshipit-source-id: 4134145c406f2d1fdeaddf9a7d244c1bd25ec60b

Summary: Some versions of gcc miscompile the code in the parallel-map exception test. Work around the miscompile by adding an extra log line.

Reviewed By: Orvid

Differential Revision: D26509079

fbshipit-source-id: 1474437d96aeabad0294bcb395b7baec229eb9ba

Summary: The CMake function `check_cxx_compiler_flag` is snake-case rather than uppercase.

Reviewed By: vitaut

Differential Revision: D26573998

fbshipit-source-id: 91d5e2cea933d35967d013b2344e50f52fb6ba77

Summary: Tweak a test that relies on hardcoding the expected result of `std::exception().what()` instead to call that and use the result without hardcoding it.

Differential Revision: D26573226

fbshipit-source-id: 3802d12186c96b864e2680d99f5cfca185235f8d

Reviewed By: andriigrynenko

Differential Revision: D26572552

fbshipit-source-id: 035e1313cd4dec070555b5cd41efc58a0b4a948a

Summary:
Revise `EventBaseLocal` interface to export the suite `emplace`, `try_emplace`, `emplace_with`, and `try_emplace_with`, and adding initializer-list overloads for completeness.

This mirrors the interface of `DelayedInit`.

Differential Revision: D26262870

fbshipit-source-id: 951e2f781a1b5546d96adb2e77aa1cadfbff2d6b

Reviewed By: yfeldblum

Differential Revision: D25514804

fbshipit-source-id: 7e6ad5c6063d2cf5047f656c0f2336b34576c60b

Summary: EventBaseLoopController does this already. This way it's safe to destroy FiberManager even if some local callback is left in the executor.

Differential Revision: D26563934

fbshipit-source-id: 0845e5c532634fa9c53694cd67e983886cea82de

Summary:
A portability header for `<filesystem>` backed up by `<experimental/filesystem>`.

The only thing it does is figure out the right header to include and the right namespace to import so that folly code which already may assume C++17 can use names via `folly::fs`.

And polyfills `lexically_normal`.

Reviewed By: Orvid

Differential Revision: D26453988

fbshipit-source-id: 7a7b2e7a312f5d0438d60bc866b109240d096520

Summary: Technically speaking the contract of isRunning() is unclear on whether it can be used as a synchronization mechanism; it could, e.g. be used solely for diagnostics purposes, in which case memory_order_relaxed would be enough. But it is used at least to implement waitUntilRunning(), so memory_order_acquire/memory_order_release is needed to achieve proper synchronization.

Reviewed By: yfeldblum

Differential Revision: D26436697

fbshipit-source-id: 2be07c473e1cb286e6044fa677304816921e888d

Summary:
Currently, the first thing you want to do when getting a context from the pool is
reset it, because you don't know the state the last user left it in. This is bug
prone, because you may forget to reset the context. Instead, the pool should
handle the resetting logic, so it cannot be forgotten.

This diff adds a `Resetter` to the pool, and the reset function is called every
time a context is returned to the pool. I chose to reset when returning to the
pool instead of when retrieving a context because that gives the context a
chance to free any resources it doesn't expect to need in the future before it
potentially becomres idle.

Reviewed By: Cyan4973

Differential Revision: D26439710

fbshipit-source-id: 8b9afa3db2fda4d167f0fd5307791e8a8a4f283c

Summary: We already cache the contexts in the context pool. We don't also need to cache the contexts in the `Codec` when the user keeps the codec around. This will allow us to better utilize the contexts, and potentially reduce memory usage.

Reviewed By: felixhandte

Differential Revision: D26356275

fbshipit-source-id: 6d80326f469de61094478c73b6fe557eebcd7a22

Summary: Mark the param to `folly::type_info_of` param as `[[maybe_unused]]` since, when RTTI is unavailable, it is unused.

Differential Revision: D26537864

fbshipit-source-id: 3fd0a1bb0d3ecbbcdc5e07f153e321a0b8537478

Summary: Let `FOLLY_MAYBE_UNUSED` which wraps `[[maybe_unused]]` support MSVC even when the latter misreports `__cplusplus`.

Differential Revision: D26537706

fbshipit-source-id: 37554b7f9d27332088b72de41e53a5b85ef46327

Summary: This permits removal of the `_getptd()`-based implementation of `folly::uncaught_exceptions` sometimes used on windows, in favor of forwarding to `std::uncaught_exceptions`.

Reviewed By: akrieger

Differential Revision: D26509011

fbshipit-source-id: b8fd14df91fcd849420d3c52328cb080b528245b

Reviewed By: andriigrynenko

Differential Revision: D26497717

fbshipit-source-id: 35307fce7f9e85882f8f9e5a6bdfe651122727a3

Summary: Ensure all the tests in `folly/lang/test/` are in the cmake build. For disambiguation, prefix them all with `lang_`.

Reviewed By: Orvid

Differential Revision: D26496837

fbshipit-source-id: 185a7990c2c7f535123229a07436e901dca6ae7f

Summary: This returns a char* so correct return is nullptr, not 0.

Reviewed By: usikder

Differential Revision: D26122982

fbshipit-source-id: 8c62520403fce966f83939be7cc03f6dc6be4aee