Summary: Fix MemoryIdler::flushLocalMallocCaches() to use the correct type when querying
jemalloc's "opt.narenas" mallctl.

Reviewed By: @jobenber

Differential Revision: D2414309

Summary: Allow users to access the stack pointer and stack size

Reviewed By: @alikhtarov

Differential Revision: D2407252

Summary: makeFutureWith executes a function and returns a future containing
set to the function's return value. This diff adds a specialisation
for the case when the function returns some type Future<T>. Instead
of returning Future<Future<T>>, makeFutureWith now just passes
on the future that the function returned, which may or may not have
a value set at this time. In case the function throws,
makeFutureWith returns a Future<T> containing the exception.

With this diff, the following two lines produce equivalent results:
  auto f1 = makeFutureWith(func);
  auto f2 = makeFuture().then(func);
except for the fact that f2 requires an additional temporary
Future<Unit> to be created and destroyed.

Reviewed By: @fugalh

Differential Revision: D2388335

Summary: Erasing Cpp2Worker happens in acceptor thread but when evb is
destroyed and drains its callback queue, the callbacks may access the
worker. This moves erasing Cpp2Worer to its own evb thread, after all
callbacks are run.

Reviewed By: @yfeldblum

Differential Revision: D2398089

Summary: Universally update to jemalloc 4.0.0.

Update MALLOC_CONF/malloc_conf configuration, and use the
"arena.<i>.purge" mallctl (replaces "arenas.purge").

Reviewed By: @​bmaurer

Differential Revision: D2355602

Summary: Provide a folly interface for logging the client supported TLS signature algorithms.

Reviewed By: @siyengar

Differential Revision: D2374528

This reverts commit 37f45ee5 which should not have been pushed.

Reviewed By: @​sarals, @siyengar, @​dpittman, @​alephbax

Summary: Contains only the central implementation of the TokenBucket, without moving other files to use the new implementation.

Reviewed By: @yfeldblum

Differential Revision: D2329892

Summary: Projects depending on benchmark library fail to build with clang with the following error
```
./folly/Benchark.h:263:16: error: inline asm not supported yet: don't know how to handle tied indirect register inputs
```

I am not an expert here, just tried to take an approach similar to the MSVC compiler clause

Reviewed By: @andralex, @yfeldblum

Differential Revision: D2368670

Summary: [Folly] Fix the example for folly::static_function_deleter.

The problem is that

    int BIO_free(BIO*)

is not directly compatible.

So have two examples. One using `RSA_free` which is compatible, and one making a compatible wrapper around `BIO_free`.

Reviewed By: @Gownta

Differential Revision: D2381125

Summary: Recently default timeout logic was added to HHWheelTimer interface. This diff adds a unit test exercising this new logic.

Reviewed By: @djwatson

Differential Revision: D2376834

Summary: Don't automatically generate the main function for this test.

Reviewed By: @Gownta

Differential Revision: D2375814

Summary: - Mark all methods in AsyncUDPSocket as virtual
- Create new gmock AsyncUDPSocket: MockAsyncUDPSocket

Reviewed By: @yfeldblum

Differential Revision: D2370668

Summary: [Folly] Rename fixup from TDelayedDestruction to DelayedDestruction.

Just affects comments.

Reviewed By: @haijunz

Differential Revision: D2373364

Summary: [Folly] Rename fixup from TNotificationQueue to NotificationQueue.

    hg locate 'folly/**/*.{h,cpp}' | xargs perl -pi -e 's/TNotificationQueue/NotificationQueue/g'

Reviewed By: @haijunz

Differential Revision: D2373334

Summary: [Folly] static_function_deleter.

So you can write this:

    using BIO_deleter = folly::static_function_deleter<BIO, &BIO_free>;
    auto buf = std::unique_ptr<BIO, BIO_deleter>(BIO_new(BIO_s_mem()));
    buf = nullptr;

In place of this:

    struct BIO_deleter {
      void operator()(BIO* bio) {
        BIO_free(bio);
      }
    };
    auto buf = std::unique_ptr<BIO, BIO_deleter>(BIO_new(BIO_s_mem()));
    buf = nullptr;

Reviewed By: @alandau

Differential Revision: D2364544

Summary: Running ./configure on Fedora 21 failed until I installed some
more prerequisite packages.  Add them to the list in README.md.

Reviewed By: @markisaa

Differential Revision: D2369708

Summary: Currently HHWheelTimer requires providing an explicit timeout value every time a new timeout is scgeduled. This change adds an optional "default timeout" parameter. With this parameter set, HHWheelTimer can be used the same way as AsyncTimeoutSet. Variable timeout functionality is still available even if the default parameter is set.

Reviewed By: @djwatson

Differential Revision: D2366783

Summary: Per Boris's report.

Preparing to re-push this after fixing downstream errors.

Reviewed By: @ot

Differential Revision: D2361333

Summary: clang-3.7 upstream warns that these calls prevent a copy elision
(-Wpessimizing-move).

Reviewed By: @​mzlee

Differential Revision: D2366951

Summary: BREAKING CHANGE! Deprecate Singleton::get() and replace it with Singleton::try_get() that return smart pointer that allows to manage lifetime of a reference and prevents it from being deleted if the reference is still in use

Reviewed By: @chipturner

Differential Revision: D2268791

Summary: This reverts commit aee49bc8a945db9ad6c0ef6a9598ca5c44a4dc8c.

Temporarily reverting the previous diff to investigate possible downstream regression.

Reviewed By: @yfeldblum

Differential Revision: D2359864

Summary: Per Boris's report.

Reviewed By: @yfeldblum

Differential Revision: D2357299

Summary: Originally #247, which was closed before being merged, this doesn't change the behavior of the formatting function now.
Closes #297

Reviewed By: @yfeldblum

Differential Revision: D2340712

Pulled By: @sgolemon

Summary: [Folly] EnvVarSaver.

    TEST(SomeBigClassTest, ChangesEnvVars) {
      folly::EnvVarSaver saver;
      setenv("USER", "root", 1);
      BigClass().doSomethingWithUser();
    }

Reviewed By: @Gownta

Differential Revision: D2354679

Summary: This adds support to Acceptor to be able to switch between
multiple protcols when SSL is being negotiated using MSG_PEEK.

The motivation for this is to be able to try out multiple protocols.

Reviewed By: @djwatson

Differential Revision: D2327946

Summary: Fixes compiling crc32c function on older hardware. (I believe this problem was introduced by 7ec1fc0e).

Also ensures is in final library since the API suggests it should be. This should correct a number of build problems for other projects like wdt and hhvm when they are compiled using a shared folly library rather than borrowing sources at compile time.
Closes #296

Reviewed By: @yfeldblum

Differential Revision: D2340699

Pulled By: @sgolemon

Summary: MSVC spotted this very suspicious line when I was running it with /analyze, and complained about using an unitialized variable. This variable isn't used anywhere in the function, and is initializing itself with itself. I have no idea how this compiled in the first place, but apparently it does.
This removes the line in question.
Closes #295

Reviewed By: @yfeldblum, @paulbiss

Differential Revision: D2340688

Pulled By: @sgolemon

Summary: This uses `<random>` to implement it, because there is no `/dev/urandom` on Windows.
Closes #250

Reviewed By: @yfeldblum

Differential Revision: D2282887

Pulled By: @sgolemon

Summary: [Folly] Extract folly/io/async/test/RequestContextTest.cpp (from Thrift).

Reviewed By: @haijunz

Differential Revision: D2350908

Summary: When we FATAL with a "Double registration of singleton" error, make it abundantly clear why such a thing could have happended so that the developer doesn't feel lost.

Reviewed By: @chipturner

Differential Revision: D2345728

Summary: Today it's hardcoded to use the flag, which could result
in some problem. We should allow callers to choose.

Reviewed By: @djwatson

Differential Revision: D2345036

Summary: Remove recently introduced dependency on boost_thread by breaking EventBaseLocal off to its own rule

Introduced in D2203063

Reviewed By: @​mzlee

Differential Revision: D2301943

Summary: [Folly] Fix AtomicHashArray::defaultConfig SIOF.

May fix https://github.com/facebook/folly/issues/300.

Reviewed By: @paulbiss

Differential Revision: D2343162

Summary: - This mirrors the behavior in AsyncServerSocket

Reviewed By: @yfeldblum

Differential Revision: D2338665