Summary:
Remove the logic that clamps the log level to DFATAL in debug builds.

This behavior logically makes some sense, but results in subtle and potentially
confusing behavior changes.  It seems confusing that after calling
`setLevel(LogLevel::MAX_LEVEL)` on a log category, calling `getLevel()` on that
object would return `DFATAL` rather than `MAX_LEVEL`.  We also weren't clamping
the level to DFATAL consistently: `setLevel()` would clamp the value, but on
construction the initial level was still set to MAX_LEVEL rather than DFATAL.
This resulted in some issues when implementing `LoggerDB::getConfig()` since we
could not consistently detect if a log category was using the default level
settings.  Rather than fix this inconsistency it seems better to simply remove
this clamping behavior.

This means that it is possible for users to disable DFATAL log messages even in
debug builds if they really want to.  Previously this was only allowed in
release builds.  This protection doesn't really seem all that
valuable--presumably most developers won't really want to do this, and if they
really do request this configuration it doesn't seem all that bad to honor it.

Reviewed By: bolinfest, yfeldblum

Differential Revision: D6200569

fbshipit-source-id: 83ef8e2e4d3b61bc5b105038cbe3132979e9ac67

Summary:
Add a method to LogHandler to return its current configuration.  This will
make it possible to query the LoggerDB for its current configuration state.

Reviewed By: bolinfest

Differential Revision: D6200563

fbshipit-source-id: 2b8b9752bbeb26c8aac28d1a73b7e2312fd198c8

Summary:
Update the LoggerDB to track a list of LogHandlers by name, and
LogHandlerFactories by handler type.

This will be needed to support updating the LoggerDB configuration from a
LogConfig object.

Reviewed By: bolinfest

Differential Revision: D6200562

fbshipit-source-id: e365b4e0df65aa5aaa34e118eb3cee9c9c45cb05

Summary:
[Folly] Synchronize coupled caches in `folly::threadlocal_detail::StaticMeta`.

The caches should be set together, and only together, because they are coupled. This prevents bugs where one function that sets one cache but not the other cache is inlined into the caller in one module, and another function that reads both caches is inlined into the caller in another module.

Reviewed By: djwatson

Differential Revision: D6435175

fbshipit-source-id: 846c4972b40e525f2c04da6e6609c2ad54f674c0

Summary:
Add a new LogHandlerFactory interface for creating LogHandler objects from a
LogHandlerConfig.

Also add an initial FileHandlerFactory implementation capable of creating
LogHandler objects that write to a file descriptor.

Reviewed By: bolinfest

Differential Revision: D6200567

fbshipit-source-id: 14b86fc14ad475223aa4b57d45c40638b48c7594

Summary:
Just testing that ElfFile returns expected error code and message for
non-ELF files.

Depends on D6410210

Reviewed By: yfeldblum

Differential Revision: D6418365

fbshipit-source-id: aaab3b9f3ca1a12d384ae98a1772b7f640115192

Summary:
We need to keep LockedPtr returned by rlock/wlock alive for the entire duration
of the loop. Else we are working on a snapshot of the data structure, which
might not be what we want.

Reviewed By: aary, ot, luciang

Differential Revision: D6426613

fbshipit-source-id: 294c0a7b9d756f3116d5cdb3a7b3678f693e0754

Summary:
Replace all assert() checks in the folly logging code with
FOLLY_SAFE_DCHECK().

Reviewed By: yfeldblum

Differential Revision: D6422885

fbshipit-source-id: 5cb12dd59f2fe5d346f15b9d317abede8654a879

Summary:
_mm_cvtsi64_si128 and _mm_cvtsi128_si64 intrinsics are available on
x86_64 only, so do not enable SSE instructions when compiling for 32 bits.

Reviewed By: yfeldblum

Differential Revision: D6422271

fbshipit-source-id: a3c5cf3f2ee4b47d5a1edcf93cfff3d1f3fe907e

Summary:
[Folly] Remove some custom test `main()`'s in `folly/experimental/symbolizer/`.

In one of the two cases, we need to do some extra work to give `ElfFile` knowledge of the program path.

Reviewed By: ot, Orvid, philippv, luciang

Differential Revision: D6420517

fbshipit-source-id: 11c694a8342abb6fff05c368f3cda2e6287ccb4b

Summary:
This seems to fix issue in the attached task.

`context_` is set in `scheduleTimeout()` and never gets reset. So when an AsyncTimeout object is held across requests by objects like `IOThreadPool0` in I/O Thread Pool used by wangle/acceptor/ConnectionManager, RequestContext object created for the last request is leaked until `IOThreadPool0` handles another request and overrides `context_`. In the issue described in attached task, unit test has single request, next request never comes in, RequestContext doesn't end until test service stops, logging feature relying on RequestContext dtor doesn't get called in time, and thus unit test fails.

Reviewed By: yfeldblum

Differential Revision: D6402268

fbshipit-source-id: 200c6d358dfa6d7d9aa68ab05f6f1c7f4117b0ec

Summary:
Use FOLLY_SAFE_DCHECK() rather than assert() to avoid compiler errors about
this variable being unused in builds where assert() statements are disabled.

Reviewed By: bolinfest, yfeldblum

Differential Revision: D6421978

fbshipit-source-id: ab1019fa1fa420856e1bf3bc39c441662c4efd75

Summary: Now that we're using C++14 more broadly, we should use it to improve generator performance. This speeds up a microbenchmark >5x by removing type erasure.

Reviewed By: philippv

Differential Revision: D6398730

fbshipit-source-id: 5809058a3b5ff0e66fd4b1e8954698944e1a7d09

Summary:
When loading cert/key pair, order matters:
(a) Wrong key will fail to load if a cert is loaded;
(b) Wrong cert will succeed to load even if a private key is loaded.

So this diff adds:
(1) SSLContext::checkPrivateKey() -- must call for case (b).
(2) SSLContext::loadCertKeyPairFromBufferPEM() -- use this if one loads both cert and key. Guaranteed to throw if cert/key mismatch.

Reviewed By: yfeldblum

Differential Revision: D6416280

fbshipit-source-id: 8ae370883d46e9b5afb69c506c09fbf7ba82b1b9

Summary:
ElfFile in init() assumes file is long enough to contain complete
ElfHeader, which is 64 bytes long. This is not true for valid non-ELF files like
short scripts. They shouldn't cause crash.

Reviewed By: myreg

Differential Revision: D6410210

fbshipit-source-id: 28fd017d8de17c431d7d006a1655ade8a95994bd

Summary: Use newest hazptr hotness in concurrenthashmap.  Shaves ~10% off of the single-thread find performance.

Reviewed By: magedm

Differential Revision: D6259947

fbshipit-source-id: 7ecf99d38fdf8e311fca3313137e0fca5af3f165

Summary: This would save some function calls and reduce code size if we know we will use jemalloc.

Reviewed By: yfeldblum

Differential Revision: D6411810

fbshipit-source-id: 9db471678e473945f5f00354c0a1285ecdb91de6

Summary:
Move `folly::hasher` closer to feature parity with `std::hash`.
This is in order to replace some instances of `folly::hash::hash_combine(...)` with `folly::Hash()(...)` (`std::hash` is the identity for integers, which makes it an unsafe default for more sophisticated hash data structures, including open-addressing hash tables).

The implementation is similar to `libstdc++`'s implementation, in that we handle separately the `0` case, because `0` and `-0` have different binary representations but are equal according to `operator==`, and hash the bytes otherwise. It is probably a little faster than `libstdc++`'s implementation, that delegates a out-of-line Murmur hash routine for arbitrary buffers, while this uses a minimal inlineable machine word hashing routine.

Reviewed By: yfeldblum

Differential Revision: D6410713

fbshipit-source-id: 86d9e4ed8da04fffe283949825852e539ec7d5cf

Summary:
[Folly] `QueuedImmediateExecutor` tweaks.

* Add a leaky meyers singleton instance.
* Make the thread-local queue an instance variable. Callers which want the singleton thread-local queue can use the singleton executor instance instance, for the same effect.
* Simplify the body of `add`, and perform the thread-local lookup only once per invocation.

Reviewed By: djwatson

Differential Revision: D6399067

fbshipit-source-id: 03904885a70c4b943141bd83868414d27232fd6a

Summary:
AsyncSocket will return isDetachable() -> false if there is a read callback set.
So,
 - ignore event registration status
 - unregister the events when detaching and
 - register them back when attaching again.

Reviewed By: afrind, eduardo-elizondo

Differential Revision: D6331787

fbshipit-source-id: fca4e6101f2d5666edbafa09116aa50f34cb084d

Summary:
Fix #673
Closes https://github.com/facebook/folly/pull/715

Reviewed By: meyering

Differential Revision: D6394731

Pulled By: yfeldblum

fbshipit-source-id: fb92828d0c633278601125dbc64cadf0cdbaa7fe

Summary:
Currently folly::ThreadLocal[Ptr] is pretty heavy-weight for a get():

1) call instance(), take a static init guard, branch
2) call getThreadEntry, check if thread_local is not null, branch
3) check if id < threadEntry->capacity, branch
4) Finally, return threadEntry->elements[id]

If we have real thread_locals, we can do better by caching the capacity directly,
combining all three checks:

1) checkif id < threadLocalCapacityCheck, branch.  If not, do slow path.
2) return threadEntry->elements[id].  Threadentry is never null if capacity > 0, and
    instance() setup work is called during the first getThreadEntry call when threadlocalcapacity == 0.

Reviewed By: yfeldblum

Differential Revision: D6379878

fbshipit-source-id: 4fc7564bbb2f319d65875124026aef28d910ef06

Summary: dogscience

Reviewed By: yfeldblum

Differential Revision: D6387604

fbshipit-source-id: 3e33da3e3c6801d6e2186637c8fb6adaf1eef6f6

Summary:
Without -Werror Apple LLVM will print a warning about -Wshadow-local
and -Wshadow-compatible-local but won't fail.  This causes problems for
downstream code which uses -Werror.

Reviewed By: meyering

Differential Revision: D6375771

fbshipit-source-id: 4ac3a4ff46bdb76f1f07b0c8ba479a7d404c8c9b

Summary:
Add a new LogConfig class to represent the current configuration state of the
LoggerDB.  This also includes code to parse config strings in both JSON and a
simpler more human-writable format (intended primarily for use in command line
arguments).

I generally expect the human-writable format to be used mainly to configure log
levels.  It also supports configuring log handler details as well, but the
format for this data is slightly more cumbersome and will probably be harder
for people to easily remember.

The parsing code is intentionally kept as part of the 'init' library rather
than the core 'logging' library so that other libraries that simply wish to log
messages do not need to depend on it.  For instance, this would allow the folly
JSON library to use the logging library without causing a circular dependency.

Reviewed By: bolinfest

Differential Revision: D6200560

fbshipit-source-id: e4e3c7f941808251b6c7bcbbdac0210118675fb0

Summary:
Benchmark appears to be memory bound, and is affected by different cpus.
Add a new benchmark that does more work.

Reviewed By: yfeldblum

Differential Revision: D6380904

fbshipit-source-id: 5cfbaab70379aa4a2923d957e1147d8486deeff7

Summary: Broken, found via build failures in D6259947

Reviewed By: yfeldblum

Differential Revision: D6374215

fbshipit-source-id: 3a438b5416683c602966c2f029c6b7787acdccaa

Summary:
This reverts commit db55b944ca63e565997094c11b90c4ebe98531ce

bypass-lint

Differential Revision: D6366352

fbshipit-source-id: e25906409186b077ef9117aa524cc7c86314ae12

Summary:
As it stood currently folly::get_default() would unnecessarily
construct a value into the third parameter, which was unnecessary in the fast
path where the element was found in the map

Reviewed By: yfeldblum

Differential Revision: D6366352

fbshipit-source-id: db55b944ca63e565997094c11b90c4ebe98531ce

Summary: When specializing traits you should not use the libcpp inline std namespace, but rather the plain one.

Reviewed By: yfeldblum

Differential Revision: D6373787

fbshipit-source-id: d62ef9ccbf90c6f952db5f93f2377c920e68a809

Summary: MSVC has a bug that causes it to think that `once_flag` in this context resolves to the `using` statement further up, rather than `detail::once_flag`.

Reviewed By: yfeldblum

Differential Revision: D6376501

fbshipit-source-id: 481dbd75ed21f3d519bd920258fa743f314668ad

Summary:
[Folly] No need for a wrapping structure for `posixTimeToDuration`.

We can just use a variant of tag dispatch. In this variant, we pass to `posixTimeToDuration` a default-initialized value of the desired return type and we write overload templates for each possible variant. The argument is used purely for overload resolution and return type deduction, not for its runtime value. It is slightly different from tag dispatch because we do not use separate types which are purely tag types.

Reviewed By: simpkins

Differential Revision: D6371572

fbshipit-source-id: 1987dee31fceec8733caa61495e96489dbf1ca39

Summary:
Adds a hazptr_retire and domain::retire methods to hazptr.  They both allocate memory instead of inheriting.

This will make implementation of google's cell proposal trivial, vs. D4754972 which felt overwraught.

http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2017/p0561r0.html

Reviewed By: magedm

Differential Revision: D6361162

fbshipit-source-id: 9f259f434139f960483b2ab7c5190d3807edcf52

Summary: There is no guarantee that this code will finish in 6 seconds (nor other threads will release singleton in 5 seconds), especially in ASAN mode. Though I don't have better idea, this diff will relax the condition in ASAN mode to make the unit-test less flaky.

Reviewed By: yfeldblum

Differential Revision: D6371692

fbshipit-source-id: 58dd15cc0b3273719314c8b323ba88ee47e8ff61

Summary:
Add folly::to() conversions to convert between std::chrono::duration or
std::chrono::time_point types and struct timespec or struct timeval types.

To conform to the behavior of the existing arithmetic-to-arithmetic
conversions, this code performs proper overflow checking and throws a
`ConversionError` on overflow.  This unfortunately does make the code rather
complicated compared to a non-checking implementation.

Conversions between some unusual duration types is not implemented yet, and
will fail at compile time if someone tries to use it.  This happens for
durations where neither the numerator nor the denominator of the ratio is 1.
For instance, 7/13ths of a second.

Reviewed By: yfeldblum

Differential Revision: D6356700

fbshipit-source-id: 9dce8ab8f32d8c18089f32c7176a8abf3c3f11f7

Summary:
In the past, these calls were required.

Used to solve static destruction ordering issue.  Any static object that uses RequestContext must call this function in its constructor.

That is when we were using `static folly::ThreadLocal<std::shared_ptr<RequestContext>>`, which was non-leaky.

The problem being addressed is when we have some code of the form:

```lang=c++
void doWork() {
  static EventBase eb;
}
```

But now we are using `SingletonThreadLocal<std::shared_ptr<RequestContext>>`, which is leaky.

So the issue that these calls were there to address seems to have been resolved.

Reviewed By: yfeldblum

Differential Revision: D6332597

fbshipit-source-id: c6aba6620ef2fb3a344ea20f56c8b9c0cdf42c70

Summary:
This is required to avoid new GCC 7 -Wimplict-fallthrough warnings.
We also update to use the C++17 [[fallthrough]] attribute if supported.

Reviewed By: yfeldblum

Differential Revision: D6367140

fbshipit-source-id: d5380983cb300f944df9c2885d0faa0155994be7

Summary: This unit-test takes too long to finish in ASAN mode from heavily loaded system (more than 10 minutes).

Reviewed By: yfeldblum

Differential Revision: D6362111

fbshipit-source-id: b097eff60f88ace4fb869132598806700804e267

Summary:
Seems to be causing rendering error:

{F114885894}

Created from Diffusion's 'Open in Editor' feature.

Reviewed By: lskuff

Differential Revision: D6358188

fbshipit-source-id: 3c437007b425bcadb79a71807ad92d520a70e8cb

Summary:
With ASAN enabled (actually, only with ASAN *and* its detect_stack_use_after_return=1 option),
the addTaskFinally test would fail.  This adapts to accommodate the larger stack offsets.
Also, use EXPECT_GT and EXPECT_LT rather than EXPECT_TRUE.

Reviewed By: yfeldblum

Differential Revision: D6353666

fbshipit-source-id: 39e04caffa7b24cde97c749686c7e651a071dcec