Summary:
Added missing steps for registering and deregistering per-thread hazptr_priv structures with the default domain.
Updated tests to detect the error

Reviewed By: djwatson, davidtgoldblatt

Differential Revision: D6965459

fbshipit-source-id: 6eb6a5981c1d22518f318ace7387a3941a013352

Summary:
It's common to use UBSAN with ASAN at the same time, and we'll merge them eventually too. As a result we can check for ASAN defined macros to conditionally compile code for UBSAN.

We still have a few build modes that use standalone UBSAN, and unlike other sanitizers, UBSAN doesn't define a magic macro that we could use. In the past we added UNDEFINED_SANITIZER to our compiler flags, but I'd like to go away from it towards a more standard solution (eg check for ASAN).

Reviewed By: yfeldblum, luciang

Differential Revision: D6931054

fbshipit-source-id: 4b298d1cd72cecd693bf99eda309568ef019c544

Summary: Starting with C++11, implicit generation of copy/move constructors/operators in presence of user-defined destructors/constructors/operators is deprecated

Reviewed By: yfeldblum

Differential Revision: D6875337

fbshipit-source-id: a623a3ce2e7f5559d2de2324565c8995682085c7

Summary:
The `bulk_insert()` method left duplicates from the input range when the
container was empty.

Also, `bulk_insert()` relies on `std::inplace_merge()` to be stable, but
`libc++`s version of `std::inplace_merge()` was buggy until [August 2017](https://github.com/llvm-mirror/libcxx/commit/25a78dcd773ce509469a2b892adbeac0c230cdb9).
Since it is fixed upstream, and it is a library bug, I've not mitigated the bug.

Reviewed By: yfeldblum

Differential Revision: D6943195

fbshipit-source-id: 136caa52c21a9e0dbc09b0a47ad57e99d2fde380

Summary:
Clang warns about formatting when we were using `stringPrintf` to print
the result of overflowing an unsigned 64-bit value.

Closes issue #757.
Closes https://github.com/facebook/folly/pull/765

Reviewed By: simpkins

Differential Revision: D6934612

Pulled By: yfeldblum

fbshipit-source-id: df8c29c0ca8a47aca99007b5182529df17a27149

Summary:
Add a domain argument to frestanding RCU functions.
Perhaps slightly less awkward than using the domain, especially
for retire() and rcu_reader, but nothing that can't already
be done with domain calls directly.

Reviewed By: yfeldblum

Differential Revision: D6878941

fbshipit-source-id: 2397bc8c18a4175793932687e7c3c59f2825caa2

Reviewed By: yfeldblum

Differential Revision: D6893567

fbshipit-source-id: 054e109e3c3c3b2b20ac6ffe8f2ee05335e3dd89

Summary:
[Folly] A `transparent` adapter to facilitate the `is_transparent` protocol.

Compare types with member type or member alias `is_transparent` may be used to enable heterogeneous lookup in ordered containers which support it, such as `std::set` and `std::map` as of C++14, and `folly::sorted_vector_set` and `folly::sorted_vector_map`.

In some cases, nothing more is required than to add that `is_transparent` member to an existing compare type - `folly::transparent` may facilitate such cases.

Example:

```lang=c++
using TransparentStringPieceLess = folly::transparent<std::less<folly::StringPiece>>>;
std::map<std::string, float, /* Compare = */ TransparentStringPieceLess> m;
m["somekey"] = 12.;
m.find(folly::StringPiece("somekey"))->second = 13.; // heterogeneous lookup
```

To demonstrate the example, rewrite the `folly::sorted_vector_set` and `folly::sorted_vector_map` transparent-lookup tests.

Reviewed By: ot

Differential Revision: D6921862

fbshipit-source-id: 306ff1f6f178ae29ac80ee7cd28595b80fc393d1

Summary: Implement some OpenSSL 1.1.0 specific methods, so that code written with OpenSSL 1.1.0 can be compiled with OpenSSL 1.0.2

Reviewed By: yfeldblum

Differential Revision: D6930234

fbshipit-source-id: 257893787cecdfeaf9df13dce0ada46c15990699

Summary: Fix a case of null pointer dereference when ~hazptr_array() calls ~hazptr_holder(). For speed hazptr_array doesn't fill the domain_ field of hazptr_holder. The common case is that elements of the array are taken from and returned to the thread cache. If ~hazptr_array() doesn't find enough space in the thread cache, then it calls ~hazptr_holder() with a null domain_. The fix is to set domain_ to &default_hazptr_domain() before calling ~hazptr_holder() in such a case.

Reviewed By: yfeldblum

Differential Revision: D6928705

fbshipit-source-id: 9c756dce2854a2d1869a0712c4e6ed30f5df6b30

Summary: [Folly] Work around UB `signed-integer-overflow` in `Histogram`.

Reviewed By: igorsugak

Differential Revision: D6922381

fbshipit-source-id: 96ab6ffae9509ce8138f1741ea976226dec597c8

Summary: [Folly] Fix UB `signed-integer-overflow` in tracing test.

Reviewed By: igorsugak

Differential Revision: D6922450

fbshipit-source-id: 0d9a338fdcf205987d88875aac4112666ca32b65

Summary: Needed to update folly in the open source build of HHVM.

Reviewed By: yfeldblum

Differential Revision: D6913772

fbshipit-source-id: 50e540fe30651c991dc198ede66d4816a85f68db

Summary:
During program destruction we destroy registered log handlers (to allow them to
flush state if desired).  However we intentionally leak the LoggerDB object
itself and the LogCategory objects it contains, so it will still be safe if
static destruction code that runs later still tries to log messages.

However, when run with `ASAN_OPTIONS` set to `detect_leaks=1` ASAN complains
about the fact that we leak the LoggerDB singleton.

This changes the code to store the LoggerDB address in a static variable, to
prevent ASAN from complaining that we leak it.

Reviewed By: yfeldblum

Differential Revision: D6920772

fbshipit-source-id: 62f6837053ba4a538a89b4b61f9b1c722f253d33

Summary: The free function hazptr_cleanup reclaims all reclaimable objects retired to a domain before this call.

Reviewed By: djwatson

Differential Revision: D6876253

fbshipit-source-id: 54a6b98291712286597561e971fc9b6ac9c4d6f0

Summary:
[Folly] Do nothing in `spin_pause_until` if `spin_max` is zero.

Rather than go through the loop at all, if the instruction is not to spin - encoded as `spin_max <= 0` - then return early.

Reviewed By: nbronson

Differential Revision: D6910431

fbshipit-source-id: fa94510e7ec1ebdd7bc3f1a407dcf70e1e153b40

Summary: This dict holds strong reference to all the loops and leads to leak if new loop is created and used, change it to weakref to avoid it.

Reviewed By: yfeldblum

Differential Revision: D6892775

fbshipit-source-id: a628a3743876ef0580acce5fb9f283da573cf459

Summary:
This updates the code to create a default LogHandler that logs to stderr when
the main LoggerDB singleton is first created.  With this change users will get
reasonable log behavior even if they never call `folly::initLogging()` (log
messages will go to stderr instead of being ignored).

This new initialization code can potentially run very early during program
initialization (before `main()` starts and before folly `Singleton`
initialization) if some code in the program logs messages before main.  However
this current initialization code should be safe to run before main.

For compilers that support weak symbols (clang and gcc), I have defined
`initializeLoggerDB()` as a weak symbol, allowing users to override this
behavior at link time on a per-program basis should they choose to.

Reviewed By: yfeldblum

Differential Revision: D6893207

fbshipit-source-id: 13b72a01aa383c4611204df88ff8dad93abcc7b2

Summary: With the mobile thread_local changes, these assertions are no longer correct.

Reviewed By: magedm

Differential Revision: D6915065

fbshipit-source-id: 42072ed638b3dca56cba65ca0cff5e4d7aef2cb3

Summary:
Change `LoggerDB::get()` to a reference instead of a pointer since this
function can never return null.

Reviewed By: yfeldblum

Differential Revision: D6893206

fbshipit-source-id: af47063918a79c851fd39b838d6c63755166e033

Summary:
LogCategory objects store pointers to their parent and sibling categories.  The
address of each LogCategory therefore matters, and we cannot allow moving them
to another address.

Explicitly delete the move constructor and assignment operator to ensure that
no-one ever accidentally tries to move a LogCategory object.

Reviewed By: yfeldblum

Differential Revision: D6893208

fbshipit-source-id: 32f36c5992e850cc6e90d02c7c74fb758e9f745b

Summary:
When compiling with clang, also specify -Wno-nullability-completeness

We currently use this flag for internal Facebook builds using buck.  This
brings the CMake and buck compilation command lines closer in sync.  Even
though the folly code currently does not produce any nullability completeness
warnings, we are likely to break this in the future since we do not have
this warning enabled internally.

It also seems like `-Wnullability-completeness` results in code that cannot be
compiled successfully across various clang versions.  Recent versions of clang
complain about missing nullability specifiers on `std::unique_ptr` arguments,
while slightly older clang versions produce compilation errors if nullability
specifiers are listed on anything but raw pointers.

Reviewed By: igorsugak

Differential Revision: D6902930

fbshipit-source-id: 0b8cd5eca17cf8f9deda6b0c32f5838489542128

Summary:
Unit tests using the IPv6 loopback would fail on Travis with `[::1]:0: Cannot assign requested address`.

The issue is discussed at length here: https://github.com/travis-ci/travis-ci/issues/8711

Reviewed By: jstrizich

Differential Revision: D6914321

fbshipit-source-id: c247c6760dddcd875a67f7b31419bde974514f19

Summary:
Added a function to Observer::Snapshot that allows the user to get the shared_ptr held in a snapshot. This is to make it easier to integrate Observers with other Api's that expect shared_ptrs.

(Note: this ignores all push blocking failures!)

Reviewed By: andriigrynenko

Differential Revision: D6909245

fbshipit-source-id: 4fe108e04f43a5a250bd7bdc3031994833d73a60

Summary:
[Folly] Hack around MSVC not supporting `constexpr` storage.

GCC and Clang do, but MSVC does not, support:

```lang=c++
extern int const num;
constexpr int const num = 3;
```

Work around it.

Reviewed By: elsteveogrande

Differential Revision: D6903413

fbshipit-source-id: 3a0957816d8d82718fb6187dc1803df1605f80aa

Summary:
The refcounting logic is correct, however, decreasing the next node's
refcount only happens in ~NodeT, after going through hazptr bulkReclaim.

The current default for bulkReclaim is 1k, which exceeds the value of
refcount uint8_t.  Increase size to uint16_t so that we don't wrap refcount
before bulkReclaim calls the node destructors.

Reviewed By: magedm

Differential Revision: D6900530

fbshipit-source-id: 9b2530dc6cd0609b4795b909fd46aeffbb1ae821

Summary:
Now that we have Travis CI builds enabled and passing, add a link to the build
status to the README.

Reviewed By: yfeldblum

Differential Revision: D6903206

fbshipit-source-id: 643b72a63a4113e00ff1602b1c3638384115b23d

Summary:
Allow retrieving sub-tree from folly::dynamic using JSON pointer syntax (RFC6901). This adds new overload for the `get_ptr` method, retrieving raw pointer to a sub-tree, or returning `nullptr` if element is not found.

The `folly::dynamic` implementation traverses the path specified by JSON pointer string dynamically every time, since the underlying object may change, and JSON pointer traversal depends on the underlying object structure. E.g. "123" could be a key name for `dynamic::object` and index in `dynamic::array`.

Reviewed By: yfeldblum

Differential Revision: D6790515

fbshipit-source-id: bb6ea10acb83673e87721cf1e5a02506b44bb273

Summary: Debian 8.6 comes with a CMake version that is too old for folly. Add commands to update cmake.

Reviewed By: snarkmaster

Differential Revision: D6842797

fbshipit-source-id: 22f4c407f82a92c3fa90a5c2de334cf8b92e584b

Summary:
Unfortunately it looks like D6880256 was not sufficient to make
function_scheduler_test pass reliably on Travis CI builds.  This just disables
function_scheduler_test on open source CMake-based builds for now.

Ideally in the future we should try to convert most of the function scheduler
to use proper synchronization mechanisms (like folly::Baton) for synchronizing
between threads, rather than sleeping and expecting the other thread to have
completed its work before we wake up.

Reviewed By: yfeldblum

Differential Revision: D6891889

fbshipit-source-id: 9ee98f5807e1d90f083bd942ba98df6fad656ba6

Summary:
The XLOG() macros compute the log category based on the source file name, as
defined by `__FILE__`.  Both gcc and clang set the `__FILE__` contents to the
file name passed in on the command line, so if they are invoked with absolute
path names `__FILE__` will contain an absolute path.  This is normally
undesirable for `XLOG()` behavior.  Furthermore, CMake always invokes the
compiler with absolute path names, so it is difficult to avoid this situation
in open source builds.

This adds support for a `FOLLY_XLOG_STRIP_PREFIXES` macro.  This can be defined
to a string containing a colon-separated list of directory prefixes to strip
from `__FILE__` before using it to compute the log category name.  If
`FOLLY_XLOG_STRIP_PREFIXES` is not defined then the code continues to use
`__FILE__` as-is.

Reviewed By: yfeldblum

Differential Revision: D6868773

fbshipit-source-id: a5c3276d8d22d3b27ae621700a75b1362d049292

Summary:
Many of the tests in FunctionSchedulerTest appear to frequently fail on
Travis CI builds.  It looks like this is probably due to timing issues where
events do not fire within the expected time interval.

This diff increases the `timeFactor` value to hopefully make the tests more
reliable, at the expense of making them take longer to run.

It would be nicer if we had a better solution here that did not involve
potentially unreliable timing-dependent checks.  Increasing `timeFactor` seems
like the easiest solution in the short-term though, and avoids just disabling
the tests entirely.

Reviewed By: yfeldblum

Differential Revision: D6880256

fbshipit-source-id: 59da3b20f2bb9c982834752942a984e3f91442fd

Summary:
Most of the tests in FunctionSchedulerTest sleep for lengths of time based on
the timeFactor constant set at the top of the file.  However a couple of places
that need to sleep for non-integral timeFactor durations used manual `usleep()`
calls instead.  This made the tests break if you change the `timeFactor` value.

This changes `delay()` to accept floating point numbers, and updates the tests
that were using hard-coded `usleep()` calls to use `delay()` instead.

Reviewed By: yfeldblum

Differential Revision: D6880258

fbshipit-source-id: 0cfdf84d3ba27aa2acd1f5210becbc7306b857bf

Summary:
Most of the tests in FunctionSchedulerTest update a variable in one thread (the
FunctionScheduler thread) and then check its value in the main thread.  These
previously were not using any synchronization between the threads.  This fixes
them to use `std::atomic` to ensure proper synchronization.

Reviewed By: wez

Differential Revision: D6880254

fbshipit-source-id: facfbdf06312f731c64389fb4382b5cb078b76b9

Summary:
Hi! I noticed this repo could use some documentation to help new contributors.

**why make this change?:**
Regarding the `CONTRIBUTING.md` -> Including instructions for new contributors to open source is a courtesy that will foster a welcoming and safe open source community at Facebook.

Regarding the Code of Conduct ->

Facebook Open Source provides a Code of Conduct statement for all
projects to follow, to promote a welcoming and safe open source community.

Exposing the COC via a separate markdown file is a standard being
promoted by Github via the Community Profile in order to meet their Open
Source Guide's recommended community standards.

As you can see, adding this file will improve [the folly community profile](https://github.com/facebook/folly/community)
checklist and increase the visibility of our COC.

Closes https://github.com/facebook/folly/pull/719

Reviewed By: flarnie

Differential Revision: D6881011

Pulled By: simpkins

fbshipit-source-id: a6f0988ac2b49f81bb13d3219095105b7b27e66f

Summary: Add a constexpr implementation of `strcmp()`

Reviewed By: yfeldblum

Differential Revision: D6881816

fbshipit-source-id: 08a9e5ab9ef4a422ff6d004ec13b8ec8c71b0d6f

Summary:
- Use `FOLLLY_HAS_FEATURE(cxx_constexpr_string_builtins)` to check for the
  availability of `__builtin_strlen()` on clang, as recommended by the clang
  documentation (https://clang.llvm.org/docs/LanguageExtensions.html)
- Only use `std::strlen()` on gcc, rather than having checks to exclude all
  other compilers.  `strlen()` happens to be constexpr on gcc, but this is not
  required by the standard and cannot be assumed to be true in general.

Reviewed By: yfeldblum

Differential Revision: D6881815

fbshipit-source-id: 88c614419ae390464893087513dc764b0bfddfb8

Summary:
This test has been failing in travis because it is missing test/certs/tests-cert.pem
Mark it as a content dir

Reviewed By: yfeldblum

Differential Revision: D6884000

fbshipit-source-id: c7d42a4b23334781de28f1586ecbde04aea74ff2

Summary:
Update the build steps in the README to build using cmake instead of autoconf
on Linux platforms.

I haven't updated the instructions for Mac OS X or the
`bootstrap-osx-homebrew.sh` script primarily because I have not tested the
steps on Mac.  It sounds like the build is close to working on Mac, but
currently fails due to missing compiler flags to disable some warnings on
clang.

Reviewed By: akushner

Differential Revision: D6780814

fbshipit-source-id: 9c9c211c306d5b35235549ae451bb620098f37e3

Summary:
Update the CMake build process to always invoke test programs from the root of
the repository.  This makes the CMake build consistent with how we run tests
internally at Facebook.

Reviewed By: yfeldblum

Differential Revision: D6880257

fbshipit-source-id: 4c27cc643faa21d9c26faf8c77f6700ec5fe6393