Summary:
The breakage has been fixed, so bring back the manifest, but only the Linux one, because the Mac version is failing due to another issue.

Also to make it easier to debug issues on GitHub Actions separate out the dependencies build from Mononoke and rust-shed builds.

Reviewed By: krallin

Differential Revision: D21448412

fbshipit-source-id: 68d89c858d1692727a7fd66bca114920e6dfb4dc

Summary:
callWithExtractedKey previously incorrectly translated
std::pair<X&,Y&>&& into X&&, Y&&, when reference collapsing rules call
instead for X& and Y&.  This diff fixes the problem, adds unit tests,
and also targets the 2-argument std::pair constructor when possible.
This means that callWithExtractedKey can be safely used on top of
libc++'s std::unordered_map::emplace without disabling the key-finding
optimization in that method.  (That lib's emplace is clever enough to
be able to search on the key before constructing a node, but does not
include the logic to unwrap the std::pair 3-arg piecewise constructor.)

Reviewed By: yfeldblum

Differential Revision: D21419822

fbshipit-source-id: d3c61e45d7fe7d8c6805342bbe63a8c866f4e306

Summary: Add comments about the performance of frequent removal and mention clear() in the comment about the timing of destruction of removed key-value pairs.

Reviewed By: davidtgoldblatt

Differential Revision: D21455366

fbshipit-source-id: be9f80136260f2de3b68d6fff54ccdadc5544046

Summary:
This class makes it easier to spawn an unbounded number of tasks
over time within a given scope and then later wait for all of
those tasks to complete. e.g. during shutdown.

Reviewed By: kirkshoop

Differential Revision: D21391413

fbshipit-source-id: c91c66aa61a8ae30e04d0608426037d8adba14d7

Summary:
Change folly/logging's AsyncFileWriter code to avoid calling writevFull() on
Windows.  On POSIX system we use this function to minimize the number of
system calls we make.  However, on Windows there is no `writev()` function
provided by the system, so folly emulates this by performing multiple separate
`write()` calls.  Given that we are just going to perform multiple `write()`
calls there is no reason to construct the iobuf data structure here.

Differential Revision: D21367424

fbshipit-source-id: 8f6ba9d5fb36318a33d293a416008cd7fa03cb70

Summary:
Update the `AsyncLogWriter::performIO()` API to take the vector of messages by
const reference, rather than by modifiable pointer.  This parameter is purely
an input parameter, and shouldn't be modified by the `performIO()` code, so it
makes more sense to take this as const reference.

Differential Revision: D21367425

fbshipit-source-id: 4bbf3a58dd48bb620101ec3982b189bea656b301

Summary: The first portion of the task (before the first suspension point) was run inline.

Reviewed By: yfeldblum

Differential Revision: D21460275

fbshipit-source-id: 31d50e15d5890fbc7127dfff424a4e814dcb2320

Summary: [Folly] `copy_to_unique_ptr`, `copy_to_shared_ptr` - type-deducing wrappers around `make_unique` and `make_shared`.

Reviewed By: vitaut

Differential Revision: D21332222

fbshipit-source-id: f41dac0fcfa972118d3a721194c941e14511a80c

Summary:
fboss-oss build links to hash that corresponds to tag v4.4.0 released on Jan 11 2016

```
repo_url = https://git.kernel.org/pub/scm/linux/kernel/git/shemminger/iproute2.git
rev = 92a0236a3cdf3438000834121b7ea8a09f1f52b1
```

The change is to update the iproute2 version to ```v4.12.0``` (July 5 2017) to match with the version used internally to Facebook

Reviewed By: shri-khare

Differential Revision: D21411714

fbshipit-source-id: fac606396e284193bb7199cf60d2601594bfa5f0

Reviewed By: wez

Differential Revision: D21417542

fbshipit-source-id: f7f0726e84db3dc903bf11c9045a9975a3e90234

Summary:
The `FOLLY_XLOG_STRIP_PREFIXES` macro tells folly::logging how to strip off
the absolute portion of the path when automatically computing log category
names from the source file name.

Previously this was set in FollyCompilerUnix.cmake, which caused it to only be
applied on Linux/Mac platforms, and only for targets that explicitly used
`apply_folly_compile_options_to_target()`.

This changes the code to apply this definition with a directory property, so
it affects all targets under the top-level directory, and it is also applied
on Windows.

Reviewed By: wez

Differential Revision: D21333014

fbshipit-source-id: 9256e14ab4fd03d19f28957e7ca298c930a22b18

Summary:
Original commit changeset: f1dc8bd17150

This accidentally enabled the openssl internal session cache which can use
excessive memory.

Reviewed By: jalopezsilva, mingtaoy

Differential Revision: D21433514

fbshipit-source-id: 91358f6fd2decc0d18eff7ca8194b677f3c8629c

Summary:
Remove the `HANGING` tag from this test, so that it is run by default in the
CMake build.

Differential Revision: D21367512

fbshipit-source-id: 4bb707c83005ee8cd39ed2ce6b82b7f0c53f2c88

Summary:
This puts the `writeFileAtomic()` function behind and `#ifndef _WIN32` check,
to disable it on Windows.

While this function currently compiles on Windows, it does not behave
correctly: it often fails with "Permission denied" errors when trying to use
folly's emulated `fchmod()` function.  Additionally, even the `fchmod()`
succeeds (or is skipped) it fails if the destination path already exists, as
`rename()` on Windows does not allow replacing an existing file.  While we
could attempt to replace the existing file, doing so would not be atomic.  We
might be able to achieve this with transactional NTFS features if the
underlying file system is NTFS, but according to Microsoft these APIs are
discouraged and may be removed in future versions of Windows.

Therefore for now simply disable `writeFileAtomic()` on Windows.

Differential Revision: D21367511

fbshipit-source-id: 50a5a1004feacc8bdb0a24421bea339962e1d546

Summary:
Add a CMakeLists.txt file to build the logging example program in CMake
builds.

This also disables the `AsyncFileWriter.fork` test on Windows, since `fork()`
isn't available on Windows.

Reviewed By: yfeldblum

Differential Revision: D21332968

fbshipit-source-id: b578df92576e79d34b7fe2bf79dc8f30e0fd6d6b

Summary:
D8941725 changed the default log level from WARNING to INFO.  This updates the
comments in the example logging program to reflect that.

Reviewed By: yfeldblum

Differential Revision: D21332969

fbshipit-source-id: 88b206bf4dc891d8868a34ea02d91cc79b7c5c48

Summary:
D21364132 accidentally broke this; we can't run the fetcher
for projects for which we pulled the build out of cache, because there
is no source to update in that case.

This commit adjusts the logic so that we write out a marker file
to indicate that we installed a build from cache and to look for
that file being present to gate the new update logic.

Reviewed By: lnicco

Differential Revision: D21419122

fbshipit-source-id: 304670848add22531d88549d66f22c40ff255140

Summary: Return 0 from PollIoBackend::eb_event_add if the event type is not set

Reviewed By: danobi

Differential Revision: D21416811

fbshipit-source-id: 5825e473dca06d33017785e4d73947aade5f554c

Summary:
Add warnings for pathological cases:
- Too large retired lists
- Thread cache overflow

Move the warning for executor backlog to a separate function for clearer profiling.

Reviewed By: davidtgoldblatt

Differential Revision: D21314798

fbshipit-source-id: e07d6182309a0940adbdf76774a9c44fb1aa200c

Summary: Giving SSLContext ownership over the new session callback attached to SSL_CTX, so that it may do additional work with the new session (i.e. forward the session to its corresponding abstracted SSLSession).

Reviewed By: yfeldblum

Differential Revision: D21021333

fbshipit-source-id: f1dc8bd17150d45498ae1c22ef4eed1d2ca00c01

Summary: Adding SSLContext members to SSL_CTX ex data to access them inside callbacks attached to SSL_CTX (e.g. session callbacks)

Reviewed By: yfeldblum, mingtaoy

Differential Revision: D21021353

fbshipit-source-id: 6aa3995f7d719ca7e87bad798876a92dd5765b86

Summary:
Clarify that the RequestContext is captured at the time the coroutine is
started (e.g. due to co_awaiting it or calling .start()) rather than at
the time the coroutine is called.

Reviewed By: andriigrynenko

Differential Revision: D21353947

fbshipit-source-id: 40ec7b25d84e96a611a1f63b6184a0d6a016948d

Summary:
Migrating buck integration to use configurations, which is more performant and is the primary supported way to run Pyre.
Mypy is no longer supported at Facebook, so mypy buck integration is covered in this migration as well.

Reviewed By: grievejia

Differential Revision: D21398810

fbshipit-source-id: 12c0cd395e243703f4380b31527d0c202d58cec2

Summary:
On Windows, implement `readvFull()` and `writevFull()` by calling `read()` and
`write()` repeatedly, rather than `readv()` and `writev()`.

Windows does not provide native implementations of `readv()` and `writev()`,
so the code was previously using implementations provided by folly in
`folly/portability/SysUio.cpp`.  The folly implementations attempted to lock
the file to provide similar atomicity guarantees that `readv()` and `writev()`
provide on POSIX.  However, callers of `readvFull()` and `writevFull()` don't
care about this atomicity (since they are willing to do multiple `readv()` and
`writev()` calls if necessary).  Therefore this locking behavior is not
needed.  Locking in this case is undesirable because it adds extra overhead,
and fails in some cases.  For instance, attempting to lock stderr blocks for 9
seconds before eventually failing.

Differential Revision: D21367446

fbshipit-source-id: b2ae3c9c5da977402336c750d3d21ba9825527d9

Summary: Use zerocopy only if all the IOBufs in the chain have a SharedInfo

Reviewed By: knekritz

Differential Revision: D21342852

fbshipit-source-id: 458c32ef336b76ed0dc226ea670b7c04c1427bee

Summary: Add session resumption functionality in AsyncSSLSocket for the new abstracted SSLSession. Add logic to resume sessions correctly no matter which API is used, allowing for a piecewise transition to using the abstracted SSLSession.

Reviewed By: mingtaoy

Differential Revision: D20600380

fbshipit-source-id: bb81d5be95ccaa6486a7431391676f3e5c0b5b8e

Summary: I temporarily enabled Clang's `-Wdocumentation` flag, which reports things like using an `param` for a parameter that doesn't exist. [Full explanation with examples here](https://clang.llvm.org/docs/DiagnosticsReference.html#wdocumentation).

Reviewed By: yfeldblum

Differential Revision: D21396362

fbshipit-source-id: d7b34bc14af7bb1b0901a8f195c02674eb587fbc

Summary: The oss jobs are causing problems to the developers now, lets disable them temporarily.

Reviewed By: StanislavGlebik

Differential Revision: D21400898

fbshipit-source-id: f7a3567056633d9eef98a8d05a37cd029c9e506c

Summary:
While compiling EdenFS on Windows, for every single file being compiled,
cl.exe complains that 2 incompatible /std option are being passed in on the
command line, a c++latest, and a c++17. The former is added by this file,
while the later is added by cmake via CMAKE_CXX_STANDARD.

I'm not exactly sure why the default is set to c++latest, but it's guaranteed
to always force cl.exe to generate a warning, as it simply cannot be set via
CMAKE_CXX_STANDARD.

A better fix to this would be to set CMAKE_CXX_STANDARD to 17, but since
folly still supports cmake 3.0, that didn't know about c++17, this can't
be done.

Reviewed By: yfeldblum

Differential Revision: D21392360

fbshipit-source-id: fc0e29ed89adf0a2c0b0193e7bd5f033463ed376

Summary: Add support for detecting x86_64 vectorized AES (VAES) and vectorized carry-less multiplication (VPCLMULQDQ) CPU flags.

Reviewed By: yfeldblum

Differential Revision: D21271564

fbshipit-source-id: 08be20c07d6dfb562b8a0c1e86b77a9a416445fb

Summary:
When the commit hash changed in fbsource, we would correctly decide
that we'd need to rebuild first-party projects but we would incorrectly skip
running the fetcher.update method.  This would mean that we'd not perform the
shipit run and that our shipit tree would diverge from the source tree.

This commit resolves this by performing the fetcher.update but ignoring
the source update status in this case.

Reviewed By: xavierd

Differential Revision: D21364131

fbshipit-source-id: b4001e549c7d3f27aa4a21b19893c9bb7c0f6d1f

Summary:
We didn't do a great job of recognizing that we'd need to
build a project when one of its dependencies had changed: we relied
chiefly on the dependency hash for this and could fail to handle
changes in individual source files.

This commit helps to improve this situation by checking to see if
any installed files in the dependencies of a manifest are newer than
the most recent built time of a given manifest.  If so, we'll trigger
a build.  We try to be reasonably smart about deciding when to trigger
a cmake reconfigure if it looks like cmake files in the deps have
been changed.

Reviewed By: xavierd

Differential Revision: D21364132

fbshipit-source-id: 7534496e10d1f532aa9cf865900ace84a8785327

Summary:
Only run cmake reconfigure for .cmake, .cmake.in and CMakeLists.txt
files changes.

Previously we would reconfigure for any change to a file with a path that
matched `cmake` which could result in false positives in cases where
you may be iterating on .py or .c files in shared cmake directories.

This also reclassifies non-cmake files under fbcode_builder/CMake as source
files so that we run cmake for those; previously they would cause a
reconfigure and build, now they just cause a build.

Reviewed By: xavierd

Differential Revision: D21364133

fbshipit-source-id: a1231f657d6c6056b269656c677d3449d8715cf6

Summary:
Our linter really wants to include formatting changes unrelated
to my diff stack.

This is a formatting only change to avoid clouding my diffs; no functional
effect.

Reviewed By: xavierd

Differential Revision: D21364519

fbshipit-source-id: 7670dd4154e788f593f256aabdfdeef6d17aeec4

Summary:
Previously getdeps would remove the entire top-level `CMakeFiles` directory
from the build output when it wanted to invalidate the CMake cache.  This
directory is used to keep all of the compiled object files for any libraries
or executables defined in the top-level CMakeLists.txt file.  Blowing away
this directory forces all of these sources to be re-compiled, even if this was
not necessary.  This is particularly problematic for folly, which compiles all
of its source files via rules in the top-level CMakeLists.txt target file.

I did have the code still blow away the CMake error and output logs in this
directory: in the past I have seen situations where CMake would not update
these files on new CMake runs if they already existed.

Reviewed By: wez

Differential Revision: D21360668

fbshipit-source-id: 6fcd1a8e371d756114fbab60d8636be8cd5f8978

Summary:
Update the manifest file for folly to indicate a dependency on lz4.
folly does not require lz4 be available, but it will use it if it is found at
configure time.

getdeps is unfortunately not strict about providing projects only with the
dependencies that they require at build time.  This causes it to sometimes
make lz4 available to folly (if you are also building another project that
requires lz4), and sometimes not.  This ends up causing changes in
folly-config.h depending on which projects you are building, forcing all of
the folly sources to be recompiled.

In the future we perhaps should update getdeps to consistently only pass in
include directories for dependencies actually listed in the manifest file.
However, specifying that folly depends on lz4 also works to mitigate this
particular issue for now, and it is also generally desirable to build folly
with lz4 support.

Reviewed By: wez

Differential Revision: D21359995

fbshipit-source-id: aaf61671b7750d6c47e3613c732d220b3311b5ba

Summary: [Folly] `c_array`, a container for returning C arrays from `constexpr` functions. May be used until C++17 may be assumed, since C++17 specifies `constexpr` member functions of `std::array` which make `std::array` suitable for the purpose and a better choice since it  can house zero-sized arrays.

Reviewed By: markisaa

Differential Revision: D21382680

fbshipit-source-id: 56a6528bb05988c1f1c8916f9267575783dc6276

Summary: fix D21371986's sadness (SEGV  for `folly::make_exception<TException>().class_name()`)

Reviewed By: yfeldblum

Differential Revision: D21373766

fbshipit-source-id: 0dd951fd9393e73c5c38670167312c3d2e619fb5

Summary:
Sorry about the churn (& time spent on this), but this backs out the change in D20671620.

Despite best efforts it seems I've missed something and the issue runs deeper than TSAN observing thread exit.

My change has possibly only altered the timing of events, and has exposed another related failure during testpilot runs that didn't occur before, so I think it's best to return to baseline.

Reviewed By: yfeldblum

Differential Revision: D21310561

fbshipit-source-id: 369865f185415123ea8fddf6007115b87ea8fe82

Summary:
In the initial stages of the windows port we had
problems building rocksdb on windows, so we disabled it.

These days we're able to build it and detect it--we even
require it in the cmake code, but hadn't gotten around
to telling the rest of the code that we can use it.

This commit re-enables it in the build but leaves sqlite
as the default engine until we're able to perform some
benchmarking.

Rocksdb itself has some build issues on Windows; it doesn't
use cmake to locate dependencies, so even though we built
snappy it doesn't know how to find it without modifying the
source:
https://github.com/facebook/rocksdb/blob/master/thirdparty.inc#L4

For that reason, we disable the use of Snappy in the Windows build.
However, in the version of rocksdb that we were using, it would
default to trying to use Snappy even though it wasn't compiled in
and throw an exception.

I've upgraded to a newer version of rocksdb that will simply not
use compression if no compression was enabled at build time.

Given that we mostly store relatively small objects, I'm assuming
that the lack of compression is fine for now.

Reviewed By: xavierd

Differential Revision: D21319896

fbshipit-source-id: 2a2d06d4bd5382706e9220f9b4a2de99dc18311d