Summary: Ubuntu 18.04 zstd version is too old for the folly buikld

Reviewed By: Croohand

Differential Revision: D32693530

fbshipit-source-id: 9de4b8d1eac1043b329016e487241e27bf6bb554

Summary:
cmake 3.10 on ubuntu 18.04 reserves the Python3::Interpreter target so that the FBPythonBinary.cmake rules can't definite it when missing

This breaks builds using the Python3::Interpreter rule, notably the python-six manifest which in turn is used by fbthrift

The fix is to use the getdeps manifest cmake manifest instead on .deb platforms.  Once CI is on ubuntu 20.04 everywhere we could revisit

Reviewed By: Croohand

Differential Revision: D32693523

fbshipit-source-id: 8521d904fba588941ab1efd0148f9d24a350e849

Summary: Useful to be able to see what packages would be installed without doing it, and to see rpm vs deb packages

Reviewed By: Croohand

Differential Revision: D32693528

fbshipit-source-id: a01257e7192f2f0299d57f6f8f7ee93452a6f3e4

Summary:
Fixes warnings like the following:

```
folly\io\async\EventBase.cpp(725): warning C5219: implicit conversion from '_Rep' to 'double', possible loss of data
        with
        [
            _Rep=__int64
        ]
folly\io\async\EventBase.cpp(738): warning C5219: implicit conversion from '_Rep' to 'double', possible loss of data
        with
        [
            _Rep=__int64
        ]
folly\io\async\EventBase.cpp(740): warning C5219: implicit conversion from 'size_t' to 'double', possible loss of data
```

Reviewed By: luciang

Differential Revision: D31330566

fbshipit-source-id: 86a2cd10fdad53dce9c0ee0704b26bb1bce2bf18

Summary: No place which defines this nominally owns the name.

Reviewed By: Gownta

Differential Revision: D32632858

fbshipit-source-id: ab5ee161b7c80f08a177c752c4bfe693cf9c9de3

Summary:
Makes `folly::AsyncioExecutor` an abstract class and adds new `folly::python::setExecutorForLoop()` (C++) and `folly.executor.set_executor_for_loop()` (Python) functions. These control what is returned by `folly::python::getExecutor()` and `folly.executor.get_executor()` and hence allow installing a custom system for handling native callbacks on the Python event loop.

In this implementation we only allow the executor for a given Python event loop to be set _once_ before being explicitly cleared. An alternative which was discussed was returning the previous executor when setting a new one, allowing the caller to stack them. My rationale for not doing this is I think this makes it too easy to accidentally "lose" callbacks registered with an existing executor.

Reviewed By: vladima, pranavtbhat

Differential Revision: D32486564

fbshipit-source-id: 4c9ad4e5f2b7b89f1edafdb4eddd9ccacb23635f

Summary: For invoking things given possible-qualified names. Differs from invoke-free as the latter accepts an optional list of namespaces scope to include in name lookup but does not permit qualified names and does not admit a lambda variation.

Reviewed By: Gownta

Differential Revision: D32670267

fbshipit-source-id: f9b9e33bd1e3cfa1c191eb8c91b3f50bb0e9734e

Reviewed By: Gownta

Differential Revision: D32747533

fbshipit-source-id: f3fe004797eeaf22ae1db85f27b547dc7ad7cfbc

Summary: Hand to do just one

Reviewed By: zpao

Differential Revision: D32693529

fbshipit-source-id: ac27c79f8a43c540e3b0836eba3a40bd61dd725b

Summary: As in title, skip lookup in `WeakKeyDictionary` for repeated `get_fiber_manager` calls on the same event loop

Reviewed By: pranavtbhat

Differential Revision: D32566214

fbshipit-source-id: 7a9305ef88538ae4ef422283bc1b8a4140b8ea33

Summary: Update to new version of upload-artifact without branch name as that's a separate repo that our dynamic branch settings should not apply to.

Reviewed By: zpao

Differential Revision: D32717494

fbshipit-source-id: 968908d1072d6d4550705e7df0b3828528e057dc

Summary: Pick up the changes from master to main change

Reviewed By: zpao

Differential Revision: D32693532

fbshipit-source-id: 8cf68e3310b4d2daf71fb10eee7a801c6eabbd48

Summary: Branches updated some time ago, update this to match

Reviewed By: zpao

Differential Revision: D32693531

fbshipit-source-id: 02da79e3fb5fcd3b204be4605521a151b977ce2c

Summary:
Before removing `getX509()` from `AsyncTransportCertificate` we need to migrate
all the callsites to `OpenSSLTransportCertificate`.

Differential Revision: D30487611

fbshipit-source-id: 50deea4904b7b6a6c08c6d4fd0c635a4bc8774cb

Summary: Match overloads of upstream Invoke action to ease migration

Reviewed By: ispeters

Differential Revision: D32706320

fbshipit-source-id: d9564f63e534e3651433806fcf4f1fff7730fcc5

Summary: Mononoke has decided to no longer support getdeps builds. This diff deletes all mentions to those I could find.

Reviewed By: StanislavGlebik

Differential Revision: D32469441

fbshipit-source-id: a11b3f1d5c57ef20792deb0e7f164e655f62648c

Summary:
Conversion for cases when we have a `folly::any_badge<A, B>` and want to call some function that accepts `folly::any_badge<A, B, C>`.
```
void call_superset(folly::any_badge<A, B, C>);
void call_subset(folly::any_badge<A, B> badges) { call_superset(badges); }

// somewhere in A class
call_subset(folly::badge<A>{});

// somewhere in B class
call_subset(folly::badge<B>{});
```

Also establish bidirectional conversions between `folly::badge<A>` and `folly::any_badge<A>`.

Reviewed By: yfeldblum

Differential Revision: D32697212

fbshipit-source-id: 2f55d0b5015e133166fcdd6d9c25a0e396c2a151

Summary: Missed this when switching from spaces to underscores in distro names

Reviewed By: krallin

Differential Revision: D32693527

fbshipit-source-id: 658504d76c96defd3f1a217ada8b05f99c841ede

Summary: Update it

Reviewed By: krallin

Differential Revision: D32641645

fbshipit-source-id: 77ad7a2771804ac9d32dc38982635b45863af4dc

Reviewed By: Gownta

Differential Revision: D32669969

fbshipit-source-id: 4e0cd2cf73ba695720dd41416d0d04a3996385da

Summary: Wrap more of poisoning/unpoisoning and fiber stack switching.

Reviewed By: luciang

Differential Revision: D31635307

fbshipit-source-id: 9a6f9adbefe6fd995cfa864e0498102e237e6351

Summary:
Use inline variables to avoid the function-pointer trick.

Use explicit default constructors per the types in the standard.

Import the names from the standard in C++17.

This allows declaring a function with tags from one namespace and calling the function with tags from the other namespace in C++17.

Reviewed By: vitaut

Differential Revision: D32588945

fbshipit-source-id: 0ece26ebf7f9f6acf9abb8b008327f06f84091f2

Summary:
`__folly_memcpy` already behaved as `memmove` for n <= 256B.

For n > 256B and overlapping buffers `__folly_memcpy` did some work
- it determined the size is large enough for 128B AVX loads/stores
- it has already copied the first 128B and the last 128B in YMM registers

but then discarded it and fell back on `memmove`.

Instead of wasting this work, forward copy (dst < src) or backward copy (dst > src).
- use unaligned loads + aligned stores, but not non-temporal stores
- for dst < src forward copy in 128 byte batches:
-- unaligned load the first 32 bytes & last 4 x 32 bytes
-- forward copy (unaligned load + aligned stores) 4 x 32 bytes at a time
-- unaligned store the first 32 bytes & last 4 x 32 bytes
- for dst > src backward copy in 128 byte batches:
-- unaligned load the first 4 x 32 bytes & last 32 bytes
-- backward copy (unaligned load + aligned stores) 4 x 32 bytes at a time
-- unaligned store the first 4 x 32 bytes & last 32 bytes

Reviewed By: yfeldblum

Differential Revision: D31915389

fbshipit-source-id: 2c0197b2bddc102a7fb8f70a6f43e79ac994dc73

Summary: Before this change Try always reports as copy constructible then fires a static_assert if you try to instantiate that constructor for a non-copy-constructible type, which breaks logic like that inside std::vector::resize.

Reviewed By: yfeldblum

Differential Revision: D32567774

fbshipit-source-id: 4e3e35a1f823ac9ccb5e2d648015fddccb14cfbf

Summary: That is an old glibc. The shims are no longer required.

Reviewed By: Orvid

Differential Revision: D32632173

fbshipit-source-id: d6ce675c6f16b896a950363b411bec66cedca966

Summary: renable after python update from previous diff took us to a Big Sur compatible version

Reviewed By: HarveyHunt

Differential Revision: D32619730

fbshipit-source-id: 50cefe8a2448b7e12d2f666b4570790bfbbe2cdf

Summary:
Update to a newer python that builds on macOS Big Sur, make eden_scm depend on it and use it from PATH

python 3.8 requires libffi which is reference via its pkg-config setup,  however python's pkg-config libffi detection is broken (https://bugs.python.org/issue34823) with the documented workaround requiring an environment variable to be passed into its ./configure step, which is why this change also adds a feature to AutoconfBuilder

With the updated python in place I was able to remove disable_env_override_pkgconfig = 1
disable_env_override_path = 1 from the eden_scm config so that it actually uses the pkg-configs, PATHs and other settings from its dependencies. This should make future python and other dependency upgrades much simpler.

Reviewed By: HarveyHunt

Differential Revision: D32231261

fbshipit-source-id: a2b6addbe22f38e3d71618c802d2c6f836fdd86c

Summary: A macro which expands to `constexpr` under C++17. Can be used with `if` and lambdas in cases where `constexpr` is an optional hint but not necessary for the program to compile.

Reviewed By: Mizuchi

Differential Revision: D32607012

fbshipit-source-id: 56a6d8679ab2add4f88c89642166f1ed4b956089

Summary:
Fixes this warning:

```
folly\experimental\bser\Load.cpp(109) : warning C4866: compiler may not enforce left-to-right evaluation order for call to 'folly::dynamic::operator[]'
folly\experimental\bser\Load.cpp(131) : warning C4866: compiler may not enforce left-to-right evaluation order for call to 'folly::dynamic::operator[]'
folly\experimental\bser\Load.cpp(136) : warning C4866: compiler may not enforce left-to-right evaluation order for call to 'folly::dynamic::operator[]'
```

Reviewed By: luciang

Differential Revision: D31320782

fbshipit-source-id: 9e42176436d7ebf22af38b0589ccdc75e020b771

Reviewed By: Gownta

Differential Revision: D31824054

fbshipit-source-id: 101f8daa3018e89c1857d4d2980e97bc8487ca21

Reviewed By: yfeldblum

Differential Revision: D32364716

fbshipit-source-id: f36356d5818adebe5bbbe219044c81282ab008a6

Summary: I found this useful to see where manifests resolve to when they are already built

Reviewed By: HarveyHunt

Differential Revision: D32644308

fbshipit-source-id: b796cb67f154177fbac1093290d97fe5aab6b20c

Summary: Fix FLAKE8 lint about redefintion of 'getdeps' variable

Reviewed By: HarveyHunt

Differential Revision: D32644576

fbshipit-source-id: 9c3b4ef54d2278238c3dc88a4969e34e24976a3d

Summary: This fixes some missing cert errors when running eden_scm tests against our own build of openssl on macOS

Reviewed By: HarveyHunt

Differential Revision: D32617808

fbshipit-source-id: efc34341c2c9b578f9c1c92e0678dfb16ef9112e

Summary: System openssl is preferred on Linux

Reviewed By: HarveyHunt

Differential Revision: D32617807

fbshipit-source-id: 30913858b7f4172cea3d819a63d57563b3da64aa

Summary: I don't think the spaces were breaking anything, but many years on unix means they set my spidey sense tingling when looking at the build logs.

Reviewed By: StanislavGlebik

Differential Revision: D32619583

fbshipit-source-id: 8d5bc35f085faced20457645f1c1430371a3c578

Summary: Just replace it with `__GLIBCXX__`.

Reviewed By: andrewjcg

Differential Revision: D32624136

fbshipit-source-id: fa3ae4b4e7c3b6c05793a25cc99a93a0d0c18c7f

Summary: A lambda-expression parallel to `FOLLY_CREATE_MEMBER_INVOKER_SUITE`.

Reviewed By: Gownta

Differential Revision: D32601861

fbshipit-source-id: 8bb5876d800215346b19c42090a1cc13b6aa9042

Summary: In generic algorithms parameterized by a mutex type, it may be the case that a shared lock is desired if the mutex is a shared mutex and a unique lock is desired otherwise. This allows a wider range of mutex type arguments rather than just permitting shared mutex types. A hybrid lock may be used with read-only access to data; when the mutex type is shared, the lock would be optimized to a shared lock, but when the mutex type is non-shared, the lock would fall back to a unique lock. This may be helpful since writes may dominate reads in some cases and unique locks on non-shared mutex types may be faster than unique locks on shared mutex types.

Reviewed By: Gownta

Differential Revision: D29624498

fbshipit-source-id: 9caaf9c413fd6ce7ea318906362c3c4330f8d343

Summary: Lets make this faster

Reviewed By: Croohand

Differential Revision: D32614135

fbshipit-source-id: 2285bdb89c2ad982667261800f327f851a6cfcec