release-3baa2a3: Fix loom logger build when using libcxx

Summary: Include missing header

Reviewed By: BurntBrunch

Differential Revision: D6714224

fbshipit-source-id: b745d1403b965463241632fddc70eccaebc770f3

release-a9c75c9: Typos in comments and log messages

Summary:
No code changes, no testing required.

alligned -> aligned
allignment -> alignment
completly -> completely
conseptually -> conceptually
decendents -> descendants
indefinetly -> indefinitely
dimention -> dimension
doesnt -> doesn't
safegaurd -> safeguard
intialization -> initialization
hierachy -> hierarchy
happend -> happened
gaurd -> guard
programatically -> programmatically
initalized -> initialized
immidiately -> immediately
occured -> occurred
unkown -> unknown
neccessary -> necessary
neccesarily -> necessarily
occuring -> occurring
comoponent -> component
propogate -> propagate
recieved -> received
referece -> reference
perfomance -> performance
recieving -> receiving
subsquently -> subsequently
scoll -> scroll
suprisingly -> surprisingly
targetting -> targeting
tranform -> transform
symetrical -> symmetrical
wtih -> with
Closes https://github.com/facebook/react-native/pull/17578

Differential Revision: D6718791

Pulled By: shergin

fbshipit-source-id: 4ab79c1131ec5971d35a0c7199eba7ec0a0918ad

release-a7529f4: Restore sample app target

Summary:
Restore the open source sample target to restore the Travis build.

I'm not sure the python code can parse this data yet, so I'm not gonna flip the
target over to have processmetadata as well.

Differential Revision: D6717041

fbshipit-source-id: 516054861117f7000cd9bbd9c9ab03055515e5dc

release-45a2858: Fix potential issue in profiling thread sampling

Summary:
Fix for 2 potential issues with Sampling Profiler during `flushStackTraces`.

1) Tracing the profiling thread is potentially unsafe. It's is theoretically possible to cause a deadlock while taking a sample while the profiling thread is in flushing stacks. Any code path which uses a mutex inside `SamplingProfiler::flushStackTraces` has a potential to cause the deadlock. The easiest way to do it is to add some logging to the flush code.
  Why it happens:
  While we are in a critical section inside flushStackTraces protected by a mutex and we increment the cpu clock we end up in `sigprof_handler` if it crashes with a `SIGSEGV` for example our `sigcatch_handler` will catch that but then it will return to
  `SamplingProfiler::flushStackTraces` instead of `sigprof_handler` leaving the mutex locked for the process lifetime. That's because we have the same catch handler for profiling and flush.
  I didn't find any code paths which can cause this, but have this behavior as a potential threat makes me uncomfortable.
2) Interrupting flush may lead to profiling stack overflows, but I am not entirely sure if it's the cause of overflows we see on slower Dalvik devices.

Reviewed By: BurntBrunch

Differential Revision: D6678947

fbshipit-source-id: e594ff6bba1424cd915c753089cd7537580875df

release-42ea147: Move robolectric3_test to fbandroid/tools/build_defs

Reviewed By: ttsugriy

Differential Revision: D6687995

fbshipit-source-id: ac88e314f46e5b9b16c2e99f31e1dbfffae905ac

release-ca10c2e: Move ForkJail into own target

Summary:
ForkJail's super useful in its own right, shouldn't be limited to just Loom.

This is an updated version of D6159922 that doesn't break fblite - libforkjail.so is now correctly merged, and accounted for in fblite's library-stripping lists.

Reviewed By: hlu1

Differential Revision: D6624129

fbshipit-source-id: 393a63641d7dd4a62ad8df2d9ff343f32662ea20

release-f3e28c7: Skylarkify OSS compatibility layer.

Summary: This is necessary to keep internal and OSS file names in sync.

Reviewed By: BurntBrunch

Differential Revision: D6667034

fbshipit-source-id: 1dffe26b09f853f820ecc57b4764c28081033462

release-dba88ed: Revert D6159922: Move ForkJail into own target

Summary:
This reverts commit fd519aeac69230fe72f88b6eccfb2c87d450d01d

bypass-lint

Differential Revision: D6159922

fbshipit-source-id: af1f29dad128ed9e1975661ba6c9299aec2f0698

release-cc0fe4a: Fix paths in generate_sources.sh.

Summary: - Fixed the path names in the script.

Reviewed By: BurntBrunch

Differential Revision: D6608672

fbshipit-source-id: f7e42f70e9768cc848c6f6667738dce7fcbc3d02

release-84a4075: Move ForkJail into own target

Summary: ForkJail's super useful in its own right, shouldn't be limited to just Loom.

Reviewed By: BurntBrunch

Differential Revision: D6159922

fbshipit-source-id: fd519aeac69230fe72f88b6eccfb2c87d450d01d