Summary:
When closing sockets on retry, we should use `shutdownSocketSet_`
if it is set (so socket will properly be removed from the set).

Also I changed all `::close()` and `::shutdown()` calls to `*NoInt()`
version.

Test Plan:
fbconfig unicorn/test:basic_compression_test && fbmake opt -j32
while _bin/unicorn/test/test:basic_compression_test; do echo DONE; done

Reviewed By: mcduff@fb.com

Subscribers: trunkagent, folly-diffs@, yzhan, yfeldblum

FB internal diff: D1821104

Tasks: 4752579, 6123510

Signature: t1:1821104:1423001050:5bc09ffdd6666c2884ea82dbd70831a56513cfc9

Blame Revision: D1795120

Summary:
before:

E0203 12:32:34.548096 3354863 [admonitor-local] Singleton.cpp:72] Singleton of type N8facebook12configerator22ConfigeratorStaticDataE has a living reference at destroyInstances time; beware! Raw pointer is 0x7fd6ccc81000. It is very likely that some other singleton is holding a shared_ptr to it. Make dependencies between these singletons are properly defined.

after:

E0203 13:48:09.013022 3913115 Singleton.cpp:72] Singleton of type facebook::configerator::ConfigeratorStaticData has a living reference at destroyInstances time; beware! Raw pointer is 0x7f6f7dc4c000. It is very likely that some other singleton is holding a shared_ptr to it. Make dependencies between these singletons are properly defined.

Test Plan: run it

Reviewed By: chip@fb.com

Subscribers: folly-diffs@, yfeldblum

FB internal diff: D1822466

Signature: t1:1822466:1423000686:345f40fa706701476256a7157468521bc69166a0

Summary: This was only used in a unittest, every real use case uses AsyncServerSocket, and decides on its own to do SSL or not.

Test Plan: fbconfig -r thrift/lib/cpp/test; fbmake runtests

Reviewed By: alandau@fb.com

Subscribers: fugalh, trunkagent, doug, alandau, bmatheny, ssl-diffs@, mshneer, jsedgwick, folly-diffs@

FB internal diff: D1806807

Signature: t1:1806807:1422396209:02333c736e1ef10963e3fe0b4ed6ecf6122475bb

Summary:
Originally meant as a kill switch.  It's turned on in everything except adfinder, so we can probably remove this flag after running a couple adfinder canaries

I have no idea who should review this in ads team, guessed and added some people

Test Plan: https://our.intern.facebook.com/intern/tupperware/canary/view/?experiment_id=32937

Reviewed By: haijunz@fb.com

Subscribers: trunkagent, doug, ruibalp, nli, alandau, bmatheny, wormhole-diffs@, mshneer, folly-diffs@

FB internal diff: D1809255

Tasks: 5645329

Signature: t1:1809255:1422490068:5b4551e6c126d86cac8f8e7b06da6474e6b8a25a

Summary: This diff introduces 5 seconds wait time to let other threads release Singleton which may be temporarily locked. This helps prevent most of "Singleton object alive after destruction" warnings in cases where weak_ptr API is used correctly. Abusive use of folly::Singletons, where dependencies between singletons are not properly defined will still cause a warning.

Test Plan: unit test

Reviewed By: chip@fb.com

Subscribers: trunkagent, folly-diffs@

FB internal diff: D1808371

Signature: t1:1808371:1422487261:573eb40b6a260e428d96be476659335250c7ea76

Summary:
AsyncSocket sets CLOEXEC if it connects an address, but won't
if it uses a connected fd. This sets CLOEXEC for such fd.

Test Plan: fbconfig folly/io/async/test && fbmake runtests

Reviewed By: davejwatson@fb.com

Subscribers: trunkagent, folly-diffs@, zacm

FB internal diff: D1809300

Signature: t1:1809300:1422900997:40fcc84506582ac67076fd975a36d094522c35c2

Summary:
[Folly] EventBase::runInEventLoopThreadAndWait.

Useful for when some code needs to be run in the event loop thread, but another thread needs to trigger the code and then wait for it to be done.

Test Plan:
Unit tests:
* `folly/io/async/test/EventBaseTest.cpp`

Reviewed By: davejwatson@fb.com

Subscribers: trunkagent, folly-diffs@, brettp, dougw

FB internal diff: D1810764

Signature: t1:1810764:1422900654:7ff0aa7feb2792266f620b344cf8a1110a09f7ef

Summary:
Fixing following the issues raised by nsuke on github
following his diff (https://github.com/facebook/folly/pull/112)

Test Plan: visual inspection, tested build on arch linux

Reviewed By: davejwatson@fb.com

Subscribers: folly-diffs@

FB internal diff: D1807378

Tasks: 5909189

Signature: t1:1807378:1422403093:ca024f396a27ecd9b304c57643f0107764a38ca1

Summary:
fetch_and and fetch_or are actually required in some instances,
since the operation they perform isn't reversible.  This also adds some
other missing functions, ^= and fetch_{add,sub,xor}.

Test Plan: use from other code's unit tests

Reviewed By: mssarang@fb.com

Subscribers: yfeldblum, folly-diffs@

FB internal diff: D1798924

Signature: t1:1798924:1422892374:a9859039075ddcac54f75259b9bc6a29bb963a09

Summary: This reverts commit 90da92b05762f8032560d0d6a66237ab2772d7f2.

Test Plan: compiles

Reviewed By: tjackson@fb.com

Subscribers: lins, anca, folly-diffs@, yfeldblum

FB internal diff: D1817376

Tasks: 6118752

Summary:
There was a race reading `callback_` in `maybeCallback` and setting `callback_` in `setCallback`. This diff reworks the state machine to make this unpossible. To avoid the explosion of states due to the cross-product of has-interrupt-handler/has-been-interrupted/etc. I introduce a separate lock for setting interrupt handler and interruption, since this is primarily orthogonal. Other attributes (active, for example) are still atomic variables, and while somewhat tied into the state machine logically (e.g. transitioning from Armed to Done only happens when active) they are mostly independent, keeping the state machine simple (and probably faster).

I think it may even be possible to do some things cheaper. In some states, we may not need to protect the writing of `callback_` and `result_`. But we'd need to enforce some ordering so I'm not going to try to tackle that. But that could be some speedup if we can do it cheaply.

Test Plan:
Builds and all existing tests pass.

Reviewed By: rockyliu4@fb.com

Subscribers: yfeldblum, stepan, trunkagent, exa, folly-diffs@, jsedgwick

FB internal diff: D1807854

Tasks: 6087856

Signature: t1:1807854:1422656713:25b62706cd7952b2dde06dab08074f8030db456b

Summary:
Add a description of the fault-after-unlock problem that can
occur when a synchronization object is used to guard its own
destruction.

Test Plan: comment change only

Reviewed By: alikhtarov@fb.com

Subscribers: trunkagent, folly-diffs@

FB internal diff: D1808446

Signature: t1:1808446:1422481173:83c0369e2277cd5b8cf6377fcd92b132019611a7

Summary: That is, for all inputs except r-values.

Test Plan:
Run tests
@override-test-failures

Reviewed By: ajaymenon@fb.com

Subscribers: trunkagent, folly-diffs@

FB internal diff: D1807789

Signature: t1:1807789:1422410689:442d50ab17f8e3b08e34f33318a4dc7f23b4c6cc

Summary:
This speeds up TLS handshake and might be a factor of liger perf regression.
The enabling is guarded by an #ifdef. The condition itself is defined in an
openssl patch.

Test Plan: folly unit tests pass. Tried on devices as well and from tcpdump data was sent before the new ticket was received, essentially speeded up the handshake process.

Reviewed By: subodh@fb.com

Subscribers: trunkagent, kmdent, seanc, benyluo, ssl-diffs@, ranjeeth, folly-diffs@

FB internal diff: D1806856

Tasks: 5284979

Signature: t1:1806856:1422494521:0a048ea9001da13b5d698b5a764d1e66dcbedc99

Test Plan: Unit tests

Reviewed By: ajaymenon@fb.com

Subscribers: trunkagent, folly-diffs@

FB internal diff: D1808023

Signature: t1:1808023:1422410709:8db2d73d5b4c0c2eab563643e5fa1557ebfd4730

Summary:
Included climits and limits.h due to non-conflicting changes,
clean this up.

Test Plan: compiled

Reviewed By: kelarini@fb.com

Subscribers: trunkagent, folly-diffs@

FB internal diff: D1810133

Signature: t1:1810133:1422483209:0826d2198769efed86607de6c5b28ce0bc4ac04b

Summary: should have been using fulfilTry anyways. and fulfilTry needs to take the Try by value so fix that too

Test Plan: unit

Reviewed By: hans@fb.com

Subscribers: fugalh, folly-diffs@, jsedgwick

FB internal diff: D1808923

Tasks: 6098987

Signature: t1:1808923:1422478981:d4b9727394339c996ebccb7841b94e0c7b2bffb4

Summary: Now CHAR_BIT will be found

Test Plan: ran multifeed and folly

Reviewed By: seanc@fb.com

Subscribers: seanc, folly-diffs@

FB internal diff: D1810027

Signature: t1:1810027:1422480946:4568c5190e318ff097d33b3d0434593b72f8205e

Summary: ^

Test Plan: fbmake

Reviewed By: woo@fb.com

Subscribers: folly-diffs@

FB internal diff: D1810061

Signature: t1:1810061:1422481560:089c9186834f2bf87ac6d10b71d53784299b7992

Blame Revision: D1806632

Summary:
We do a lot of copying of TransportInfo in proxygen, and in most cases
the SSL structs don't change after connection establishment. We could
cut down on memory usage by sharing these huge strings. This is
especially true with SPDY where all streams belonging to a session could
share these fields.

Facebook:

Test Plan: Unit tests, will canary

Reviewed By: afrind@fb.com

Subscribers: fugalh, bmatheny, ssl-diffs@, folly-diffs@, jsedgwick, woo

FB internal diff: D1807557

Tasks: 5343753

Signature: t1:1807557:1422472932:53038345fca620632097586fb9e410bca8fe748d

Summary: Normally I would use the folly::to<> function, but it would cause a circular dependency, so I just added a static cast

Test Plan: Ran on iOS and ran the folly tests

Reviewed By: seanc@fb.com

Subscribers: lucian, mpawlowski, marcelo, tudorb, aalexandre, seanc, folly-diffs@

FB internal diff: D1806632

Signature: t1:1806632:1422416646:b8104f18f90eb7457f2f358428f2bd800f8f1db5

Summary: as above, there were less of these than I expected, nice

Test Plan: unit

Reviewed By: hans@fb.com

Subscribers: trunkagent, folly-diffs@, jsedgwick

FB internal diff: D1789332

Tasks: 5949939

Signature: t1:1789332:1421878033:d7c2979a77b51a5257b8bcd910ad9296ca1aa7e0

Summary:
std::to_string doesn't exist on Android so don't use it.

Facebook: Did a sync to fbandroid and confirmed liger compiles with this fix.

Test Plan: existing tests

Reviewed By: ranjeeth@fb.com

Subscribers: trunkagent, folly-diffs@, shikong, kmdent, fma, pgriess

FB internal diff: D1808037

Signature: t1:1808037:1422410556:d78e0633a1554254b1a1f25bef49a4550a1817c6

Summary:
See D1785572. Check out the new Future test and the commented portion of wait(Duration) for the fix

The test only fails a few times out of a hundred before the fix, but hasn't failed yet after

Test Plan: futures unit, wait for contbuild

Reviewed By: hans@fb.com

Subscribers: trunkagent, rushix, fbcode-common-diffs@, hero-diffs@, cold-storage-diffs@, adamsyta, zhuohuang, darshan, micha, folly-diffs@, lins, tingy, hannesr, jsedgwick

FB internal diff: D1803526

Tasks: 5940008, 6059995

Signature: t1:1803526:1422309486:3613c59a708ecac312d241723828763feb2a57aa

Summary: Reciprocal of readFile

Test Plan: fbmake runtests

Reviewed By: tudorb@fb.com

Subscribers: trunkagent, net-systems@, folly-diffs@

FB internal diff: D1807551

Signature: t1:1807551:1422410565:f5bda294deb788da9f3881c19bb20cfa1f588c09

Summary:
Fixing following the issues raised by nsuke on github
following his diff (https://github.com/facebook/folly/pull/112)

Test Plan: visual inspection, tested build on arch linux

Reviewed By: davejwatson@fb.com

Subscribers: folly-diffs@

FB internal diff: D1807378

Tasks: 5909189

Signature: t1:1807378:1422403093:ca024f396a27ecd9b304c57643f0107764a38ca1

Summary: no need to be mutable, since it only calls AsyncUDPSocket::getFD() which is a const method

Test Plan: just compiled -- simple change

Reviewed By: afrind@fb.com

Subscribers: trunkagent, doug, folly-diffs@

FB internal diff: D1802765

Signature: t1:1802765:1422307972:eab6ab4b16407df6e5034cef78c79f45cd9fa46f

Summary:
Make python scripts compatible woth both python 3 and python 2.
Added ignore rule for test/glog directory.

Facebook:
While verifying some issues reported on github i noticed that
folly does not build when python 3 is the default (at least on arch
linux). As it turns out it is fairly easy to make our python files
compatible with both python 2 and python 3.
I also added test/glog*/* to gitignore so that it is easier to work
on folly outside of facebook.

Test Plan:
unit tests in fb envirnoment and unit test on arch linux
(python 3.4.2, gcc 4.9.2)

Reviewed By: njormrod@fb.com

Subscribers: trunkagent, folly-diffs@

FB internal diff: D1801368

Signature: t1:1801368:1422294777:4e7b6b5634e8dccd849194e171c9a342bd1cb8b8

Summary: print the value which failed conversion to aid debugging.

Test Plan: run code which throws in to<> and contbuild

Reviewed By: tudorb@fb.com, andrei.alexandrescu@fb.com

Subscribers: trunkagent, folly-diffs@

FB internal diff: D1786294

Signature: t1:1786294:1421795912:00cc10923990e5aac0a15eedec09deb8e8d470d1

Summary: All is good now.

Test Plan: Run all the tests.

Reviewed By: hans@fb.com

Subscribers: jsedgwick, trunkagent, fugalh, folly-diffs@

FB internal diff: D1758272

Signature: t1:1758272:1421889405:a015d30783715e106a1e6667e971f7cf47c8392d

Summary: Not sure why we'd need that? This avoids a move, I think.

Test Plan:
Run all the tests. Also unbreaks this code, which used to work at some
point.
https://phabricator.fb.com/diffusion/FBCODE/browse/master/taoThriftService/TaoServiceHandler.cpp;a27bc2ef36cd671d65ae0fd2cc1fb1f823e68ae4$246

Reviewed By: hans@fb.com

Subscribers: trunkagent, folly-diffs@, jsedgwick

FB internal diff: D1800619

Signature: t1:1800619:1422053903:f9e65a0be3d820a9a9989b3cf5dfaf2a61e2e900

Test Plan: g-unittest

Reviewed By: philipp@fb.com

Subscribers: folly-diffs@

FB internal diff: D1801665

Signature: t1:1801665:1422093104:eecf86ea8c1ab39dcb3d01442ed66ae45ef3bede

Summary: I'm in unfamiliar territory, so shout if I'm doing something dumb.  Perhaps it's a bad assumption that if the ipv4 port is free that the ipv6 port is also free?

Test Plan: g-unittest

Reviewed By: davejwatson@fb.com

Subscribers: trunkagent, ps, bmatheny, folly-diffs@

FB internal diff: D1795120

Signature: t1:1795120:1422034693:bd315023ab6cd9e9bda12161d05dd781dc401546

Summary: Now all connections are always scheduled with connectionManager's default timeout.  This diff enables us to specify the different timeouts for each managed connections.

Test Plan: prospect testing

Reviewed By: davejwatson@fb.com

Subscribers: trunkagent, fugalh, folly-diffs@, jsedgwick

FB internal diff: D1797193

Tasks: 5343753

Signature: t1:1797193:1422034092:ed67b96efe8af8f8d1355d3f86fb1289daafb178

Summary: Future service interfaces similar to finagle.  Service creators for client, filters

Test Plan: Unittests included - also sets up a simple pipeline to test a full stack client/server.

Reviewed By: hans@fb.com

Subscribers: jsedgwick, trunkagent, njormrod, folly-diffs@, doug, fugalh

FB internal diff: D1573086

Tasks: 5002456

Signature: t1:1573086:1421970698:328453c4a980bb6950fc9aeed6a2b6d9819c20db

Summary:
Subject:

This reverts commit d8eeb9a12f932b74d268c74ce7ce2610dc14b152.

Test Plan: Watch contbuild?

Reviewed By: njormrod@fb.com

Subscribers: trunkagent, doug, fbcode-common-diffs@, hero-diffs@, cold-storage-diffs@, adamsyta, zhuohuang, darshan, micha, folly-diffs@, lins, tingy, hannesr, jsedgwick

FB internal diff: D1797163

Tasks: 5940008

Signature: t1:1797163:1421959040:c9d528367fed2de2f7dafdac9416aa01bdb47df9

Summary:
This also adds a dependency from all folly::Singletons on RequestContext. Better fix would probaly be to switch RequestContext to be managed by folly::Singleton, however that would require us to switch all singletons which use it to be folly::Singletons too.

Facebook:
Making sure we initialize SR singletons when OBCClient singleton is created, so that SR is destoyed after OBCClient.

Test Plan:
fbconfig -r --sanitize=address --platform-all=gcc-4.8.1-glibc-2.17 dragon/
fbmake runtests

Reviewed By: meyering@fb.com, davejwatson@fb.com

Subscribers: trunkagent, folly-diffs@, meyering

FB internal diff: D1796847

Tasks: 5986816

Signature: t1:1796847:1421954531:e9ec177d5289a33c12fe0947426d89963c8e262c

Summary:
Histogram::getPercentileBucketIdx() and getPercentileEstimate() are
implemented using template methods of HistogramBuckets.

This updates Instantiations.cpp to also instantiate the versions of
these template methods that are required by Histogram<int64_t>.  Without
this, anyone using these methods was required to also include
Histogram-defs.h, or otherwise they may get link errors.  This makes it
so they no longer need to include Histogram-defs.h

The C++ syntax required is pretty horrendous.  We could potentially
avoid this by refactoring the code so that these methods are no longer
templates (and require that the Bucket class provide methods for this
purpose).  At the moment adding instantiations seems simpler, and
doesn't change any public APIs.

Test Plan:
Successfully linked code using
Histogram<int64_t>::getPercentileEstimate() without including
Histogram-defs.h

Reviewed By: ldbrandy@fb.com, meyering@fb.com

Subscribers: trunkagent, doug, net-systems@, exa, folly-diffs@

FB internal diff: D1786346

Signature: t1:1786346:1421429734:da17b7ea326c64a1e158fec3b87dc49c3db3d848

Summary:
This splits out some code from the existing getIov() method, to support
appending to an existing iovec array, rather than always returning a new
one.

I have a use case where I have multiple IOBuf chains to write at once,
preceded by some header data.  appendToIov() makes it easier to populate
a single iovec array with this data, and then perform a single writev()
call.

Test Plan: Updated the unit tests to exercise appendToIov()

Reviewed By: jasmeetbagga@fb.com

Subscribers: trunkagent, doug, exa, net-systems@, folly-diffs@

FB internal diff: D1783090

Tasks:

Signature: t1:1783090:1421450476:d5c89d6e1adbd2f2239057238e020b302df09f23

Blame Revision: