This CL adds these classes but does not change any functonality or interface
yet. This is in preparation for future CLs. To be used for this:
https://codereview.webrtc.org/2000163002/
RTCCertificateGenerator is meant to replace DtlsIdentityStoreInterface and
implementations. In order to continue to support mocking and to help with the
transition, RTCCertificateGenerator gets an interface that it implements (just
like the store has both interface and impl).
PeerConnectionFactoryInterface::CreatePeerConnection will take an
RTCCertificateGeneratorInterface instead of DtlsIdentityStoreInterface. As to
not break Chromium, both versions of CreatePeerConnection need to exist for a
transition period. This will be done by wrapping a store into a generator
wrapper - RTCCertificateGeneratorStoreWrapper.
BUG=webrtc:5707, webrtc:5708
R=hta@webrtc.org, tommi@chromium.org, tommi@webrtc.org
Review URL: https://codereview.webrtc.org/2001103002 .
Cr-Commit-Position: refs/heads/master@{#12879}
This is needed as there are targets such as newlib_pnacl/remoting_client_plugin_newlib.pexe that depend on rtc_base_approved but don't need TaskQueue. We could implement support for TaskQueue in nacl using ppapi types, but it looks like there isn't a need for it. Libevent isn't supported for nacl either, so I'm introducing a layer on top of rtc_base_approved for TaskQueue. It's conceivable that this target will morph into a target that holds other threading primitives such as platform_thread and possibly socket related operations, which is also an area that we currently #ifdef out for nacl in a few places.
Functional change: Removes the "is_nacl" check.
R=phoglund@webrtc.org
Review-Url: https://codereview.webrtc.org/2001913002
Cr-Commit-Position: refs/heads/master@{#12878}
Reason for revert:
Seems like this CL cause
DtlsTransportChannelTest.TestReceiveClientHelloBeforeRemoteFingerprint
DtlsTransportChannelTest.TestReceiveClientHelloBeforeWritable
to consistently fail on Win DrMemory Full and for
DtlsTransportChannelTest.TestReceiveClientHelloBeforeRemoteFingerprint
DtlsTransportChannelTest.TestReceiveClientHelloBeforeWritable
to consistently fail on Linux Memcheck
Original issue's description:
> Change initial DTLS retransmission timer from 1 second to 50ms.
>
> This will help ensure a timely DTLS handshake when there's packet
> loss. It will likely result in spurious retransmissions (since the
> RTT is usually > 50ms), but since exponential backoff is still used,
> there will at most be ~4 extra retransmissions. For a time-sensitive
> application like WebRTC this seems like a reasonable tradeoff.
>
> R=juberti@chromium.org, juberti@webrtc.org, pthatcher@webrtc.org
>
> Committed: https://crrev.com/1e435628366fb9fed71632369f05928ed857d8ef
> Cr-Commit-Position: refs/heads/master@{#12853}
TBR=pthatcher@webrtc.org,juberti@webrtc.org,juberti@chromium.org,deadbeef@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review-Url: https://codereview.webrtc.org/2002403002
Cr-Commit-Position: refs/heads/master@{#12864}
This will help ensure a timely DTLS handshake when there's packet
loss. It will likely result in spurious retransmissions (since the
RTT is usually > 50ms), but since exponential backoff is still used,
there will at most be ~4 extra retransmissions. For a time-sensitive
application like WebRTC this seems like a reasonable tradeoff.
R=juberti@chromium.org, juberti@webrtc.org, pthatcher@webrtc.org
Review URL: https://codereview.webrtc.org/1981463002 .
Cr-Commit-Position: refs/heads/master@{#12853}
This helps a lot on Android devices where the user threads can be scheduled with low priority when the app is in the background, causing spurious significantly delayed before a packet can be read from the socket. With this patch the timestamp is taken by the kernel when the packet actually arrives.
R=juberti@chromium.orgTBR=juberti@webrtc.org
BUG=webrtc:5773
Review URL: https://codereview.webrtc.org/1944683002 .
Cr-Commit-Position: refs/heads/master@{#12850}
if the network monitor detects it after the native code does.
Also set the network cost for ethernet, wifi, unknown, cellular network type to be 0, 10, 50, 900,
so that unknown networks will have lower precedence than known networks with low cost (like Wifi) but higher precedence than known networks with high cost.
And third, infer network type based on limited name matching in Android if there is no network monitor or network monitor did not find the type.
BUG=webrtc:5890
R=pthatcher@chromium.org, pthatcher@webrtc.org
Review URL: https://codereview.webrtc.org/1976683003 .
Cr-Commit-Position: refs/heads/master@{#12833}
ACCESS_ON is an alias of GUARDED_BY but should take thread-like object instead of mutex,
RUN_ON is an alias of EXCLUSIVE_LOCKS_REQUIRED to annotate function always run on same
thread/task_queue
RTC_DCHECK_RUN_ON - creates an object that allows use of annotated variables and functions, and adds a run-time DCHECK given thread/queue is current.
R=tommi@webrtc.org
Review-Url: https://codereview.webrtc.org/1981893002
Cr-Commit-Position: refs/heads/master@{#12812}
We're now ready https://codereview.webrtc.org/1984503002/ downstream,
so make sure we can enable libevent but still choose which libevent
implementation to use. This follows the common pattern where an enable_
flag controls whether we should use the feature at all, whereas build_
controls if we should use the dependency from our DEPS file or
something else.
NOTRY=True
Review-Url: https://codereview.webrtc.org/1980003002
Cr-Commit-Position: refs/heads/master@{#12772}
New task queueing primitive for async tasks: TaskQueue.
TaskQueue is a new way to asynchronously execute tasks sequentially
in a thread safe manner with minimal locking. The implementation
uses OS supported APIs to do this that are compatible with async IO
notifications from things like sockets and files.
This class is a part of rtc_base_approved, so can be used by both
the webrtc and libjingle parts of the WebRTC library. Moving forward,
we can replace rtc::Thread and webrtc::ProcessThread with this implementation.
NOTE: It should not be assumed that all tasks that execute on a TaskQueue,
run on the same thread. E.g. on Mac and iOS, we use GCD dispatch queues
which means that tasks might execute on different threads depending on
what's the most efficient thing to do.
TBR=perkj@webrtc.org,phoglund@webrtc.org
Review-Url: https://codereview.webrtc.org/1984503002
Cr-Commit-Position: refs/heads/master@{#12749}
Reason for revert:
sigh. Have to revert again as there seems to have have been some change made for pnacl and CrOS.
Original issue's description:
> Reland of New task queueing primitive for async tasks: TaskQueue. (patchset #1 id:1 of https://codereview.webrtc.org/1935483002/ )
>
> New task queueing primitive for async tasks: TaskQueue.
> TaskQueue is a new way to asynchronously execute tasks sequentially
> in a thread safe manner with minimal locking. The implementation
> uses OS supported APIs to do this that are compatible with async IO
> notifications from things like sockets and files.
>
> This class is a part of rtc_base_approved, so can be used by both
> the webrtc and libjingle parts of the WebRTC library. Moving forward,
> we can replace rtc::Thread and webrtc::ProcessThread with this implementation.
>
> NOTE: It should not be assumed that all tasks that execute on a TaskQueue,
> run on the same thread. E.g. on Mac and iOS, we use GCD dispatch queues
> which means that tasks might execute on different threads depending on
> what's the most efficient thing to do.
>
> TBR=perkj@webrtc.org
>
> Committed: https://crrev.com/65d1f2aba216d077c6d22488f03e56984aef1c68
> Cr-Commit-Position: refs/heads/master@{#12737}
TBR=perkj@webrtc.org,phoglund@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review-Url: https://codereview.webrtc.org/1981573002
Cr-Commit-Position: refs/heads/master@{#12738}
New task queueing primitive for async tasks: TaskQueue.
TaskQueue is a new way to asynchronously execute tasks sequentially
in a thread safe manner with minimal locking. The implementation
uses OS supported APIs to do this that are compatible with async IO
notifications from things like sockets and files.
This class is a part of rtc_base_approved, so can be used by both
the webrtc and libjingle parts of the WebRTC library. Moving forward,
we can replace rtc::Thread and webrtc::ProcessThread with this implementation.
NOTE: It should not be assumed that all tasks that execute on a TaskQueue,
run on the same thread. E.g. on Mac and iOS, we use GCD dispatch queues
which means that tasks might execute on different threads depending on
what's the most efficient thing to do.
TBR=perkj@webrtc.org
Review-Url: https://codereview.webrtc.org/1927133004
Cr-Commit-Position: refs/heads/master@{#12737}
Without this, some toolchains may fail to build base/checks.cc
because errno is undefined.
NOTRY=true
Review-Url: https://codereview.webrtc.org/1971513002
Cr-Commit-Position: refs/heads/master@{#12696}
Reason for revert:
Breaks user code. Said code needs to stop using scoped_ptr!
Original issue's description:
> Remove webrtc/base/scoped_ptr.h
>
> BUG=webrtc:5520
>
> NOTRY=True
>
> Committed: https://crrev.com/65fc62e9dd8a8716db625aaef76ab92f542ecc5a
> Cr-Commit-Position: refs/heads/master@{#12684}
TBR=tommi@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:5520
Review-Url: https://codereview.webrtc.org/1965063003
Cr-Commit-Position: refs/heads/master@{#12686}
Instead, use a neat trick with union to ensure that we have a T only
when we're supposed to (and just a bunch of unused memory otherwise).
This is how std::optional behaves, so it makes sense for us to do the
same (and it's convenient, too, since we don't have to pay for the
default-constructed T, and we support types that don't have default
constructors).
Doing this became possible recently when we dropped support for MSVC
2013, which didn't support unions containing non-trivial types.
Review-Url: https://codereview.webrtc.org/1896833004
Cr-Commit-Position: refs/heads/master@{#12664}
Reason for revert:
Breaks downstream gtest usage.
Original issue's description:
> Remove the rtc_relative_path GYP variable and similar defines
>
> This is a reland of https://codereview.webrtc.org/1903553003/ but with
> the SRTP changes removed, since they're needed downstream.
>
> The defines that can be used to alter the include paths for Expat and gtest
> are no longer needed in WebRTC or Chromium. Remove them to simplify GYP.
>
> Removed defines:
> EXPAT_RELATIVE_PATH
> GTEST_RELATIVE_PATH
>
> They're all set in the Chromium build so this shouldn't affect Chromium:
> https://code.google.com/p/chromium/codesearch#chromium/src/third_party/libjingle/libjingle.gyp
>
> BUG=webrtc:4256
> NOTRY=True
> NOPRESUBMIT=True
> TBR=perkj@webrtc.org
>
> Committed: https://crrev.com/081254f2c62037d016f9fc961764c6f01cb095da
> Cr-Commit-Position: refs/heads/master@{#12536}
TBR=perkj@webrtc.org,kjellander@webrtc.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=webrtc:4256
Review-Url: https://codereview.webrtc.org/1945803003
Cr-Commit-Position: refs/heads/master@{#12622}
But keep #including scoped_ptr.h in .h files, so as not to break
WebRTC users who expect those .h files to give them rtc::scoped_ptr.
BUG=webrtc:5520
Review-Url: https://codereview.webrtc.org/1937693002
Cr-Commit-Position: refs/heads/master@{#12581}
And redefine rtc::Buffer as
using Buffer = BufferT<uint8_t>;
(In the long run, I'd like to remove the type alias and rename the
template to just rtc::Buffer, but that requires all current users of
Buffer to start saying Buffer<uint8_t> instead, and since Buffer is
used in the API, we can't do that in one step.)
The immediate reason for the new template is that we'd like to use
BufferT<int16_t> in the AudioDecoder interface.
BUG=webrtc:5801
Review-Url: https://codereview.webrtc.org/1929903002
Cr-Commit-Position: refs/heads/master@{#12564}
The problem with gmock is worked around by commenting out any other override declarations in classes using gmock.
NOPRESUBMIT=True
BUG=webrtc:3970
Review-Url: https://codereview.webrtc.org/1921653002
Cr-Commit-Position: refs/heads/master@{#12563}
Reason for revert:
Reverting this temporarily while I figure out the issues with the Chrome on android GN debug build.
Original issue's description:
> New task queueing primitive for async tasks: TaskQueue.
> TaskQueue is a new way to asynchronously execute tasks sequentially
> in a thread safe manner with minimal locking. The implementation
> uses OS supported APIs to do this that are compatible with async IO
> notifications from things like sockets and files.
>
> This class is a part of rtc_base_approved, so can be used by both
> the webrtc and libjingle parts of the WebRTC library. Moving forward,
> we can replace rtc::Thread and webrtc::ProcessThread with this implementation.
>
> NOTE: It should not be assumed that all tasks that execute on a TaskQueue,
> run on the same thread. E.g. on Mac and iOS, we use GCD dispatch queues
> which means that tasks might execute on different threads depending on
> what's the most efficient thing to do.
TBR=perkj@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review-Url: https://codereview.webrtc.org/1935483002
Cr-Commit-Position: refs/heads/master@{#12562}
TaskQueue is a new way to asynchronously execute tasks sequentially
in a thread safe manner with minimal locking. The implementation
uses OS supported APIs to do this that are compatible with async IO
notifications from things like sockets and files.
This class is a part of rtc_base_approved, so can be used by both
the webrtc and libjingle parts of the WebRTC library. Moving forward,
we can replace rtc::Thread and webrtc::ProcessThread with this implementation.
NOTE: It should not be assumed that all tasks that execute on a TaskQueue,
run on the same thread. E.g. on Mac and iOS, we use GCD dispatch queues
which means that tasks might execute on different threads depending on
what's the most efficient thing to do.
Review-Url: https://codereview.webrtc.org/1919733002
Cr-Commit-Position: refs/heads/master@{#12561}
This CL adds the ability to convert RTCCertificate objects to and from
PEM string representations of it (its private key and certificate).
The RTCCertificate being a wrapper of SSLIdentity, this is where the
meat is.
Changes:
- SSLIdentity::PrivateKeyToPEMString() added. It together with the
already existing SSLCertificate::ToPEMString() yields both private
key and certificate PEM strings, both of which are required
parameters to SSLIdentity::FromPEMStrings().
- Its only implementation, OpenSSLIdentity::PrivateKeyToPemString().
- SSLIdentity::PublicKeyToPEMString() added, used by tests.
- sslidentity_unittest.cc updated:
* FromPEMStringsRSA and FromPEMStringsEC updated.
* CloneIdentityRSA and CloneIdentityECDSA added.
- RTCCertificate::To/FromPem added, using new class RTCCertificatePem.
- rtccertificate_unittest.cc: New test CloneWithPemSerialization.
- Renamed rtc_unittests.cc to rtccertificate_unittest.cc to match
convention.
BUG=webrtc:5794, chromium:581354
Review-Url: https://codereview.webrtc.org/1898383003
Cr-Commit-Position: refs/heads/master@{#12546}
Moves logic of default SocketServer from MessageQueue to SocketServer
Review-Url: https://codereview.webrtc.org/1891293002
Cr-Commit-Position: refs/heads/master@{#12541}
- Places most ObjC code into webrtc/sdk/objc instead.
- New gyp targets to build, strip and export symbols for dylib.
- Removes old script used to generate dylib.
BUG=
Review URL: https://codereview.webrtc.org/1903663002
Cr-Commit-Position: refs/heads/master@{#12524}
Any file that uses the RTC_DISALLOW_* macros should #include
"webrtc/base/constructormagic.h", but a shocking number of them don't.
This causes trouble when we try to wean files off of #including
scoped_ptr.h, since a bunch of files get their constructormagic macros
only from there.
Rather than fixing these errors one by one as they turn up, this CL
simply ensures that every file in the WebRTC tree that uses the
RTC_DISALLOW_* macros #includes "webrtc/base/constructormagic.h".
BUG=webrtc:5520
Review URL: https://codereview.webrtc.org/1917043005
Cr-Commit-Position: refs/heads/master@{#12509}
This propagated into various other places. Also had to #include headers that
were implicitly pulled by "scoped_ptr.h".
BUG=webrtc:5520
Review URL: https://codereview.webrtc.org/1920043002
Cr-Commit-Position: refs/heads/master@{#12501}
There's also one fixup that could neither be done before or after
landing the type alias CL: changing the FunctorMessageHandler template
specialization for rtc::scoped_ptr<ReturnT> to be for
std::unique_ptr<ReturnT> instead. (Having one specialization for
scoped_ptr and one for unique_ptr doesn't work when the types are the
same, and having just the scoped_ptr specialization doesn't work when
scoped_ptr is a type alias. I'm sure there's some template
metaprogramming magic that could solve the problem, but it's easy
enough to just make the change in this CL instead.)
BUG=webrtc:5520
Review URL: https://codereview.webrtc.org/1797463002
Cr-Commit-Position: refs/heads/master@{#12499}
