Once the buffer returned by Windows is not newly allocated, it may contain
legacy images from previous capturing attempts. This usually is not a problem,
as implementations other than ScreenCapturerWinDirectx paint each pixel on the
frame. But due to the one capturer per monitor design of
ScreenCapturerWinDirectx, part of the frame may not be covered by any
DxgiOutputDuplicator, and cause the legacy image to be shown.
So a very simple fix is to clear the DesktopFrame in DxgiFrame.
BUG=708766
Review-Url: https://codereview.webrtc.org/2827983007
Cr-Commit-Position: refs/heads/master@{#17847}
The key change of this CL is to merge ScreenCapturerWinDirectx::frames_ and
contexts_ into a new DxgiFrame class. So consumers of DxgiDuplicateController
does not need to maintain two objects. DxgiDuplicateController::Duplicate*()
functions are also updated to accept DxgiFrame parameter instead of
SharedDesktopFrame + Context. The advantages of this change are,
1. Once the screen resolution changes or an existing monitor has been removed,
DxgiFrame can automatically reset the frame without needing to return a capture
failure.
2. Remove public APIs of DxgiDuplicatorController. Some public APIs are not
needed anymore, i.e. consumers of DxgiDuplicatorController do not need to take
care about these internal states anymore. It also helps to remove several lock
acquiements.
3. Reduce the complexity of ScreenCapturerWinDirectx.
But the disadvantage is, instead of a boolean value,
DxgiDuplicateController::Duplicate*() now return an enumeration. Clients need to
use the enumeration to decide whether the error can be recovered or not.
This change also removes a duplicating logic in ScreenCapturerWinDirectx. i.e.
ResolutionChangeDetector, DxgiDuplicateController now takes care of the screen
resolution changes.
I have verified the scenarios with and without SharedMemoryFactory, also the
Desktop capture API example. So far no regression is detected.
BUG=704205
Review-Url: https://codereview.webrtc.org/2788863006
Cr-Commit-Position: refs/heads/master@{#17795}
The root cause is because the offset used in DxgiOutputDuplicator should always
depend on the position of the monitor in the system instead of the offset in the
target frame. Otherwise, once switching between two monitors with different
screen size, the updated region in the context would base on the old monitor,
and cause the copied regions to be out of the source DesktopFrame.
This issue also impacts the SpreadContextChange() function, the updated region
stores in the Context should also only depend on the position of the monitor.
BUG=chromium:706797
Review-Url: https://codereview.webrtc.org/2801433002
Cr-Commit-Position: refs/heads/master@{#17548}
I have verified this change on my laptop, which cannot reproduce the issue anymore.
BUG=704205
Review-Url: https://codereview.webrtc.org/2781253002
Cr-Commit-Position: refs/heads/master@{#17494}
Once the first capture attempt failed, we should keep the
context->updated_region unchanged for next attempt. This change can (partially?)
fix issue 704205.
BUG=704205
Review-Url: https://codereview.webrtc.org/2780093002
Cr-Commit-Position: refs/heads/master@{#17456}
This change adds a DesktopCapturerId namespace, and attaches an int to each
DesktopFrame. ScreenCapturerWinGdi and ScreenCapturerWinDirectx now actively set
this field to differentiate themselves.
BUG=679523, 650926
Review-Url: https://codereview.webrtc.org/2759493002
Cr-Original-Commit-Position: refs/heads/master@{#17329}
Committed: 41e3d9ff3b
Review-Url: https://codereview.webrtc.org/2759493002
Cr-Commit-Position: refs/heads/master@{#17347}
Reason for revert:
I suspect that this CL breaks Chromium WebRTC FYI bots. (Thanks kjellander@ for spotting.) The added dep in BUILD.gn would be the problem.
Example:
https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.webrtc.fyi%2FLinux_Builder%2F15058%2F%2B%2Frecipes%2Fsteps%2Fcompile%2F0%2Fstdout
FAILED: newlib_pnacl/obj/third_party/webrtc/api/libjingle_peerconnection_api/mediaconstraintsinterface.o
/b/c/goma_client/gomacc ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/pnacl-clang++ -MMD -MF newlib_pnacl/obj/third_party/webrtc/api/libjingle_peerconnection_api/mediaconstraintsinterface.o.d -DNACL_TC_REV=5dfe030a71ca66e72c5719ef5034c2ed24706c43 -DV8_DEPRECATION_WARNINGS -DUSE_OPENSSL_CERTS=1 -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DCHROMIUM_BUILD -DENABLE_MEDIA_ROUTER=1 -DFIELDTRIAL_TESTING_ENABLED -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE=2 -DNDEBUG -DNVALGRIND -DWEBRTC_RESTRICT_LOGGING -DEXPAT_RELATIVE_PATH -DHAVE_SCTP -DENABLE_EXTERNAL_AUTH -DHAVE_WEBRTC_VIDEO -DHAVE_WEBRTC_VOICE -DLOGGING_INSIDE_WEBRTC -DUSE_WEBRTC_DEV_BRANCH -DFEATURE_ENABLE_VOICEMAIL -DEXPAT_RELATIVE_PATH -DGTEST_RELATIVE_PATH -DNO_MAIN_THREAD_WRAPPING -DNO_SOUND_SYSTEM -DWEBRTC_CHROMIUM_BUILD -DWEBRTC_POSIX -I../.. -Inewlib_pnacl/gen -I../../third_party/webrtc_overrides -I../../third_party -fno-strict-aliasing -Wno-builtin-macro-redefined -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -fcolor-diagnostics -Wall -Werror -Wextra -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-deprecated-register -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -O2 -fno-ident -fdata-sections -ffunction-sections -g0 -fvisibility=hidden -fvisibility-inlines-hidden -std=gnu++11 -fno-rtti -fno-exceptions -c ../../third_party/webrtc/api/mediaconstraintsinterface.cc -o newlib_pnacl/obj/third_party/webrtc/api/libjingle_peerconnection_api/mediaconstraintsinterface.o
In file included from ../../third_party/webrtc/api/mediaconstraintsinterface.cc:11:
In file included from ../../third_party/webrtc/api/mediaconstraintsinterface.h:27:
In file included from ../../third_party/webrtc/api/peerconnectioninterface.h:77:
In file included from ../../third_party/webrtc/api/dtmfsenderinterface.h:16:
In file included from ../../third_party/webrtc/api/mediastreaminterface.h:33:
In file included from ../../third_party/webrtc/media/base/mediachannel.h:28:
../../third_party/webrtc/base/socket.h:18:10: fatal error: 'sys/socket.h' file not found
#include <sys/socket.h>
^
1 error generated.
Original issue's description:
> Add DesktopCapturerId and attach it to DesktopFrame
>
> This change adds a DesktopCapturerId namespace, and attaches an int to each
> DesktopFrame. ScreenCapturerWinGdi and ScreenCapturerWinDirectx now actively set
> this field to differentiate themselves.
>
> BUG=679523, 650926
>
> Review-Url: https://codereview.webrtc.org/2759493002
> Cr-Commit-Position: refs/heads/master@{#17329}
> Committed: 41e3d9ff3bTBR=sergeyu@chromium.org,zijiehe@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=679523, 650926
Review-Url: https://codereview.webrtc.org/2767003002
Cr-Commit-Position: refs/heads/master@{#17336}
This change adds a DesktopCapturerId namespace, and attaches an int to each
DesktopFrame. ScreenCapturerWinGdi and ScreenCapturerWinDirectx now actively set
this field to differentiate themselves.
BUG=679523, 650926
Review-Url: https://codereview.webrtc.org/2759493002
Cr-Commit-Position: refs/heads/master@{#17329}
I cannot even imagine this change is useful. But it consistently reduces average
capture time by 0.375% (4.07 -> 4.055), and average encode time by 0.313%
(8.042 -> 8.016) without other impacts. Considering this is a one-line change,
it's worthy to be added.
BUG=679523
Review-Url: https://codereview.webrtc.org/2743233002
Cr-Commit-Position: refs/heads/master@{#17235}
A bug has been reported to complaint the ScreenCapturerWinDirectx cannot capture
the first frame, which is used in "Report an issue" page.
A simple solution here is to skip the first frame.
This change also removes the friend relationship between
DxgiDuplicatorController / DxgiAdapterDuplicator / DxgiOutputDuplicator, which
is not really necessary.
BUG=682112
Review-Url: https://codereview.webrtc.org/2703123002
Cr-Commit-Position: refs/heads/master@{#16815}
This is a trivial change to remove duplicate logic, i.e. fallback capturer, from
ScreenCapturerWinMagnifier.
BUG=webrtc:7215
Review-Url: https://codereview.webrtc.org/2704943002
Cr-Commit-Position: refs/heads/master@{#16781}
This change adds a ResolutionChangeDetector to help dxgi components, say
DxgiDuplicatorController and DxgiTexture to detect resolution changes.
BUG=684162
Review-Url: https://codereview.webrtc.org/2682913002
Cr-Commit-Position: refs/heads/master@{#16654}
BitBlt returns a BOOL value, which should be taken care in ScreenCapturerWinGdi.
Meanwhile, this change also replaces assert() / abort() with RTC_DCHECK() /
RTC_CHECK() / RTC_NOTREACHED().
This change cannot fix the bug, the reason of the issue is still unknown, but it
is still the right thing to do.
In ScreenCapturerIntegrationTest, each frame will be captured at most 600 times.
Since the test case fails, which means the ScreenCapturerWinGdi consistently
returns a white frame for 600 times under a certain state. With this change,
instead of returning white frame, ScreenCapturerWinGdi will return a temporary
error. But I do not think a ScreenCapturerWinGdi can automatically recover by
retrying.
BUG=webrtc:6843
Review-Url: https://codereview.webrtc.org/2553353002
Cr-Commit-Position: refs/heads/master@{#15465}
A trivial change to remove a deprecated comment.
BUG=chromium:314516
Review-Url: https://codereview.webrtc.org/2553283002
Cr-Commit-Position: refs/heads/master@{#15454}
To support rotation in DirectX capturer, several other changes are also
required.
1. Removing AddRect in RotateDesktopFrame, this is a performance improvement.
DxgiOutputDuplicator creates a rotated DesktopRegion, which can be directly
add to updated_region.
2. DxgiOutputDuplicator::SourceRect() is not accurate, the rectangle in source
is controlled by |offset| or |rotation_| + |offset|, instead of desktop_rect().
3. The |region| in DxgiTexture::CopyFrom() is not accurate. It needs an
unrotated DesktopRegion which offsets by |offset| instead of desktop_rect(). To
avoid generating both rotated and unrotated updated_region, this parameter has
been removed. This impacts DxgiTextureStagning performance a little bit (1.5ms).
Refer to bug for details.
BUG=webrtc:6646
Review-Url: https://codereview.webrtc.org/2530303002
Cr-Commit-Position: refs/heads/master@{#15308}
In DxgiOutputDuplicator, we need to convert between a monitor based coordinate
and a entire screen based coordinate. i.e. Copying an updated area from a
monitor (an output in DirectX API) to the entire screen frame (DesktopFrame).
But DxgiOutputDuplicator always assumes the coordinate is based on screen frame.
So we only need to convert a rectange in updated_region to monitor based
coordinate when copying data from texture_. But in last_frame_, the data are
always based on screen coordinate.
So fixes are both required in line 167 and line 180. In the previous one, we do
not need to convert the DesktopRect, which is already screen based, into screen
based coordinate. In the late one, we do not need to convert the DesktopRect at
all. So after these two changes, DxgiOutputDuplicator::TargetRect() function can
be removed.
Flickers of DirectX capturer can happen on any devices, but a virtual machine
can easily reproduce it. While on a regular high performance machine, it's
harder, but not totally impossible, to reproduce the issue.
BUG=314516
Review-Url: https://codereview.webrtc.org/2495143002
Cr-Commit-Position: refs/heads/master@{#15075}
A tiny but critical change to avoid a crash failure in DirectX capturer.
A good news is this failure is caught by ScreenCapturer integration tests.
BUG=314516
Review-Url: https://codereview.webrtc.org/2494893002
Cr-Commit-Position: refs/heads/master@{#15046}
This change removes ScreenCapturer and WindowCapturer from WebRTC.
BUG=webrtc:6513
Review-Url: https://codereview.webrtc.org/2490063002
Cr-Commit-Position: refs/heads/master@{#15033}
D3D_FEATURE_LEVEL_11_0 is not offically documented on MSDN to be a requirement
of DXGI duplicator APIs. So instead of using D3D_FEATURE_LEVEL_11_0 as a
requirement in D3dDevice::Initialize(), this change adds a
ScreenCapturerWinDirectx::SupportedFeatureLevel() function to retrieves minimum
and maximium for further reference (in HostTraits to control the experiment).
BUG=314516
Review-Url: https://codereview.webrtc.org/2468083002
Cr-Commit-Position: refs/heads/master@{#14962}
This change replaces all GetWindowList / GetScreenList with GetScreenList,
SelectWindow / SelectScreen with SelectSource, and BringSelectedWindowToFront
with FocusOnSelectedSource in WebRTC.
BUG=webrtc:6513
Review-Url: https://codereview.webrtc.org/2479553006
Cr-Commit-Position: refs/heads/master@{#14960}
This change copies ScreenCapturerDifferWrapper to a new
DesktopCapturerDifferWrapper, and adds DesktopCapturer::CreateWindowCapturer and
DesktopCapturer::CreateScreenCapturer functions to replace
WindowCapturer::Create and ScreenCapturer::Create.
BUG=webrtc:6513
Committed: https://crrev.com/b763e39beba92b45baa09542f949daabbe6258a3
Review-Url: https://codereview.webrtc.org/2468753002
Cr-Original-Commit-Position: refs/heads/master@{#14880}
Cr-Commit-Position: refs/heads/master@{#14903}
Reason for revert:
Prevents WebRTC rolls into Chrome.
https://build.chromium.org/p/chromium.linux/builders/Blimp%20Linux%20%28dbg%29/builds/14848/steps/compile/logs/stdio
The reason for reverting is: Breaks
https://build.chromium.org/p/chromium.linux/builders/Blimp%20Linux%20%28dbg%2...
[881/894] SOLINK ./libcontent.so
FAILED: libcontent.so libcontent.so.TOC
../../third_party/webrtc/modules/desktop_capture/desktop_capturer.cc:45: error:
undefined reference to
'webrtc::DesktopCapturer::CreateRawWindowCapturer(webrtc::DesktopCaptureOptions
const&)'
../../third_party/webrtc/modules/desktop_capture/desktop_capturer.cc:56: error:
undefined reference to
'webrtc::DesktopCapturer::CreateRawScreenCapturer(webrtc::DesktopCaptureOptions
const&)'
clang: error: linker command failed with exit code 1 (use -v to see invocation)
ninja: build stopped: subcommand failed.
Original issue's description:
> Add CreateWindowCapturer() and CreateScreenCapturer() in DesktopCapturer
>
> This change copies ScreenCapturerDifferWrapper to a new
> DesktopCapturerDifferWrapper, and adds DesktopCapturer::CreateWindowCapturer and
> DesktopCapturer::CreateScreenCapturer functions to replace
> WindowCapturer::Create and ScreenCapturer::Create.
>
> BUG=webrtc:6513
>
> Committed: https://crrev.com/b763e39beba92b45baa09542f949daabbe6258a3
> Cr-Commit-Position: refs/heads/master@{#14880}
TBR=sergeyu@chromium.org,zijiehe@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:6513
Review-Url: https://codereview.webrtc.org/2471773002
Cr-Commit-Position: refs/heads/master@{#14884}
This change copies ScreenCapturerDifferWrapper to a new
DesktopCapturerDifferWrapper, and adds DesktopCapturer::CreateWindowCapturer and
DesktopCapturer::CreateScreenCapturer functions to replace
WindowCapturer::Create and ScreenCapturer::Create.
BUG=webrtc:6513
Review-Url: https://codereview.webrtc.org/2468753002
Cr-Commit-Position: refs/heads/master@{#14880}
Enabled the plugin and cleaned up all issues it found, mainly virtual
destructors not being marked as override.
BUG=webrtc:163
Review-Url: https://codereview.webrtc.org/2436503004
Cr-Commit-Position: refs/heads/master@{#14793}
To ensure this change won't break Chromium, this is the first change, to add a
new CaptureFrame() function, and let Capture(DesktopRegion) and CaptureFrame()
call each other. So both a legacy consumer or a legacy implementation won't be
broken.
BUG=https://bugs.chromium.org/p/webrtc/issues/detail?id=6513
Review-Url: https://codereview.webrtc.org/2409833002
Cr-Commit-Position: refs/heads/master@{#14635}
We can use ScreenCapturerDifferWrapper if needed, otherwise ScreenCapturer does
not need to calculate updated region itself, setting to entire screen is enough.
BUG=633802
Review-Url: https://codereview.webrtc.org/2348803003
Cr-Commit-Position: refs/heads/master@{#14377}
1. Remove legacy screen-saver-blocking logic
2. tls_index_ is not a good choice, we can use thread-static
3. ScreenCapturerHelper is not designed for this scenario
4. Disable this capturer on 2+ monitors system
BUG=638802
Review-Url: https://codereview.webrtc.org/2319383002
Cr-Commit-Position: refs/heads/master@{#14342}
1. It looks like ComPtr cannot work well with vector::emplace_back, I got a
consistent crash on one of my machine, but not the other. Move constructor
should have no impact to lvalue reference, but I may be wrong here. The
impact here is ComPtr released before it should be. So a simple solution is to
use copy instead of reference. The D3dDevice is a collection of reference
counted pointers (Microsoft::WRL::ComPtr), there is almost no extra cost.
2. Actively set several fields in D3D11_TEXTURE2D_DESC to avoid potential break
if there are some platform changes later.
3. AcquireNextFrame returns both a DXGI_OUTDUPL_FRAME_INFO with
AccumulatedFrames and an IDXGIResource. But there is no comment in MSDN to
ensure IDXGIResource won't be nullptr if AccumulatedFrames > 0. Adding an extra
check in DxgiOutputDuplicator makes it a safer.
BUG=314516
Review-Url: https://codereview.webrtc.org/2345163002
Cr-Commit-Position: refs/heads/master@{#14341}
This change is to add an DesktopCapturerOptions accurate_updated_region() with
default value as false to indicate whether a pixel-wise differentiation is
required. And ScreenCapturer::Create() function will wrap the implementation
with ScreenCapturerDifferWrapper.
Chromoting will use this option to filter out unchanged frames.
BUG=314516
Review-Url: https://codereview.webrtc.org/2314323002
Cr-Commit-Position: refs/heads/master@{#14248}
DirectX capturer won't be able to update the DesktopFrame for the second
ScreenCapturerWinDirectx instance, if AcquireNextFrame() returns timeout or
unchanged error. Current solution uses |last_frame| of the second
ScreenCapturerWinDirectx instance, which also does not contain the updated frame
between this and last AcquireNextFrame() calls. Considering following situation,
(C1: capturer 1, C2: capturer 2, update: screen updated, next AcquireNextFrame()
call will return a new frame, Fx: a frame x)
update -> C2.capture returns F1 -> update -> C1.capture returns F2 ->
C2.capture unchanged
So using F1 to update the last capture frame is not correct, we need to use F2.
Refer to design doc https://goo.gl/hU1ifG for a detail description.
The change also makes DxgiDuplicatorController work with 2+ DesktopFrame queue.
Now TwoDirectxCapturers test can pass.
BUG=314516
Review-Url: https://codereview.webrtc.org/2299663003
Cr-Commit-Position: refs/heads/master@{#14077}
Current DirectX capturer cannot capture multiple video cards or monitors. But
according to DXGI desktop duplication sample (https://goo.gl/An0L9l), we can
capture multiple video cards and monitors by duplicating them one by one. So
instead of one IDXGIOutputDuplication instance, this change creates an
IDXGIOutputDuplication instance for each monitor, and merge the output into
one DesktopFrame.
Several other changes are also included,
1. Add supports to DXGI_OUTDUPL_DESC.DesktopImageInSystemMemory. When this flag
is true, we won't copy its content to staging buffer.
2. Capture one monitor instead of entire screen.
Above changes make the logic complex. But with some refactor work, the logic is
not disordered. Please refer to the doc @ https://goo.gl/hU1ifG.
BUG=314516
Review-Url: https://codereview.webrtc.org/2099123002
Cr-Commit-Position: refs/heads/master@{#13684}
When Windows is switching display mode, DirectX based capturer may not be able
to create a new IDXGIOutputDuplication instance, which is expected. So it should
return a temporary error instead of a permanent error.
BUG=
Review-Url: https://codereview.webrtc.org/2092543003
Cr-Commit-Position: refs/heads/master@{#13279}
Previously raw pointers were used for owned DesktopFrame instances.
Updated all screen and window capturer implementations to use
std::unique_ptr<>.
Also includes some other cleanups in the capturers:
- s/NULL/nullptr
- moved default initializers to class definition.
BUG=webrtc:5950
Review-Url: https://codereview.webrtc.org/1988783003
Cr-Commit-Position: refs/heads/master@{#13058}
Reason for revert:
Broke chromium builder
Original issue's description:
> Use std::unique_ptr<> to pass frame ownership in DesktopCapturer impls.
>
> Previously raw pointers were used for owned DesktopFrame instances.
> Updated all screen and window capturer implementations to use
> std::unique_ptr<>.
>
> Also includes some other cleanups in the capturers:
> - s/NULL/nullptr
> - moved default initializers to class definition.
>
> BUG=webrtc:5950
>
> Committed: https://crrev.com/4a627a8c13554d12412cabb8f751caee6e61ee32
> Cr-Commit-Position: refs/heads/master@{#13032}
TBR=wez@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:5950
Review-Url: https://codereview.webrtc.org/2030333003
Cr-Commit-Position: refs/heads/master@{#13033}
Previously raw pointers were used for owned DesktopFrame instances.
Updated all screen and window capturer implementations to use
std::unique_ptr<>.
Also includes some other cleanups in the capturers:
- s/NULL/nullptr
- moved default initializers to class definition.
BUG=webrtc:5950
Review-Url: https://codereview.webrtc.org/1988783003
Cr-Commit-Position: refs/heads/master@{#13032}
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}
