Bug: none
Change-Id: I99aab901c2d43c98e603deaa60cc2803be656a67
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/377780
Auto-Submit: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#43928}
This is to ensure that the device_scale_factor stored per frame is the
latest one according to the system settings and not an old value.
Bug: chromium:383946052
Change-Id: I11e6342201678451554e6883ded48999fd996743
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/377541
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Palak Agarwal <agpalak@google.com>
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43924}
Since the device_scale_factor is usually exposed as a float in chromium,
we want to keep it same here for consistency.
Bug: chromium:383946052
Change-Id: I8d055ca0fcac623f59dcf96eb3cee15efc23b2ba
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/376700
Commit-Queue: Palak Agarwal <agpalak@google.com>
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43869}
In all the other functions `monitor_id` is a global id from range [0..total monitor count), but each DxgiAdapterDuplicator always assumes that it gets a local id from range [0..adapter monitor count).
Bug: chromium:395807060
Change-Id: I4bb232ee5d83f09859534f813111446763fe9fc9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/376840
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43867}
When using GDI to capture windows, not only will the target window be
captured, but all owned windows of the target window will also be
captured, and the captured results of the owned windows will be drawn
onto the captured result of target window.
GDI (BitBlt or PrintWindow) cannot capture windows with WS_EX_LAYERED
attribute, so when the owned window has this attribute, the area of the
owned window in the result is black.
After modification, if the owned window has the WS_EX_LAYERED attribute,
it will not be captured or drawn.
Bug: webrtc:394380765
Change-Id: I5f0c7d31809a353377b0aa52452634b46247af5f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/376360
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Joe Downing <joedow@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#43851}
This is a reland of commit e20fbb00d0e0219b710da24664e81a10b12c703a
which failed due to CHECK failing on monitor_id in
DxgiDuplicatorController::GetDeviceScaleFactor(). This has now been
fixed in this CL by removing the check for greater than 0 and instead
returning std::nullopt.
Original change's description:
> Get DeviceScaleFactor for the captured monitor/screen
>
> Accesses DeviceScaleFactor using the windows API
> GetScaleFactorForMonitor and adds it to the DesktopFrame. In a follow-up
> CL, this value is propagated to
> DesktopCaptureDevice::Core::OnCaptureResult where it is added to the
> frame metadata.
>
> In a follow-up CL, add RegisterScaleChangeEvent to get notified whenever
> the device scale factor changes.
>
> Design doc: go/expose-captured-surface-resolution
>
> Bug: chromium:383946052
> Change-Id: I363af33c569419d95ddf31a0cc2f9cecf6fb0c7b
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/374344
> Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
> Reviewed-by: Mark Foltz <mfoltz@chromium.org>
> Commit-Queue: Palak Agarwal <agpalak@google.com>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#43827}
Bug: chromium:383946052
Change-Id: Iffcc889b9a560695302f71623e3e929ecb2489fb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/376061
Commit-Queue: Palak Agarwal <agpalak@google.com>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43832}
This reverts commit e20fbb00d0e0219b710da24664e81a10b12c703a.
Reason for revert: Breaks WebRTC roll to Chromium, see:
https://chromium-review.googlesource.com/c/chromium/src/+/6218060
Example of error: https://ci.chromium.org/ui/p/chrome/builders/ci/win-arm64-rel-ready/51821/test-results?sortby=&groupby=
Original change's description:
> Get DeviceScaleFactor for the captured monitor/screen
>
> Accesses DeviceScaleFactor using the windows API
> GetScaleFactorForMonitor and adds it to the DesktopFrame. In a follow-up
> CL, this value is propagated to
> DesktopCaptureDevice::Core::OnCaptureResult where it is added to the
> frame metadata.
>
> In a follow-up CL, add RegisterScaleChangeEvent to get notified whenever
> the device scale factor changes.
>
> Design doc: go/expose-captured-surface-resolution
>
> Bug: chromium:383946052
> Change-Id: I363af33c569419d95ddf31a0cc2f9cecf6fb0c7b
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/374344
> Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
> Reviewed-by: Mark Foltz <mfoltz@chromium.org>
> Commit-Queue: Palak Agarwal <agpalak@google.com>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#43827}
Bug: chromium:383946052
Change-Id: I3065b278939ca0e686ee6da0f0721082bc0c99e8
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/375902
Owners-Override: Mirko Bonadei <mbonadei@webrtc.org>
Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43829}
Accesses DeviceScaleFactor using the windows API
GetScaleFactorForMonitor and adds it to the DesktopFrame. In a follow-up
CL, this value is propagated to
DesktopCaptureDevice::Core::OnCaptureResult where it is added to the
frame metadata.
In a follow-up CL, add RegisterScaleChangeEvent to get notified whenever
the device scale factor changes.
Design doc: go/expose-captured-surface-resolution
Bug: chromium:383946052
Change-Id: I363af33c569419d95ddf31a0cc2f9cecf6fb0c7b
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/374344
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Reviewed-by: Mark Foltz <mfoltz@chromium.org>
Commit-Queue: Palak Agarwal <agpalak@google.com>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43827}
When capturing only one display, it is unnecessary to use
DoDuplicateAll, which allocates bitmap image memory for a rectangle
encompassing all screens and captures all displays. In this case, I
switch to DoDuplicateOne.
I have added an int parameter to GetNumFrameCaptured and
EnsureFrameCaptured to distinguish which display's skip behavior is
currently being executed.
After the modification, when capturing a single monitor in a
multi-monitor environment, only the bitmap image memory of the size of
the captured monitor will be allocated, rather than for all monitors.
Bug: webrtc:391914255
Change-Id: Iee56796c50282beaf1dbeef47f5937fe7fe94a05
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/375181
Reviewed-by: Joe Downing <joedow@chromium.org>
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#43822}
Checks if the DirtyRegionMode property is present in GraphicsCaptureSession and logs a boolean histogram with the result.
Detecting support for this property means that the WGC API supports
dirty regions and it can be utilized to improve the capture
performance and the existing zero-herz support.
See also https://issues.chromium.org/issues/347991512 for more details
on how to detect support for dirty regions in WGC.
Bug: chromium:40259177
Change-Id: Ia316c4ece54bd93cfef1fa23c199675c64143f76
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/362240
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#43015}
This is a reland of commit 844225a76a98aa3be5aca09c19ab72a5e7b6c38a
Original change's description:
> Fix 'Image will be cropped if WindowCapturerWinGdi used'
>
> Bug: webrtc:15719
> Change-Id: I7daf8ee5b90fbe9f1246f1d99211ffa0d8a19f73
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/330780
> Reviewed-by: Alexander Cooper <alcooper@chromium.org>
> Commit-Queue: Alexander Cooper <alcooper@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#41503}
Bug: webrtc:15719
Change-Id: Idbb2f4dcc8811d3b2b763a49adc7a57535b3d1b2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/334380
Reviewed-by: Tomas Gunnarsson <tommi@webrtc.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Tomas Gunnarsson <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#42666}
When refactoring the WGC capture path, the check of a closed source
had been placed at a level where the notification of a closed source
was left without being detected since the error message was never
provided to the main WgcCapturerWin::CaptureFrame() which sends the
error message up to the client.
This trivial change ensures that WgcCapturerWin::CaptureFrame() returns
with DesktopCapturer::Result::ERROR_PERMANENT as soon as
WgcCaptureSession::OnItemClosed() has been triggered and it will
ensure that the WGC capture session stops and that any attached
MediaStreamTrack will signal its onended event as expected.
Bug: chromium:330863510
Change-Id: I57e44df417c33efa0595fc277cac5429cf539b26
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/344420
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#41963}
The code uses IDXGIOutput1::DuplicateOutput for screen capture and
it allows only DXGI_FORMAT_B8G8R8A8_UNORM texture format, which
works on most monitor cases except HDR monitor.
HDR mointor returns type of DXGI_FORMAT_R16G16B16A16_FLOAT.
These two types of DXGI_FORMAT_B8G8R8A8_UNORM and
DXGI_FORMAT_R16G16B16A16_FLOAT are all formats that DuplicateOutput
returns based on Windows OS team.
The fix is to add allowed format of DXGI_FORMAT_R16G16B16A16_FLOAT.
Manually repro the issue and validated the fix.
Bug: chromium:40787684
Change-Id: I0a7be38b14a06261d631d2db172f12725edbbf1f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/339621
Reviewed-by: Mark Foltz <mfoltz@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#41749}
Bug: webrtc:15719
Change-Id: I7daf8ee5b90fbe9f1246f1d99211ffa0d8a19f73
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/330780
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#41503}
This CL is a follow-up of work done in
https://webrtc-review.googlesource.com/c/src/+/323882 where the goal
was to reduce the amount of FrameDropped error logs in
WebRTC.DesktopCapture.Win.WgcCaptureSessionGetFrameResult.
The previous work avoids FrameDropped logs for a minimized window
being captured with WGC but we still se a large amount of these error
(or rather warning) logs. See [1] which comes from Canary.
This CL does two different things to improve the situation:
1) It adds kFramePoolEmpty to the existing
GetFrameResult::kFrameDropped enum to point out that the warning
comes from the frame pool not being able to return a valid new frame.
It also makes it more clear that it does not cause an outer/final
error as WgcCapturerResult::kFrameDropped. We still keep the inner
GetFrameResult::kFrameDropped but it is only produced when the frame
pool returns NULL and our external queue is empty. Hence, a real
frame-drop error. Note that, it is still easy to provoke
kFramePoolEmpty simply by asking for a high resolution at a high rate.
The example in [2] comes from a 4K screen @30fps. Hence, we have not
fixed anything yet.
2) It also increases the size of the internal frame pool from 1 to 2.
This does lead to an almost zero rate of kFramePoolEmpt
warnings at the expense of a slightly reduced max capture rate. BUT,
with 1 as size, we can "see" a higher max capture rate but it is not
a true rate since it comes with a high rate of kFramePoolEmpty
errors. Hence, we "emulate" a high capture rate by simply feeding
copies of the last frame that we had stored in the external queue.
Using 2 leads to a more "true" rate of what we actually can capture
in terms of *new* frames and also a substantially lower rate of
kFramePoolEmpty.
In addition, with 1 as size, if we ask at a too high rate and provide
a copy of the last frame, our CPU adaptation will not reduce its rate
since we think that things are OK when it is actually not.
Also, the samples in [3] and [4] both use 2 as numberOfBuffers
as well.
Let me also mention that with this small change, I a have not been
able to provoke any kFramePoolEmpty error messages.
Finally, geDisplayMedia can be called called with constraints where
min and max framerate is defined. The mechanism which maintains the
min rate is implemented via the RequestRefreshFrame API and it can
be sent to the source (DesktopCaptureDevice) back to back with a
previous timer interrupt for a capture request. Without this change,
these RRFs were also a source of a large amount of
kFramePoolEmpty error logs. With 2 buffers instead; this is no
longer the case.
[1] https://screenshot.googleplex.com/7sfv6HdGXLwyxdj
[2] https://paste.googleplex.com/4795680001359872
[3] https://github.com/robmikh/Win32CaptureSample/blob/master/Win32CaptureSample/SimpleCapture.cpp
[4] https://learn.microsoft.com/en-us/windows/uwp/audio-video-camera/screen-capture#add-the-screen-capture-capability
Bug: chromium:1314868
Change-Id: I73b823b31a993fd2cd6e007b212826dfe1a80012
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/325521
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#41079}
Bug: webrtc:1314868
Change-Id: Ia743d17d61d7d8ffc44030b5691efef1c7ed7991
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/324305
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#40994}
Without this change a FrameDropped sample will be added to
WebRTC.DesktopCapture.Win.WgcCaptureSessionGetFrameResult at the
current capture rate as long as a captured window is minimized.
Bug: webrtc:1314868
Change-Id: I9b68675486642e7ca25674df689c207ac94a206e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/323882
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#40969}
Makes it more clear that a certain API is only supported in Windows 11.
Bug: webrtc:15451
Change-Id: Ic3abfb2cbf0e30f9cb722ac843876f41279bf200
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/323161
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40931}
Only has an effect on Windows versions higher than 2104 (10.0.20348.0).
Bug: webrtc:15451
Change-Id: I3ca48c88a6c2b9b87d43805fcb2ade444cd90480
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/318060
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40721}
This change fixes a minor issue where we previosuly assumed that the
following was true:
RTC_DCHECK_EQ(map_info.RowPitch, current_frame->stride())
It turns out that this is not always the case when sharing a window
where the stride can sometimes be a few bytes smaller than the
rowpitch.
The code is behind a command-line flag and no tests are affected.
Given limited review resources I therefore plan to bypass the CQ.
I know that it is not recommended but the change has been tested
locally on two different Windows platforms and it does avoid an
existing crash.
Code-Review: alcooper@chromium.org
No-Try: true
Bug: chromium:1421242
Change-Id: I01e7105a6f9fca7ce1349a57635dd373c28d160b
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/309342
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40308}
This CL adds a light-weight detection of "frame content has changed
since last frame" to an existing pass where bytes are copied from a
texture to a DesktopFrame. The resulting boolean can then later be
used to bypass a full detection of if the content is static or not.
As a result, we only check for static content for a small fraction of
all captured WGC frames and this reduces the total load when 0Hz
is enabled for WGC.
Both WGC and 0Hz support for WGC is still behind a flag.
Bug: chromium:1421242
Change-Id: If9e3721c60a244a3919758fe861d56d4b54cb039
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/308821
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40299}
This is a partial revert of the previously landed CL in
https://webrtc-review.googlesource.com/c/src/+/301200.
I noticed when I used `getDisplayMedia` on my private Windows laptop
in combination with WGC that the captured screen was distorted and
did only contain tilted (~35 degrees) lines in all sorts of colors.
The issue only happened for one particular screen resolution
3000 x 2000 and 200% scale. Changing to 100% scale instead resolved
the issue.
I tried many other resolutions but could only trigger for the one
above with 200% scaling.
Next, I bisected and found [1] which led to [2] which contains my own
change in https://webrtc-review.googlesource.com/c/src/+/301200.
The only part that could affect the video frame was the part which
did `CopyPixelsFrom` so I reverted that part and it solved the issue
on my private Windows laptop.
I did not dig deeper into why this particular resolution triggered
the distortion but deiced to revert to avoid more reports.
[1] b4c2a7fcf5..ff848b7a43
[2] a9a2957dbc
Bug: chromium:1428592
Change-Id: I328e77840cd3ca6871254cdf06500bdc616b0c36
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/306600
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#40147}
* Removes a ~60Hz thread-wakeup signal caused by the FrameArrived event
* Initial power measurements shows a reduced power consumption
* No increase in time to first captured packet found
Bug: chromium:1428592
Change-Id: Ia23b5db0c87e70e5b0d6919394494a24d8944493
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/301200
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39861}
Before:
No attempt was made to figure out of the cursor was embedded into the
captured video frame when using DXGI on Windows as screen capturer.
Instead the cursor is superimposed on the frame by an external mouse
and cursor composer.
After:
We now check if the display adapter supports embedding the mouse
cursor and if so use it as is and thereby avoid adding it independently.
Bug: chromium:1421656
Change-Id: Ie07fe13e1c8f9583769961328bb41fbc689cd8e0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/299241
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39742}
Logs if the mouse cursor has moved and if so if it is embedded in the
captured DXGI frame or not.
Bug: chromium:1421656
Change-Id: Ic8fa8a5a3c020ec28b04064c765b1c204accec1c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/298564
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#39644}
* Avoids alloc/dealloc for each captured frame.
* Reduces time to capture first frame.
* Improves performance in terms of max FPS.
Bug: chromium:1412584
Change-Id: Ie16519ad788165c9553451ecea5adff12cd15eea
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/293582
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39384}
I am updating Chrome Remote Desktop to apply a scale factor when using
curtain mode (i.e. a loopback RDP session) and I've found that while
the changes are applied and the desktop is scaled, DXGI stops
producing frames.
This is essentially the same issue as crbug.com/1307357 except this
issue is occurring when the DPI is changed rather than the desktop
size.
The fix is to look at the effective DPI for the source being
captured (or the primary monitor when capturing the full desktop)
and then signaling an environment change when the DPI differs.
Bug: webrtc:14894,b:154733991
Change-Id: Id768d4a384434ba59e7396bc919d0ba30d0f6acc
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/292791
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Joe Downing <joedow@google.com>
Cr-Commit-Position: refs/heads/main@{#39305}
If we are notified of the destruction of the window before a
CaptureFrame call can fail, then we may end up attempting to destroy the
underlying WGC object inside it's own event handler. This can be
problematic, as the class itself may want to run other code. Instead,
we just unsubscribe and signal that any future CaptureFrame calls should
reject.
This also removes setting "is_capture_started_=false" in the item closed
handler, as all that served to do is cause the WgcCapturerWin code to
attempt to restart the capturer, and somewhat muddies up our metrics.
Bug: chromium:1413005
Change-Id: Ibccb7a2e7ce531ba80b4b331b9bc2cda0ff75f4e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/292762
Auto-Submit: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Mark Foltz <mfoltz@chromium.org>
Commit-Queue: Mark Foltz <mfoltz@chromium.org>
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Cr-Commit-Position: refs/heads/main@{#39275}
This records high level errors, or success, encountered across the entire capture flow in the DXGI based capturer.
Using the same style as for WebRTC.DesktopCapture.Win.WgcCapturerResult
Bug: chromium:1400204
Change-Id: I7096d1790d7c2a23bbe29761b7dbf40426ce1e6a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291707
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39259}
This is a follow up to this CL
https://webrtc-review.googlesource.com/c/src/+/269223
which unintentionally prevents the WindowCapturerWinGdi from returning
permanent errors.
Bug: webrtc:14265
Change-Id: I8eb9f8852fb6247a045d32e407ebdd5f45e6aa9d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/271880
Reviewed-by: Mark Foltz <mfoltz@chromium.org>
Commit-Queue: Mark Foltz <mfoltz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#37847}
This overload was removed in C++20.
Bug: chromium:1284275
Change-Id: I67a25ae23fa111e4972d1b207f1c078da13d86a3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/269440
Reviewed-by: Mark Foltz <mfoltz@chromium.org>
Auto-Submit: Peter Kasting <pkasting@chromium.org>
Commit-Queue: Peter Kasting <pkasting@chromium.org>
Cr-Commit-Position: refs/heads/main@{#37630}
Consumers expect the frame to be valid if Result::SUCCESS is delivered.
If the frame is nullptr, we should deliver ERROR_TEMPORARY instead.
Bug: webrtc:14265
Change-Id: If94a3ead38d7657d7b90bbe046256be697312216
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/269223
Commit-Queue: Mark Foltz <mfoltz@chromium.org>
Reviewed-by: Mark Foltz <mfoltz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#37590}
This reverts commit dd32562f242b247aed8add4efecaf3e20c623b9a.
Reason for revert: Updated the original change to dynamically load
the CoreMessaging.dll instead of statically linking with the .lib.
Original change's description:
> Revert "Wait for frames to arrive in WgcCapturer instead of returning nothing."
>
> This reverts commit 93bb3051490253d56dc1cdab4701b91138a151c3.
>
> Reason for revert: It breaks a test while rolling into Chromium,
> see https://webrtc-review.googlesource.com/c/src/+/261780/21#message-4a96e33bfb475f19a618be82bbe72951b23085ef for details.
>
> Original change's description:
> > Wait for frames to arrive in WgcCapturer instead of returning nothing.
> >
> > We're seeing a high instance of "first capture failed" in Chromium when
> > using WGC. We can reduce this by waiting for frames to arrive if there
> > are none in the frame pool instead of returning a temporary error.
> >
> > I've set the maximum time to wait for a frame to 50ms. If no frame
> > arrives before 50ms has elapsed, we will return a temporary error.
> > Added a new test, FirstCaptureSucceeds, to verify that this is working
> > as expected.
> >
> > As part of this I updated the name of the `kCreateFreeThreadedFailed`
> > enum value to `kCreateFramePoolFailed`. The value remains the same
> > since they both report failures in frame pool creation.
> >
> > I also increased `kNumBuffers` from 1 to 2, so that the frame pool can
> > store two frames. This should prevent us from having to wait on the
> > event as frequently. This will increase the latency between capture
> > and display, however. High frame rate applications should not be
> > noticeably affected.
> >
> > Additionally, we uncovered a bug in the OS that prevents window capture
> > when there are displays attached, but none of them are active. Added
> > a new check to `IsWgcSupported` to cover this scenario.
> >
> > Finally, some issues with other WGC tests blocked moving the TryBots
> > to a newer version of Windows. This CL fixes those issues and updates
> > the TryBot configuration.
> >
> > bug: chromium:1314868
> > Change-Id: Id9c4d5ee98621e682ef04864c3848d50e761cdb7
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/261780
> > Reviewed-by: Alexander Cooper <alcooper@chromium.org>
> > Reviewed-by: Jeremy Leconte <jleconte@webrtc.org>
> > Commit-Queue: Austin Orion <auorion@microsoft.com>
> > Reviewed-by: Jeremy Leconte <jleconte@google.com>
> > Cr-Commit-Position: refs/heads/main@{#37404}
>
> Change-Id: If237df4826fe20b6fe2ca4b57253623321bf33c5
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/267460
> Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
> Owners-Override: Mirko Bonadei <mbonadei@webrtc.org>
> Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
> Auto-Submit: Mirko Bonadei <mbonadei@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#37408}
Change-Id: I6cc2becd9ed363782ab2f326f58d9401bc8fb820
Bug: chromium:1314868
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/267902
Reviewed-by: Jeremy Leconte <jleconte@webrtc.org>
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Austin Orion <auorion@microsoft.com>
Cr-Commit-Position: refs/heads/main@{#37470}
This reverts commit 93bb3051490253d56dc1cdab4701b91138a151c3.
Reason for revert: It breaks a test while rolling into Chromium,
see https://webrtc-review.googlesource.com/c/src/+/261780/21#message-4a96e33bfb475f19a618be82bbe72951b23085ef for details.
Original change's description:
> Wait for frames to arrive in WgcCapturer instead of returning nothing.
>
> We're seeing a high instance of "first capture failed" in Chromium when
> using WGC. We can reduce this by waiting for frames to arrive if there
> are none in the frame pool instead of returning a temporary error.
>
> I've set the maximum time to wait for a frame to 50ms. If no frame
> arrives before 50ms has elapsed, we will return a temporary error.
> Added a new test, FirstCaptureSucceeds, to verify that this is working
> as expected.
>
> As part of this I updated the name of the `kCreateFreeThreadedFailed`
> enum value to `kCreateFramePoolFailed`. The value remains the same
> since they both report failures in frame pool creation.
>
> I also increased `kNumBuffers` from 1 to 2, so that the frame pool can
> store two frames. This should prevent us from having to wait on the
> event as frequently. This will increase the latency between capture
> and display, however. High frame rate applications should not be
> noticeably affected.
>
> Additionally, we uncovered a bug in the OS that prevents window capture
> when there are displays attached, but none of them are active. Added
> a new check to `IsWgcSupported` to cover this scenario.
>
> Finally, some issues with other WGC tests blocked moving the TryBots
> to a newer version of Windows. This CL fixes those issues and updates
> the TryBot configuration.
>
> bug: chromium:1314868
> Change-Id: Id9c4d5ee98621e682ef04864c3848d50e761cdb7
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/261780
> Reviewed-by: Alexander Cooper <alcooper@chromium.org>
> Reviewed-by: Jeremy Leconte <jleconte@webrtc.org>
> Commit-Queue: Austin Orion <auorion@microsoft.com>
> Reviewed-by: Jeremy Leconte <jleconte@google.com>
> Cr-Commit-Position: refs/heads/main@{#37404}
Change-Id: If237df4826fe20b6fe2ca4b57253623321bf33c5
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/267460
Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Owners-Override: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Auto-Submit: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37408}
We're seeing a high instance of "first capture failed" in Chromium when
using WGC. We can reduce this by waiting for frames to arrive if there
are none in the frame pool instead of returning a temporary error.
I've set the maximum time to wait for a frame to 50ms. If no frame
arrives before 50ms has elapsed, we will return a temporary error.
Added a new test, FirstCaptureSucceeds, to verify that this is working
as expected.
As part of this I updated the name of the `kCreateFreeThreadedFailed`
enum value to `kCreateFramePoolFailed`. The value remains the same
since they both report failures in frame pool creation.
I also increased `kNumBuffers` from 1 to 2, so that the frame pool can
store two frames. This should prevent us from having to wait on the
event as frequently. This will increase the latency between capture
and display, however. High frame rate applications should not be
noticeably affected.
Additionally, we uncovered a bug in the OS that prevents window capture
when there are displays attached, but none of them are active. Added
a new check to `IsWgcSupported` to cover this scenario.
Finally, some issues with other WGC tests blocked moving the TryBots
to a newer version of Windows. This CL fixes those issues and updates
the TryBot configuration.
bug: chromium:1314868
Change-Id: Id9c4d5ee98621e682ef04864c3848d50e761cdb7
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/261780
Reviewed-by: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Jeremy Leconte <jleconte@webrtc.org>
Commit-Queue: Austin Orion <auorion@microsoft.com>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
Cr-Commit-Position: refs/heads/main@{#37404}
