If the get_hooked procedure is called during the unhooked signal
the source would still report that it's hooked because gc->capturing
hasn't been updated yet.
In `parse_decl_string`, it jumps to `fail` label if an error occurs.
However, if the lexer encountered an error for example, `cfp.error_list`
may be empty.
The blake2_static target has no specific FOLDER property set and thus
would appear ungrouped in Visual Studio's Solution Explorer.
With this change both the default blake2 target as well as
blake2_static appear in their own "blake2" subdirectory.
When a unique path is set up as the location for user settings,
profiles, or scene collections, migrating the files from one computer
to another will lead to confusing error messages as the original paths
might not exist on the new machine.
The default directories for configuration files are already created by
this point, so using those paths as a fallback should enable OBS Studio
to also create the corresponding fallback settings files.
The frontend has set a log handler which locks the mutex in `LogString`.
If `os_dlopen` fails in the child process, it calls `blog` and attempted
to lock the mutex that may have already been locked by another thread
before `fork()`. This change prevents the child process from locking the
mutex, resolving the potential deadlock.
AMD VCN uses a different surface alignment compared to other GPUs for
encoding, which required an upstream fix in Mesa, libva, FFmpeg, and
the VCN firmware.
This commit adds the FFmpeg patches to the flatpak build of OBS which
is currently using FFmpeg version n7.1, however the patches were applied
upstream after that release. libva API version 1.21 or higher is also
required, which is satisfied by using KDE runtime 6.8.
The incorrect 1920x1088 HEVC output resolution (instead of 1920x1080)
resulted in streaming disconnects for Twitch Enhanced Broadcsating
Linux users. Local file recording when using AMD VAAPI HEVC also
outputs 1088 instead of 1080. This commit fixes both cases.
Please refer to [1], [2], and [3] for context. The patches are taken
from [2].
Note that the fix is applicable to flatpak builds only. Standalone and
Debian package builds depend on the host library versions and will
still have the problem until a version of FFmpeg with the fix is used.
[1]: https://patchwork.ffmpeg.org/project/ffmpeg/list/?series=11222
[2]: https://patchwork.ffmpeg.org/project/ffmpeg/list/?series=13168
[3]: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10985
Log the codec level being used by AMF-based encoders after
ffmpeg_opts have been parsed. Users could have manually changed the
codec level so query the level via AMF then determine the string for
logging.
The default "level" setting was being used for each codec (AVC, HEVC,
AV1) supported by AMF. For example, all HEVC encoders were using
level 6.2 and this caused some playback devices to reject the
bitstream for decode because the device reported a maximum decode
level lower than 6.2.
Add functionality to determine the best match for the codec level
instead of relying on the defaults.
The NonCheckableButton class was previously used for situations where a
*checkable* button was meant to not be checkable by the *user*, but only
through code. This was useful mostly as a styling tool (and to confuse
developers like me as to what it actually did).
The disadvantage is that such a button - a button that is actually
checkable (has the checkable attribute) but ignores the user - still
gets recognized by QAccessible as a checkable button, which has
accessibility issues when used like a normal button (see QTBUG-110737).
We can still get the styling effect on *actually* uncheckable buttons by
giving them a class, so this widget should not be necessary.
Checkable buttons are troublesome because QAccessible interprets them as
checkboxes, and doesn't pass the "clicked" signal to them (see
QTBUG-110737). Instead, we only get the "toggled" signal, which also
gets triggered by things other than user input (i.e., the button getting
changed by the program).
Making them uncheckable means that they behave like normal buttons
again. To style them, we can just add a class and address them that way.
Note that uncheckable here means *actually* not checkable - unlike the
cursed NonCheckableButton class previously used that makes a button
that *is* checkable ignore the user input (and only allow check state
changes via code, effectively making it solely a styling tool).
Add function that tries to reset v4l2loopback output for module versions
from 0.12.5 to 0.12.7. If successful, then set flag that STREAMON and
STREAMOFF are necessary each time the device is opened/closed.
If a frame has a width or height of zero, this value will make it into
libobs/media-io/video-frame.c:video_frame_init and cause linesizes or
heights to be zero, which will result in a bmalloc(0) call and OBS will
crash.
Instead of letting the call stack get that far, check the frame width
and height here at the source, log an error, and return early if the
frame width or height are zero.
Vulkan Direct Display uses extensions:
VK_KHR_display/VK_EXT_direct_mode_display
This workflow allows creating swapchains that arn't active on the
desktop, so they don't have HWNDs.
Avoid trying to create swap_data when a HWND can't be found.
Fixes#11581
Certain draw calls were creating/destroying vertex buffers. Every call.
That's kind of not a great thing to do, so instead use the new
gs_draw_quadf() function to optimize rendering and reduce the need for
swapping vertex buffers.
Also uses a shader for DrawStripedLine so it does not have to split it
up into separate draw calls.
Sprites were unnecessarily updating a dynamic vertex buffer every single
draw call. This creates static vertex buffers for common vertex buffer
dimensions/uvs and only uses a dynamic vertex buffer when the texture
coordinates actually change.
Porbably not a significant perf change considering the very few number
of traingles we draw per frame but still nice to do.
Optimizes certain 4x4 matrix functions by minimizing unnecessary math
operations. This ends up optimizing gs_matrix_transform() and
gs_matrix_rotate() functions and makes positional/rotational operations
a bit more optimal and requiring much less math than they would
otherwise.
If the crop values combined are larger than the width or height of the
source, an integer overflow will occur.
This fix converts the width/height values to int, and then clamps
any negative values to 0.
- Update ingest URLs to restore streaming functionality
- Consolidate free and premium member entries into a single entry
- Add more_info_link and stream_key_link
Without this setting many multi-line macro definitions in the code-base
would have been reformatted. As many of those definitions are formatted
in a bespoke way to ensure legibility, changing their format would be
detrimental to that specific goal.
It appears that in conjunction with other changes that happened around
the time the removal of the transparent border got merged, it became
necessary again due to those other changes. Without it, only the
selected and focussed changes have a border, making the content jump
around when selected and focussed.
Reverts part of 5fa4ea44d0b364bf9be18816c2d9b35b96bab939.
