On Raspberry Pi 3 via the VC4 driver in firmware KMS mode, none of the
found configs match the desired format, causing the function to fall through
without any config being selected.
Fix by first iterating over the found configs, and if no match exists,
don't exclude the non-matching configs. This should fix RPI3 and possibly other
targets without breaking targets that have a matching native format (such as RPI4).
This allows you to bind surfaceless contexts on a background thread to, for
example, load assets in a separate context, for platforms that have different
requirements about sharing surfaces, etc.
Martin's notes on the matter:
"Here's a patch that enables passing NULL windows to SDL_GL_MakeCurrent, if
the involved APIs allow it. Currently, this is only the case for EGL, and
even then only if some specific extensions are present (which they usually
are).
If "surfaceless" contexts are not supported, SDL_GL_MakeCurrent continues to
generate an error (albeit with a more specific error message than it used to),
so this should not break anything that wasn't broken before."
(Please see https://bugzilla.libsdl.org/show_bug.cgi?id=3695 for more
discussion.)
Fixes Bugzilla #3695.
This behavior matches SDL_RecreateWindow and makes it less likely that
another piece of code (e.g. a DestroyWindowFramebuffer implementation)
will attempt to use or free the stale surface pointer.
From hmk:
"When scaling is enabled (e.g. via SDL_RenderSetLogicalSize, size not equal
to window size), mouse motion events are also scaled. Small motions are
rounded up (SDL_max() when the value after scaling is less than 1), while
larger motions are truncated by the floating point -> integer conversion.
https://hg.libsdl.org/SDL/file/b18197f9bf9d/src/render/SDL_render.c#l658
The end result feels something like mouse reverse mouse acceleration + angle
snapping at low speeds, but less consistent (amount of truncation & rounding
depends on how fast the mouse is moved) and potentially much worse if the
scaling factor is large. This pretty much makes it useless for anything
where you need precise mouse aiming (think of games). I suspect this is why
aiming gets so terrible in some games that let you use scaling to reduce the
render resolution (e.g. Ion Fury).
With 4x4 scaling, I can reproduce a situation where it takes three fast flicks
of the mouse across the pad to undo one slow sweep across the pad. In other
words, extreme reverse acceleration. This does not happen when scaling is
disabled.
Furthermore, any game that uses relative mouse motion events for 3D camera
rotation probably wants the raw mouse deltas and not a value that depends on
scaling and resolution and rounding and truncation. Ideal camera rotation
just takes mouse input, multiplies it by sensitivity, and adds it to the
angle-in-radians or whatever measure is used for yaw & pitch. Pixels and
screen resolution or window dimensions should not be a part of the equation
at all, even if it could be implemented without rounding errors.
[...]
This [patch] completely eliminates angle snapping for me, and makes
sensitivity consistent. In other words, it's completely usable for, say,
aiming in a first person shooter."
Partially fixes Bugzilla #4811.
Caleb Cornett's comments:
"A few weeks ago, Alex added a partial Metal API to SDL2:
https://hg.libsdl.org/SDL/rev/22c8e7cd8d38
I noticed it was missing a few features that would help Metal become a
first-class citizen in SDL, so I went ahead and wrote them! Here are the new
APIs:
1. SDL_WINDOW_METAL flag for SDL_CreateWindow(). This allows the programmer
to specify that they intend to create a window for use with SDL_MetalView.
The flag is used to ensure correct usage of the API and to prevent
accidentally defaulting to OpenGL on iOS.
2. SDL_Metal_GetLayer(). This function takes a SDL_MetalView and returns a
pointer to the view's backing CAMetalLayer. This simplifies things
considerably, since in the current version of the SDL_Metal API the
programmer is required to bridge-cast a SDL_MetalView handle to an NSView or
UIView (depending on the platform) and then extract the layer from there.
SDL_Metal_GetLayer automatically handles all of that, making the operation
simple and cross-platform.
3. SDL_Metal_GetDrawableSize(). This function already exists in the current
SDL_Metal API (and is used behind-the-scenes for SDL_Vulkan_GetDrawableSize
on Apple platforms) but was not publicly exposed. My patch exposes this
function for public use. It works just like you'd expect.
Tested on macOS 10.14 and iOS 12.4."
Fixes Bugzilla #4796.
This does not account for scrollbars nor margins. But is much better then returning the full display size when not running fullscreen, but for example in an iframe.
The first terminator is for input parameters. The second terminator was for the
output parameters.
If an error occurs when calling MakeThreadHighPriority(), e.g. a bad thread id,
then the reply from connection_send_with_reply_and_block() will be null.
Anthony Pesch's notes on his patch:
"Currently, the WASAPI backend creates a stream in shared mode and sets the
device's callback size to be half of the shared stream's total buffer size.
This works, but doesn't coordinate will with the actual hardware. The hardware
will raise an interrupt after every period which in turn will signal the
object being waited on inside of WaitDevice. From my empirical testing, the
callback size was often larger than the period size and not a multiple of it,
which resulted in poor latency when trying to time an application based on the
audio callback. The reason for this looked something like:
* The device's callback would be called and and the audio buffer was filled.
* WaitDevice would be called.
* The hardware would raise an interrupt after one period.
* WaitDevice would resume, see that a a full callback had not been played and
then wait again.
* The hardware would raise an interrupt after another period.
* WaitDevice would resume, see that a full callback + some extra amount had
been played and then it would again call our callback and this process would
repeat.
The effect of this is that the pacing between subsequent callbacks is poor -
sometimes it's called very quickly, sometimes it's called very late.
By matching the callback's size to the stream's period size, the pacing of
calls to the user callback is improved substantially. I didn't write an actual
test for this, but my use case for this was my Dreamcast emulator
(https://redream.io) which uses the audio callback to help drive the emulation
speed. Without this change and with the default shared stream buffer (which
has a period of ~10ms) I would get frame times that were between ~3-30
milliseconds; after this change I get frame times of ~11-22 milliseconds.
Note, this patch also has a change that removes passing a duration to the
Initialize call. It seems that the default duration used (when 0 is passed)
does typically match up with the duration returned by GetDevicePeriod, however
the Initialize docs say:
> To set the buffer to the minimum size required by the engine thread, the
> client should call Initialize with the hnsBufferDuration parameter set to 0.
> Following the Initialize call, the client can get the size of the resulting
> buffer by calling IAudioClient::GetBufferSize.
This change isn't strictly required, but I made it to hopefully rule out
another source of unexpected latency."
Fixes Bugzilla #4592.
If called from background threads, use Grand Central Dispatch to use the
main thread instead. On the main thread, just call them directly.
Fixes Bugzilla #4932.
The warnings were produced by GCC 9.2.x for x86_64-linux-gnu or
i386-pc-msdosdjgpp targets.
Most of the fixes involve changing the type of a variable rather than
the format specifier. For many of the affected test conuter variables,
a basic int seems sufficient.
Some format specifier warnings still remain for cases where changing
type or casting seemed inappropriate. Those warnings will probably
require some new format specifier macros (e.g. SDL_PRIu32).
