This is effectively unused, as the window handles that we pass to the
GLInterface are window handles for the frame which isn't ever a real
toplevel window. Host_UpdateTitle is what actually sets the proper title
on the render window.
Now that MainNoGUI is properly architected and GLX doesn't need to
sometimes craft its own windows sometimes which we have to thread back
into MainNoGUI, we don't need to thread the window handle that GLX
creates at all.
This removes the reference to pass back here, and the g_pWindowHandle
always be the same as the window returned by Host_GetRenderHandle().
A future cleanup could remove g_pWindowHandle entirely.
We now have two cases: the GLX window is parented into a frame, or it's
parented into the MainNoGUI host. In both cases, the GLX window should
be locked to the size of the parent, so just sync it up based on that.
Our existing code was relying on the GLX backend to create the GLX
window properly, and for the rest of the code to patch that up, sort
of. If we rely on Host_GetRenderHandle() returning a valid window, we
can do a lot better about this.
Create a simple window inside MainNoGUI to make this happen.
Move to one display. There's no reason to have two displays here -- the
comment stated that one should touch GLX and one should touch window
events, and that they should be touched from different threads, but the
current code wasn't this careful.
Just use one Display connection.
Now, the only supported EGL platform is Android. We might eventually add
back support for EGL/X11 or EGL/Wayland, but it will have to be
architected differently.
Yes, this is a fancy new feature, but our Wayland support was
particularly bitrotten, and ideally this would be handled by a platform
layer like SDL. If not, we can always add this back in when GLInterface
has caught up. We might be able to even support wxWidgets and GL
together with subsurfaces!
The JIT block compare code didn't set the same options for the PPCAnalyzer
as the actual JIT did, which made the PPC side of the JIT block viewer stop
at the first branch instead of the end of the block.
This class loads all the common PP shader configuration options and passes those options through to a inherited class that OpenGL or D3D will have.
Makes it so all the common code for PP shaders is in VideoCommon instead of duplicating the code across each backend.
Prior to this after painting the hex values, it would increment the curAddress by 32. This is not only a bug, but unnecessary, since the OnMouseDownL and OnScrollWheel functions should be the only things to handle address incrementing for scrolling purposes.
Just use the Windows API to accomplish the same thing (this is what is done in Update()).
This makes the backing window handle the correct data-type for Windows for easier use in function calls.
This was actually never used as far as I can tell. There was no wx event handling done whatsoever for the global ID, So this is basically a dead function.
Restructure our build system so we have multiple targets. Right now we
only build MainNoGUI if we are using X11, since that's the only truly
supported backend: the OS X code actually doesn't compile, according to
comments made on IRC.
We want to have two build targets: Main and MainNoGUI, and this code
will be linked against in both cases, so the ifdef isn't enough for
this case.
Just append to a vector of strings, and then convert it after the fact.
This was used to send fake button or key events to every backend to
implement freelook. When I ported these to be based on wx events in
commit 5248211, I didn't realize that these were now unused, so I didn't
remove them. Do that now.
This moves the Gekko disassembler to Common where it should be. Having it in the Bochs disassembly Externals is incorrect.
Unlike the PowerPC disassembler prior however, this one is updated to have an API that is more fitting for C++. e.g. Not needing to specify a string buffer and size. It does all of this under the hood.
This modifies all the DebuggingInterfaces as necessary to handle this.
Fixes a bug where "Use Fullscreen" would initialize into exclusive fullscreen regardless of the borderless fullscreen setting.
Also relieves the need for the video renderer to check the borderless fullscreen setting each time.
This can happen during shutdown.
As long as we don't call event.Skip() the CFrame won't be closed yet, so even if we can't veto the shutdown will still happen in the correct order.
It was only used for Windows XP and lower.
This also bumps the _WIN32_WINNT define in the stdafx precompiled headers to set the minimum version as Windows Vista.
The hack was needed because the Nvidia 3D Vision heuristics are documented to only support surfaces that are the same size as the backbuffer. This would be the case if you enabled the hack and selected the "Auto (Window Size)" internal resolution.
However, on recent drivers the same effect is achieved by selecting the "Auto (Multiple)" internal resolution. Therefore the hack is no longer required.
Also have the renderer remember its own fullscreen state. This is done to prevent a case where we exit exclusive fullscreen through the configuration and a focus shift at the same time. In this case the renderer would fail to detect that the fullscreen state was changed.
This ensures the transition from/to exclusive mode happens while the RenderFrame is fullscreen.
This prevents fullscreen loops and relieves us of having to restore the window size after we exit fullscreen.
It's not used outside of the class, and it also shouldn't be modified
outside of it either (considering it holds all the blr instructions inserted by the user).
We can't use RendererHasFocus for this purpose because of some issues
with exclusive fullscreen, and the new RendererHasFocus implementation
didn't work for non-Render to Main Window cases, since the renderer
window wasn't managed by wx.
This reverts commit ff918df889.
This changed it from "RendererHasFocus" to "UIHasFocus", which is
wrong. Specifically, it broke for non-Render to Main Window cases where
the renderer window isn't managed by wx. It also broke the pending
exclusive fullscreen support, which checks this function to determine if
the renderer is on top so it can full-screen it.
We'll add a new hook, "UIHasFocus", in the next commit.
- Gets rid of the underscore in some function names
- Appends the 'm_' prefix to class member variables.
Now it isn't a pain to see what is being modified in terms of state. Especially in the OnPaint call.
The logic of RendererHasFocus was extremely convoluted beforehand and
was too brittle and broke in certain circumstances, like when the
GCPad window was active. Simplify it and leave a comment explaining the
new logic. This fixes the GCPad window not working when Background Input
is unchecked.
Lessens the restrictions on the searching in the code view.
Now typing out the full 8 digit hex number isn't needed. For example, you don't need to type 000000FF to go to FF, you just literally type FF.
Also makes JumpToAddress a boolean function to remain consistent with the DSP code view.
This will also change the address search box to have a red background if
either an invalid hex number is given, or if it's longer than 8 characters
Jasper St. Pierre