pablomk7 / citra Goto Github PK

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Android

Description of Issue

En menos de 30 min de juego el teléfono estaba para freír un huevo encima, avisándome hasta el propio teléfono de la alta temperatura.

Expected Behavior

A ver si ahora después de la que habéis liado todos los programadores de 3ds vais a hacer una buena chapuza con nuestros dispositivos. Porque con los otros emuladores puedo jugar horas sin problema

Reproduction Steps

.

Log File

.

System Configuration

Oneplus nord 3 5g

Issue

Hi, I have this texture bug in Citra android, I have already tried changing the rom and emulator and it keeps happening. I'd appreciate your help.

Issue

I think that the emulator telemetry should be removed as it's a privacy problem since the original developers had to stop development. The things that are being really difficult to remove telemetry are the opengl and vulkan renderers. I got some help from one of the suyu commits: https://git.suyu.dev/suyu/suyu/commit/22f8b858a285468030f8f163c5ec31b80efb0ad3

These are the places with telemetry:
src/common/telemetry.cpp and h and from cmakelists.txt
src/core/core.cpp
src/core/core.h
src/core/telemetry_session.cpp and h and from cmakelists.txt
src/core/hle/service/apt/apt.cpp
src/core/loader/ncch.cpp
src/video_core/renderer_opengl/gl_driver.cpp
src/video_core/renderer_vulkan/renderer_vulkan.cpp
src/video_core/renderer_vulkan/vk_instance.h
src/video_core/renderer_opengl/renderer_opengl.cpp
src/video_core/renderer_opengl/gl_texture_runtime.cpp
src/video_core/renderer_opengl/pica_to_gl.h
src/video_core/renderer_vulkan/vk_instance.cpp
src/video_core/renderer_opengl/gl_driver.h
src/citra/citra.cpp
src/webservice/telemetry_json.cpp and h and cmakelists.txt
src/citra_qt/main.cpp
src/citra_qt/compatdb.h and cpp
src/citra_qt/configuration/configure_web.cpp .h and .ui

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Citra 0c2f076

Description of Issue

Citra crashes almost instantly when starting full screen or a few seconds after enlarging the window. Happened with both Smash Bros. 3DS and Pokemon Alpha Sapphire.

Device: Macbook Pro 13 inch M2, 16GB Ram

Expected Behavior

Citra should crash when enlarging the screen are starting full screen mode.

Reproduction Steps

Open Citra, start a game, switch to full screen or enlarge the window.

Log File

citra_log.txt

System Configuration

CPU:M2
GPU/Driver:M2/ Vulkan
RAM:16 GB
OS:MacOS Sonoma 14.4

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

There should be a way to toggle Pretendo in the UI, because Nintendo Network is gone, and there should be a UI setting for Pretendo, or a guide on how to setup Pretendo.

Why would this feature be useful?

This would be useful, because this fork is mainly centered around online play, and with Nintendo Network gone, there would be an easy way for users to play online with Pretendo if there is a guide/UI option for it

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

build from master + latest release
UPDATE: I have checked and it boots properly in r8433057, but the regression occurs for r775ceac onwards

Description of Issue

On an M3 Pro, is_moltenvk in vk_instance.cpp is not set to true in vk_instance.cpp. I'm not sure the why this is occurring, but the line
const bool is_moltenvk = (driver_id == vk::DriverIdKHR::eMoltenvk);
is always evaluating to false.
This is causing perVertexKHR to be generated in SPIR-V and resulting in the following crash:
[mvk-error] SPIR-V to MSL conversion error: PerVertexKHR decoration is not supported in MSL.
libc++abi: terminating due to uncaught exception of type vk::InvalidShaderNVError: vk::Device::createGraphicsPipelineUnique: ErrorInvalidShaderNV
I am unfortunately not familiar enough with C++ to fix the issue.

The crash occurs in the loading screen of Pokemon Omega Ruby after pressing Begin Game (below).

Expected Behavior

is_moltenvk flag should be set to true, so that the incompatible SPIR-V isn't generated.

Reproduction Steps

Run Pokemon Omega Ruby on M3 Mac using given build. I have not tested with other ROMs yet nor other Mac devices.

Log File

System Configuration

CPU: M3 Pro
GPU/Driver: M3 Pro / MoltenVK 1.2.7 (I have also tried building with v1.2.8-rc1 but no success)
RAM: 18GB Unified
OS: macOS Sonoma 14.4.1

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

A way to get be able to check if there are updates and update the app from citra itself would be really nice, some apps like pcsx2 nightly do these.

Why would this feature be useful?

Giving users a way to be notified about updates and not having to re-download citra everytime a new build releases would be neat.

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

N/A

Description of Issue

Vulkan Crash in certain sceens of some games

Expected Behavior

Vulkan not Crash in certain sceens of some games

Reproduction Steps

play some games with vulkan backend

Log File

citra_log.txt

System Configuration

CPU:Ryzen 5
GPU/Driver:Nvidia GeForce RTX 2060
RAM:16GB
OS:Windows 11

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Commit 813d0c2, along with all previous versions of this fork

Description of Issue

Connections that are prematurely interrupted crash the emulator because the broken pipe is not handled appropriately.

Expected Behavior

The emulator should not fail and let the software handle the failed connection.

Reproduction Steps

This issue appears to be Linux-exclusive. Windows handles this properly.
Easiest way is viewing a friend card in Friend List, where:

• the friend is playing a title that's unknown to both Nintendo CDN and Citra.
• the friend is not playing but their favorite title is unknown.
  Expected would be "Unable to obtain title info" or the favorite disappearing.

Log File

The log doesn't contain any useful information. Here's a stack trace from GDB instead:

Thread 15 "EmuThread" received signal SIGPIPE, Broken pipe.

#0  0x00007ffff5f28f0c in __libc_sendto
(fd=78, buf=0x7ffee5f06b80, len=53, flags=0, addr=..., addrlen=0)
at ../sysdeps/unix/sysv/linux/sendto.c:27
#1  0x00005555561cef5b in Service::SOC::SOC_U::SendToImpl
(this=0x5555625cb790, holder=..., len=53, flags=0, addr_len=0, input_buff=std::vector of length 53, capacity 53 = {...}, dest_addr_buff=0x0)
at /data/0/build/citra/src/core/hle/service/soc/soc_u.cpp:1200
#2  0x00005555561cf15a in Service::SOC::SOC_U::SendToSingle (this=0x5555625cb790, ctx=...)
at /data/0/build/citra/src/core/hle/service/soc/soc_u.cpp:1235
#3  0x00005555561ef59b in Service::ServiceFramework<Service::SOC::SOC_U, Kernel::SessionRequestHandler::SessionDataBase>::Invoker
(object=0x5555625cb790, member=(void (Service::ServiceFrameworkBase::*)(class Service::ServiceFrameworkBase * const, class Kernel::HLERequestContext &)) 0x5555561cefc2 <Service::SOC::SOC_U::SendToSingle(Kernel::HLERequestContext&)>, ctx=...)
at /data/0/build/citra/src/./core/hle/service/service.h:181
#4  0x000055555615328b in Service::ServiceFrameworkBase::HandleSyncRequest
(this=0x5555625cb790, context=...)
at /data/0/build/citra/src/core/hle/service/service.cpp:186
#5  0x0000555555f041c0 in Kernel::ServerSession::HandleSyncRequest
(this=0x7fff545d04d0, thread=std::shared_ptr<Kernel::Thread> (use count 6, weak count 1) = {...})
at /data/0/build/citra/src/core/hle/kernel/server_session.cpp:97
#6  0x00005555563adbd9 in Kernel::ClientSession::SendSyncRequest (this=0x7fff545d03c0, thread=std::shared_ptr<Kernel::Thread> (empty) = {...})
at /data/0/build/citra/src/core/hle/kernel/client_session.cpp:56
#7  0x00005555563eb8f0 in Kernel::SVC::SendSyncRequest (this=0x55556237d3c0, handle=196615)
at /data/0/build/citra/src/core/hle/kernel/svc.cpp:629
#8  0x00005555563fcdb7 in Kernel::SVCWrapper<Kernel::SVC>::WrapPass<Result (Kernel::SVC::*)(unsigned int), Result>::Call<unsigned int>
(context=..., svc=(union Result (Kernel::SVC::*)(class Kernel::SVC * const, unsigned int)) 0x5555563eb70e <Kernel::SVC::SendSyncRequest(unsigned int)>)
at /data/0/build/citra/src/./core/hle/kernel/svc_wrapper.h:292
#9  0x00005555563fb383 in Kernel::SVCWrapper<Kernel::SVC>::WrapPass<Result (Kernel::SVC::*)(unsigned int), Result, unsigned int>::Call<>(Kernel::SVC&, Result (Kernel::SVC::*)(unsigned int))
(context=..., svc=(union Result (Kernel::SVC::*)(class Kernel::SVC * const, unsigned int)) 0x5555563eb70e <Kernel::SVC::SendSyncRequest(unsigned int)>)
at /data/0/build/citra/src/./core/hle/kernel/svc_wrapper.h:263
#10 0x00005555563fa18d in Kernel::SVCWrapper<Kernel::SVC>::WrapHelper<Result (Kernel::SVC::*)(unsigned int)>::Call
(context=..., svc=(union Result (Kernel::SVC::*)(class Kernel::SVC * const, unsigned int)) 0x5555563eb70e <Kernel::SVC::SendSyncRequest(unsigned int)>)
at /data/0/build/citra/src/./core/hle/kernel/svc_wrapper.h:309
#11 0x00005555563f88aa in Kernel::SVCWrapper<Kernel::SVC>::Wrap<&Kernel::SVC::SendSyncRequest> (this=0x55556237d3c0)
at /data/0/build/citra/src/./core/hle/kernel/svc_wrapper.h:26
#12 0x00005555563f4fb5 in Kernel::SVC::CallSVC (this=0x55556237d3c0, immediate=50)
at /data/0/build/citra/src/core/hle/kernel/svc.cpp:2284
#13 0x00005555563f51e5 in Kernel::SVCContext::CallSVC (this=0x5555625cad30, immediate=50)
at /data/0/build/citra/src/core/hle/kernel/svc.cpp:2305
#14 0x00005555562a1bb0 in Core::DynarmicUserCallbacks::CallSVC (this=0x5555625cad20, swi=50)
at /data/0/build/citra/src/core/arm/dynarmic/arm_dynarmic.cpp:74

System Configuration

CPU: Intel i5-6400 @ 3.3GHz
GPU/Driver: NVIDIA GeForce GTX 1050 / nvidia-dkms 550.54.14-1
RAM: 16 GiB (8 GiB used at application start)
OS: Arch Linux - rolling (Kernel: 6.7.6-arch1-1)

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

Currently as it stands citra-qt does not support Wayland natively as it gives an error:

qt.qpa.plugin: Could not find the Qt platform plugin "wayland" in ""
This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.

Available platform plugins are: xcb.

Why would this feature be useful?

It allows for better preformance when ran through Wayland directly instead of Xwayland

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Citra Canary 2760/Citra 8433057

Description of Issue

Basically, the game Inazuma Eleven 3 Bomb Blast (including therefore the Lighting Volt and Team Ogre Attacks version) has a number of problems that I have experienced:

1. The introductory game stops at 0 FPS and the audio has serious problems because it is like shutter (I attach the video of the intro).

2. Inside the game, the game goes very low FPS and also if the audio problem does not work well, if you put it at 100% it becomes worse (attached video).

P.S In advance, I apologize for my poor knowledge of English, it is not my primary language and I have tried to explain as best as possible.

Expected Behavior

I have tested with Vulkan as in OpenGL and they go in the same way.
Attached is a screenshot in OpenGL and Vulkan.

OpenGL:

Vulkan:

Reproduction Steps

These bugs appear from the moment you start the game until the game is finished, from what I understand, before Citra was closed, it already had these problems beforehand.

Log File

citra_log.txt

System Configuration

CPU: AMD Ryzen 5 3350G
GPU/Driver: NVIDIA GeForce RTX 3050
RAM: 16 GB
OS: Windows 10 Home

Issue

Will this repo continues the development?

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

It would be useful to replace https://citra-emu.org/wiki/aes-keys/ with for example https://github.com/PabloMK7/citra/tree/master/dist/dumpkeys :)

Why would this feature be useful?

The former link won't work, that's why…

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

775ceac

Description of Issue

Read the title.
Already spoken to @gpucode about this. Just wanted to put this issue here.

Log file produces this.
[ 67.221534] Debug <Critical> ./video_core/renderer_vulkan/pica_to_vk.h:operator():188: Unreachable code! Unknown cull mode 3 [ 67.221534] Debug <Critical> ./video_core/renderer_vulkan/pica_to_vk.h:operator():188: Unreachable code! Unknown cull mode 3

Expected Behavior

Doesn't crash on Vulkan.

Reproduction Steps

Set renderer to Vulkan.
Launch game.

Log File

citra_log.txt.old.zip

System Configuration

Android 13
Snapdragon 680
Adreno 610

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Description of Issue

Z-fighting in Ultra Sun/Ultra Moon

Expected Behavior

No Z-Fighting

Reproduction Steps

Go to the border between route 2 and hau'oli city

Log File

Nothing

System Configuration

CPU: Ryzen 3600
GPU/Driver: GTX 1070
RAM: 16 Go
OS: Ubuntu 24.04

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Since 9ec4954

Description of Issue

The issue comes in multiple layers.

1. BOOST_CLASS_EXPORT_KEY/SERIALIZE_EXPORT_IMPL/SERIALIZE_IMPL macros were not used for this class. These are needed for boost serialization to properly work and actually have serialization methods instantiated. If BOOST_CLASS_EXPORT_KEY/SERIALIZE_EXPORT_IMPL is not used, you will get a crash on attempting serialization. If SERIALIZE_IMPL is not used, you will not get the serialization methods compiled, and thus not actually get any kind of compile time errors from improper serialization code.
2. AsyncWakeUpCallback is a template class, meaning technically multiple different versions of AsyncWakeUpCallback exist, so in order to use BOOST_CLASS_EXPORT_KEY/SERIALIZE_EXPORT_IMPL/SERIALIZE_IMPL each possible kind of instantiated class have to have the macros used on them. However, that is problematic, as the type used in this template class is a lambda, which does not have a defined type in the first place.
3. These lambdas have captures, which would also need to be serialized (after all, in order to deserialize these, you would need to be able to recreate these lambdas from the serialized data, and that includes the captured data). Of course, this isn't actually possible to do with lambdas.

The fix likely would be to rework the RunAsync system to not take in a lambda, but rather take in a functor class implemented for each usage of RunAsync, each with serialization methods implemented.

Expected Behavior

AsyncWakeUpCallback being serialized should work and not crash.

Reproduction Steps

This is a bit hard to reproduce, as it relies on AsyncWakeUpCallback being put into a ThreadCallback at some point and that callback is not reached by the end of a frame, and therefore is still in the waiting_threads list. Pounding the savestate key when the game is constantly reading rom eventually causes the crash.

Log File

N/A

System Configuration

N/A

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Ñ - v2.2

Description of Issue

api.citra-emu.org is still used, despite that domain no longer being online. To my knowledge, the backend was not open-source, though I don't think reimplementing it would be a massive undertaking?

Expected Behavior

Functionality using Citra Web Services is either fixed or removed.

Reproduction Steps

Open the Public Room Browser, or use other functionality.

Log File

N/A

System Configuration

N/A

Issue

If this is going to function as the new repo, that would be a good idea, imo.

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

Restoring the wiki, which contained user as well as developer instructions (including build instructions).

To me there are no known backups of the wiki, but there is archive.org snapshot from minutes before it got nuked, which I would be willing to take on to restore into markdown (if no one happens to have the wiki cloned and laying around somewhere).

The original wiki did also contain dumping information, considering the "incident", it might be better to limit the amount of information restored.

Why would this feature be useful?

Having at least build instructions would be kind of neat

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

0c2f076

Description of Issue

Bug was on original Citra as well. Pokédex 3D Pro's shadows have an additional square above the expected shadow:

Expected Behavior

Shadows shouldn't include a square as there's no square to cast it.

Reproduction Steps

Open game, click Pokédex, issue should immediately appear. May need to move the camera with circle pad to get a better view.

Log File

citra_log.txt

System Configuration

CPU: AMD Ryzen 5 5600
GPU/Driver: AMD Radeon RX 6700
RAM: 32GB
OS: Windows 10

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

I wish there is a flatpak support like other similar software

Why would this feature be useful?

I hope I can run this app on Steam Deck and enjoyed it

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

Remove "shaderSampledImageArrayDynamicIndex" for Raspberry Pi 4/5

Why would this feature be useful?

Raspberry Pi 4/5 support

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

HEAD

Description of Issue

Building Citra with Clang 18.1.1 on msys2 now yields this error:

In file included from C:/msys64/home/Default/github/bravely-offline-citra/src/citra_qt/util/graphics_device_info.cpp:5:
In file included from C:/msys64/home/Default/github/bravely-offline-citra/src/./citra_qt/util/graphics_device_info.h:8:
In file included from C:/msys64/clang64/include/qt6/QtCore/QString:1:
In file included from C:/msys64/clang64/include/qt6/QtCore/qstring.h:14:
In file included from C:/msys64/clang64/include/qt6/QtCore/qchar.h:656:
In file included from C:/msys64/clang64/include/qt6/QtCore/qstringview.h:8:
In file included from C:/msys64/clang64/include/qt6/QtCore/qbytearray.h:14:
C:/msys64/clang64/include/qt6/QtCore/qbytearrayview.h:69:27: error: 'char_traits<unsigned char>' is deprecated: char_traits<T> for T not equal to char, wchar_t, char8_t, char16_t or char32_t i
s non-standard and is provided for a temporary period. It will be removed in LLVM 19, so please migrate off of it. [-Werror,-Wdeprecated-declarations]
   69 |     return qsizetype(std::char_traits<Char>::length(data));
      |                           ^
C:/msys64/clang64/include/qt6/QtCore/qbytearrayview.h:151:39: note: in instantiation of function template specialization 'QtPrivate::lengthHelperPointer<unsigned char>' requested here
  151 |               data, data ? QtPrivate::lengthHelperPointer(data) : 0) {}
      |                                       ^
C:/msys64/home/Default/github/bravely-offline-citra/src/citra_qt/util/graphics_device_info.cpp:25:34: note: in instantiation of function template specialization 'QByteArrayView::QByteArrayView
<const unsigned char *, true>' requested here
   25 |         return QString::fromUtf8(context.functions()->glGetString(GL_RENDERER));
      |                                  ^
C:/msys64/clang64/include/c++/v1/__string/char_traits.h:81:8: note: 'char_traits<unsigned char>' has been explicitly marked deprecated here
   81 | struct _LIBCPP_DEPRECATED_(
      |        ^
C:/msys64/clang64/include/c++/v1/__config:975:53: note: expanded from macro '_LIBCPP_DEPRECATED_'
  975 | #      define _LIBCPP_DEPRECATED_(m) __attribute__((__deprecated__(m)))
      |                                                     ^
1 error generated.
ninja: build stopped: subcommand failed.

And the compiler seems to now be enforcing this deprecation.

I'm not a strong coder, but I was able to get something that compiled with the assistance of Copilot AI, but it breaks Vulkan (so more investigation needed, but I get better performance on OpenGL on this set up anyway so I can live with it).

That said, as far as I can tell, at least these files need to be modified:

src/citra_qt/main.cpp
src/citra_qt/util/graphics_device_info.cpp
src/citra_qt/util/graphics_device_info.h

to rework the QString stuff. I'm not very familiar with the code to know if there are other places as well.

Expected Behavior

Successful compilation.

Reproduction Steps

Build Citra with Clang 18.1.1 (on msys2, but maybe on other OSes as well).

Log File

No logs; this is a compile issue.

System Configuration

CPU: Intel Kaby Lake
GPU/Driver: Integrated Intel HD Graphics 620
RAM: 8GB
OS: Windows 10 64-bit

Issue

Original citra installer doesn't seems to be open source and is not recovered. So why not implement an open source one in this repo

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

Previous upstream versions of Citra supported patching the romfs through the LayeredFS mod feature. It seems this continuation of the project has omitted that feature.

Why would this feature be useful?

Manually repacking dumped 3DS rom files is a pain.

.

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

8433057

Description of Issue

This is an issue that has existed in Citra for years. I verified that it's still happening in this fork, and decided to open an issue ticket for it.

It happens with all 3 graphics backends; Vulkan, OGL and even Software.

Ancient Poni Path:

Route 9:

Expected Behavior

https://youtu.be/oH-1ahAsGxQ?t=18700

https://youtu.be/oH-1ahAsGxQ?t=36818

Reproduction Steps

Simply walk into those areas and check the spots showed in the pictures above.

Log File

citra_log.txt

System Configuration

CPU: i3 2120
GPU/Driver: GIGABYTE's NVIDIA GeForce GT 1030, Driver Version 551.52
RAM: 4 GB DDR3, single channel because reasons.
OS: Windows 11 v22H2 x64, Build Number 22621.2283

pretty sure that patreon link does not work.
also the creating issue interface is different make a different community.

Also maybe consider changing the name a little bit?

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

This affects building rather than a specific build.

Description of Issue

I was trying to build citra-git based off of this PKGBUILD and ran into these issues:

1. /externals/ext-boost points to a commit not in any branch of the repository. This causes a git unload-pack error when trying to git submodule update.
2. After using a hack (local commit to update the targeted commit or something along the lines of that, I don't really have a clear idea of what I'm doing) to get past that, returns a no such file or directory error when trying to configure zycore submodule of dynarmic submodule (I think that's what it's trying to do?)
3. After commenting that entire block out, I get a bunch of CMake errors at externals/CMakeLists.txt because the directories do not contain a CMakeLists.txt file (upon inspection, they seem to be empty)

I'll come back with updates to issues 2 and 3 if/when 1 is fixed (There was a similar issue on the Yuzu mirror, those were fixed and the PKGBUILD built successfully)

Expected Behavior

Package builds (`citra-$pkgver.pkg.tar.zst is created).

Reproduction Steps

PKGBUILD.txt
Attached is the PKGBUILD I'm using.

1. have an Arch Linux (or derivative) distribution ready to build AUR packages (requires base-devel package)
2. download above PKGBUILD, run makepkg -s for 1., then uncomment lines 120-121 for 2., then comment lines 137-140 for 3.

Log File

1. log_1.txt
2. log_2.txt
3. log_3.txt

System Configuration

CPU: Ryzen 7 5800X
GPU/Driver: RX 6700 XT (amdgpu)
RAM: 16 GiB
OS: Artix Linux

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

v2.1, v.2.2

Description of Issue

I noticed that I can open and use this fork on macos and windows, but trying to extract any of these files on linux makes a weird or corrupted extraction. this, as far as i can tell, doesnt happen at all with the normal version of the nondescript fruit-based emulation program. i've tried multiple archive management programs with the same result.

Example 1:

opening windows release zip file on linux leads to weird messed up directory structure. if fixed manually the files themselves are fine, but the files instead are all in one folder with their relative paths as filenames. as you can see this doesn't happen on 7-zip running via wine.

Example 2:

Linux appimage build extracted with windows 7zip (left) works as expected. Appimage extracted with linux archive manager (right) has corrupted appimage files that do not open.

Expected Behavior

zip files extract properly regardless of operating system

Reproduction Steps

1. open and extract release archive for windows or linux on linux
2. theyre messed up

Log File

N/A

System Configuration

CPU:N/A
GPU/Driver:N/A
RAM:N/A
OS: Arch Linux x86_64

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Latest release

Description of Issue

When playing, the emulator has a small memory leak (about 1MB per second), which becomes an issue when in long sessions (for example, streaming). Tested in Pokemon Y 1.0 and Pokemon X 1.5. Latest release, latest Citra and latest Lime release build all present this issue.

Attached log uses vulkan renderer, though the issue is still present when using OpenGL. This is happening on Linux. I haven't tested on Windows. Log goes up to 1GB before closing the emu.

Expected Behavior

Emu's RAM not consuming my entire PC.

Reproduction Steps

Literally just opening a game and letting it run without doing anything else.

Though in the attached log, I play a battle, then run around for a bit, then I just sit idle.

Log File

citra_log.txt

System Configuration

CPU: AMD Ryzen 5 7600 (12) @ 5.170GHz
GPU/Driver: AMD 7600XT
RAM: 32GB DDR5
OS: Arch Linux

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

r8433057

Description of Issue

In Shin Megami Tensei IV (and its midquel Shin Megami Tensei IV Apocalypse), there's a strange issue involving transparency where certain sprites acquire a solid black background, this is easy to notice when you use a powerful offensive skill to kill demons that you find inside the game.

I remember this issue being fixed in the mainline Citra builds, but I checked things out earlier in the last nightly build of Citra, and it looks like the fix was reverted for some reason. That is affecting the builds of Citra hosted here as well, which is why I'm opening this Issue ticket.

Expected Behavior

https://youtu.be/1Qjb0iR2hGk?list=PLVe6uTyOdZXCegbuN5pZ-YxACxGbMnX2d&t=2243

Reproduction Steps

Simply go into some battle and use a flashy skill to kill enemies.

Log File

citra_log.txt

System Configuration

CPU: i3 2120
GPU/Driver: GIGABYTE's NVIDIA GeForce GT 1030, Driver Version 551.52
RAM: 4 GB DDR3, single channel because reasons.
OS: Windows 11 v22H2 x64, Build Number 22621.2283

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

ac792f7

Description of Issue

Vulkan

Expected Behavior

OpenGL

Reproduction Steps

Start a new game I guess

Log File

citra_log.txt

System Configuration

CPU:
GPU/Driver:
RAM:
OS:

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Github

Description of Issue

Why is the readme and update name/info a simple ñ?

Expected Behavior

An actual readme and info on updates.

Reproduction Steps

Open new citra github repository.
Optional open update info

Log File

N/A

System Configuration

N/A

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

8433057

Description of Issue

Running any version of Pokémon Bank results in a fatal error, the game is unable to boot.
Tested on two devices, running two different operating systems.

Was attempting to run it, to see if it'd function with the changes in this fork.

Expected Behavior

The application should, at least, boot to the menu.

Reproduction Steps

Run Pokémon Bank

Log File

citra_log.txt

System Configuration

CPU: Ryzen 7 5800X3D
GPU/Driver: NVIDIA RTX 2060
RAM: 32GB / 3600MHz
OS: Windows 11

CPU: Apple M2 Pro
GPU/Driver: Apple GPU (MoltenVK)
RAM: 16GB
OS: MacOS 14.0 Ventura

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

r8433057 aka latest available on the release page

Description of Issue

After arriving from a quest all the textures will get screwed, stuff like shadows, certain npcs and certain 3d objects will become glitchy, this gets fixed after going into a quest or just by changing the render resolution from one resolution to another.

Expected Behavior

Returning from a quest shouldn't cause textures to blow up

Reproduction Steps

1. Boot the game
2. play the game and return from quests (by using sub-quest or completing them) until you notice your shadows gets glitchy.

Log File

citra_log.txt

System Configuration

CPU: Ryzen 5 5600
GPU/Driver: Rx 6600 (Mesa 24.0.3-arch1.2)
RAM: 16 gbs
OS: EndeavourOS

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

ñightly

Description of Issue

ño ñ

Expected Behavior

ñ iñ view

Reproduction Steps

ñoñe

Log File

ñoñe

System Configuration

CPU: Ñapple Ñ1
GPU/Driver: Ñapple Ñ1
RAM: Ñeight Ñañobytes
OS: Ñapple Soñoma

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Ñ

Description of Issue

Controllers that implement their D-Pad as four buttons rather than two axes can't map their D-Pad properly on Android, as the emulator only accepts two-axis inputs for D-Pad mapping. This is a known issue from the old repo in the Android version of the emulator that has not been fixed for over three years. Affected controllers include the DualShock 3 and JoyCon Pair. The MMJ fork is able to properly map four-button D-Pads, although I'm not sure if that comes at the cost of losing two-axis D-Pad support.

Expected Behavior

When I try to map the D-Pad inputs with the affected controllers, they should map properly.

Reproduction Steps

1. Connect a DualShock 3, JoyCon Pair, or other affected controller to an Android device running the emulator, either by USB or Bluetooth.
2. Attempt to map the controller's D-Pad to the emulator's D-Pad.

Log File

N/A

System Configuration

Device: Galaxy Tab A8
Controllers: DualShock 3, JoyCon Pair

Issue

externals/soundtouch directs to https://github.com/PabloMK7/citra/blob/master/externals/soundtouch I believe it is trying to get to this link https://github.com/citra-emu/ext-soundtouch.git which is obviously now dead. I didn't see a copy of ext-soundtouch in your repos like ext-libressl.

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

The compilation of shaders in Pokémon X/Y is extremely long, which causes freezes in game.

Why would this feature be useful?

optimize shader compilation with Pokémon X/Y

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

775ceac

Description of Issue

Exactly what is typed on the title.
This has been an issue on Citra for months.
@gpucode I'm putting this issue here. So people don't make it again.

Expected Behavior

Rotating the screen shouldn't crash the app.

Reproduction Steps

Rotate the screen 3 to 10 times from landscape to portrait and portrait to landscape.
It's kinda random when it'll crash.

Log File

A log file is not needed.

System Configuration

Snapdragon 680
Adreno 610
Android 12

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

d9a1157

Description of Issue

When rotating the screen quickly a fair few number of times (with either Vulkan or OpenGLES), the FPS counter will update much quicker than originally intended.

Expected Behavior

The FPS counter should remain at the fixed intended rate, not increase when rotating the screen.

Reproduction Steps

1. Open a game on Citra
2. Check that the "Show FPS" option is enabled in the sidebar -> Overlay Options -> Show FPS
3. Rotate the screen multiple times

Log File

citra_log.txt

System Configuration

Device: Samsung Galaxy S23
CPU: Snapdragon 8 Gen 2
GPU/Driver: Adreno 740
RAM: 8GB
OS: One UI 6.1

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

r8433057

Description of Issue

Whenever Citra is toggled to fullscreen mode, it causes the program to crash.

Expected Behavior

Fullscreen should work just fine and allow the user to play games without crashing

Reproduction Steps

1. Open Citra
2. Select a game
3. Go to View > Fullscreen and toggle fullscreen
4. Program crashes.

Log File

Not sure how to pull the log file from Linux. If anyone knows, please let me know and I can upload a log.

System Configuration

Using 256GB LCD Steam Deck

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

Now that experimental online support has been added, is it possible to allow connecting to Pretendo?

Why would this feature be useful?

This would allow online play after the Nintendo wireless connections close, not come with a ban risk, and be better legally as Pretendo is an open source project. It would also allow playing games with a real 3DS, without forcing you to use Nintendo's servers

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

Source (Windows, MSVC)

Description of Issue

When attempting to view the Friends list from the home menu with Pretendo IPS patches in place, the emulation hangs. This only occurs, however, if both the FRD and HTTP LLE modules are active; leaving only one or the other on fails to generate a friend code but otherwise does nothing. Additionally, without the Pretendo patch for TitleID 0004013000002902 the system throws an error code 001-0803 and refuses to connect.

Although others have been able to connect to Pretendo and create a valid friend code with Citra, this renders it functionally impossible to do so since (among others) the two listed LLE modules are necessary to establish an online connection (according to the archived 2.1 release)

Expected Behavior

Attempting to visit the Friends list with patches in place to connect to Pretendo Network and LLE modules enabled causes emulation to hang at 0FPS until stopped

Reproduction Steps

1. Copy Pretendo patches to 'mods' folder and select Pretendo network using Nimbus
2. Enable LLE patches FRD and HTTP
3. Ensure a valid Mii is present
4. Open Friends list after booting home menu

Log File

citra_log.txt.old.txt

System Configuration

CPU: AMD Ryzen 7 3800X
GPU/Driver: RTX 3060 Ti / 552.12
RAM: 32GB 3200Mhz
OS: Windows 10

Is there an existing issue for this?

• I have searched the existing issues

What feature are you suggesting?

I want to suggest a feature where you'd be able to change the RTC during gameplay. This is how it would look like in action. This is from the DS emulator no$zoomer.

https://streamable.com/40bo2h

Why would this feature be useful?

It would add more fun to gameplay for the users of the emulator.

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

0c2f076 (MSVC, msys2 build doesn't seem to be affected?)

Description of Issue

I have the Polish language added as the 2nd place for "Preferred languages" within language settings in Windows with nothing but the "Basic typing" pack installed (so I could get the keyboard layout), and the English language installed in 1st place and set as the "display language". However, Citra seems to ignore this, and uses the 2nd language (Polish) to display the UI, when in reality it should display English.

Expected Behavior

Citra UI should be in English.

Reproduction Steps

Add another language to Windows within the Settings app and set it as 2nd place under Preferred languages. Citra will display the language in 2nd place instead of the language set as the display language and the language that's in 1st place.

Log File

Log file isn't needed

System Configuration

CPU: AMD Ryzen 7 3700X
GPU/Driver: NVIDIA GeForce RTX 3060
RAM: 32GB DDR4
OS: Windows 11 Pro 23H2 (build 22631.3447)

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

0c2f076

Description of Issue

During dialogue that shows the speakers's 3d models, the game will show the first background that was used in this type of dialogue, since the game launched.
This bug only occurs when Internal Resolution is set to 2x or higher.

Expected Behavior

This is how the background is supposed to look (Internal Resolution is set to 1x)

Reproduction Steps

(1- make sure that Internal Resolution is set to 2x or higher.)
2- initiate any kind of dialogue that shows the speakers's 3d models.
3- initiate another one of this kind of dialogue (make sure it's in a different location). You will notice that the background is the same as the one in step 2.

In this save file, speaking with any of the characters in the room will initiate this kind of dialogue.
0006d400.zip

Log File

citra_log.txt.old.txt

System Configuration

CPU: Exynos 2400
GPU/Driver: Xclipse 940
RAM: 12GB
OS: Android 14

Issue

When I play pokemon alpha saphir with analog joystick mode the character moves tiptoes when I put my joystick to down or right directions. To fix my issue I just edit this line

Is there an existing issue for this?

• I have searched the existing issues

Affected Build(s)

d9a1157

Description of Issue

The audio skips slightly when rotating the screen (small segments get missed).

Expected Behavior

The audio should be consistent when rotating the screen.

Reproduction Steps

1. Open a game and ensure audio playback is working
2. Rotate the screen

Log File

citra_log.txt

System Configuration

Device: Samsung Galaxy S23
CPU: Snapdragon 8 Gen 2
GPU/Driver: Adreno 740
RAM: 8GB
OS: One UI 6.1