mainly doing this to reduce memory footprint; we all know how nice ankerl::unordered_dense is
in theory 4x faster - in practice these maps arent that "hot" anyways so not likely to have much perf gained
i just want to reduce mem fragmentation to ease my porting process, plus it helps other platforms as well (ahem weak Mediatek devices) :)
Signed-off-by: lizzie <lizzie@eden-emu.dev>
Co-authored-by: Caio Oliveira <caiooliveirafarias0@gmail.com>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3442
Reviewed-by: DraVee <dravee@eden-emu.dev>
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Co-authored-by: lizzie <lizzie@eden-emu.dev>
Co-committed-by: lizzie <lizzie@eden-emu.dev>
basically a check runs that depends on the instance being created, but instance isnt created yet so check fails
Signed-off-by: lizzie <lizzie@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3134
Reviewed-by: Caio Oliveira <caiooliveirafarias0@gmail.com>
Reviewed-by: crueter <crueter@eden-emu.dev>
Co-authored-by: lizzie <lizzie@eden-emu.dev>
Co-committed-by: lizzie <lizzie@eden-emu.dev>
Some genius decided to put the entire MainWindow class into main.h and
main.cpp, which is not only horrific practice but also completely
destroys clangd beyond repair. Please, just don't do this.
(this will probably merge conflict to hell and back)
Also, fixes a bunch of issues with Ryujinx save data link:
- Paths with spaces would cause mklink to fail
- Add support for portable directories
- Symlink detection was incorrect sometimes(????)
- Some other stuff I'm forgetting
Furthermore, when selecting "From Eden" and attempting to save in Ryujinx, Ryujinx would destroy the link for... some reason? So to get around this we just copy the Eden data to Ryujinx then treat it like a "From Ryujinx" op
Signed-off-by: crueter <crueter@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2929
Reviewed-by: Lizzie <lizzie@eden-emu.dev>
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
This adds an action to the Game List context menu that lets users link
save data from Eden to Ryujinx, or vice versa.
Unfortunately, this isn't so simple to deal with due to the way Ryujinx's saves work. Ryujinx stores its saves in the... config directory... in `bis/user/save`. Unlike Yuzu, however, it doesn't store things by TitleID, instead it's just a bunch of directories from 000...01 to 000...0f and so on. The way it *maps* TitleID to SaveID is via `imkvdb.arc` in `bis/system/save/8000000000000000/0/` and also an identical copy in the `1` directory for... some reason. `imkvdb.arc` is handled by `FlatMapKeyValueStore` in LibHac, which, as the name implies, is a key-value storage system that `imkvdb.arc`, and seemingly `imkvdb.arc` alone, uses. The way this class is written is really weird, almost as if it's designed to accommodate more types of kvdbs... but for now we can safely assume that there aren't gonna be any other `kvdb` implementations added to HorizonNX.
Regardless, the file format is ridiculously simple so I didn't actually need to do a deep dive into C# code... of which I can basically only read Avalonia. A simple `xxd` on the `imkvdb.arc` is all that's needed, and here's everything that matters:
- The `IMKV` magic header (4 bytes)
- 8 bytes that don't really have anything useful to us, except for a size byte (presumably a `u32`) strewn at offset `0x08` from the start of the file, which is useless to us
- Then we start the `IMEN` list. I don't know what the `IM` stands for, but `IMEN` is just, well, an ENtry. Offsets shown are relative to the start of the `IMEN` header.
* 4-byte `IMEN` magic header at 0x0
* 8 bytes of filler data. It contains two `0x40` bytes, but I'm not really sure what they do
* TitleID (u64) at `0xC`, for example `00a0 df10 501f 0001` for Legends: Arceus (the byte order is swapped)
* 0x38 bytes of filler starting at offset 0x14
* SaveID (u64) at `0x4C`, for example `0a00 0000 0000 0000` for my Legends: Arceus save
* 0x38 bytes of filler starting at offset 0x54
Full example for Legends: Arceus:
```
000001b0: 494d 454e 4000 0000 4000 0000 00a0 df10 IMEN@...@.......
000001c0: 501f 0001 0100 0000 0000 0000 0000 0000 P...............
000001d0: 0000 0000 0000 0000 0000 0000 0100 0000 ................
000001e0: 0000 0000 0000 0000 0000 0000 0000 0000 ................
000001f0: 0000 0000 0000 0000 0000 0000 0a00 0000 ................
00000200: 0000 0000 0000 0000 0000 0000 0000 0000 ................
00000210: 0000 0000 0100 0000 0000 0000 0000 0000 ................
00000220: 0000 0000 0000 0000 0000 0000 0000 0000 ................
00000230: 0000 0000 0000 0000 0000 0000 494d 454e ............IMEN
```
Ultimately, the size of the `IMEN` sits at 0x8C or 140 bytes. With this knowledge reading all the TitleID -> SaveID pairs is basically free, and outside of validation and stuff is like 15 lines of relevant code. Some interesting caveats, though:
- There are two entries for some TitleIDs for... some reason? Ignoring the second one seems to work though.
- Within each save directory, there are directories `0` and `1`... and only `0` ever seems used??? It's where Ryujinx points you to for save, so I just chose to use that.
Once everything is parsed, the rest of the implementation is extremely trivial:
- When the user requests a Ryujinx link, match the current program_id to the corresponding SaveID in `imkvdb`
- If it doesn't exist, just error out (save data is probably nonexistent)
- If it does though, give the user the option to use Eden's current save data OR Ryujinx's current save data.
Old save data is deleted depending on which one you chose.
Signed-off-by: crueter <crueter@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2815
Reviewed-by: Lizzie <lizzie@eden-emu.dev>
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Previously, if the user had their NAND in a nonstandard location,
profiles.dat would be read from the standard Eden path and thus return
effectively garbage data. What this would result in is:
- The Qt profile manager would be completely nonfunctional
- "Open Save Data Location" would put you into the completely wrong
place
- Games would read from incorrect locations for their saves
To solve this, I made it so that profiles.dat is re-read *after*
QtConfig initializes. It's not the perfect solution, but it works.
Additionally, this adds an orphaned profiles finder:
- walks through the save folders in nand/user/save/000.../
- for each subdirectory, checks to see if profiles.dat contains a
corresponding UUID
- If not, the profile is "orphaned". It may contain legit save data, so
let the user decide how to handle it (famous last words)
- Empty profiles are just removed. If they really matter, they're
instantly recreated anyways.
The orphaned profiles check runs right *after* the decryption keys
check, but before the game list ever gets populated
Signed-off-by: crueter <crueter@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2678
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Compilation and CMake fixes for both Windows on ARM and clang-cl, meaning Windows can now be built on both MSVC and clang on both amd64 and aarch64.
Compiling on clang is *dramatically* faster so this should be useful for CI.
Co-authored-by: crueter <crueter@eden-emu.dev>
Co-authored-by: crueter <crueter@crueter.xyz>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/348
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Reviewed-by: crueter <crueter@eden-emu.dev>
Co-authored-by: lizzie <lizzie@eden-emu.dev>
Co-committed-by: lizzie <lizzie@eden-emu.dev>
Completely replaces vcpkg with CPM for all "system" dependencies. Primarily needed for Android and Windows. Also uses my OpenSSL CI for those two platforms.
In theory, improves configure and build time by a LOT and makes things much easier to manage
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/250
Reviewed-by: Lizzie <lizzie@eden-emu.dev>
Includes citron, sudachi, yuzu
currently broken, because the eden dir is always made early?
Signed-off-by: swurl <swurl@swurl.xyz>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/91
Co-authored-by: swurl <swurl@swurl.xyz>
Co-committed-by: swurl <swurl@swurl.xyz>
Resolves a case on Windows where an unmounted bitlocker protected volume containing an assigned game directory would crash Yuzu at start.
May also resolve cases where a disconnected SMB volume causes similar crashes (needs testing)
This also covers std::span, which does not have a const iterator.
Also renames IsSTLContainer to IsContiguousContainer to explicitly convey its semantics.
Clang (rightfully) warns that we are checking for the existence of
pointer to something just allocated on the stack, which is always true.
Instead, check whether GetModuleFileNameW failed.
Co-authored-by: Mai M <mathew1800@gmail.com>
This formats all copyright comments according to SPDX formatting guidelines.
Additionally, this resolves the remaining GPLv2 only licensed files by relicensing them to GPLv2.0-or-later.
These functions allow to construct a string view from an input buffer, avoiding the copy done by the non string view counterparts. However, callers must be cognizant of the viewed buffer's lifetime to avoid a use-after-free.
To keep the TAS inputs synced to the game speed even through lag spikes and loading zones, deeper access is required.
First, the `TAS::UpdateThread` has to be executed exactly once per frame. This is done by connecting it to the service method the game calls to pass parameters to the GPU: `Service::VI::QueueBuffer`.
Second, the loading time of new subareas and/or kingdoms (SMO) can vary. To counteract that, the `CPU_BOOST_MODE` can be detected: In the `APM`-interface, the call to enabling/disabling the boost mode can be caught and forwarded to the TASing system, which can pause the script execution if neccessary and enabled in the settings.
First of all, TASing requires a script to play back. The user can select the parent directory at `System -> Filesystem`, next to an option to pause TAS during loads: This requires a "hacky" setup deeper in the code and will be added in the last commit.
Also, Hotkeys are being introduced: CTRL+F5 for playback start/stop, CTRL+F6 for re-reading the script and CTRL+F7 for recording a new script.
It became apparent that logging can continuously spam errors that trigger file flushing.
Since committing the files to disk is an expensive operation, this causes unnecessarily high disk usage.
As such, we will revert Flush() to the previous behavior and add a Commit() member function in the event that this behavior is needed.
lizzie
kleidis
crueter
Caio Oliveira
crueter
Briar
t895
Merry
Liam
BreadFish64
liamwhite
FearlessTobi
Squall-Leonhart
xcfrg
Mario
bunnei
german77
Morph
lat9nq
ameerj
Lioncash
Creak
MonsterDruide1