The read thread could call Reset, which in turn tried to join the read
thread, leading to a SIGABRT. This manifested as Dolphin consistently
crashing when disconnecting a GC adapter and having a chance of crashing
a few seconds after connecting a GC adapter.
Now that patches and codes are enabled on a case by case basis, remove patcher code blocking codes entirely in hardcore mode, and reword the warning to be more accurate.
This apparently didn't compile on macOS six years ago before c++20, but
it should be fine by now.
While I'm at it, make the constants upper case per convention.
If the host is in hardcore mode, all joining players will be set to hardcore mode; if not, all joining players will be set to softcore. This ensures all players have the same settings and remain synchroized.
The primary focus of this PR is the Eternal Darkness patch which fixes hanging at startup, which prior to this fix makes Eternal Darkness unplayable in hardcore. The MHTri patch was added as well simply because it could be.
GXAbortFrame() is problematic for Dolphin because it first writes
PI_FIFO_RESET (for which we discard our internal fifo), then disables CP
reads (for which we execute pending commands in the GP fifo in emulated
memory). I don't know whether there is a race condition on hardware, but
there is one for us. Avoid this by also doing a GPU sync here.
The value being stored must be loaded into a register. In the case of an
immediate value, this means it must be materialized. The value is
eventually byteswapped before performing the store.
This can be simplified for the value 0 for two reasons:
- ARM64 has a dedicated zero register, so does not need to be
materialized.
- Byteswapping zero is still zero, so we can skip this step.
We could skip byteswapping for other values by immediately materializing
the byteswapped value in a register, but the benefits are not so clear
there (if the value needs to be materialized anyway, it is better to do
it up front).
Before:
0x5280001b mov w27, #0x0 ; =0
0xb9404fba ldr w26, [x29, #0x4c]
0x12881862 mov w2, #-0x40c4 ; =-16580
0x0b020342 add w2, w26, w2
0x5ac00b61 rev w1, w27
0xb8226b81 str w1, [x28, x2]
After:
0xb9404fbb ldr w27, [x29, #0x4c]
0x12881862 mov w2, #-0x40c4 ; =-16580
0x0b020362 add w2, w27, w2
0xb8226b9f str wzr, [x28, x2]
The game calls GXSetDrawDone and then switches the GP fifo without first
waiting for the draw done interrupt to arrive. Before
e96960e2a6, Dolphin would not execute the
draw done command and potentially also skip other commands in the old GP
fifo. Since that commit, Dolphin executes the remaining commands on the
old GP fifo just before disabling reads for switching, but because
PixelEngineManager::RaiseEvent() enforces a minimum delay of 500 cycles
for the draw done interrupt, it arrives after the game has switched to
the new GP fifo which seems to trigger the deadlock.
This patch replaces the call to GXSetDrawDone by a call to GXDrawDone
which does the same but also waits for the interrupt.
Removes the EFBAccessEnable=false explicit game overrides for:
GT6 (Terminator 3: The Redemption)
GXB (SSX3) [deleted - no other configuration]
RTH (Tony Hawk's Downhill Jam)
SNC (SONIC COLOURS) [deleted - no other configuration]
Makes Graphics -> Hacks -> Skip EFB Access from CPU enabled by default. Some GPU drivers stall when EFB access occurs in games where EFB is not used. Most games that require this setting set to 'true' already have this defined in their game inis.
Adjusts the Texture Cache settings/slider to 512 samples which equates
to the 'middle' option in the UI.
If using the 'Safe' Texture Cache Accuracy mode, the left screen will only render bloom in 2P modes.
If using the 'Fast' Texture Cache Accuracy mode, the Intro Movie will
drop frames near the end when the "Shadow the Hedgehog" game title
letters appear. This mode also can result in artifacts/bleed from bloom
effects in some situations.
`std::erase` is a replacement for the remove-erase idiom.
Changes to `OpenModeToAndroid` inadvertently revealed that the prior implementation had UB (potentially deleting the end iterator). This is now fixed.
Three bugs specific to older Wii games:
- The size difference between high-pass and biquad filter was not
accounted for, causing wiimote related fields to be corrupted.
- Wiimote sample buffer pointers were advanced by 32 samples per
millisecond instead of 6 samples. Usually hidden by the first bug.
- PB updates on Wii were being byte-swapped twice, but I've not actually
found any Wii games that make use of PB updates.
This fixes wiimote audio in at least the following games:
- Excite Truck
- Ice Age 2: The Meltdown
- Kororinpa: Marble Mania
- Rapala Tournament Fishing
- Shrek the Third
- Super Monkey Ball: Banana Blitz
- Tiger Woods PGA Tour 07
- WarioWare: Smooth Moves (issue 11725)
- Wing Island
Recently we have been getting some requests to make the existing vsync
setting available in the Android GUI:
https://bugs.dolphin-emu.org/issues/13650https://forums.dolphin-emu.org/Thread-vsync-toggle-for-android
I don't quite understand why enabling the vsync setting is helpful when
Android already enforces vsync, but I guess having the option available
doesn't hurt. I'm putting the setting under Advanced, unlike in
DolphinQt, since there's no clear reason why the typical user would want
to use this setting.
Fixes LIT (https://bugs.dolphin-emu.org/issues/13635). The text does not include normals, but has lighting enabled. With the previous default of (0, 0, 0), lighting was always black (as dot(X, (0, 0, 0)) is always 0). It seems like the normal from the map in the background (0, 0, 1) is re-used.
LIT also has the vertex color enabled while vertex color is not specified, the same as SMS's debug cubes; the default MissingColorValue GameINI value of solid white seems to work correctly in this case.
No need to materialize the immediate if it is zero, we can just use WZR.
Before:
mov w27, #0x0 ; =0
str w27, [x29, #0x1178]
After:
str wzr, [x29, #0x1178]
In VolumeVerifier.cpp, constructing a `std::string_view` of the volume's GameID is unnecessary, as `std::`(`ranges::`)`binary_search` supports heterogeneous lookup. The usage in GameFile.cpp is a perfect example.
In OGLConfig.cpp, `std::views::reverse` is used rather than sorting using `std::ranges::greater` in order to parallel other instances of reverse iteration in the function.
In DSPCore.cpp, there were two `std::fill` uses that could be simplified using `std::fill_n`. Due to their proximity with other `std::fill` algorithms being modernized with ranges, I chose to make these examples into the rare `std::ranges::fill_n`.
In StringUtil.h, the lambdas wrapping `Common::ToLower(char)` and `Common::ToUpper(char)` were only necessary due to the function names being overloaded.
Many games call GXSetGPFifo() without first waiting for the GP to finish
consuming outstanding commands in the previous GP fifo. Normally,
Dolphin runs OpcodeDecoding in 1000-cycle time slices. In that time
frame, GXSetGPFifo() has probably completed and the GP read pointer now
points to entirely new memory. If the last GP fifo copy ended in an
incomplete command, the new GP fifo would most likely desync for a
while. To avoid all this, give the GP a time slice right now to copy the
remaining data from the previous GP fifo.
Indexed XF loads specify the number of 32-bit words (generally floats, but light data has some integers) to load, not the number of bytes. This was only a mistake in the fifo analyzer text; the actual implementation already loaded words.
The light LIT fifolog from https://bugs.dolphin-emu.org/issues/13635 has position data at physical address 11ae3180. This works fine when using the memory viewer in physical mode, but the corresponding virtual address (91ae3180) previously didn't show anything in effective mode. It works fine now though.
This shouldn't affect playback of fifologs as everything in there uses physical addresses; this only impacts the memory viewer.
This logic was copied from CBoot::SetupBAT.
This reverts the revert commit bc67fc97c3,
except for the changes in BaseConfigLoader.cpp, which caused the bug
that made us revert 72cf2bdb87. PR 12917
contains an improved change to BaseConfigLoader.cpp, which can be merged
(or rejected) independently.
A few changes have also been made based on review comments.
`EFBEmulateFormatChanges = True` properly emulates the blur in the pause screen.
`ImmediateXFBEnable = False` to prevent epilepsy-inducing FMVs and main menu.
`EFBEmulateFormatChanges = True` properly emulates the blur in the pause screen.
`ImmediateXFBEnable = False` to prevent epilepsy-inducing FMVs and main menu.
Found this bug while testing; if I manually edit the config files while Dolphin is closed I was able to get debug and hardcore on at the same time, this resolves that.
AchievementManager::SetHardcoreMode now handles the (non-Qt) settings disabled by hardcore mode, instead of doing this on the Qt layer. Also ensured Init/Enable Achievements paths run this code, fixing the bug wherein the player can manipulate things when achievements are disabled that persist when turned back on.
This reverts commit 6dad8f8372.
Our bundled zlib-ng uses zlib compat mode, making it override system zlib.
System zlib-ng will not, and all its functions will be prefixed with zng_.
Therefore the two aren't actually compatible.
The use of any anti-aliasing will cause graphical corruption, which occurs on the ground.
MaxAnisotropy is supposed to be for [Video_Enhancements], but also it doesn't cause any of the issues in game so it is being removed rather than set under the correct heading.
Added a comment on why these values are set.
These codes are actually Gecko codes so setting that correctly.
There were errors which pop up when running this game as the comments for AR codes need to be set with hash(#) instead of asterisk(*). I replaced those anyways.
Removed the url in the comment since it's inaccessible and not on wayback machine either.
Also corrected the GameID, the wrong one is listed.
Dolphin has been using edge-to-edge rendering for a little while now,
but it has required a bit of manual work. Now that edge-to-edge is
becoming something expected of apps in Android 15, there's a nicer API
we can use.
Tested on Android 8, 11 and 13, with no changes in behavior noted.
So far, Dolphin hasn't been able to run on Android devices that use a
16 KiB page size. 16 KiB page sizes are a very new Android feature that
no phones have shipped with so far, so we're still compatible with the
phones that exist out there, but let's get this fixed before phones
start shipping with 16 KiB page sizes.
Because I couldn't get Android Studio's emulator to work, I haven't been
able to confirm that this change actually makes Dolphin fully compatible
with devices that use a 16 KiB page size. But I have confirmed that this
doesn't break anything on a regular 4 KiB page size device.
Prevent potential issues when creating the Graphics window (and thus
calling PopulateBackendInfo) while the core state is Stopping, like we
already do while it's Starting or Running.
Remove the PopulateBackendInfoFromUI function, which had a single caller
(GraphicsWindow::OnBackendChanged) and checked that the core wasn't
running or starting before calling PopulateBackendInfo.
Move the core state check into PopulateBackendInfo and have
OnBackendChanged call that instead. This guarantees the check is
performed by all callers of PopulateBackendInfo, preventing
potential reintroduction of the crash fixed in 3d4ae63f if another call
to PopulateBackendInfo is added.
As of the previous commit the only other caller of PopulateBackendInfo
is Core::Init shortly before s_state is set to Starting, so it will
always pass the check and so maintain its current behavior.
Remove the second of two calls to PopulateBackendInfo during emulation
startup. This call was originally the only one, but b214e0e added an
earlier call to handle the backend being changed by the GameINI.
The PR discussion doesn't explain why the original call was left in; I
suspect it was just overlooked.
As a bonus, this removes one of the extra copies of the "Video Info" On
Screen Display message at startup when using OpenGL.
Fix a crash when opening the Graphics window for the first time during
emulation startup when the backend is Vulkan, D3D11, or D3D12.
Don't call PopulateBackendInfo() from the Host thread when the core is
starting up. First, the function has already been called in Core::Init()
so we don't need to again. More importantly, PopulateBackendInfo() calls
g_video_backend->InitBackendInfo(), and the Vulkan and D3D
implementations of those functions load and then unload libraries (and
their associated function pointers) which are potentially in use by
other threads.
This crash was reliably reproducible with the following steps:
1) Select an affected backend.
2) Enable "Compile Shaders Before Starting"
3) Delete the cached shaders (but not the .uidcache file) for the game
you're testing.
4) Close and reopen Dolphin.
5) Start the game.
6) While the game is still booting or compiling shaders, open the
Graphics window for the first time in that Dolphin session.
Fixes https://bugs.dolphin-emu.org/issues/13634.
Before, Dolphin would randomly crash when updating the cheat search when automatic refresh was enabled. (Having a large number of addresses listed, e.g. by starting with an "any value" search, may contribute). The crash was due to QTableWidget::item returning nullptr in RefreshGUICurrentValues, presumably due to the table being resized on the UI thread while the emulated CPU thread was updating the values. I've fixed this by pausing the CPU thread for the entirety of OnNextScanClicked; this eliminated crashes in my testing.
With 12 uses of `JoinStrings` in the codebase vs 36 uses of `fmt::join`, fmtlib's range adapter for string concatenation with delimiters is clearly the preferred option.
Migrating `Common::CaseInsensitiveLess` to StringUtil.h will hopefully discourage rolling one's own solution in the future for case-insensitive associative containers when this (quite robust!) solution already exists.
`Common::CaseInsensitiveStringCompare::IsEqual` was removed in favor of using the `Common::CaseInsensitiveEquals` function.
The `a.size() != b.size()` condition in `Common::CaseInsensitiveEquals` can be removed, since `std::ranges::equal` already checks this condition (confirmed in libc++).
There were three distinct mechanisms for signaling breakpoint changes in DolphinQt, and the wiring had room for improvement. The behavior of these signals has been consolidated into the new `Host::PPCBreakpointsChanged` signal, which can be emitted from anywhere in DolphinQt to properly update breakpoints everywhere in DolphinQt.
This improves a few things:
- For the `CodeViewWidget` and `MemoryViewWidget`, signals no longer need to propagate through the `CodeWidget` and `MemoryWidget` (respectively) to reach their destination (incoming or outgoing).
- For the `BreakpointWidget`, by self-triggering from its own signal, it no longer must manually call `Update()` after all of the emission sites.
- For the `BranchWatchDialog`, it now has one less thing it must go through the `CodeWidget` for, which is a plus.
includes the following:
- org.DolphinEmu.dolphin-emu.metainfo.xml.in this file contains general
information and metadata that is used by most linux based app stores
(kde discover, gnome software, flathub.org, etc) to showcase dolphin.
- fill_release_node.sh a script to automatically fill in the release
information in the metainfo file when building the flatpak.
- SDL2.json manifest to build the vendored version of SDL from the
Exports directory, it's only temporarily needed until the kde runtime
is updated with the current SDL2 version and should be deleted after.
- org.DolphinEmu.dolphin-emu.yml the manifest itself.
Arbitrary Mipmap Detection doesn't work when GPU Texture Decoding is
enabled, so disable GPU Texture Decoding for games where the .ini
enables Arbitrary Mipmap Detection.
Co-authored-by: jeremyie <73066289+McAchi@users.noreply.github.com>
If texture dumping is enabled, notify the user on emulation startup
using an On Screen Display message.
Also notify the user when texture dumping is toggled.
Addresses https://bugs.dolphin-emu.org/issues/12445.
With this change, a different message is displayed if starting a game with RetroAchievements fails due to the Dolphin version being blocked as opposed to failing because the game hash is unsupported.
Normally, the asserts added in 34b0a6ea90 are only triggered when
something actually went wrong in Dolphin. But there is one exception:
In FallBackToInterpreter, we flush all registers regardless of whether
they're discarded. This is fine as long as none of the discarded
registers are inputs to the instruction that the interpreter will run.
To avoid false positive asserts, this change adds a parameter to Flush
that controls whether to skip the asserts for discarded registers.
Additionally, an assert for discarded registers is added to
Arm64FPRCache::Flush. (Previously JitArm64 asserted for GPRs (and CRs)
only, whereas Jit64 asserted both for GPRs and FPRs. I most likely
didn't think of FPRs when writing 34b0a6ea90.)
When the inverted depth range is unsupported and zRange is greater than farZ then min_depth becomes a negative value and far_depth will then exceed a depth of 1.0 (which is outside the scope of most backends and greater than GX_MAX_DEPTH of the console).
This happens when the backend supports depth clamping the min_depth is not clamped to zero.
2023-09-28 04:17:56 +01:00
1304 changed files with 56240 additions and 48129 deletions
- The codebase currently uses C++20, though not all compilers support all C++20 features.
- See CMakeLists.txt "Enforce minimium compiler versions" for the currently supported compilers.
- See CMakeLists.txt "Enforce minimum compiler versions" for the currently supported compilers.
- Use the [nullptr](https://en.cppreference.com/w/cpp/language/nullptr) type over the macro `NULL`.
- If a [range-based for loop](https://en.cppreference.com/w/cpp/language/range-for) can be used instead of container iterators, use it.
- Obviously, try not to use `goto` unless you have a *really* good reason for it.
@ -227,7 +227,7 @@ Summary:
## <aname="cpp-code-functions"></a>Functions
- If a function parameter is a pointer or reference and its value or data isn't intended to be changed, please mark that parameter as `const`.
- Functions that specifically modify their parameters should have the respective parameter(s) marked as a pointer so that the variables being modified are syntaxically obvious.
- Functions that specifically modify their parameters should have the respective parameter(s) marked as a pointer so that the variables being modified are syntactically obvious.
Tygyh