summaryrefslogtreecommitdiff
path: root/src/core/arm
diff options
context:
space:
mode:
authorcomex <comexk@gmail.com>2023-07-01 15:00:39 -0700
committercomex <comexk@gmail.com>2023-07-15 12:00:28 -0700
commitd7c532d8894ce806c9af13b8dd3eec975642b348 (patch)
tree1353e6d1776384575019db28987e23237552a9de /src/core/arm
parent04868ab9da2049429808a265dc50696f6300b2b4 (diff)
Fixes and workarounds to make UBSan happier on macOS
There are still some other issues not addressed here, but it's a start. Workarounds for false-positive reports: - `RasterizerAccelerated`: Put a gigantic array behind a `unique_ptr`, because UBSan has a [hardcoded limit](https://stackoverflow.com/questions/64531383/c-runtime-error-using-fsanitize-undefined-object-has-a-possibly-invalid-vp) of how big it thinks objects can be, specifically when dealing with offset-to-top values used with multiple inheritance. Hopefully this doesn't have a performance impact. - `QueryCacheBase::QueryCacheBase`: Avoid an operation that UBSan thinks is UB even though it at least arguably isn't. See the link in the comment for more information. Fixes for correct reports: - `PageTable`, `Memory`: Use `uintptr_t` values instead of pointers to avoid UB from pointer overflow (when pointer arithmetic wraps around the address space). - `KScheduler::Reload`: `thread->GetOwnerProcess()` can be `nullptr`; avoid calling methods on it in this case. (The existing code returns a garbage reference to a field, which is then passed into `LoadWatchpointArray`, and apparently it's never used, so it's harmless in practice but still triggers UBSan.) - `KAutoObject::Close`: This function calls `this->Destroy()`, which overwrites the beginning of the object with junk (specifically a free list pointer). Then it calls `this->UnregisterWithKernel()`. UBSan complains about a type mismatch because the vtable has been overwritten, and I believe this is indeed UB. `UnregisterWithKernel` also loads `m_kernel` from the 'freed' object, which seems to be technically safe (the overwriting doesn't extend as far as that field), but seems dubious. Switch to a `static` method and load `m_kernel` in advance.
Diffstat (limited to 'src/core/arm')
-rw-r--r--src/core/arm/arm_interface.cpp4
-rw-r--r--src/core/arm/arm_interface.h2
2 files changed, 3 insertions, 3 deletions
diff --git a/src/core/arm/arm_interface.cpp b/src/core/arm/arm_interface.cpp
index aa0eb9791..0c012f094 100644
--- a/src/core/arm/arm_interface.cpp
+++ b/src/core/arm/arm_interface.cpp
@@ -217,8 +217,8 @@ void ARM_Interface::Run() {
}
}
-void ARM_Interface::LoadWatchpointArray(const WatchpointArray& wp) {
- watchpoints = &wp;
+void ARM_Interface::LoadWatchpointArray(const WatchpointArray* wp) {
+ watchpoints = wp;
}
const Kernel::DebugWatchpoint* ARM_Interface::MatchingWatchpoint(
diff --git a/src/core/arm/arm_interface.h b/src/core/arm/arm_interface.h
index d5f2fa09a..3d866ff6f 100644
--- a/src/core/arm/arm_interface.h
+++ b/src/core/arm/arm_interface.h
@@ -186,7 +186,7 @@ public:
virtual void SaveContext(ThreadContext64& ctx) const = 0;
virtual void LoadContext(const ThreadContext32& ctx) = 0;
virtual void LoadContext(const ThreadContext64& ctx) = 0;
- void LoadWatchpointArray(const WatchpointArray& wp);
+ void LoadWatchpointArray(const WatchpointArray* wp);
/// Clears the exclusive monitor's state.
virtual void ClearExclusiveState() = 0;