diff --git a/core/shader/recompiler/Recompiler.cpp b/core/shader/recompiler/Recompiler.cpp
index 7e5883eeb..770bcef49 100644
--- a/core/shader/recompiler/Recompiler.cpp
+++ b/core/shader/recompiler/Recompiler.cpp
@@ -290,6 +290,14 @@ struct SourceEntry {
 
 namespace {
 
+class PrimedSharedMutex : public std::shared_mutex {
+public:
+    PrimedSharedMutex() {
+        lock();
+        unlock();
+    }
+};
+
 struct ResourceProgram {
     explicit ResourceProgram(const RecompileRequest& request, ShaderPreparationContext* preparation = nullptr) : program(std::make_unique<IrProgram>(PrepareResourceProgram(request, preparation))), plan(std::make_shared<const IrResourcePlan>(ResourceMaterializer{}.ExtractPlan(*program))) {}
 
@@ -318,7 +326,7 @@ bool FailureMemo() {
 }
 
 std::shared_ptr<SourceEntry> getSource(const RecompileRequest& request, ShaderPreparationContext* preparation = nullptr) {
-    static std::shared_mutex mutex;
+    static PrimedSharedMutex mutex;
     // Entries whose code hashes alike share a bucket; the code comparison picks the right one.
     static std::unordered_map<std::vector<std::uint64_t>, std::vector<std::shared_ptr<SourceEntry>>, SourceKeyHash> sources;
     struct SourceKeyStorage {};
@@ -587,7 +595,7 @@ struct PreparedBindingPlan {
     std::once_flag ready;
     std::exception_ptr failure;
     DescriptorBindingPlan bindings;
-    std::shared_mutex mutex;
+    PrimedSharedMutex mutex;
     std::map<std::vector<std::uint32_t>, std::shared_ptr<PreparedModuleEntry>> modules;
 };
 
@@ -618,7 +626,7 @@ std::shared_ptr<PreparedBindingPlan> preparedBindingPlan(const CompiledVariant&
     }
     key.push_back(exports.size());
     for (const auto mapping : exports) key.push_back(mapping);
-    static std::shared_mutex mutex;
+    static PrimedSharedMutex mutex;
     static std::map<std::vector<std::uint64_t>, std::shared_ptr<PreparedBindingPlan>> plans;
     std::shared_ptr<PreparedBindingPlan> plan;
     {
@@ -647,7 +655,7 @@ std::shared_ptr<const SpecializedModule> specializeModule(const CompiledShaderAr
     key.clear();
     key.push_back(artifact.variantId);
     for (const auto& constant : constants) key.push_back((static_cast<std::uint64_t>(constant.id) << 32u) | constant.value);
-    static std::shared_mutex mutex;
+    static PrimedSharedMutex mutex;
     static std::map<std::vector<std::uint64_t>, std::shared_ptr<SpecializedModuleEntry>> modules;
     std::shared_ptr<SpecializedModuleEntry> entry;
     {
diff --git a/docs/dev/TechnicalDebt.md b/docs/dev/TechnicalDebt.md
index 0c72b62a7..cd6228162 100644
--- a/docs/dev/TechnicalDebt.md
+++ b/docs/dev/TechnicalDebt.md
@@ -7,7 +7,7 @@
   - libgcc_s_seh-1.dll
   - libstdc++-6.dll
   - libwinpthread-1.dll
-- winpthreads initializes a statically initialized rwlock on its first lock and fails a concurrent first lock with `EINVAL`, which `std::shared_mutex` and `std::shared_timed_mutex` ignore, so that lock does not exclude. Guest rwlocks ([Pthread.hpp](../../core/libs/prx/libkernel/Pthread/include/Pthread.hpp)) are locked once when created. The host `std::shared_mutex` objects in [Recompiler.cpp](../../core/shader/recompiler/Recompiler.cpp) and [GuestWriteWatch.cpp](../../core/libs/prx/libc/src/GuestWriteWatch.cpp) are not, and can still race on their first contended use.
+- winpthreads initializes a statically initialized rwlock on its first lock and fails a concurrent first lock with `EINVAL`, which `std::shared_mutex` and `std::shared_timed_mutex` ignore, so that lock does not exclude. Guest rwlocks ([Pthread.hpp](../../core/libs/prx/libkernel/Pthread/include/Pthread.hpp)) and the host `std::shared_mutex` objects in [Recompiler.cpp](../../core/shader/recompiler/Recompiler.cpp) are locked once when created. A new host `std::shared_mutex` used on Windows needs the same.
 
 ### Silent stubs