[HLE] Fix guest-thread sync and boot for Unreal Engine titles (#102)

* [HLE] Fix guest-thread sync and boot for Unreal Engine titles

Silent Hill: The Short Message (and other UE titles) now boot the full
engine thread graph instead of hanging early. Four related fixes:

- pthread cond/mutex semantics: retain a signal raised with no waiter as
  pending, and key block/wake on the state's identity rather than a
  resolved address that could differ between lock and unlock. This ends
  the ~1.5M-call cond_wait busy-spin.

- Warm HLE type initializers and force-JIT their methods on a host thread
  at Freeze(). A .cctor or first-time JIT running on a guest thread's
  hijacked stack fail-fasts the CLR as "Invalid Program: attempted to
  call a UnmanagedCallersOnly method from managed code".

- Guest thread scheduling: pump after a wake so a readied thread actually
  runs, add a dispatcher thread for when every guest thread is parked,
  and make the pump-depth guard an atomic CAS.

- Route mutex/rwlock lock/unlock off the non-blocking leaf-import fast
  path so a contended lock can deschedule its guest thread.

Ported from the unreal-boot-fixes branch.

* [HLE] Keep mutex/rwlock unlock on the leaf-import fast path

The previous change routed all mutex/rwlock lock and unlock NIDs off the
leaf fast path so a contended lock could deschedule its guest thread. But
unlock never blocks, and taking it off the fast path made it slow enough
that Demon's Souls' job workers livelocked in a guest spinlock (millions
of mutex_unlock calls, no import progress, main thread stuck in
sceKernelWaitEventFlag).

Only *lock* needs to leave the leaf path. Restore the four unlock NIDs
(mutex + rwlock) so guest spinlocks stay cheap, while lock/rd/wrlock
remain off it for the blocking case Silent Hill needs.

* [HLE] Gate pthread_mutex_lock guest-thread blocking (fixes Demon's Souls)

Re-enabling cooperative deschedule on a contended pthread_mutex_lock
regressed Demon's Souls: its job workers run on libSceFiber, and blocking
a guest thread mid-fiber left sceFiberSwitch returning ESRCH followed by
a null fiber-context deref (0xC0000005). Bisect confirmed the pthread
change as the cause; the game reaches the same point as before it once
the block is skipped.

Gate the block behind SHARPEMU_MUTEX_LOCK_BLOCKING (off by default) so
contended locks fall through to the synchronous host-thread wait. The
rest of the pthread fixes (cond_wait pending signals, identity wake keys)
are unaffected.

src/SharpEmu.Core/Cpu/Native/DirectExecutionBackend.cs

@@ -2782,11 +2782,10 @@ public sealed unsafe partial class DirectExecutionBackend : INativeCpuBackend, I
 		{
 			return;
 		}
-		if (_guestThreadPumpDepth != 0)
+		if (Interlocked.CompareExchange(ref _guestThreadPumpDepth, 1, 0) != 0)
 		{
 			return;
 		}
-		_guestThreadPumpDepth++;
 		try
 		{
 			for (int i = 0; i < 8; i++)
@@ -2832,7 +2831,7 @@ public sealed unsafe partial class DirectExecutionBackend : INativeCpuBackend, I
 		}
 		finally
 		{
-			_guestThreadPumpDepth--;
+			Volatile.Write(ref _guestThreadPumpDepth, 0);
 		}
 	}
 
@@ -2875,9 +2874,18 @@ public sealed unsafe partial class DirectExecutionBackend : INativeCpuBackend, I
 			}
 		}
 
-		if (wakeCount != 0 && _logGuestThreads)
+		if (wakeCount != 0)
 		{
-			Console.Error.WriteLine($"[LOADER][INFO] guest_threads.wake key={wakeKey} count={wakeCount}");
+			if (_logGuestThreads)
+			{
+				Console.Error.WriteLine($"[LOADER][INFO] guest_threads.wake key={wakeKey} count={wakeCount}");
+			}
+
+			// Pump or the readied thread waits for an import dispatch that never comes.
+			if (_cpuContext is { } wakeContext)
+			{
+				Pump(wakeContext, "wake");
+			}
 		}
 
 		return wakeCount;
@@ -4422,6 +4430,26 @@ public sealed unsafe partial class DirectExecutionBackend : INativeCpuBackend, I
 			return;
 		}
 		_stallWatchdogStop = false;
+
+		// Drives woken threads when every guest thread is parked (nothing dispatches then).
+		var dispatcherThread = new Thread(new ThreadStart(delegate
+		{
+			while (!_stallWatchdogStop)
+			{
+				Thread.Sleep(1);
+				WakeExpiredBlockedGuestThreads();
+				if (Volatile.Read(ref _readyGuestThreadCount) > 0 && _cpuContext is { } dispatchContext)
+				{
+					Pump(dispatchContext, "dispatcher");
+				}
+			}
+		}))
+		{
+			IsBackground = true,
+			Name = "SharpEmu-GuestThreadDispatcher"
+		};
+		dispatcherThread.Start();
+
 		long num = (long)((double)stallWatchdogSeconds * Stopwatch.Frequency);
 		_stallWatchdogThread = new Thread(new ThreadStart(delegate
 		{
@@ -4450,6 +4478,15 @@ public sealed unsafe partial class DirectExecutionBackend : INativeCpuBackend, I
 					MarkExecutionProgress();
 					continue;
 				}
+				if (IsExpectedBlockingImportStall(out var blockingNid, out var blockingName))
+				{
+					Console.Error.WriteLine(
+						$"[LOADER][WARN] No import progress for {stallWatchdogSeconds}s while waiting in {blockingName} ({blockingNid}); continuing.");
+					LogStallWatchdogSnapshot();
+					Console.Error.Flush();
+					MarkExecutionProgress();
+					continue;
+				}
 				if (Interlocked.Exchange(ref _stallWatchdogTriggered, 1) != 0)
 				{
 					continue;
@@ -4485,6 +4522,38 @@ public sealed unsafe partial class DirectExecutionBackend : INativeCpuBackend, I
 		return false;
 	}
 
+	// A thread parked in a blocking wait is idle by design, not stalled.
+	private bool IsExpectedBlockingImportStall(out string nid, out string name)
+	{
+		nid = string.Empty;
+		name = string.Empty;
+		var cpuContext = _cpuContext;
+		if (cpuContext is null)
+		{
+			return false;
+		}
+
+		var importAddress = cpuContext.Rip & 0xFFFFFFFFFFFFFFF0uL;
+		foreach (var entry in _importEntries)
+		{
+			if (entry.Address != importAddress)
+			{
+				continue;
+			}
+
+			nid = entry.Nid;
+			name = _moduleManager.TryGetExport(nid, out var export)
+				? $"{export.LibraryName}:{export.Name}"
+				: nid;
+			return nid is
+				"Op8TBGY5KHg" or // pthread_cond_wait
+				"27bAgiJmOh0" or // pthread_cond_timedwait
+				"fzyMKs9kim0";   // sceKernelWaitEqueue
+		}
+
+		return false;
+	}
+
 	private void StopStallWatchdog()
 	{
 		_stallWatchdogStop = true;