Add API to query max MMCSS threads on Windows

By default Windows limits the number of MMCSS threads to 32.
This can cause problems on modern systems with >= 32 cores,
when Ardour uses many process and I/O threads.

So far this is just a first step (query API)

libs/ardour/luabindings.cc

@@ -22,6 +22,7 @@
 
 #include <glibmm.h>
 
+#include "pbd/cpus.h"
 #include "pbd/history_owner.h"
 #include "pbd/stateful_diff_command.h"
 #include "pbd/openuri.h"
@@ -609,6 +610,9 @@ LuaBindings::common (lua_State* L)
 		.addConst ("UseGroup", PBD::Controllable::GroupControlDisposition(PBD::Controllable::UseGroup))
 		.endNamespace ()
 
+		.addFunction ("hardware_concurrency", hardware_concurrency)
+		.addFunction ("max_mmcss_threads_per_process", max_mmcss_threads_per_process)
+
 		.endNamespace (); // PBD
 
 	luabridge::getGlobalNamespace (L)

libs/pbd/cpus.cc

@@ -39,6 +39,27 @@
 #include <ardourext/pthread.h>  // Gets us 'PTW32_VERSION'
 #endif
 
+int32_t
+max_mmcss_threads_per_process ()
+{
+#ifdef PLATFORM_WINDOWS
+	DWORD dwType = REG_DWORD;
+	DWORD dwSize = 4;
+	int32_t rv   = 32;
+	HKEY hKey;
+	if (ERROR_SUCCESS == RegOpenKeyExA (HKEY_LOCAL_MACHINE, "Software\\Microsoft\\Windows NT\\CurrentVersion\\Multimedia\\SystemProfile", 0, KEY_READ, &hKey)) {
+		if (ERROR_SUCCESS == RegQueryValueExA (hKey, "MaxThreadsPerProcess", 0, &dwType, (LPBYTE)&rv, &dwSize)) {
+			if (dwType == REG_DWORD && dwSize == 4) {
+				return rv;
+			}
+		}
+	}
+	return 32;
+#else
+	return INT32_MAX;
+#endif
+}
+
 uint32_t
 hardware_concurrency()
 {

libs/pbd/pbd/cpus.h

@@ -23,4 +23,4 @@
 #include "pbd/libpbd_visibility.h"
 
 LIBPBD_API extern uint32_t hardware_concurrency ();
-
+LIBPBD_API extern int32_t max_mmcss_threads_per_process ();