Implement malloc_getcpu for amd64 and arm64 macOS

This enables per CPU arena on MacOS

configure.ac

@@ -510,6 +510,23 @@ typedef unsigned __int32 uint32_t;
       else
         AC_MSG_ERROR([cannot determine number of significant virtual address bits])
       fi
+      AC_CACHE_CHECK([rdtscp support],
+		     [je_cv_rdtscp],
+		     AC_RUN_IFELSE([AC_LANG_PROGRAM(
+[[
+#include <stdint.h>
+]],
+[[
+      unsigned int dx;
+      asm volatile("rdtscp" : "=d"(dx) ::);
+      return 0;
+]])],
+      [je_cv_rdtscp=yes],
+      [je_cv_rdstcp=no],
+      [je_cv_rdtscp=no]))
+      if test "x${je_cv_rdtscp}" = "xyes"; then
+        AC_DEFINE([HAVE_RDTSCP], 1, [])
+      fi
     fi
     ;;
   *)

include/jemalloc/internal/jemalloc_internal_inlines_a.h

@@ -14,6 +14,15 @@ malloc_getcpu(void) {
 	return GetCurrentProcessorNumber();
 #elif defined(JEMALLOC_HAVE_SCHED_GETCPU)
 	return (malloc_cpuid_t)sched_getcpu();
+#elif defined(HAVE_RDTSCP)
+	unsigned int ax, cx, dx;
+	asm volatile("rdtscp" : "=a"(ax), "=d"(dx), "=c"(cx) ::);
+	return (malloc_cpuid_t)(dx & 0xfff);
+#elif defined(__aarch64__) && defined(__APPLE__)
+	/* Other oses most likely use tpidr_el0 instead */
+	uintptr_t c;
+	asm volatile("mrs %x0, tpidrro_el0" : "=r"(c) :: "memory");
+	return (malloc_cpuid_t)(c & (1 << 3) - 1);
 #else
 	not_reached();
 	return -1;