Add support for multiple microarchitectures in single binary

 * compiler -march= parameter is changed from native to corei7
   so code is always genereted with instructions which are available
   on the Nehalem microarchitecture (up to SSE4.2)

 * compiler -mtune= parameter is added so code is optimized for
   corei7-avx which equals to Sandy Bridge microarchitecture

 * set of macros is added which allows run-time detection of available
   cpu instructions (e.g. clib_cpu_supports_avx())

 * set of macros is added which allows us to clone graph node funcitons
   where cloned function is optmized for different microarchitecture
   Those macros are using following attributes:
     __attribute__((flatten))
     __attribute__((target("arch=core-avx2)))

   I.e. If applied to foo_node_fn() macro will generate cloned
   functions foo_node_fn_avx2() and foo_node_fn_avx512() (future)
   It will also generate function void * foo_node_fn_multiarch_select()
   which detects available instruction set and returns pointer to the
   best matching function clone.

Change-Id: I2dce0ac92a5ede95fcb56f47f3d1f3c4c040bac0
Signed-off-by: Damjan Marion <damarion@cisco.com>

1 files changed, 27 insertions, 0 deletions

diff --git a/vlib/vlib/node.h b/vlib/vlib/node.h
index 2caede6e411..9b33a0a3ae0 100644
--- a/vlib/vlib/node.h
+++ b/vlib/vlib/node.h
@@ -40,6 +40,7 @@
 #ifndef included_vlib_node_h
 #define included_vlib_node_h
 
+#include <vppinfra/cpu.h>
 #include <vppinfra/longjmp.h>
 #include <vppinfra/timing_wheel.h>
 #include <vlib/trace.h>		/* for vlib_trace_filter_t */
@@ -149,6 +150,32 @@ static void __vlib_add_node_registration_##x (void)                     \
 }                                                                       \
 __VA_ARGS__ vlib_node_registration_t x 
 
+#if CLIB_DEBUG > 0
+#define VLIB_NODE_FUNCTION_CLONE_TEMPLATE(arch, fn)
+#define VLIB_NODE_FUNCTION_MULTIARCH_CLONE(fn)
+#define VLIB_NODE_FUNCTION_MULTIARCH(node, fn)
+#else
+#define VLIB_NODE_FUNCTION_CLONE_TEMPLATE(arch, fn, tgt)		\
+  uword									\
+  __attribute__ ((flatten))						\
+  __attribute__ ((target (tgt)))					\
+  CLIB_CPU_OPTIMIZED							\
+  fn ## _ ## arch ( struct vlib_main_t * vm,				\
+                   struct vlib_node_runtime_t * node,			\
+                   struct vlib_frame_t * frame)				\
+  { return fn (vm, node, frame); }
+
+#define VLIB_NODE_FUNCTION_MULTIARCH_CLONE(fn)				\
+  foreach_march_variant(VLIB_NODE_FUNCTION_CLONE_TEMPLATE, fn)
+
+#define VLIB_NODE_FUNCTION_MULTIARCH(node, fn)				\
+  VLIB_NODE_FUNCTION_MULTIARCH_CLONE(fn)				\
+  CLIB_MULTIARCH_SELECT_FN(fn, static inline)				\
+  static void __attribute__((__constructor__))				\
+  __vlib_node_function_multiarch_select_##node (void)			\
+  { node.function = fn ## _multiarch_select(); }
+#endif
+
 always_inline vlib_node_registration_t *
 vlib_node_next_registered (vlib_node_registration_t * c)
 {