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>

3 files changed, 5 insertions, 0 deletions

diff --git a/vpp/app/l2t_ip6.c b/vpp/app/l2t_ip6.c
index ac24886a..8a1e23a9 100644
--- a/vpp/app/l2t_ip6.c
+++ b/vpp/app/l2t_ip6.c
@@ -274,6 +274,8 @@ static VLIB_REGISTER_NODE (sw6_ip6_node) = {
   },
 };
 
+VLIB_NODE_FUNCTION_MULTIARCH (sw6_ip6_node, ip6_l2t_node_fn)
+
 static clib_error_t *
 l2tp_config (vlib_main_t * vm, unformat_input_t * input)
 {
diff --git a/vpp/app/l2t_l2.c b/vpp/app/l2t_l2.c
index 5c8327d5..f6385a5a 100644
--- a/vpp/app/l2t_l2.c
+++ b/vpp/app/l2t_l2.c
@@ -250,3 +250,5 @@ VLIB_REGISTER_NODE (l2t_l2_node) = {
   },
 };
 
+VLIB_NODE_FUNCTION_MULTIARCH (l2t_l2_node, l2t_l2_node_fn)
+
diff --git a/vpp/app/version.c b/vpp/app/version.c
index 588aadf2..22bec400 100644
--- a/vpp/app/version.c
+++ b/vpp/app/version.c
@@ -59,6 +59,7 @@ show_vpe_version_command_fn (vlib_main_t * vm,
       _("Compiler", "%s", vpe_compiler);
       _("CPU model name", "%U", format_cpu_model_name);
       _("CPU microarchitecture", "%U", format_cpu_uarch);
+      _("CPU flags", "%U", format_cpu_flags);
       _("Current PID", "%d", getpid());
 #if DPDK > 0
       _("DPDK Version", "%s", rte_version());