[PATCH 2/6] target/i386: Use tcg gvec for gen_op_movo

[Richard Henderson]

Low hanging fruit, using gvec to move 16 bytes.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/i386/cpu.h           | 4 ++--
 target/i386/tcg/translate.c | 7 +++----
 2 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 81e5abed86..dbc9a99a3b 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1587,8 +1587,8 @@ typedef struct CPUArchState {
     float_status mmx_status; /* for 3DNow! float ops */
     float_status sse_status;
     uint32_t mxcsr;
-    ZMMReg xmm_regs[CPU_NB_REGS == 8 ? 8 : 32];
-    ZMMReg xmm_t0;
+    ZMMReg xmm_regs[CPU_NB_REGS == 8 ? 8 : 32] QEMU_ALIGNED(16);
+    ZMMReg xmm_t0 QEMU_ALIGNED(16);
     MMXReg mmx_t0;
 
     uint64_t opmask_regs[NB_OPMASK_REGS];
diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
index b7972f0ff5..c1f1f6f66b 100644
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -23,6 +23,7 @@
 #include "disas/disas.h"
 #include "exec/exec-all.h"
 #include "tcg/tcg-op.h"
+#include "tcg/tcg-op-gvec.h"
 #include "exec/cpu_ldst.h"
 #include "exec/translator.h"
 
@@ -2753,10 +2754,8 @@ static inline void gen_sto_env_A0(DisasContext *s, int 
offset)
 
 static inline void gen_op_movo(DisasContext *s, int d_offset, int s_offset)
 {
-    tcg_gen_ld_i64(s->tmp1_i64, cpu_env, s_offset + offsetof(ZMMReg, 
ZMM_Q(0)));
-    tcg_gen_st_i64(s->tmp1_i64, cpu_env, d_offset + offsetof(ZMMReg, 
ZMM_Q(0)));
-    tcg_gen_ld_i64(s->tmp1_i64, cpu_env, s_offset + offsetof(ZMMReg, 
ZMM_Q(1)));
-    tcg_gen_st_i64(s->tmp1_i64, cpu_env, d_offset + offsetof(ZMMReg, 
ZMM_Q(1)));
+    int xmm_ofs = offsetof(ZMMReg, ZMM_X(0));
+    tcg_gen_gvec_mov(MO_64, d_offset + xmm_ofs, s_offset + xmm_ofs, 16, 16);
 }
 
 static inline void gen_op_movq(DisasContext *s, int d_offset, int s_offset)
-- 
2.34.1