[Qemu-devel] [PULL 2/2] tcg/i386: omit a few REXW prefixes in softmmu code

[Richard Henderson <rth@twiddle.net>]

From: Aurelien Jarno <aurelien@aurel32.net>

When computing the TLB address we are likely to mask out the high
32-bits by using shr + and. We can use 32-bit instructions in that
case. This saves 2 bytes per TLB access.

Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Message-Id: <1437306632-20655-1-git-send-email-aurelien@aurel32.net>
Signed-off-by: Richard Henderson <rth@twiddle.net>
---
 tcg/i386/tcg-target.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/tcg/i386/tcg-target.c b/tcg/i386/tcg-target.c
index d2adbc4..9187d34 100644
--- a/tcg/i386/tcg-target.c
+++ b/tcg/i386/tcg-target.c
@@ -1178,8 +1178,8 @@ static inline void tcg_out_tlb_load(TCGContext *s, TCGReg addrlo, TCGReg addrhi,
     const TCGReg r0 = TCG_REG_L0;
     const TCGReg r1 = TCG_REG_L1;
     TCGType ttype = TCG_TYPE_I32;
-    TCGType htype = TCG_TYPE_I32;
-    int trexw = 0, hrexw = 0;
+    TCGType tlbtype = TCG_TYPE_I32;
+    int trexw = 0, hrexw = 0, tlbrexw = 0;
     int s_mask = (1 << (opc & MO_SIZE)) - 1;
     bool aligned = (opc & MO_AMASK) == MO_ALIGN || s_mask == 0;
 
@@ -1189,12 +1189,15 @@ static inline void tcg_out_tlb_load(TCGContext *s, TCGReg addrlo, TCGReg addrhi,
             trexw = P_REXW;
         }
         if (TCG_TYPE_PTR == TCG_TYPE_I64) {
-            htype = TCG_TYPE_I64;
             hrexw = P_REXW;
+            if (TARGET_PAGE_BITS + CPU_TLB_BITS > 32) {
+                tlbtype = TCG_TYPE_I64;
+                tlbrexw = P_REXW;
+            }
         }
     }
 
-    tcg_out_mov(s, htype, r0, addrlo);
+    tcg_out_mov(s, tlbtype, r0, addrlo);
     if (aligned) {
         tcg_out_mov(s, ttype, r1, addrlo);
     } else {
@@ -1203,12 +1206,12 @@ static inline void tcg_out_tlb_load(TCGContext *s, TCGReg addrlo, TCGReg addrhi,
         tcg_out_modrm_offset(s, OPC_LEA + trexw, r1, addrlo, s_mask);
     }
 
-    tcg_out_shifti(s, SHIFT_SHR + hrexw, r0,
+    tcg_out_shifti(s, SHIFT_SHR + tlbrexw, r0,
                    TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS);
 
     tgen_arithi(s, ARITH_AND + trexw, r1,
                 TARGET_PAGE_MASK | (aligned ? s_mask : 0), 0);
-    tgen_arithi(s, ARITH_AND + hrexw, r0,
+    tgen_arithi(s, ARITH_AND + tlbrexw, r0,
                 (CPU_TLB_SIZE - 1) << CPU_TLB_ENTRY_BITS, 0);
 
     tcg_out_modrm_sib_offset(s, OPC_LEA + hrexw, r0, TCG_AREG0, r0, 0,
-- 
2.4.3