tci: Use 32-bit signed offsets to loads/stores

Since the change to tcg_exit_req, the first insn of every TB is
a load with a negative offset from env.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Signed-off by: Stefan Weil <sw@weilnetz.de>

tcg/tci/tcg-target.c

@@ -513,7 +513,7 @@ static void tcg_out_ld(TCGContext *s, TCGType type, TCGReg ret, TCGReg arg1,
         tcg_out_op_t(s, INDEX_op_ld_i64);
         tcg_out_r(s, ret);
         tcg_out_r(s, arg1);
-        assert(arg2 == (uint32_t)arg2);
+        assert(arg2 == (int32_t)arg2);
         tcg_out32(s, arg2);
 #else
         TODO();
@@ -636,7 +636,7 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args,
     case INDEX_op_st_i64:
         tcg_out_r(s, args[0]);
         tcg_out_r(s, args[1]);
-        assert(args[2] == (uint32_t)args[2]);
+        assert(args[2] == (int32_t)args[2]);
         tcg_out32(s, args[2]);
         break;
     case INDEX_op_add_i32:

tci.c

@@ -182,7 +182,7 @@ static tcg_target_ulong tci_read_i(uint8_t **tb_ptr)
     return value;
 }
 
-/* Read constant (32 bit) from bytecode. */
+/* Read unsigned constant (32 bit) from bytecode. */
 static uint32_t tci_read_i32(uint8_t **tb_ptr)
 {
     uint32_t value = *(uint32_t *)(*tb_ptr);
@@ -190,6 +190,14 @@ static uint32_t tci_read_i32(uint8_t **tb_ptr)
     return value;
 }
 
+/* Read signed constant (32 bit) from bytecode. */
+static int32_t tci_read_s32(uint8_t **tb_ptr)
+{
+    int32_t value = *(int32_t *)(*tb_ptr);
+    *tb_ptr += sizeof(value);
+    return value;
+}
+
 #if TCG_TARGET_REG_BITS == 64
 /* Read constant (64 bit) from bytecode. */
 static uint64_t tci_read_i64(uint8_t **tb_ptr)
@@ -550,7 +558,7 @@ tcg_target_ulong tcg_qemu_tb_exec(CPUArchState *cpustate, uint8_t *tb_ptr)
         case INDEX_op_ld8u_i32:
             t0 = *tb_ptr++;
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             tci_write_reg8(t0, *(uint8_t *)(t1 + t2));
             break;
         case INDEX_op_ld8s_i32:
@@ -563,25 +571,25 @@ tcg_target_ulong tcg_qemu_tb_exec(CPUArchState *cpustate, uint8_t *tb_ptr)
         case INDEX_op_ld_i32:
             t0 = *tb_ptr++;
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             tci_write_reg32(t0, *(uint32_t *)(t1 + t2));
             break;
         case INDEX_op_st8_i32:
             t0 = tci_read_r8(&tb_ptr);
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             *(uint8_t *)(t1 + t2) = t0;
             break;
         case INDEX_op_st16_i32:
             t0 = tci_read_r16(&tb_ptr);
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             *(uint16_t *)(t1 + t2) = t0;
             break;
         case INDEX_op_st_i32:
             t0 = tci_read_r32(&tb_ptr);
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             *(uint32_t *)(t1 + t2) = t0;
             break;
 
@@ -818,7 +826,7 @@ tcg_target_ulong tcg_qemu_tb_exec(CPUArchState *cpustate, uint8_t *tb_ptr)
         case INDEX_op_ld8u_i64:
             t0 = *tb_ptr++;
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             tci_write_reg8(t0, *(uint8_t *)(t1 + t2));
             break;
         case INDEX_op_ld8s_i64:
@@ -829,43 +837,43 @@ tcg_target_ulong tcg_qemu_tb_exec(CPUArchState *cpustate, uint8_t *tb_ptr)
         case INDEX_op_ld32u_i64:
             t0 = *tb_ptr++;
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             tci_write_reg32(t0, *(uint32_t *)(t1 + t2));
             break;
         case INDEX_op_ld32s_i64:
             t0 = *tb_ptr++;
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             tci_write_reg32s(t0, *(int32_t *)(t1 + t2));
             break;
         case INDEX_op_ld_i64:
             t0 = *tb_ptr++;
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             tci_write_reg64(t0, *(uint64_t *)(t1 + t2));
             break;
         case INDEX_op_st8_i64:
             t0 = tci_read_r8(&tb_ptr);
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             *(uint8_t *)(t1 + t2) = t0;
             break;
         case INDEX_op_st16_i64:
             t0 = tci_read_r16(&tb_ptr);
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             *(uint16_t *)(t1 + t2) = t0;
             break;
         case INDEX_op_st32_i64:
             t0 = tci_read_r32(&tb_ptr);
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             *(uint32_t *)(t1 + t2) = t0;
             break;
         case INDEX_op_st_i64:
             t0 = tci_read_r64(&tb_ptr);
             t1 = tci_read_r(&tb_ptr);
-            t2 = tci_read_i32(&tb_ptr);
+            t2 = tci_read_s32(&tb_ptr);
             *(uint64_t *)(t1 + t2) = t0;
             break;