tcg: Disambiguate qemu_ld32u with 32-bit and 64-bit outputs.

Some targets (e.g. Alpha and MIPS64) need to keep 32-bit operands
sign-extended in 64-bit registers (regardless of the "real" sign
of the operand).  For that, we need to be able to distinguish
between a 32-bit load with a 32-bit result and a 32-bit load with
a given extension to a 64-bit result.  This distinction already
exists for the ld* loads, but not the qemu_ld* loads.

Reserve qemu_ld32u for 64-bit outputs and introduce qemu_ld32 for
32-bit outputs.  Adjust all code generators to match.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>

11 files changed, 37 insertions, 24 deletions

diff --git a/tcg/README b/tcg/README
index fe8c3d5d8f..3433908460 100644
--- a/tcg/README
+++ b/tcg/README
@@ -378,13 +378,17 @@ instructions.
 qemu_ld8s t0, t1, flags
 qemu_ld16u t0, t1, flags
 qemu_ld16s t0, t1, flags
+qemu_ld32 t0, t1, flags
 qemu_ld32u t0, t1, flags
 qemu_ld32s t0, t1, flags
 qemu_ld64 t0, t1, flags
 
-Load data at the QEMU CPU address t1 into t0. t1 has the QEMU CPU
-address type. 'flags' contains the QEMU memory index (selects user or
-kernel access) for example.
+Load data at the QEMU CPU address t1 into t0. t1 has the QEMU CPU address
+type. 'flags' contains the QEMU memory index (selects user or kernel access)
+for example.
+
+Note that "qemu_ld32" implies a 32-bit result, while "qemu_ld32u" and
+"qemu_ld32s" imply a 64-bit result appropriately extended from 32 bits.
 
 * qemu_st8 t0, t1, flags
 qemu_st16 t0, t1, flags
diff --git a/tcg/arm/tcg-target.c b/tcg/arm/tcg-target.c
index cf005b1a83..abd2187c9b 100644
--- a/tcg/arm/tcg-target.c
+++ b/tcg/arm/tcg-target.c
@@ -1488,7 +1488,7 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld(s, COND_AL, args, 1 | 4);
         break;
-    case INDEX_op_qemu_ld32u:
+    case INDEX_op_qemu_ld32:
         tcg_out_qemu_ld(s, COND_AL, args, 2);
         break;
     case INDEX_op_qemu_ld64:
@@ -1597,7 +1597,7 @@ static const TCGTargetOpDef arm_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "x", "X" } },
     { INDEX_op_qemu_ld16u, { "r", "x", "X" } },
     { INDEX_op_qemu_ld16s, { "r", "x", "X" } },
-    { INDEX_op_qemu_ld32u, { "r", "x", "X" } },
+    { INDEX_op_qemu_ld32, { "r", "x", "X" } },
     { INDEX_op_qemu_ld64, { "d", "r", "x", "X" } },
 
     { INDEX_op_qemu_st8, { "x", "x", "X" } },
diff --git a/tcg/hppa/tcg-target.c b/tcg/hppa/tcg-target.c
index 98e997c9ee..f9ae898fff 100644
--- a/tcg/hppa/tcg-target.c
+++ b/tcg/hppa/tcg-target.c
@@ -863,7 +863,7 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld(s, args, 1 | 4);
         break;
-    case INDEX_op_qemu_ld32u:
+    case INDEX_op_qemu_ld32:
         tcg_out_qemu_ld(s, args, 2);
         break;
 
@@ -923,7 +923,7 @@ static const TCGTargetOpDef hppa_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L" } },
-    { INDEX_op_qemu_ld32u, { "r", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L" } },
     { INDEX_op_qemu_ld64, { "r", "r", "L" } },
 
     { INDEX_op_qemu_st8, { "L", "L" } },
@@ -935,7 +935,7 @@ static const TCGTargetOpDef hppa_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L", "L" } },
-    { INDEX_op_qemu_ld32u, { "r", "L", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L", "L" } },
     { INDEX_op_qemu_ld64, { "r", "r", "L", "L" } },
 
     { INDEX_op_qemu_st8, { "L", "L", "L" } },
diff --git a/tcg/i386/tcg-target.c b/tcg/i386/tcg-target.c
index ef16e964e4..e684b33f55 100644
--- a/tcg/i386/tcg-target.c
+++ b/tcg/i386/tcg-target.c
@@ -1197,7 +1197,7 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld(s, args, 1 | 4);
         break;
-    case INDEX_op_qemu_ld32u:
+    case INDEX_op_qemu_ld32:
         tcg_out_qemu_ld(s, args, 2);
         break;
     case INDEX_op_qemu_ld64:
@@ -1281,7 +1281,7 @@ static const TCGTargetOpDef x86_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L" } },
-    { INDEX_op_qemu_ld32u, { "r", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L" } },
     { INDEX_op_qemu_ld64, { "r", "r", "L" } },
 
     { INDEX_op_qemu_st8, { "cb", "L" } },
@@ -1293,7 +1293,7 @@ static const TCGTargetOpDef x86_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L", "L" } },
-    { INDEX_op_qemu_ld32u, { "r", "L", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L", "L" } },
     { INDEX_op_qemu_ld64, { "r", "r", "L", "L" } },
 
     { INDEX_op_qemu_st8, { "cb", "L", "L" } },
diff --git a/tcg/mips/tcg-target.c b/tcg/mips/tcg-target.c
index 566e35272b..f7ed0baedd 100644
--- a/tcg/mips/tcg-target.c
+++ b/tcg/mips/tcg-target.c
@@ -1298,7 +1298,7 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld(s, args, 1 | 4);
         break;
-    case INDEX_op_qemu_ld32u:
+    case INDEX_op_qemu_ld32:
         tcg_out_qemu_ld(s, args, 2);
         break;
     case INDEX_op_qemu_ld64:
@@ -1371,7 +1371,7 @@ static const TCGTargetOpDef mips_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "L", "lZ" } },
     { INDEX_op_qemu_ld16u, { "L", "lZ" } },
     { INDEX_op_qemu_ld16s, { "L", "lZ" } },
-    { INDEX_op_qemu_ld32u, { "L", "lZ" } },
+    { INDEX_op_qemu_ld32, { "L", "lZ" } },
     { INDEX_op_qemu_ld64, { "L", "L", "lZ" } },
 
     { INDEX_op_qemu_st8, { "SZ", "SZ" } },
@@ -1383,7 +1383,7 @@ static const TCGTargetOpDef mips_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "L", "lZ", "lZ" } },
     { INDEX_op_qemu_ld16u, { "L", "lZ", "lZ" } },
     { INDEX_op_qemu_ld16s, { "L", "lZ", "lZ" } },
-    { INDEX_op_qemu_ld32u, { "L", "lZ", "lZ" } },
+    { INDEX_op_qemu_ld32, { "L", "lZ", "lZ" } },
     { INDEX_op_qemu_ld64, { "L", "L", "lZ", "lZ" } },
 
     { INDEX_op_qemu_st8, { "SZ", "SZ", "SZ" } },
diff --git a/tcg/ppc/tcg-target.c b/tcg/ppc/tcg-target.c
index 01c02eaff2..0b80a20146 100644
--- a/tcg/ppc/tcg-target.c
+++ b/tcg/ppc/tcg-target.c
@@ -1661,7 +1661,7 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld(s, args, 1 | 4);
         break;
-    case INDEX_op_qemu_ld32u:
+    case INDEX_op_qemu_ld32:
         tcg_out_qemu_ld(s, args, 2);
         break;
     case INDEX_op_qemu_ld64:
@@ -1775,7 +1775,7 @@ static const TCGTargetOpDef ppc_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L" } },
-    { INDEX_op_qemu_ld32u, { "r", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L" } },
     { INDEX_op_qemu_ld64, { "r", "r", "L" } },
 
     { INDEX_op_qemu_st8, { "K", "K" } },
@@ -1787,7 +1787,7 @@ static const TCGTargetOpDef ppc_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L", "L" } },
-    { INDEX_op_qemu_ld32u, { "r", "L", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L", "L" } },
     { INDEX_op_qemu_ld64, { "r", "L", "L", "L" } },
 
     { INDEX_op_qemu_st8, { "K", "K", "K" } },
diff --git a/tcg/ppc64/tcg-target.c b/tcg/ppc64/tcg-target.c
index 4eb5612cf1..4d6e68c306 100644
--- a/tcg/ppc64/tcg-target.c
+++ b/tcg/ppc64/tcg-target.c
@@ -1520,6 +1520,7 @@ static void tcg_out_op (TCGContext *s, TCGOpcode opc, const TCGArg *args,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld (s, args, 1 | 4);
         break;
+    case INDEX_op_qemu_ld32:
     case INDEX_op_qemu_ld32u:
         tcg_out_qemu_ld (s, args, 2);
         break;
@@ -1648,6 +1649,7 @@ static const TCGTargetOpDef ppc_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L" } },
     { INDEX_op_qemu_ld32u, { "r", "L" } },
     { INDEX_op_qemu_ld32s, { "r", "L" } },
     { INDEX_op_qemu_ld64, { "r", "L" } },
diff --git a/tcg/sparc/tcg-target.c b/tcg/sparc/tcg-target.c
index d6a59fe3bf..6d8410c65c 100644
--- a/tcg/sparc/tcg-target.c
+++ b/tcg/sparc/tcg-target.c
@@ -1324,7 +1324,10 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld(s, args, 1 | 4);
         break;
+    case INDEX_op_qemu_ld32:
+#if TCG_TARGET_REG_BITS == 64
     case INDEX_op_qemu_ld32u:
+#endif
         tcg_out_qemu_ld(s, args, 2);
         break;
 #if TCG_TARGET_REG_BITS == 64
@@ -1480,8 +1483,9 @@ static const TCGTargetOpDef sparc_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L" } },
-    { INDEX_op_qemu_ld32u, { "r", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L" } },
 #if TCG_TARGET_REG_BITS == 64
+    { INDEX_op_qemu_ld32u, { "r", "L" } },
     { INDEX_op_qemu_ld32s, { "r", "L" } },
 #endif
 
diff --git a/tcg/tcg-op.h b/tcg/tcg-op.h
index f7d11b0633..bafac2bb6a 100644
--- a/tcg/tcg-op.h
+++ b/tcg/tcg-op.h
@@ -2075,9 +2075,9 @@ static inline void tcg_gen_qemu_ld16s(TCGv ret, TCGv addr, int mem_index)
 static inline void tcg_gen_qemu_ld32u(TCGv ret, TCGv addr, int mem_index)
 {
 #if TARGET_LONG_BITS == 32
-    tcg_gen_op3i_i32(INDEX_op_qemu_ld32u, ret, addr, mem_index);
+    tcg_gen_op3i_i32(INDEX_op_qemu_ld32, ret, addr, mem_index);
 #else
-    tcg_gen_op4i_i32(INDEX_op_qemu_ld32u, TCGV_LOW(ret), TCGV_LOW(addr),
+    tcg_gen_op4i_i32(INDEX_op_qemu_ld32, TCGV_LOW(ret), TCGV_LOW(addr),
                      TCGV_HIGH(addr), mem_index);
     tcg_gen_movi_i32(TCGV_HIGH(ret), 0);
 #endif
@@ -2086,9 +2086,9 @@ static inline void tcg_gen_qemu_ld32u(TCGv ret, TCGv addr, int mem_index)
 static inline void tcg_gen_qemu_ld32s(TCGv ret, TCGv addr, int mem_index)
 {
 #if TARGET_LONG_BITS == 32
-    tcg_gen_op3i_i32(INDEX_op_qemu_ld32u, ret, addr, mem_index);
+    tcg_gen_op3i_i32(INDEX_op_qemu_ld32, ret, addr, mem_index);
 #else
-    tcg_gen_op4i_i32(INDEX_op_qemu_ld32u, TCGV_LOW(ret), TCGV_LOW(addr),
+    tcg_gen_op4i_i32(INDEX_op_qemu_ld32, TCGV_LOW(ret), TCGV_LOW(addr),
                      TCGV_HIGH(addr), mem_index);
     tcg_gen_sari_i32(TCGV_HIGH(ret), TCGV_LOW(ret), 31);
 #endif
diff --git a/tcg/tcg-opc.h b/tcg/tcg-opc.h
index 9797f637e4..674c73a527 100644
--- a/tcg/tcg-opc.h
+++ b/tcg/tcg-opc.h
@@ -251,9 +251,9 @@ DEF2(qemu_ld16s, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 DEF2(qemu_ld16s, 1, 2, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 #endif
 #if TARGET_LONG_BITS == 32
-DEF2(qemu_ld32u, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
+DEF2(qemu_ld32, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 #else
-DEF2(qemu_ld32u, 1, 2, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
+DEF2(qemu_ld32, 1, 2, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 #endif
 #if TARGET_LONG_BITS == 32
 DEF2(qemu_ld64, 2, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
@@ -288,6 +288,7 @@ DEF2(qemu_ld8u, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 DEF2(qemu_ld8s, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 DEF2(qemu_ld16u, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 DEF2(qemu_ld16s, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
+DEF2(qemu_ld32, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 DEF2(qemu_ld32u, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 DEF2(qemu_ld32s, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
 DEF2(qemu_ld64, 1, 1, 1, TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS)
diff --git a/tcg/x86_64/tcg-target.c b/tcg/x86_64/tcg-target.c
index 52812890de..3892f7586c 100644
--- a/tcg/x86_64/tcg-target.c
+++ b/tcg/x86_64/tcg-target.c
@@ -1238,6 +1238,7 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args,
     case INDEX_op_qemu_ld16s:
         tcg_out_qemu_ld(s, args, 1 | 4);
         break;
+    case INDEX_op_qemu_ld32:
     case INDEX_op_qemu_ld32u:
         tcg_out_qemu_ld(s, args, 2);
         break;
@@ -1412,6 +1413,7 @@ static const TCGTargetOpDef x86_64_op_defs[] = {
     { INDEX_op_qemu_ld8s, { "r", "L" } },
     { INDEX_op_qemu_ld16u, { "r", "L" } },
     { INDEX_op_qemu_ld16s, { "r", "L" } },
+    { INDEX_op_qemu_ld32, { "r", "L" } },
     { INDEX_op_qemu_ld32u, { "r", "L" } },
     { INDEX_op_qemu_ld32s, { "r", "L" } },
     { INDEX_op_qemu_ld64, { "r", "L" } },