[RFC-PATCH-for-11.0 v3 00/21] accel/tcg: Remove most MO_TE uses in cpu_ld/st_code/data()

[RFC-PATCH-for-11.0 v3 00/21] accel/tcg: Remove most MO_TE uses in cpu_ld/st_code/data()

[Philippe Mathieu-Daudé]

Quick attempt to remove MO_TE uses for fixed-endianness targets.

v3: convert all CODE uses, almost all DATA uses (except PPC/MIPS).
v2: convert all but mips targets and remove unexplicit API.

Superseeded: <20251120201919.8460-1-philmd@linaro.org>

Philippe Mathieu-Daudé (21):
  accel/tcg: Add endianness variants of cpu_ld{uw,l,q}_code()
  target/alpha: Use little-endian variant of cpu_ldl_code()
  target/loongarch: Use little-endian variant of cpu_ldl_code()
  target/sparc: Use big-endian variant of cpu_ldl_code()
  target/s390x: Use big-endian variant of cpu_ld{uw,l}_code()
  target/riscv: Use little-endian variant of cpu_ld{l,q}_code()
  target/ppc: Replace cpu_ldl_code() by explicit endianness variants
  target/mips: Replace cpu_ld{uw,l}_code() by explicit endianness
    variants
  accel/tcg: Remove non-explicit endian cpu_ld*_code() helpers
  target/hexagon: Use little-endian variant of cpu_ld/st_data*()
  target/tricore: Use little-endian variant of cpu_ld/st_data*()
  target/rx: Use little-endian variant of cpu_ld/st_data*()
  target/m68k: Use little-endian variant of cpu_ld/st_data*()
  target/s390x: Use little-endian variant of cpu_ld/st_data*()
  target/sparc: Use little-endian variant of cpu_ld/st_data*()
  target/i386: Use little-endian variant of cpu_ld/st_data*()
  target/hppa: Use little-endian variant of cpu_ld/st_data*()
  target/riscv: Use little-endian variant of cpu_ld/st_data*() for
    vector
  target/sh4: Replace cpu_stl_data() by explicit endianness variants
  target/mips: Use big-endian variant of cpu_ld/st_data*() for MSA
  accel/tcg: Remove non-explicit endian cpu_ld/st*_data*() helpers

 include/accel/tcg/cpu-ldst.h         |  79 +++++---------
 target/hexagon/macros.h              |   6 +-
 target/i386/ops_sse.h                |  12 +--
 target/i386/tcg/seg_helper.h         |  12 +--
 linux-user/vm86.c                    |   4 +-
 target/alpha/mem_helper.c            |   2 +-
 target/hexagon/op_helper.c           |   6 +-
 target/hppa/op_helper.c              |  44 ++++----
 target/i386/tcg/mem_helper.c         |   8 +-
 target/i386/tcg/mpx_helper.c         |  28 ++---
 target/i386/tcg/seg_helper.c         |  16 +--
 target/i386/tcg/system/excp_helper.c |   8 +-
 target/i386/tcg/system/svm_helper.c  |  48 ++++-----
 target/loongarch/tcg/tcg_cpu.c       |   2 +-
 target/m68k/fpu_helper.c             |  12 +--
 target/m68k/op_helper.c              |  88 ++++++++--------
 target/mips/tcg/msa_helper.c         |  51 +++++----
 target/mips/tcg/system/tlb_helper.c  |  22 +++-
 target/ppc/tcg-excp_helper.c         |  16 +--
 target/riscv/translate.c             |   2 +-
 target/riscv/vector_helper.c         |  12 +--
 target/riscv/zce_helper.c            |   4 +-
 target/rx/helper.c                   |  14 +--
 target/rx/op_helper.c                |   6 +-
 target/s390x/tcg/mem_helper.c        |  54 +++++-----
 target/s390x/tcg/vec_helper.c        |   8 +-
 target/sh4/op_helper.c               |   6 +-
 target/sparc/int32_helper.c          |   2 +-
 target/sparc/ldst_helper.c           |   6 +-
 target/tricore/op_helper.c           | 152 +++++++++++++--------------
 30 files changed, 360 insertions(+), 370 deletions(-)

-- 
2.51.0

[RFC-PATCH-for-11.0 v3 01/21] accel/tcg: Add endianness variants of cpu_ld{uw, l, q}_code()

[Philippe Mathieu-Daudé]

Allow target with fixed-endianness to not use the MO_TE definition.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 include/accel/tcg/cpu-ldst.h | 43 +++++++++++++++++++++++++++++++-----
 1 file changed, 37 insertions(+), 6 deletions(-)

diff --git a/include/accel/tcg/cpu-ldst.h b/include/accel/tcg/cpu-ldst.h
index 0de7f5eaa6b..e4ec4e7d367 100644
--- a/include/accel/tcg/cpu-ldst.h
+++ b/include/accel/tcg/cpu-ldst.h
@@ -481,25 +481,56 @@ static inline uint32_t cpu_ldub_code(CPUArchState *env, abi_ptr addr)
     return cpu_ldb_code_mmu(env, addr, oi, 0);
 }
 
-static inline uint32_t cpu_lduw_code(CPUArchState *env, abi_ptr addr)
+static inline uint32_t cpu_lduw_le_code(CPUArchState *env, abi_ptr addr)
 {
     CPUState *cs = env_cpu(env);
-    MemOpIdx oi = make_memop_idx(MO_TEUW, cpu_mmu_index(cs, true));
+    MemOpIdx oi = make_memop_idx(MO_LEUW, cpu_mmu_index(cs, true));
     return cpu_ldw_code_mmu(env, addr, oi, 0);
 }
 
-static inline uint32_t cpu_ldl_code(CPUArchState *env, abi_ptr addr)
+static inline uint32_t cpu_ldl_le_code(CPUArchState *env, abi_ptr addr)
 {
     CPUState *cs = env_cpu(env);
-    MemOpIdx oi = make_memop_idx(MO_TEUL, cpu_mmu_index(cs, true));
+    MemOpIdx oi = make_memop_idx(MO_LEUL, cpu_mmu_index(cs, true));
     return cpu_ldl_code_mmu(env, addr, oi, 0);
 }
 
-static inline uint64_t cpu_ldq_code(CPUArchState *env, abi_ptr addr)
+static inline uint64_t cpu_ldq_le_code(CPUArchState *env, abi_ptr addr)
 {
     CPUState *cs = env_cpu(env);
-    MemOpIdx oi = make_memop_idx(MO_TEUQ, cpu_mmu_index(cs, true));
+    MemOpIdx oi = make_memop_idx(MO_LEUQ, cpu_mmu_index(cs, true));
     return cpu_ldq_code_mmu(env, addr, oi, 0);
 }
 
+static inline uint32_t cpu_lduw_be_code(CPUArchState *env, abi_ptr addr)
+{
+    CPUState *cs = env_cpu(env);
+    MemOpIdx oi = make_memop_idx(MO_BEUW, cpu_mmu_index(cs, true));
+    return cpu_ldw_code_mmu(env, addr, oi, 0);
+}
+
+static inline uint32_t cpu_ldl_be_code(CPUArchState *env, abi_ptr addr)
+{
+    CPUState *cs = env_cpu(env);
+    MemOpIdx oi = make_memop_idx(MO_BEUL, cpu_mmu_index(cs, true));
+    return cpu_ldl_code_mmu(env, addr, oi, 0);
+}
+
+static inline uint64_t cpu_ldq_be_code(CPUArchState *env, abi_ptr addr)
+{
+    CPUState *cs = env_cpu(env);
+    MemOpIdx oi = make_memop_idx(MO_BEUQ, cpu_mmu_index(cs, true));
+    return cpu_ldq_code_mmu(env, addr, oi, 0);
+}
+
+#if TARGET_BIG_ENDIAN
+# define cpu_lduw_code        cpu_lduw_be_code
+# define cpu_ldl_code         cpu_ldl_be_code
+# define cpu_ldq_code         cpu_ldq_be_code
+#else
+# define cpu_lduw_code        cpu_lduw_le_code
+# define cpu_ldl_code         cpu_ldl_le_code
+# define cpu_ldq_code         cpu_ldq_le_code
+#endif
+
 #endif /* ACCEL_TCG_CPU_LDST_H */
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 02/21] target/alpha: Use little-endian variant of cpu_ldl_code()

[Philippe Mathieu-Daudé]

Alpha instructions are always stored in little-endian order.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/alpha/mem_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/alpha/mem_helper.c b/target/alpha/mem_helper.c
index 2113fe33ae2..d04d086b59d 100644
--- a/target/alpha/mem_helper.c
+++ b/target/alpha/mem_helper.c
@@ -30,7 +30,7 @@ static void do_unaligned_access(CPUAlphaState *env, vaddr addr, uintptr_t retadd
     cpu_restore_state(env_cpu(env), retaddr);
 
     pc = env->pc;
-    insn = cpu_ldl_code(env, pc);
+    insn = cpu_ldl_le_code(env, pc);
 
     env->trap_arg0 = addr;
     env->trap_arg1 = insn >> 26;                /* opcode */
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 03/21] target/loongarch: Use little-endian variant of cpu_ldl_code()

[Philippe Mathieu-Daudé]

LoongArch instructions are always stored in little-endian order.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/loongarch/tcg/tcg_cpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/loongarch/tcg/tcg_cpu.c b/target/loongarch/tcg/tcg_cpu.c
index 9d077c56d9d..3f4b5f9258e 100644
--- a/target/loongarch/tcg/tcg_cpu.c
+++ b/target/loongarch/tcg/tcg_cpu.c
@@ -140,7 +140,7 @@ static void loongarch_cpu_do_interrupt(CPUState *cs)
     }
 
     if (update_badinstr) {
-        env->CSR_BADI = cpu_ldl_code(env, env->pc);
+        env->CSR_BADI = cpu_ldl_le_code(env, env->pc);
     }
 
     /* Save PLV and IE */
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 04/21] target/sparc: Use big-endian variant of cpu_ldl_code()

[Philippe Mathieu-Daudé]

SPARC instructions are always stored in big-endian order.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/sparc/int32_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/sparc/int32_helper.c b/target/sparc/int32_helper.c
index b29f693a6bf..a777d17a01c 100644
--- a/target/sparc/int32_helper.c
+++ b/target/sparc/int32_helper.c
@@ -151,7 +151,7 @@ void sparc_cpu_do_interrupt(CPUState *cs)
         if (!env->fsr_qne) {
             env->fsr_qne = FSR_QNE;
             env->fq.s.addr = env->pc;
-            env->fq.s.insn = cpu_ldl_code(env, env->pc);
+            env->fq.s.insn = cpu_ldl_be_code(env, env->pc);
         }
         env->pc = env->npc;
         env->npc = env->npc + 4;
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 05/21] target/s390x: Use big-endian variant of cpu_ld{uw, l}_code()

[Philippe Mathieu-Daudé]

S390x instructions are always stored in big-endian order.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/s390x/tcg/mem_helper.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/target/s390x/tcg/mem_helper.c b/target/s390x/tcg/mem_helper.c
index 24675fc818d..84d901c2008 100644
--- a/target/s390x/tcg/mem_helper.c
+++ b/target/s390x/tcg/mem_helper.c
@@ -2437,7 +2437,7 @@ void HELPER(ex)(CPUS390XState *env, uint32_t ilen, uint64_t r1, uint64_t addr)
         tcg_s390_program_interrupt(env, PGM_SPECIFICATION, GETPC());
     }
 
-    insn = cpu_lduw_code(env, addr);
+    insn = cpu_lduw_be_code(env, addr);
     opc = insn >> 8;
 
     /* Or in the contents of R1[56:63].  */
@@ -2449,10 +2449,10 @@ void HELPER(ex)(CPUS390XState *env, uint32_t ilen, uint64_t r1, uint64_t addr)
     case 2:
         break;
     case 4:
-        insn |= (uint64_t)cpu_lduw_code(env, addr + 2) << 32;
+        insn |= (uint64_t)cpu_lduw_be_code(env, addr + 2) << 32;
         break;
     case 6:
-        insn |= (uint64_t)(uint32_t)cpu_ldl_code(env, addr + 2) << 16;
+        insn |= (uint64_t)(uint32_t)cpu_ldl_be_code(env, addr + 2) << 16;
         break;
     default:
         g_assert_not_reached();
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 06/21] target/riscv: Use little-endian variant of cpu_ld{l, q}_code()

[Philippe Mathieu-Daudé]

RISC-V instructions are always stored in little-endian order
(see "Volume I: RISC-V Unprivileged ISA" document, chapter
'Instruction Encoding Spaces and Prefixes': "instruction fetch
in RISC-V is little-endian").

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/riscv/translate.c  | 2 +-
 target/riscv/zce_helper.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index e1f4dc5ffd0..847481a9b41 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -1182,7 +1182,7 @@ static uint32_t opcode_at(DisasContextBase *dcbase, target_ulong pc)
     CPUState *cpu = ctx->cs;
     CPURISCVState *env = cpu_env(cpu);
 
-    return cpu_ldl_code(env, pc);
+    return cpu_ldl_le_code(env, pc);
 }
 
 #define SS_MMU_INDEX(ctx) (ctx->mem_idx | MMU_IDX_SS_WRITE)
diff --git a/target/riscv/zce_helper.c b/target/riscv/zce_helper.c
index 55221f5f375..992e2f964e0 100644
--- a/target/riscv/zce_helper.c
+++ b/target/riscv/zce_helper.c
@@ -44,10 +44,10 @@ target_ulong HELPER(cm_jalt)(CPURISCVState *env, uint32_t index)
 
     if (xlen == 32) {
         t0 = base + (index << 2);
-        target = cpu_ldl_code(env, t0);
+        target = cpu_ldl_le_code(env, t0);
     } else {
         t0 = base + (index << 3);
-        target = cpu_ldq_code(env, t0);
+        target = cpu_ldq_le_code(env, t0);
     }
 
     return target & ~0x1;
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 07/21] target/ppc: Replace cpu_ldl_code() by explicit endianness variants

[Philippe Mathieu-Daudé]

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/ppc/tcg-excp_helper.c | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

diff --git a/target/ppc/tcg-excp_helper.c b/target/ppc/tcg-excp_helper.c
index edecfb85725..b2d36ab11b1 100644
--- a/target/ppc/tcg-excp_helper.c
+++ b/target/ppc/tcg-excp_helper.c
@@ -424,22 +424,10 @@ G_NORETURN void powerpc_checkstop(CPUPPCState *env, const char *reason)
     cpu_loop_exit_noexc(cs);
 }
 
-/* Return true iff byteswap is needed to load instruction */
-static inline bool insn_need_byteswap(CPUArchState *env)
-{
-    /* SYSTEM builds TARGET_BIG_ENDIAN. Need to swap when MSR[LE] is set */
-    return !!(env->msr & ((target_ulong)1 << MSR_LE));
-}
-
 uint32_t ppc_ldl_code(CPUArchState *env, target_ulong addr)
 {
-    uint32_t insn = cpu_ldl_code(env, addr);
-
-    if (insn_need_byteswap(env)) {
-        insn = bswap32(insn);
-    }
-
-    return insn;
+    return FIELD_EX64(env->msr, MSR, LE) ? cpu_ldl_le_code(env, addr)
+                                         : cpu_ldl_be_code(env, addr);
 }
 
 #if defined(TARGET_PPC64)
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 08/21] target/mips: Replace cpu_ld{uw, l}_code() by explicit endianness variants

[Philippe Mathieu-Daudé]

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/mips/tcg/system/tlb_helper.c | 22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

diff --git a/target/mips/tcg/system/tlb_helper.c b/target/mips/tcg/system/tlb_helper.c
index 566924b079e..415c214c5b4 100644
--- a/target/mips/tcg/system/tlb_helper.c
+++ b/target/mips/tcg/system/tlb_helper.c
@@ -997,18 +997,30 @@ static void set_hflags_for_handler(CPUMIPSState *env)
     }
 }
 
+static uint16_t lduw_code(CPUMIPSState *env, target_ulong addr)
+{
+    return mips_env_is_bigendian(env) ? cpu_lduw_be_code(env, addr)
+                                      : cpu_lduw_le_code(env, addr);
+}
+
+static uint32_t ldl_code(CPUMIPSState *env, target_ulong addr)
+{
+    return mips_env_is_bigendian(env) ? cpu_ldl_be_code(env, addr)
+                                      : cpu_ldl_le_code(env, addr);
+}
+
 static inline void set_badinstr_registers(CPUMIPSState *env)
 {
     if (env->insn_flags & ISA_NANOMIPS32) {
         if (env->CP0_Config3 & (1 << CP0C3_BI)) {
-            uint32_t instr = (cpu_lduw_code(env, env->active_tc.PC)) << 16;
+            uint32_t instr = (lduw_code(env, env->active_tc.PC)) << 16;
             if ((instr & 0x10000000) == 0) {
-                instr |= cpu_lduw_code(env, env->active_tc.PC + 2);
+                instr |= lduw_code(env, env->active_tc.PC + 2);
             }
             env->CP0_BadInstr = instr;
 
             if ((instr & 0xFC000000) == 0x60000000) {
-                instr = cpu_lduw_code(env, env->active_tc.PC + 4) << 16;
+                instr = lduw_code(env, env->active_tc.PC + 4) << 16;
                 env->CP0_BadInstrX = instr;
             }
         }
@@ -1020,11 +1032,11 @@ static inline void set_badinstr_registers(CPUMIPSState *env)
         return;
     }
     if (env->CP0_Config3 & (1 << CP0C3_BI)) {
-        env->CP0_BadInstr = cpu_ldl_code(env, env->active_tc.PC);
+        env->CP0_BadInstr = ldl_code(env, env->active_tc.PC);
     }
     if ((env->CP0_Config3 & (1 << CP0C3_BP)) &&
         (env->hflags & MIPS_HFLAG_BMASK)) {
-        env->CP0_BadInstrP = cpu_ldl_code(env, env->active_tc.PC - 4);
+        env->CP0_BadInstrP = ldl_code(env, env->active_tc.PC - 4);
     }
 }
 
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 09/21] accel/tcg: Remove non-explicit endian cpu_ld*_code() helpers

[Philippe Mathieu-Daudé]

All uses were converted to the explicit cpu_ld*_{be,le}_code()
helpers, no need for the non-explicit versions anymore.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 include/accel/tcg/cpu-ldst.h | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/include/accel/tcg/cpu-ldst.h b/include/accel/tcg/cpu-ldst.h
index e4ec4e7d367..a3125fc9026 100644
--- a/include/accel/tcg/cpu-ldst.h
+++ b/include/accel/tcg/cpu-ldst.h
@@ -523,14 +523,4 @@ static inline uint64_t cpu_ldq_be_code(CPUArchState *env, abi_ptr addr)
     return cpu_ldq_code_mmu(env, addr, oi, 0);
 }
 
-#if TARGET_BIG_ENDIAN
-# define cpu_lduw_code        cpu_lduw_be_code
-# define cpu_ldl_code         cpu_ldl_be_code
-# define cpu_ldq_code         cpu_ldq_be_code
-#else
-# define cpu_lduw_code        cpu_lduw_le_code
-# define cpu_ldl_code         cpu_ldl_le_code
-# define cpu_ldq_code         cpu_ldq_le_code
-#endif
-
 #endif /* ACCEL_TCG_CPU_LDST_H */
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 10/21] target/hexagon: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the Hexagon target using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=hexagon; \
    end=le; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/hexagon/macros.h    | 6 +++---
 target/hexagon/op_helper.c | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/target/hexagon/macros.h b/target/hexagon/macros.h
index 088e5961ab7..6c2862a2320 100644
--- a/target/hexagon/macros.h
+++ b/target/hexagon/macros.h
@@ -519,9 +519,9 @@ static inline TCGv gen_read_ireg(TCGv result, TCGv val, int shift)
 #define fLOAD(NUM, SIZE, SIGN, EA, DST) MEM_LOAD##SIZE##SIGN(DST, EA)
 #else
 #define MEM_LOAD1 cpu_ldub_data_ra
-#define MEM_LOAD2 cpu_lduw_data_ra
-#define MEM_LOAD4 cpu_ldl_data_ra
-#define MEM_LOAD8 cpu_ldq_data_ra
+#define MEM_LOAD2 cpu_lduw_le_data_ra
+#define MEM_LOAD4 cpu_ldl_le_data_ra
+#define MEM_LOAD8 cpu_ldq_le_data_ra
 
 #define fLOAD(NUM, SIZE, SIGN, EA, DST) \
     do { \
diff --git a/target/hexagon/op_helper.c b/target/hexagon/op_helper.c
index e2e80ca7efa..08db1e9c56b 100644
--- a/target/hexagon/op_helper.c
+++ b/target/hexagon/op_helper.c
@@ -77,13 +77,13 @@ static void commit_store(CPUHexagonState *env, int slot_num, uintptr_t ra)
         cpu_stb_data_ra(env, va, env->mem_log_stores[slot_num].data32, ra);
         break;
     case 2:
-        cpu_stw_data_ra(env, va, env->mem_log_stores[slot_num].data32, ra);
+        cpu_stw_le_data_ra(env, va, env->mem_log_stores[slot_num].data32, ra);
         break;
     case 4:
-        cpu_stl_data_ra(env, va, env->mem_log_stores[slot_num].data32, ra);
+        cpu_stl_le_data_ra(env, va, env->mem_log_stores[slot_num].data32, ra);
         break;
     case 8:
-        cpu_stq_data_ra(env, va, env->mem_log_stores[slot_num].data64, ra);
+        cpu_stq_le_data_ra(env, va, env->mem_log_stores[slot_num].data64, ra);
         break;
     default:
         g_assert_not_reached();
-- 
2.51.0

Re: [RFC-PATCH-for-11.0 v3 10/21] target/hexagon: Use little-endian variant of cpu_ld/st_data*()

[Brian Cain]

[-- Attachment #1: Type: text/plain, Size: 2816 bytes --]

On Fri, Nov 21, 2025 at 8:13 PM Philippe Mathieu-Daudé <philmd@linaro.org>
wrote:

> We only build the Hexagon target using little endianness order,
> therefore the cpu_ld/st_data*() definitions expand to the little
> endian declarations. Use the explicit little-endian variants.
>
> Mechanical change running:
>
>   $ tgt=hexagon; \
>     end=le; \
>     for op in data mmuidx_ra; do \
>       for ac in uw sw l q; do \
>         sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
>                   $(git grep -l cpu_ target/${tgt}/); \
>       done;
>       for ac in w l q; do \
>         sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
>                   $(git grep -l cpu_ target/${tgt}/); \
>       done;
>     done
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
>

Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com>


>  target/hexagon/macros.h    | 6 +++---
>  target/hexagon/op_helper.c | 6 +++---
>  2 files changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/target/hexagon/macros.h b/target/hexagon/macros.h
> index 088e5961ab7..6c2862a2320 100644
> --- a/target/hexagon/macros.h
> +++ b/target/hexagon/macros.h
> @@ -519,9 +519,9 @@ static inline TCGv gen_read_ireg(TCGv result, TCGv
> val, int shift)
>  #define fLOAD(NUM, SIZE, SIGN, EA, DST) MEM_LOAD##SIZE##SIGN(DST, EA)
>  #else
>  #define MEM_LOAD1 cpu_ldub_data_ra
> -#define MEM_LOAD2 cpu_lduw_data_ra
> -#define MEM_LOAD4 cpu_ldl_data_ra
> -#define MEM_LOAD8 cpu_ldq_data_ra
> +#define MEM_LOAD2 cpu_lduw_le_data_ra
> +#define MEM_LOAD4 cpu_ldl_le_data_ra
> +#define MEM_LOAD8 cpu_ldq_le_data_ra
>
>  #define fLOAD(NUM, SIZE, SIGN, EA, DST) \
>      do { \
> diff --git a/target/hexagon/op_helper.c b/target/hexagon/op_helper.c
> index e2e80ca7efa..08db1e9c56b 100644
> --- a/target/hexagon/op_helper.c
> +++ b/target/hexagon/op_helper.c
> @@ -77,13 +77,13 @@ static void commit_store(CPUHexagonState *env, int
> slot_num, uintptr_t ra)
>          cpu_stb_data_ra(env, va, env->mem_log_stores[slot_num].data32,
> ra);
>          break;
>      case 2:
> -        cpu_stw_data_ra(env, va, env->mem_log_stores[slot_num].data32,
> ra);
> +        cpu_stw_le_data_ra(env, va, env->mem_log_stores[slot_num].data32,
> ra);
>          break;
>      case 4:
> -        cpu_stl_data_ra(env, va, env->mem_log_stores[slot_num].data32,
> ra);
> +        cpu_stl_le_data_ra(env, va, env->mem_log_stores[slot_num].data32,
> ra);
>          break;
>      case 8:
> -        cpu_stq_data_ra(env, va, env->mem_log_stores[slot_num].data64,
> ra);
> +        cpu_stq_le_data_ra(env, va, env->mem_log_stores[slot_num].data64,
> ra);
>          break;
>      default:
>          g_assert_not_reached();
> --
> 2.51.0
>
>
>

[-- Attachment #2: Type: text/html, Size: 3693 bytes --]

[RFC-PATCH-for-11.0 v3 11/21] target/tricore: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the TriCore target using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=tricore; \
    end=le; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/tricore/op_helper.c | 152 ++++++++++++++++++-------------------
 1 file changed, 76 insertions(+), 76 deletions(-)

diff --git a/target/tricore/op_helper.c b/target/tricore/op_helper.c
index 2c8281a67e0..88b50bc1a9b 100644
--- a/target/tricore/op_helper.c
+++ b/target/tricore/op_helper.c
@@ -2451,84 +2451,84 @@ static bool cdc_zero(uint32_t *psw)
 
 static void save_context_upper(CPUTriCoreState *env, uint32_t ea)
 {
-    cpu_stl_data(env, ea, env->PCXI);
-    cpu_stl_data(env, ea+4, psw_read(env));
-    cpu_stl_data(env, ea+8, env->gpr_a[10]);
-    cpu_stl_data(env, ea+12, env->gpr_a[11]);
-    cpu_stl_data(env, ea+16, env->gpr_d[8]);
-    cpu_stl_data(env, ea+20, env->gpr_d[9]);
-    cpu_stl_data(env, ea+24, env->gpr_d[10]);
-    cpu_stl_data(env, ea+28, env->gpr_d[11]);
-    cpu_stl_data(env, ea+32, env->gpr_a[12]);
-    cpu_stl_data(env, ea+36, env->gpr_a[13]);
-    cpu_stl_data(env, ea+40, env->gpr_a[14]);
-    cpu_stl_data(env, ea+44, env->gpr_a[15]);
-    cpu_stl_data(env, ea+48, env->gpr_d[12]);
-    cpu_stl_data(env, ea+52, env->gpr_d[13]);
-    cpu_stl_data(env, ea+56, env->gpr_d[14]);
-    cpu_stl_data(env, ea+60, env->gpr_d[15]);
+    cpu_stl_le_data(env, ea, env->PCXI);
+    cpu_stl_le_data(env, ea+4, psw_read(env));
+    cpu_stl_le_data(env, ea+8, env->gpr_a[10]);
+    cpu_stl_le_data(env, ea+12, env->gpr_a[11]);
+    cpu_stl_le_data(env, ea+16, env->gpr_d[8]);
+    cpu_stl_le_data(env, ea+20, env->gpr_d[9]);
+    cpu_stl_le_data(env, ea+24, env->gpr_d[10]);
+    cpu_stl_le_data(env, ea+28, env->gpr_d[11]);
+    cpu_stl_le_data(env, ea+32, env->gpr_a[12]);
+    cpu_stl_le_data(env, ea+36, env->gpr_a[13]);
+    cpu_stl_le_data(env, ea+40, env->gpr_a[14]);
+    cpu_stl_le_data(env, ea+44, env->gpr_a[15]);
+    cpu_stl_le_data(env, ea+48, env->gpr_d[12]);
+    cpu_stl_le_data(env, ea+52, env->gpr_d[13]);
+    cpu_stl_le_data(env, ea+56, env->gpr_d[14]);
+    cpu_stl_le_data(env, ea+60, env->gpr_d[15]);
 }
 
 static void save_context_lower(CPUTriCoreState *env, uint32_t ea)
 {
-    cpu_stl_data(env, ea, env->PCXI);
-    cpu_stl_data(env, ea+4, env->gpr_a[11]);
-    cpu_stl_data(env, ea+8, env->gpr_a[2]);
-    cpu_stl_data(env, ea+12, env->gpr_a[3]);
-    cpu_stl_data(env, ea+16, env->gpr_d[0]);
-    cpu_stl_data(env, ea+20, env->gpr_d[1]);
-    cpu_stl_data(env, ea+24, env->gpr_d[2]);
-    cpu_stl_data(env, ea+28, env->gpr_d[3]);
-    cpu_stl_data(env, ea+32, env->gpr_a[4]);
-    cpu_stl_data(env, ea+36, env->gpr_a[5]);
-    cpu_stl_data(env, ea+40, env->gpr_a[6]);
-    cpu_stl_data(env, ea+44, env->gpr_a[7]);
-    cpu_stl_data(env, ea+48, env->gpr_d[4]);
-    cpu_stl_data(env, ea+52, env->gpr_d[5]);
-    cpu_stl_data(env, ea+56, env->gpr_d[6]);
-    cpu_stl_data(env, ea+60, env->gpr_d[7]);
+    cpu_stl_le_data(env, ea, env->PCXI);
+    cpu_stl_le_data(env, ea+4, env->gpr_a[11]);
+    cpu_stl_le_data(env, ea+8, env->gpr_a[2]);
+    cpu_stl_le_data(env, ea+12, env->gpr_a[3]);
+    cpu_stl_le_data(env, ea+16, env->gpr_d[0]);
+    cpu_stl_le_data(env, ea+20, env->gpr_d[1]);
+    cpu_stl_le_data(env, ea+24, env->gpr_d[2]);
+    cpu_stl_le_data(env, ea+28, env->gpr_d[3]);
+    cpu_stl_le_data(env, ea+32, env->gpr_a[4]);
+    cpu_stl_le_data(env, ea+36, env->gpr_a[5]);
+    cpu_stl_le_data(env, ea+40, env->gpr_a[6]);
+    cpu_stl_le_data(env, ea+44, env->gpr_a[7]);
+    cpu_stl_le_data(env, ea+48, env->gpr_d[4]);
+    cpu_stl_le_data(env, ea+52, env->gpr_d[5]);
+    cpu_stl_le_data(env, ea+56, env->gpr_d[6]);
+    cpu_stl_le_data(env, ea+60, env->gpr_d[7]);
 }
 
 static void restore_context_upper(CPUTriCoreState *env, uint32_t ea,
                                   uint32_t *new_PCXI, uint32_t *new_PSW)
 {
-    *new_PCXI = cpu_ldl_data(env, ea);
-    *new_PSW = cpu_ldl_data(env, ea+4);
-    env->gpr_a[10] = cpu_ldl_data(env, ea+8);
-    env->gpr_a[11] = cpu_ldl_data(env, ea+12);
-    env->gpr_d[8]  = cpu_ldl_data(env, ea+16);
-    env->gpr_d[9]  = cpu_ldl_data(env, ea+20);
-    env->gpr_d[10] = cpu_ldl_data(env, ea+24);
-    env->gpr_d[11] = cpu_ldl_data(env, ea+28);
-    env->gpr_a[12] = cpu_ldl_data(env, ea+32);
-    env->gpr_a[13] = cpu_ldl_data(env, ea+36);
-    env->gpr_a[14] = cpu_ldl_data(env, ea+40);
-    env->gpr_a[15] = cpu_ldl_data(env, ea+44);
-    env->gpr_d[12] = cpu_ldl_data(env, ea+48);
-    env->gpr_d[13] = cpu_ldl_data(env, ea+52);
-    env->gpr_d[14] = cpu_ldl_data(env, ea+56);
-    env->gpr_d[15] = cpu_ldl_data(env, ea+60);
+    *new_PCXI = cpu_ldl_le_data(env, ea);
+    *new_PSW = cpu_ldl_le_data(env, ea+4);
+    env->gpr_a[10] = cpu_ldl_le_data(env, ea+8);
+    env->gpr_a[11] = cpu_ldl_le_data(env, ea+12);
+    env->gpr_d[8]  = cpu_ldl_le_data(env, ea+16);
+    env->gpr_d[9]  = cpu_ldl_le_data(env, ea+20);
+    env->gpr_d[10] = cpu_ldl_le_data(env, ea+24);
+    env->gpr_d[11] = cpu_ldl_le_data(env, ea+28);
+    env->gpr_a[12] = cpu_ldl_le_data(env, ea+32);
+    env->gpr_a[13] = cpu_ldl_le_data(env, ea+36);
+    env->gpr_a[14] = cpu_ldl_le_data(env, ea+40);
+    env->gpr_a[15] = cpu_ldl_le_data(env, ea+44);
+    env->gpr_d[12] = cpu_ldl_le_data(env, ea+48);
+    env->gpr_d[13] = cpu_ldl_le_data(env, ea+52);
+    env->gpr_d[14] = cpu_ldl_le_data(env, ea+56);
+    env->gpr_d[15] = cpu_ldl_le_data(env, ea+60);
 }
 
 static void restore_context_lower(CPUTriCoreState *env, uint32_t ea,
                                   uint32_t *ra, uint32_t *pcxi)
 {
-    *pcxi = cpu_ldl_data(env, ea);
-    *ra = cpu_ldl_data(env, ea+4);
-    env->gpr_a[2] = cpu_ldl_data(env, ea+8);
-    env->gpr_a[3] = cpu_ldl_data(env, ea+12);
-    env->gpr_d[0] = cpu_ldl_data(env, ea+16);
-    env->gpr_d[1] = cpu_ldl_data(env, ea+20);
-    env->gpr_d[2] = cpu_ldl_data(env, ea+24);
-    env->gpr_d[3] = cpu_ldl_data(env, ea+28);
-    env->gpr_a[4] = cpu_ldl_data(env, ea+32);
-    env->gpr_a[5] = cpu_ldl_data(env, ea+36);
-    env->gpr_a[6] = cpu_ldl_data(env, ea+40);
-    env->gpr_a[7] = cpu_ldl_data(env, ea+44);
-    env->gpr_d[4] = cpu_ldl_data(env, ea+48);
-    env->gpr_d[5] = cpu_ldl_data(env, ea+52);
-    env->gpr_d[6] = cpu_ldl_data(env, ea+56);
-    env->gpr_d[7] = cpu_ldl_data(env, ea+60);
+    *pcxi = cpu_ldl_le_data(env, ea);
+    *ra = cpu_ldl_le_data(env, ea+4);
+    env->gpr_a[2] = cpu_ldl_le_data(env, ea+8);
+    env->gpr_a[3] = cpu_ldl_le_data(env, ea+12);
+    env->gpr_d[0] = cpu_ldl_le_data(env, ea+16);
+    env->gpr_d[1] = cpu_ldl_le_data(env, ea+20);
+    env->gpr_d[2] = cpu_ldl_le_data(env, ea+24);
+    env->gpr_d[3] = cpu_ldl_le_data(env, ea+28);
+    env->gpr_a[4] = cpu_ldl_le_data(env, ea+32);
+    env->gpr_a[5] = cpu_ldl_le_data(env, ea+36);
+    env->gpr_a[6] = cpu_ldl_le_data(env, ea+40);
+    env->gpr_a[7] = cpu_ldl_le_data(env, ea+44);
+    env->gpr_d[4] = cpu_ldl_le_data(env, ea+48);
+    env->gpr_d[5] = cpu_ldl_le_data(env, ea+52);
+    env->gpr_d[6] = cpu_ldl_le_data(env, ea+56);
+    env->gpr_d[7] = cpu_ldl_le_data(env, ea+60);
 }
 
 void helper_call(CPUTriCoreState *env, uint32_t next_pc)
@@ -2566,7 +2566,7 @@ void helper_call(CPUTriCoreState *env, uint32_t next_pc)
     ea = ((env->FCX & MASK_FCX_FCXS) << 12) +
          ((env->FCX & MASK_FCX_FCXO) << 6);
     /* new_FCX = M(EA, word); */
-    new_FCX = cpu_ldl_data(env, ea);
+    new_FCX = cpu_ldl_le_data(env, ea);
     /* M(EA, 16 * word) = {PCXI, PSW, A[10], A[11], D[8], D[9], D[10], D[11],
                            A[12], A[13], A[14], A[15], D[12], D[13], D[14],
                            D[15]}; */
@@ -2632,7 +2632,7 @@ void helper_ret(CPUTriCoreState *env)
         A[13], A[14], A[15], D[12], D[13], D[14], D[15]} = M(EA, 16 * word); */
     restore_context_upper(env, ea, &new_PCXI, &new_PSW);
     /* M(EA, word) = FCX; */
-    cpu_stl_data(env, ea, env->FCX);
+    cpu_stl_le_data(env, ea, env->FCX);
     /* FCX[19: 0] = PCXI[19: 0]; */
     env->FCX = (env->FCX & 0xfff00000) + (env->PCXI & 0x000fffff);
     /* PCXI = new_PCXI; */
@@ -2662,7 +2662,7 @@ void helper_bisr(CPUTriCoreState *env, uint32_t const9)
     ea = ((env->FCX & 0xf0000) << 12) + ((env->FCX & 0xffff) << 6);
 
     /* new_FCX = M(EA, word); */
-    new_FCX = cpu_ldl_data(env, ea);
+    new_FCX = cpu_ldl_le_data(env, ea);
     /* M(EA, 16 * word) = {PCXI, A[11], A[2], A[3], D[0], D[1], D[2], D[3], A[4]
                            , A[5], A[6], A[7], D[4], D[5], D[6], D[7]}; */
     save_context_lower(env, ea);
@@ -2726,7 +2726,7 @@ void helper_rfe(CPUTriCoreState *env)
       A[13], A[14], A[15], D[12], D[13], D[14], D[15]} = M(EA, 16 * word); */
     restore_context_upper(env, ea, &new_PCXI, &new_PSW);
     /* M(EA, word) = FCX;*/
-    cpu_stl_data(env, ea, env->FCX);
+    cpu_stl_le_data(env, ea, env->FCX);
     /* FCX[19: 0] = PCXI[19: 0]; */
     env->FCX = (env->FCX & 0xfff00000) + (env->PCXI & 0x000fffff);
     /* PCXI = new_PCXI; */
@@ -2744,10 +2744,10 @@ void helper_rfm(CPUTriCoreState *env)
     icr_set_ccpn(env, pcxi_get_pcpn(env));
 
     /* {PCXI, PSW, A[10], A[11]} = M(DCX, 4 * word); */
-    env->PCXI = cpu_ldl_data(env, env->DCX);
-    psw_write(env, cpu_ldl_data(env, env->DCX+4));
-    env->gpr_a[10] = cpu_ldl_data(env, env->DCX+8);
-    env->gpr_a[11] = cpu_ldl_data(env, env->DCX+12);
+    env->PCXI = cpu_ldl_le_data(env, env->DCX);
+    psw_write(env, cpu_ldl_le_data(env, env->DCX+4));
+    env->gpr_a[10] = cpu_ldl_le_data(env, env->DCX+8);
+    env->gpr_a[11] = cpu_ldl_le_data(env, env->DCX+12);
 
     if (tricore_has_feature(env, TRICORE_FEATURE_131)) {
         env->DBGTCR = 0;
@@ -2794,7 +2794,7 @@ void helper_svlcx(CPUTriCoreState *env)
     ea = ((env->FCX & MASK_FCX_FCXS) << 12) +
          ((env->FCX & MASK_FCX_FCXO) << 6);
     /* new_FCX = M(EA, word); */
-    new_FCX = cpu_ldl_data(env, ea);
+    new_FCX = cpu_ldl_le_data(env, ea);
     /* M(EA, 16 * word) = {PCXI, PSW, A[10], A[11], D[8], D[9], D[10], D[11],
                            A[12], A[13], A[14], A[15], D[12], D[13], D[14],
                            D[15]}; */
@@ -2837,7 +2837,7 @@ void helper_svucx(CPUTriCoreState *env)
     ea = ((env->FCX & MASK_FCX_FCXS) << 12) +
          ((env->FCX & MASK_FCX_FCXO) << 6);
     /* new_FCX = M(EA, word); */
-    new_FCX = cpu_ldl_data(env, ea);
+    new_FCX = cpu_ldl_le_data(env, ea);
     /* M(EA, 16 * word) = {PCXI, PSW, A[10], A[11], D[8], D[9], D[10], D[11],
                            A[12], A[13], A[14], A[15], D[12], D[13], D[14],
                            D[15]}; */
@@ -2887,9 +2887,9 @@ void helper_rslcx(CPUTriCoreState *env)
         A[13], A[14], A[15], D[12], D[13], D[14], D[15]} = M(EA, 16 * word); */
     restore_context_lower(env, ea, &env->gpr_a[11], &new_PCXI);
     /* M(EA, word) = FCX; */
-    cpu_stl_data(env, ea, env->FCX);
+    cpu_stl_le_data(env, ea, env->FCX);
     /* M(EA, word) = FCX; */
-    cpu_stl_data(env, ea, env->FCX);
+    cpu_stl_le_data(env, ea, env->FCX);
     /* FCX[19: 0] = PCXI[19: 0]; */
     env->FCX = (env->FCX & 0xfff00000) + (env->PCXI & 0x000fffff);
     /* PCXI = new_PCXI; */
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 12/21] target/rx: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the RX target using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=rx; \
    end=le; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/rx/helper.c    | 14 +++++++-------
 target/rx/op_helper.c |  6 +++---
 2 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/target/rx/helper.c b/target/rx/helper.c
index e9a7aaf610d..83cd491eb4a 100644
--- a/target/rx/helper.c
+++ b/target/rx/helper.c
@@ -68,10 +68,10 @@ void rx_cpu_do_interrupt(CPUState *cs)
         qemu_log_mask(CPU_LOG_INT, "fast interrupt raised\n");
     } else if (cpu_test_interrupt(cs, CPU_INTERRUPT_HARD)) {
         env->isp -= 4;
-        cpu_stl_data(env, env->isp, save_psw);
+        cpu_stl_le_data(env, env->isp, save_psw);
         env->isp -= 4;
-        cpu_stl_data(env, env->isp, env->pc);
-        env->pc = cpu_ldl_data(env, env->intb + env->ack_irq * 4);
+        cpu_stl_le_data(env, env->isp, env->pc);
+        env->pc = cpu_ldl_le_data(env, env->intb + env->ack_irq * 4);
         env->psw_ipl = env->ack_ipl;
         cpu_reset_interrupt(cs, CPU_INTERRUPT_HARD);
         qemu_set_irq(env->ack, env->ack_irq);
@@ -82,14 +82,14 @@ void rx_cpu_do_interrupt(CPUState *cs)
         const char *expname = "unknown exception";
 
         env->isp -= 4;
-        cpu_stl_data(env, env->isp, save_psw);
+        cpu_stl_le_data(env, env->isp, save_psw);
         env->isp -= 4;
-        cpu_stl_data(env, env->isp, env->pc);
+        cpu_stl_le_data(env, env->isp, env->pc);
 
         if (vec < 0x100) {
-            env->pc = cpu_ldl_data(env, 0xffffff80 + vec * 4);
+            env->pc = cpu_ldl_le_data(env, 0xffffff80 + vec * 4);
         } else {
-            env->pc = cpu_ldl_data(env, env->intb + (vec & 0xff) * 4);
+            env->pc = cpu_ldl_le_data(env, env->intb + (vec & 0xff) * 4);
         }
 
         if (vec == 30) {
diff --git a/target/rx/op_helper.c b/target/rx/op_helper.c
index 2b190a4b2cf..ca3e9e85fc7 100644
--- a/target/rx/op_helper.c
+++ b/target/rx/op_helper.c
@@ -217,20 +217,20 @@ void helper_scmpu(CPURXState *env)
 static uint32_t (* const cpu_ldufn[])(CPUArchState *env,
                                      abi_ptr ptr,
                                      uintptr_t retaddr) = {
-    cpu_ldub_data_ra, cpu_lduw_data_ra, cpu_ldl_data_ra,
+    cpu_ldub_data_ra, cpu_lduw_le_data_ra, cpu_ldl_le_data_ra,
 };
 
 static uint32_t (* const cpu_ldfn[])(CPUArchState *env,
                                      abi_ptr ptr,
                                      uintptr_t retaddr) = {
-    cpu_ldub_data_ra, cpu_lduw_data_ra, cpu_ldl_data_ra,
+    cpu_ldub_data_ra, cpu_lduw_le_data_ra, cpu_ldl_le_data_ra,
 };
 
 static void (* const cpu_stfn[])(CPUArchState *env,
                                  abi_ptr ptr,
                                  uint32_t val,
                                  uintptr_t retaddr) = {
-    cpu_stb_data_ra, cpu_stw_data_ra, cpu_stl_data_ra,
+    cpu_stb_data_ra, cpu_stw_le_data_ra, cpu_stl_le_data_ra,
 };
 
 void helper_sstr(CPURXState *env, uint32_t sz)
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 13/21] target/m68k: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the M68k target using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=m68k; \
    end=be; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/m68k/fpu_helper.c | 12 +++---
 target/m68k/op_helper.c  | 88 ++++++++++++++++++++--------------------
 2 files changed, 50 insertions(+), 50 deletions(-)

diff --git a/target/m68k/fpu_helper.c b/target/m68k/fpu_helper.c
index 56012863c85..f49f841d489 100644
--- a/target/m68k/fpu_helper.c
+++ b/target/m68k/fpu_helper.c
@@ -510,8 +510,8 @@ static int cpu_ld_floatx80_ra(CPUM68KState *env, uint32_t addr, FPReg *fp,
     uint32_t high;
     uint64_t low;
 
-    high = cpu_ldl_data_ra(env, addr, ra);
-    low = cpu_ldq_data_ra(env, addr + 4, ra);
+    high = cpu_ldl_be_data_ra(env, addr, ra);
+    low = cpu_ldq_be_data_ra(env, addr + 4, ra);
 
     fp->l.upper = high >> 16;
     fp->l.lower = low;
@@ -522,8 +522,8 @@ static int cpu_ld_floatx80_ra(CPUM68KState *env, uint32_t addr, FPReg *fp,
 static int cpu_st_floatx80_ra(CPUM68KState *env, uint32_t addr, FPReg *fp,
                                uintptr_t ra)
 {
-    cpu_stl_data_ra(env, addr, fp->l.upper << 16, ra);
-    cpu_stq_data_ra(env, addr + 4, fp->l.lower, ra);
+    cpu_stl_be_data_ra(env, addr, fp->l.upper << 16, ra);
+    cpu_stq_be_data_ra(env, addr + 4, fp->l.lower, ra);
 
     return 12;
 }
@@ -533,7 +533,7 @@ static int cpu_ld_float64_ra(CPUM68KState *env, uint32_t addr, FPReg *fp,
 {
     uint64_t val;
 
-    val = cpu_ldq_data_ra(env, addr, ra);
+    val = cpu_ldq_be_data_ra(env, addr, ra);
     fp->d = float64_to_floatx80(*(float64 *)&val, &env->fp_status);
 
     return 8;
@@ -545,7 +545,7 @@ static int cpu_st_float64_ra(CPUM68KState *env, uint32_t addr, FPReg *fp,
     float64 val;
 
     val = floatx80_to_float64(fp->d, &env->fp_status);
-    cpu_stq_data_ra(env, addr, *(uint64_t *)&val, ra);
+    cpu_stq_be_data_ra(env, addr, *(uint64_t *)&val, ra);
 
     return 8;
 }
diff --git a/target/m68k/op_helper.c b/target/m68k/op_helper.c
index e9c20a8e032..eac45669ce7 100644
--- a/target/m68k/op_helper.c
+++ b/target/m68k/op_helper.c
@@ -32,8 +32,8 @@ static void cf_rte(CPUM68KState *env)
     uint32_t fmt;
 
     sp = env->aregs[7];
-    fmt = cpu_ldl_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
-    env->pc = cpu_ldl_mmuidx_ra(env, sp + 4, MMU_KERNEL_IDX, 0);
+    fmt = cpu_ldl_be_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
+    env->pc = cpu_ldl_be_mmuidx_ra(env, sp + 4, MMU_KERNEL_IDX, 0);
     sp |= (fmt >> 28) & 3;
     env->aregs[7] = sp + 8;
 
@@ -48,13 +48,13 @@ static void m68k_rte(CPUM68KState *env)
 
     sp = env->aregs[7];
 throwaway:
-    sr = cpu_lduw_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
+    sr = cpu_lduw_be_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
     sp += 2;
-    env->pc = cpu_ldl_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
+    env->pc = cpu_ldl_be_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
     sp += 4;
     if (m68k_feature(env, M68K_FEATURE_EXCEPTION_FORMAT_VEC)) {
         /*  all except 68000 */
-        fmt = cpu_lduw_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
+        fmt = cpu_lduw_be_mmuidx_ra(env, sp, MMU_KERNEL_IDX, 0);
         sp += 2;
         switch (fmt >> 12) {
         case 0:
@@ -250,12 +250,12 @@ static void cf_interrupt_all(CPUM68KState *env, int is_hw)
     /* ??? This could cause MMU faults.  */
     sp &= ~3;
     sp -= 4;
-    cpu_stl_mmuidx_ra(env, sp, retaddr, MMU_KERNEL_IDX, 0);
+    cpu_stl_be_mmuidx_ra(env, sp, retaddr, MMU_KERNEL_IDX, 0);
     sp -= 4;
-    cpu_stl_mmuidx_ra(env, sp, fmt, MMU_KERNEL_IDX, 0);
+    cpu_stl_be_mmuidx_ra(env, sp, fmt, MMU_KERNEL_IDX, 0);
     env->aregs[7] = sp;
     /* Jump to vector.  */
-    env->pc = cpu_ldl_mmuidx_ra(env, env->vbr + vector, MMU_KERNEL_IDX, 0);
+    env->pc = cpu_ldl_be_mmuidx_ra(env, env->vbr + vector, MMU_KERNEL_IDX, 0);
 
     do_plugin_vcpu_interrupt_cb(cs, retaddr);
 }
@@ -270,24 +270,24 @@ static inline void do_stack_frame(CPUM68KState *env, uint32_t *sp,
         switch (format) {
         case 4:
             *sp -= 4;
-            cpu_stl_mmuidx_ra(env, *sp, env->pc, MMU_KERNEL_IDX, 0);
+            cpu_stl_be_mmuidx_ra(env, *sp, env->pc, MMU_KERNEL_IDX, 0);
             *sp -= 4;
-            cpu_stl_mmuidx_ra(env, *sp, addr, MMU_KERNEL_IDX, 0);
+            cpu_stl_be_mmuidx_ra(env, *sp, addr, MMU_KERNEL_IDX, 0);
             break;
         case 3:
         case 2:
             *sp -= 4;
-            cpu_stl_mmuidx_ra(env, *sp, addr, MMU_KERNEL_IDX, 0);
+            cpu_stl_be_mmuidx_ra(env, *sp, addr, MMU_KERNEL_IDX, 0);
             break;
         }
         *sp -= 2;
-        cpu_stw_mmuidx_ra(env, *sp, (format << 12) + (cs->exception_index << 2),
+        cpu_stw_be_mmuidx_ra(env, *sp, (format << 12) + (cs->exception_index << 2),
                           MMU_KERNEL_IDX, 0);
     }
     *sp -= 4;
-    cpu_stl_mmuidx_ra(env, *sp, retaddr, MMU_KERNEL_IDX, 0);
+    cpu_stl_be_mmuidx_ra(env, *sp, retaddr, MMU_KERNEL_IDX, 0);
     *sp -= 2;
-    cpu_stw_mmuidx_ra(env, *sp, sr, MMU_KERNEL_IDX, 0);
+    cpu_stw_be_mmuidx_ra(env, *sp, sr, MMU_KERNEL_IDX, 0);
 }
 
 static void m68k_interrupt_all(CPUM68KState *env, int is_hw)
@@ -346,49 +346,49 @@ static void m68k_interrupt_all(CPUM68KState *env, int is_hw)
         env->mmu.fault = true;
         /* push data 3 */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* push data 2 */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* push data 1 */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 1 / push data 0 */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 1 address */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 2 data */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 2 address */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 3 data */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 3 address */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, env->mmu.ar, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, env->mmu.ar, MMU_KERNEL_IDX, 0);
         /* fault address */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, env->mmu.ar, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, env->mmu.ar, MMU_KERNEL_IDX, 0);
         /* write back 1 status */
         sp -= 2;
-        cpu_stw_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stw_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 2 status */
         sp -= 2;
-        cpu_stw_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stw_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* write back 3 status */
         sp -= 2;
-        cpu_stw_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
+        cpu_stw_be_mmuidx_ra(env, sp, 0, MMU_KERNEL_IDX, 0);
         /* special status word */
         sp -= 2;
-        cpu_stw_mmuidx_ra(env, sp, env->mmu.ssw, MMU_KERNEL_IDX, 0);
+        cpu_stw_be_mmuidx_ra(env, sp, env->mmu.ssw, MMU_KERNEL_IDX, 0);
         /* effective address */
         sp -= 4;
-        cpu_stl_mmuidx_ra(env, sp, env->mmu.ar, MMU_KERNEL_IDX, 0);
+        cpu_stl_be_mmuidx_ra(env, sp, env->mmu.ar, MMU_KERNEL_IDX, 0);
 
         do_stack_frame(env, &sp, 7, oldsr, 0, env->pc);
         env->mmu.fault = false;
@@ -436,7 +436,7 @@ static void m68k_interrupt_all(CPUM68KState *env, int is_hw)
 
     env->aregs[7] = sp;
     /* Jump to vector.  */
-    env->pc = cpu_ldl_mmuidx_ra(env, env->vbr + vector, MMU_KERNEL_IDX, 0);
+    env->pc = cpu_ldl_be_mmuidx_ra(env, env->vbr + vector, MMU_KERNEL_IDX, 0);
 
     do_plugin_vcpu_interrupt_cb(cs, last_pc);
 }
@@ -784,11 +784,11 @@ void HELPER(cas2w)(CPUM68KState *env, uint32_t regs, uint32_t a1, uint32_t a2)
     int16_t l1, l2;
     uintptr_t ra = GETPC();
 
-    l1 = cpu_lduw_data_ra(env, a1, ra);
-    l2 = cpu_lduw_data_ra(env, a2, ra);
+    l1 = cpu_lduw_be_data_ra(env, a1, ra);
+    l2 = cpu_lduw_be_data_ra(env, a2, ra);
     if (l1 == c1 && l2 == c2) {
-        cpu_stw_data_ra(env, a1, u1, ra);
-        cpu_stw_data_ra(env, a2, u2, ra);
+        cpu_stw_be_data_ra(env, a1, u1, ra);
+        cpu_stw_be_data_ra(env, a2, u2, ra);
     }
 
     if (c1 != l1) {
@@ -845,11 +845,11 @@ static void do_cas2l(CPUM68KState *env, uint32_t regs, uint32_t a1, uint32_t a2,
         }
     } else {
         /* We're executing in a serial context -- no need to be atomic.  */
-        l1 = cpu_ldl_data_ra(env, a1, ra);
-        l2 = cpu_ldl_data_ra(env, a2, ra);
+        l1 = cpu_ldl_be_data_ra(env, a1, ra);
+        l2 = cpu_ldl_be_data_ra(env, a2, ra);
         if (l1 == c1 && l2 == c2) {
-            cpu_stl_data_ra(env, a1, u1, ra);
-            cpu_stl_data_ra(env, a2, u2, ra);
+            cpu_stl_be_data_ra(env, a1, u1, ra);
+            cpu_stl_be_data_ra(env, a2, u2, ra);
         }
     }
 
@@ -951,12 +951,12 @@ static uint64_t bf_load(CPUM68KState *env, uint32_t addr, int blen,
     case 0:
         return cpu_ldub_data_ra(env, addr, ra);
     case 1:
-        return cpu_lduw_data_ra(env, addr, ra);
+        return cpu_lduw_be_data_ra(env, addr, ra);
     case 2:
     case 3:
-        return cpu_ldl_data_ra(env, addr, ra);
+        return cpu_ldl_be_data_ra(env, addr, ra);
     case 4:
-        return cpu_ldq_data_ra(env, addr, ra);
+        return cpu_ldq_be_data_ra(env, addr, ra);
     default:
         g_assert_not_reached();
     }
@@ -970,14 +970,14 @@ static void bf_store(CPUM68KState *env, uint32_t addr, int blen,
         cpu_stb_data_ra(env, addr, data, ra);
         break;
     case 1:
-        cpu_stw_data_ra(env, addr, data, ra);
+        cpu_stw_be_data_ra(env, addr, data, ra);
         break;
     case 2:
     case 3:
-        cpu_stl_data_ra(env, addr, data, ra);
+        cpu_stl_be_data_ra(env, addr, data, ra);
         break;
     case 4:
-        cpu_stq_data_ra(env, addr, data, ra);
+        cpu_stq_be_data_ra(env, addr, data, ra);
         break;
     default:
         g_assert_not_reached();
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 14/21] target/s390x: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the S390x target using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=s390x; \
    end=be; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/s390x/tcg/mem_helper.c | 48 +++++++++++++++++------------------
 target/s390x/tcg/vec_helper.c |  8 +++---
 2 files changed, 28 insertions(+), 28 deletions(-)

diff --git a/target/s390x/tcg/mem_helper.c b/target/s390x/tcg/mem_helper.c
index 84d901c2008..cb990c0c983 100644
--- a/target/s390x/tcg/mem_helper.c
+++ b/target/s390x/tcg/mem_helper.c
@@ -101,7 +101,7 @@ static inline uint64_t cpu_ldusize_data_ra(CPUS390XState *env, uint64_t addr,
     case 1:
         return cpu_ldub_data_ra(env, addr, ra);
     case 2:
-        return cpu_lduw_data_ra(env, addr, ra);
+        return cpu_lduw_be_data_ra(env, addr, ra);
     default:
         abort();
     }
@@ -117,7 +117,7 @@ static inline void cpu_stsize_data_ra(CPUS390XState *env, uint64_t addr,
         cpu_stb_data_ra(env, addr, value, ra);
         break;
     case 2:
-        cpu_stw_data_ra(env, addr, value, ra);
+        cpu_stw_be_data_ra(env, addr, value, ra);
         break;
     default:
         abort();
@@ -865,7 +865,7 @@ void HELPER(srstu)(CPUS390XState *env, uint32_t r1, uint32_t r2)
             env->cc_op = 2;
             return;
         }
-        v = cpu_lduw_data_ra(env, str + len, ra);
+        v = cpu_lduw_be_data_ra(env, str + len, ra);
         if (v == c) {
             /* Character found.  Set R1 to the location; R2 is unmodified.  */
             env->cc_op = 1;
@@ -1022,7 +1022,7 @@ void HELPER(lam)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
     }
 
     for (i = r1;; i = (i + 1) % 16) {
-        env->aregs[i] = cpu_ldl_data_ra(env, a2, ra);
+        env->aregs[i] = cpu_ldl_be_data_ra(env, a2, ra);
         a2 += 4;
 
         if (i == r3) {
@@ -1042,7 +1042,7 @@ void HELPER(stam)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
     }
 
     for (i = r1;; i = (i + 1) % 16) {
-        cpu_stl_data_ra(env, a2, env->aregs[i], ra);
+        cpu_stl_be_data_ra(env, a2, env->aregs[i], ra);
         a2 += 4;
 
         if (i == r3) {
@@ -1363,7 +1363,7 @@ Int128 HELPER(cksm)(CPUS390XState *env, uint64_t r1,
 
     /* Process full words as available.  */
     for (len = 0; len + 4 <= max_len; len += 4, src += 4) {
-        cksm += (uint32_t)cpu_ldl_data_ra(env, src, ra);
+        cksm += (uint32_t)cpu_ldl_be_data_ra(env, src, ra);
     }
 
     switch (max_len - len) {
@@ -1372,11 +1372,11 @@ Int128 HELPER(cksm)(CPUS390XState *env, uint64_t r1,
         len += 1;
         break;
     case 2:
-        cksm += cpu_lduw_data_ra(env, src, ra) << 16;
+        cksm += cpu_lduw_be_data_ra(env, src, ra) << 16;
         len += 2;
         break;
     case 3:
-        cksm += cpu_lduw_data_ra(env, src, ra) << 16;
+        cksm += cpu_lduw_be_data_ra(env, src, ra) << 16;
         cksm += cpu_ldub_data_ra(env, src + 2, ra) << 8;
         len += 3;
         break;
@@ -1955,7 +1955,7 @@ void HELPER(lctlg)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
     }
 
     for (i = r1;; i = (i + 1) % 16) {
-        uint64_t val = cpu_ldq_data_ra(env, src, ra);
+        uint64_t val = cpu_ldq_be_data_ra(env, src, ra);
         if (env->cregs[i] != val && i >= 9 && i <= 11) {
             PERchanged = true;
         }
@@ -1992,7 +1992,7 @@ void HELPER(lctl)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
     }
 
     for (i = r1;; i = (i + 1) % 16) {
-        uint32_t val = cpu_ldl_data_ra(env, src, ra);
+        uint32_t val = cpu_ldl_be_data_ra(env, src, ra);
         uint64_t val64 = deposit64(env->cregs[i], 0, 32, val);
         if ((uint32_t)env->cregs[i] != val && i >= 9 && i <= 11) {
             PERchanged = true;
@@ -2028,7 +2028,7 @@ void HELPER(stctg)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
     }
 
     for (i = r1;; i = (i + 1) % 16) {
-        cpu_stq_data_ra(env, dest, env->cregs[i], ra);
+        cpu_stq_be_data_ra(env, dest, env->cregs[i], ra);
         dest += sizeof(uint64_t);
 
         if (i == r3) {
@@ -2048,7 +2048,7 @@ void HELPER(stctl)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
     }
 
     for (i = r1;; i = (i + 1) % 16) {
-        cpu_stl_data_ra(env, dest, env->cregs[i], ra);
+        cpu_stl_be_data_ra(env, dest, env->cregs[i], ra);
         dest += sizeof(uint32_t);
 
         if (i == r3) {
@@ -2065,7 +2065,7 @@ uint32_t HELPER(testblock)(CPUS390XState *env, uint64_t real_addr)
     real_addr = wrap_address(env, real_addr) & TARGET_PAGE_MASK;
 
     for (i = 0; i < TARGET_PAGE_SIZE; i += 8) {
-        cpu_stq_mmuidx_ra(env, real_addr + i, 0, MMU_REAL_IDX, ra);
+        cpu_stq_be_mmuidx_ra(env, real_addr + i, 0, MMU_REAL_IDX, ra);
     }
 
     return 0;
@@ -2324,11 +2324,11 @@ void HELPER(idte)(CPUS390XState *env, uint64_t r1, uint64_t r2, uint32_t m4)
         for (i = 0; i < entries; i++) {
             /* addresses are not wrapped in 24/31bit mode but table index is */
             raddr = table + ((index + i) & 0x7ff) * sizeof(entry);
-            entry = cpu_ldq_mmuidx_ra(env, raddr, MMU_REAL_IDX, ra);
+            entry = cpu_ldq_be_mmuidx_ra(env, raddr, MMU_REAL_IDX, ra);
             if (!(entry & REGION_ENTRY_I)) {
                 /* we are allowed to not store if already invalid */
                 entry |= REGION_ENTRY_I;
-                cpu_stq_mmuidx_ra(env, raddr, entry, MMU_REAL_IDX, ra);
+                cpu_stq_be_mmuidx_ra(env, raddr, entry, MMU_REAL_IDX, ra);
             }
         }
     }
@@ -2355,9 +2355,9 @@ void HELPER(ipte)(CPUS390XState *env, uint64_t pto, uint64_t vaddr,
     pte_addr += VADDR_PAGE_TX(vaddr) * 8;
 
     /* Mark the page table entry as invalid */
-    pte = cpu_ldq_mmuidx_ra(env, pte_addr, MMU_REAL_IDX, ra);
+    pte = cpu_ldq_be_mmuidx_ra(env, pte_addr, MMU_REAL_IDX, ra);
     pte |= PAGE_ENTRY_I;
-    cpu_stq_mmuidx_ra(env, pte_addr, pte, MMU_REAL_IDX, ra);
+    cpu_stq_be_mmuidx_ra(env, pte_addr, pte, MMU_REAL_IDX, ra);
 
     /* XXX we exploit the fact that Linux passes the exact virtual
        address here - it's not obliged to! */
@@ -2695,7 +2695,7 @@ static int decode_utf16(CPUS390XState *env, uint64_t addr, uint64_t ilen,
     if (ilen < 2) {
         return 0;
     }
-    s0 = cpu_lduw_data_ra(env, addr, ra);
+    s0 = cpu_lduw_be_data_ra(env, addr, ra);
     if ((s0 & 0xfc00) != 0xd800) {
         /* one word character */
         l = 2;
@@ -2706,7 +2706,7 @@ static int decode_utf16(CPUS390XState *env, uint64_t addr, uint64_t ilen,
         if (ilen < 4) {
             return 0;
         }
-        s1 = cpu_lduw_data_ra(env, addr + 2, ra);
+        s1 = cpu_lduw_be_data_ra(env, addr + 2, ra);
         c = extract32(s0, 6, 4) + 1;
         c = (c << 6) | (s0 & 0x3f);
         c = (c << 10) | (s1 & 0x3ff);
@@ -2730,7 +2730,7 @@ static int decode_utf32(CPUS390XState *env, uint64_t addr, uint64_t ilen,
     if (ilen < 4) {
         return 0;
     }
-    c = cpu_ldl_data_ra(env, addr, ra);
+    c = cpu_ldl_be_data_ra(env, addr, ra);
     if ((c >= 0xd800 && c <= 0xdbff) || c > 0x10ffff) {
         /* invalid unicode character */
         return 2;
@@ -2792,7 +2792,7 @@ static int encode_utf16(CPUS390XState *env, uint64_t addr, uint64_t ilen,
         if (ilen < 2) {
             return 1;
         }
-        cpu_stw_data_ra(env, addr, c, ra);
+        cpu_stw_be_data_ra(env, addr, c, ra);
         *olen = 2;
     } else {
         /* two word character */
@@ -2802,8 +2802,8 @@ static int encode_utf16(CPUS390XState *env, uint64_t addr, uint64_t ilen,
         d1 = 0xdc00 | extract32(c, 0, 10);
         d0 = 0xd800 | extract32(c, 10, 6);
         d0 = deposit32(d0, 6, 4, extract32(c, 16, 5) - 1);
-        cpu_stw_data_ra(env, addr + 0, d0, ra);
-        cpu_stw_data_ra(env, addr + 2, d1, ra);
+        cpu_stw_be_data_ra(env, addr + 0, d0, ra);
+        cpu_stw_be_data_ra(env, addr + 2, d1, ra);
         *olen = 4;
     }
 
@@ -2816,7 +2816,7 @@ static int encode_utf32(CPUS390XState *env, uint64_t addr, uint64_t ilen,
     if (ilen < 4) {
         return 1;
     }
-    cpu_stl_data_ra(env, addr, c, ra);
+    cpu_stl_be_data_ra(env, addr, c, ra);
     *olen = 4;
     return -1;
 }
diff --git a/target/s390x/tcg/vec_helper.c b/target/s390x/tcg/vec_helper.c
index 46ec4a947dd..304745c971b 100644
--- a/target/s390x/tcg/vec_helper.c
+++ b/target/s390x/tcg/vec_helper.c
@@ -45,9 +45,9 @@ void HELPER(vll)(CPUS390XState *env, void *v1, uint64_t addr, uint64_t bytes)
     if (likely(bytes >= 16)) {
         uint64_t t0, t1;
 
-        t0 = cpu_ldq_data_ra(env, addr, GETPC());
+        t0 = cpu_ldq_be_data_ra(env, addr, GETPC());
         addr = wrap_address(env, addr + 8);
-        t1 = cpu_ldq_data_ra(env, addr, GETPC());
+        t1 = cpu_ldq_be_data_ra(env, addr, GETPC());
         s390_vec_write_element64(v1, 0, t0);
         s390_vec_write_element64(v1, 1, t1);
     } else {
@@ -195,9 +195,9 @@ void HELPER(vstl)(CPUS390XState *env, const void *v1, uint64_t addr,
     probe_write_access(env, addr, MIN(bytes, 16), GETPC());
 
     if (likely(bytes >= 16)) {
-        cpu_stq_data_ra(env, addr, s390_vec_read_element64(v1, 0), GETPC());
+        cpu_stq_be_data_ra(env, addr, s390_vec_read_element64(v1, 0), GETPC());
         addr = wrap_address(env, addr + 8);
-        cpu_stq_data_ra(env, addr, s390_vec_read_element64(v1, 1), GETPC());
+        cpu_stq_be_data_ra(env, addr, s390_vec_read_element64(v1, 1), GETPC());
     } else {
         int i;
 
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 15/21] target/sparc: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the SPARC targets using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=sparc; \
    end=be; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/sparc/ldst_helper.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/target/sparc/ldst_helper.c b/target/sparc/ldst_helper.c
index 2c63eb9e036..bd0257d313a 100644
--- a/target/sparc/ldst_helper.c
+++ b/target/sparc/ldst_helper.c
@@ -1228,13 +1228,13 @@ uint64_t helper_ld_asi(CPUSPARCState *env, target_ulong addr,
             ret = cpu_ldub_data(env, addr);
             break;
         case 2:
-            ret = cpu_lduw_data(env, addr);
+            ret = cpu_lduw_be_data(env, addr);
             break;
         case 4:
-            ret = cpu_ldl_data(env, addr);
+            ret = cpu_ldl_be_data(env, addr);
             break;
         case 8:
-            ret = cpu_ldq_data(env, addr);
+            ret = cpu_ldq_be_data(env, addr);
             break;
         default:
             g_assert_not_reached();
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 16/21] target/i386: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the X86 targets using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=i386; \
    end=le; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/i386/ops_sse.h                | 12 +++----
 target/i386/tcg/seg_helper.h         | 12 +++----
 linux-user/vm86.c                    |  4 +--
 target/i386/tcg/mem_helper.c         |  8 ++---
 target/i386/tcg/mpx_helper.c         | 28 ++++++++--------
 target/i386/tcg/seg_helper.c         | 16 +++++-----
 target/i386/tcg/system/excp_helper.c |  8 ++---
 target/i386/tcg/system/svm_helper.c  | 48 ++++++++++++++--------------
 8 files changed, 68 insertions(+), 68 deletions(-)

diff --git a/target/i386/ops_sse.h b/target/i386/ops_sse.h
index a2e4d480399..213db53382c 100644
--- a/target/i386/ops_sse.h
+++ b/target/i386/ops_sse.h
@@ -2326,7 +2326,7 @@ void glue(helper_vpmaskmovd_st, SUFFIX)(CPUX86State *env,
 
     for (i = 0; i < (2 << SHIFT); i++) {
         if (v->L(i) >> 31) {
-            cpu_stl_data_ra(env, a0 + i * 4, s->L(i), GETPC());
+            cpu_stl_le_data_ra(env, a0 + i * 4, s->L(i), GETPC());
         }
     }
 }
@@ -2338,7 +2338,7 @@ void glue(helper_vpmaskmovq_st, SUFFIX)(CPUX86State *env,
 
     for (i = 0; i < (1 << SHIFT); i++) {
         if (v->Q(i) >> 63) {
-            cpu_stq_data_ra(env, a0 + i * 8, s->Q(i), GETPC());
+            cpu_stq_le_data_ra(env, a0 + i * 8, s->Q(i), GETPC());
         }
     }
 }
@@ -2369,7 +2369,7 @@ void glue(helper_vpgatherdd, SUFFIX)(CPUX86State *env,
         if (v->L(i) >> 31) {
             target_ulong addr = a0
                 + ((target_ulong)(int32_t)s->L(i) << scale);
-            d->L(i) = cpu_ldl_data_ra(env, addr, GETPC());
+            d->L(i) = cpu_ldl_le_data_ra(env, addr, GETPC());
         }
         v->L(i) = 0;
     }
@@ -2383,7 +2383,7 @@ void glue(helper_vpgatherdq, SUFFIX)(CPUX86State *env,
         if (v->Q(i) >> 63) {
             target_ulong addr = a0
                 + ((target_ulong)(int32_t)s->L(i) << scale);
-            d->Q(i) = cpu_ldq_data_ra(env, addr, GETPC());
+            d->Q(i) = cpu_ldq_le_data_ra(env, addr, GETPC());
         }
         v->Q(i) = 0;
     }
@@ -2397,7 +2397,7 @@ void glue(helper_vpgatherqd, SUFFIX)(CPUX86State *env,
         if (v->L(i) >> 31) {
             target_ulong addr = a0
                 + ((target_ulong)(int64_t)s->Q(i) << scale);
-            d->L(i) = cpu_ldl_data_ra(env, addr, GETPC());
+            d->L(i) = cpu_ldl_le_data_ra(env, addr, GETPC());
         }
         v->L(i) = 0;
     }
@@ -2415,7 +2415,7 @@ void glue(helper_vpgatherqq, SUFFIX)(CPUX86State *env,
         if (v->Q(i) >> 63) {
             target_ulong addr = a0
                 + ((target_ulong)(int64_t)s->Q(i) << scale);
-            d->Q(i) = cpu_ldq_data_ra(env, addr, GETPC());
+            d->Q(i) = cpu_ldq_le_data_ra(env, addr, GETPC());
         }
         v->Q(i) = 0;
     }
diff --git a/target/i386/tcg/seg_helper.h b/target/i386/tcg/seg_helper.h
index ea98e1a98ed..20ce47d62d9 100644
--- a/target/i386/tcg/seg_helper.h
+++ b/target/i386/tcg/seg_helper.h
@@ -40,18 +40,18 @@ int cpu_mmu_index_kernel(CPUX86State *env);
  * and use *_mmuidx_ra directly.
  */
 #define cpu_lduw_kernel_ra(e, p, r) \
-    cpu_lduw_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
+    cpu_lduw_le_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
 #define cpu_ldl_kernel_ra(e, p, r) \
-    cpu_ldl_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
+    cpu_ldl_le_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
 #define cpu_ldq_kernel_ra(e, p, r) \
-    cpu_ldq_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
+    cpu_ldq_le_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
 
 #define cpu_stw_kernel_ra(e, p, v, r) \
-    cpu_stw_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
+    cpu_stw_le_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
 #define cpu_stl_kernel_ra(e, p, v, r) \
-    cpu_stl_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
+    cpu_stl_le_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
 #define cpu_stq_kernel_ra(e, p, v, r) \
-    cpu_stq_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
+    cpu_stq_le_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
 
 #define cpu_lduw_kernel(e, p)    cpu_lduw_kernel_ra(e, p, 0)
 #define cpu_ldl_kernel(e, p)     cpu_ldl_kernel_ra(e, p, 0)
diff --git a/linux-user/vm86.c b/linux-user/vm86.c
index 5091d53fb84..4e120875a06 100644
--- a/linux-user/vm86.c
+++ b/linux-user/vm86.c
@@ -44,7 +44,7 @@ static inline int is_revectored(int nr, struct target_revectored_struct *bitmap)
 static inline void vm_putw(CPUX86State *env, uint32_t segptr,
                            unsigned int reg16, unsigned int val)
 {
-    cpu_stw_data(env, segptr + (reg16 & 0xffff), val);
+    cpu_stw_le_data(env, segptr + (reg16 & 0xffff), val);
 }
 
 void save_v86_state(CPUX86State *env)
@@ -157,7 +157,7 @@ static void do_int(CPUX86State *env, int intno)
                                        &ts->vm86plus.int21_revectored))
         goto cannot_handle;
     int_addr = (intno << 2);
-    segoffs = cpu_ldl_data(env, int_addr);
+    segoffs = cpu_ldl_le_data(env, int_addr);
     if ((segoffs >> 16) == TARGET_BIOSSEG)
         goto cannot_handle;
     LOG_VM86("VM86: emulating int 0x%x. CS:IP=%04x:%04x\n",
diff --git a/target/i386/tcg/mem_helper.c b/target/i386/tcg/mem_helper.c
index 9e7c2d80293..c15a32d60ad 100644
--- a/target/i386/tcg/mem_helper.c
+++ b/target/i386/tcg/mem_helper.c
@@ -30,8 +30,8 @@ void helper_boundw(CPUX86State *env, target_ulong a0, int v)
 {
     int low, high;
 
-    low = cpu_ldsw_data_ra(env, a0, GETPC());
-    high = cpu_ldsw_data_ra(env, a0 + 2, GETPC());
+    low = cpu_ldsw_le_data_ra(env, a0, GETPC());
+    high = cpu_ldsw_le_data_ra(env, a0 + 2, GETPC());
     v = (int16_t)v;
     if (v < low || v > high) {
         if (env->hflags & HF_MPX_EN_MASK) {
@@ -45,8 +45,8 @@ void helper_boundl(CPUX86State *env, target_ulong a0, int v)
 {
     int low, high;
 
-    low = cpu_ldl_data_ra(env, a0, GETPC());
-    high = cpu_ldl_data_ra(env, a0 + 4, GETPC());
+    low = cpu_ldl_le_data_ra(env, a0, GETPC());
+    high = cpu_ldl_le_data_ra(env, a0 + 4, GETPC());
     if (v < low || v > high) {
         if (env->hflags & HF_MPX_EN_MASK) {
             env->bndcs_regs.sts = 0;
diff --git a/target/i386/tcg/mpx_helper.c b/target/i386/tcg/mpx_helper.c
index fa8abcc4820..73d33bf5e4a 100644
--- a/target/i386/tcg/mpx_helper.c
+++ b/target/i386/tcg/mpx_helper.c
@@ -44,7 +44,7 @@ static uint64_t lookup_bte64(CPUX86State *env, uint64_t base, uintptr_t ra)
     }
 
     bde = (extract64(base, 20, 28) << 3) + (extract64(bndcsr, 20, 44) << 12);
-    bt = cpu_ldq_data_ra(env, bde, ra);
+    bt = cpu_ldq_le_data_ra(env, bde, ra);
     if ((bt & 1) == 0) {
         env->bndcs_regs.sts = bde | 2;
         raise_exception_ra(env, EXCP05_BOUND, ra);
@@ -64,7 +64,7 @@ static uint32_t lookup_bte32(CPUX86State *env, uint32_t base, uintptr_t ra)
     }
 
     bde = (extract32(base, 12, 20) << 2) + (bndcsr & TARGET_PAGE_MASK);
-    bt = cpu_ldl_data_ra(env, bde, ra);
+    bt = cpu_ldl_le_data_ra(env, bde, ra);
     if ((bt & 1) == 0) {
         env->bndcs_regs.sts = bde | 2;
         raise_exception_ra(env, EXCP05_BOUND, ra);
@@ -79,9 +79,9 @@ uint64_t helper_bndldx64(CPUX86State *env, target_ulong base, target_ulong ptr)
     uint64_t bte, lb, ub, pt;
 
     bte = lookup_bte64(env, base, ra);
-    lb = cpu_ldq_data_ra(env, bte, ra);
-    ub = cpu_ldq_data_ra(env, bte + 8, ra);
-    pt = cpu_ldq_data_ra(env, bte + 16, ra);
+    lb = cpu_ldq_le_data_ra(env, bte, ra);
+    ub = cpu_ldq_le_data_ra(env, bte + 8, ra);
+    pt = cpu_ldq_le_data_ra(env, bte + 16, ra);
 
     if (pt != ptr) {
         lb = ub = 0;
@@ -96,9 +96,9 @@ uint64_t helper_bndldx32(CPUX86State *env, target_ulong base, target_ulong ptr)
     uint32_t bte, lb, ub, pt;
 
     bte = lookup_bte32(env, base, ra);
-    lb = cpu_ldl_data_ra(env, bte, ra);
-    ub = cpu_ldl_data_ra(env, bte + 4, ra);
-    pt = cpu_ldl_data_ra(env, bte + 8, ra);
+    lb = cpu_ldl_le_data_ra(env, bte, ra);
+    ub = cpu_ldl_le_data_ra(env, bte + 4, ra);
+    pt = cpu_ldl_le_data_ra(env, bte + 8, ra);
 
     if (pt != ptr) {
         lb = ub = 0;
@@ -113,9 +113,9 @@ void helper_bndstx64(CPUX86State *env, target_ulong base, target_ulong ptr,
     uint64_t bte;
 
     bte = lookup_bte64(env, base, ra);
-    cpu_stq_data_ra(env, bte, lb, ra);
-    cpu_stq_data_ra(env, bte + 8, ub, ra);
-    cpu_stq_data_ra(env, bte + 16, ptr, ra);
+    cpu_stq_le_data_ra(env, bte, lb, ra);
+    cpu_stq_le_data_ra(env, bte + 8, ub, ra);
+    cpu_stq_le_data_ra(env, bte + 16, ptr, ra);
 }
 
 void helper_bndstx32(CPUX86State *env, target_ulong base, target_ulong ptr,
@@ -125,9 +125,9 @@ void helper_bndstx32(CPUX86State *env, target_ulong base, target_ulong ptr,
     uint32_t bte;
 
     bte = lookup_bte32(env, base, ra);
-    cpu_stl_data_ra(env, bte, lb, ra);
-    cpu_stl_data_ra(env, bte + 4, ub, ra);
-    cpu_stl_data_ra(env, bte + 8, ptr, ra);
+    cpu_stl_le_data_ra(env, bte, lb, ra);
+    cpu_stl_le_data_ra(env, bte + 4, ub, ra);
+    cpu_stl_le_data_ra(env, bte + 8, ptr, ra);
 }
 
 void helper_bnd_jmp(CPUX86State *env)
diff --git a/target/i386/tcg/seg_helper.c b/target/i386/tcg/seg_helper.c
index 667b1c38696..abce4331dfb 100644
--- a/target/i386/tcg/seg_helper.c
+++ b/target/i386/tcg/seg_helper.c
@@ -65,20 +65,20 @@ typedef struct StackAccess
 static void pushw(StackAccess *sa, uint16_t val)
 {
     sa->sp -= 2;
-    cpu_stw_mmuidx_ra(sa->env, sa->ss_base + (sa->sp & sa->sp_mask),
+    cpu_stw_le_mmuidx_ra(sa->env, sa->ss_base + (sa->sp & sa->sp_mask),
                       val, sa->mmu_index, sa->ra);
 }
 
 static void pushl(StackAccess *sa, uint32_t val)
 {
     sa->sp -= 4;
-    cpu_stl_mmuidx_ra(sa->env, sa->ss_base + (sa->sp & sa->sp_mask),
+    cpu_stl_le_mmuidx_ra(sa->env, sa->ss_base + (sa->sp & sa->sp_mask),
                       val, sa->mmu_index, sa->ra);
 }
 
 static uint16_t popw(StackAccess *sa)
 {
-    uint16_t ret = cpu_lduw_mmuidx_ra(sa->env,
+    uint16_t ret = cpu_lduw_le_mmuidx_ra(sa->env,
                                       sa->ss_base + (sa->sp & sa->sp_mask),
                                       sa->mmu_index, sa->ra);
     sa->sp += 2;
@@ -87,7 +87,7 @@ static uint16_t popw(StackAccess *sa)
 
 static uint32_t popl(StackAccess *sa)
 {
-    uint32_t ret = cpu_ldl_mmuidx_ra(sa->env,
+    uint32_t ret = cpu_ldl_le_mmuidx_ra(sa->env,
                                      sa->ss_base + (sa->sp & sa->sp_mask),
                                      sa->mmu_index, sa->ra);
     sa->sp += 4;
@@ -905,12 +905,12 @@ static void do_interrupt_protected(CPUX86State *env, int intno, int is_int,
 static void pushq(StackAccess *sa, uint64_t val)
 {
     sa->sp -= 8;
-    cpu_stq_mmuidx_ra(sa->env, sa->sp, val, sa->mmu_index, sa->ra);
+    cpu_stq_le_mmuidx_ra(sa->env, sa->sp, val, sa->mmu_index, sa->ra);
 }
 
 static uint64_t popq(StackAccess *sa)
 {
-    uint64_t ret = cpu_ldq_mmuidx_ra(sa->env, sa->sp, sa->mmu_index, sa->ra);
+    uint64_t ret = cpu_ldq_le_mmuidx_ra(sa->env, sa->sp, sa->mmu_index, sa->ra);
     sa->sp += 8;
     return ret;
 }
@@ -1887,7 +1887,7 @@ void helper_lcall_protected(CPUX86State *env, int new_cs, target_ulong new_eip,
                 pushl(&sa, env->segs[R_SS].selector);
                 pushl(&sa, env->regs[R_ESP]);
                 for (i = param_count - 1; i >= 0; i--) {
-                    val = cpu_ldl_data_ra(env,
+                    val = cpu_ldl_le_data_ra(env,
                                           old_ssp + ((env->regs[R_ESP] + i * 4) & old_sp_mask),
                                           GETPC());
                     pushl(&sa, val);
@@ -1896,7 +1896,7 @@ void helper_lcall_protected(CPUX86State *env, int new_cs, target_ulong new_eip,
                 pushw(&sa, env->segs[R_SS].selector);
                 pushw(&sa, env->regs[R_ESP]);
                 for (i = param_count - 1; i >= 0; i--) {
-                    val = cpu_lduw_data_ra(env,
+                    val = cpu_lduw_le_data_ra(env,
                                            old_ssp + ((env->regs[R_ESP] + i * 2) & old_sp_mask),
                                            GETPC());
                     pushw(&sa, val);
diff --git a/target/i386/tcg/system/excp_helper.c b/target/i386/tcg/system/excp_helper.c
index f622b5d588e..d7ea77c8558 100644
--- a/target/i386/tcg/system/excp_helper.c
+++ b/target/i386/tcg/system/excp_helper.c
@@ -90,7 +90,7 @@ static inline uint32_t ptw_ldl(const PTETranslate *in, uint64_t ra)
     if (likely(in->haddr)) {
         return ldl_p(in->haddr);
     }
-    return cpu_ldl_mmuidx_ra(in->env, in->gaddr, in->ptw_idx, ra);
+    return cpu_ldl_le_mmuidx_ra(in->env, in->gaddr, in->ptw_idx, ra);
 }
 
 static inline uint64_t ptw_ldq(const PTETranslate *in, uint64_t ra)
@@ -98,7 +98,7 @@ static inline uint64_t ptw_ldq(const PTETranslate *in, uint64_t ra)
     if (likely(in->haddr)) {
         return ldq_p(in->haddr);
     }
-    return cpu_ldq_mmuidx_ra(in->env, in->gaddr, in->ptw_idx, ra);
+    return cpu_ldq_le_mmuidx_ra(in->env, in->gaddr, in->ptw_idx, ra);
 }
 
 /*
@@ -116,9 +116,9 @@ static bool ptw_setl_slow(const PTETranslate *in, uint32_t old, uint32_t new)
     cpu_exec_end(cpu);
     /* Does x86 really perform a rmw cycle on mmio for ptw? */
     start_exclusive();
-    cmp = cpu_ldl_mmuidx_ra(in->env, in->gaddr, in->ptw_idx, 0);
+    cmp = cpu_ldl_le_mmuidx_ra(in->env, in->gaddr, in->ptw_idx, 0);
     if (cmp == old) {
-        cpu_stl_mmuidx_ra(in->env, in->gaddr, new, in->ptw_idx, 0);
+        cpu_stl_le_mmuidx_ra(in->env, in->gaddr, new, in->ptw_idx, 0);
     }
     end_exclusive();
     cpu_exec_start(cpu);
diff --git a/target/i386/tcg/system/svm_helper.c b/target/i386/tcg/system/svm_helper.c
index 505788b0e26..811c0be2023 100644
--- a/target/i386/tcg/system/svm_helper.c
+++ b/target/i386/tcg/system/svm_helper.c
@@ -30,13 +30,13 @@
 static void svm_save_seg(CPUX86State *env, int mmu_idx, hwaddr addr,
                          const SegmentCache *sc)
 {
-    cpu_stw_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, selector),
+    cpu_stw_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, selector),
                       sc->selector, mmu_idx, 0);
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, base),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, base),
                       sc->base, mmu_idx, 0);
-    cpu_stl_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, limit),
+    cpu_stl_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, limit),
                       sc->limit, mmu_idx, 0);
-    cpu_stw_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, attrib),
+    cpu_stw_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, attrib),
                       ((sc->flags >> 8) & 0xff)
                       | ((sc->flags >> 12) & 0x0f00),
                       mmu_idx, 0);
@@ -58,16 +58,16 @@ static void svm_load_seg(CPUX86State *env, int mmu_idx, hwaddr addr,
     unsigned int flags;
 
     sc->selector =
-        cpu_lduw_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, selector),
+        cpu_lduw_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, selector),
                            mmu_idx, 0);
     sc->base =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, base),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, base),
                           mmu_idx, 0);
     sc->limit =
-        cpu_ldl_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, limit),
+        cpu_ldl_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, limit),
                           mmu_idx, 0);
     flags =
-        cpu_lduw_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, attrib),
+        cpu_lduw_le_mmuidx_ra(env, addr + offsetof(struct vmcb_seg, attrib),
                            mmu_idx, 0);
     sc->flags = ((flags & 0xff) << 8) | ((flags & 0x0f00) << 12);
 
@@ -507,31 +507,31 @@ void helper_vmload(CPUX86State *env, int aflag)
 
 #ifdef TARGET_X86_64
     env->kernelgsbase =
-        cpu_ldq_mmuidx_ra(env,
+        cpu_ldq_le_mmuidx_ra(env,
                           addr + offsetof(struct vmcb, save.kernel_gs_base),
                           mmu_idx, 0);
     env->lstar =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.lstar),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.lstar),
                           mmu_idx, 0);
     env->cstar =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.cstar),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.cstar),
                           mmu_idx, 0);
     env->fmask =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sfmask),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sfmask),
                           mmu_idx, 0);
     svm_canonicalization(env, &env->kernelgsbase);
 #endif
     env->star =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.star),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.star),
                           mmu_idx, 0);
     env->sysenter_cs =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_cs),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_cs),
                           mmu_idx, 0);
     env->sysenter_esp =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_esp),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_esp),
                           mmu_idx, 0);
     env->sysenter_eip =
-        cpu_ldq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_eip),
+        cpu_ldq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_eip),
                           mmu_idx, 0);
 }
 
@@ -567,22 +567,22 @@ void helper_vmsave(CPUX86State *env, int aflag)
                  &env->ldt);
 
 #ifdef TARGET_X86_64
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.kernel_gs_base),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.kernel_gs_base),
                       env->kernelgsbase, mmu_idx, 0);
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.lstar),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.lstar),
                       env->lstar, mmu_idx, 0);
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.cstar),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.cstar),
                       env->cstar, mmu_idx, 0);
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sfmask),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sfmask),
                       env->fmask, mmu_idx, 0);
 #endif
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.star),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.star),
                       env->star, mmu_idx, 0);
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_cs),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_cs),
                       env->sysenter_cs, mmu_idx, 0);
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_esp),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_esp),
                       env->sysenter_esp, mmu_idx, 0);
-    cpu_stq_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_eip),
+    cpu_stq_le_mmuidx_ra(env, addr + offsetof(struct vmcb, save.sysenter_eip),
                       env->sysenter_eip, mmu_idx, 0);
 }
 
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 17/21] target/hppa: Use little-endian variant of cpu_ld/st_data*()

[Philippe Mathieu-Daudé]

We only build the HPPA target using little endianness order,
therefore the cpu_ld/st_data*() definitions expand to the little
endian declarations. Use the explicit little-endian variants.

Mechanical change running:

  $ tgt=hppa; \
    end=be; \
    for op in data mmuidx_ra; do \
      for ac in uw sw l q; do \
        sed -i -e "s/cpu_ld${ac}_${op}/cpu_ld${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
      for ac in w l q; do \
        sed -i -e "s/cpu_st${ac}_${op}/cpu_st${ac}_${end}_${op}/" \
                  $(git grep -l cpu_ target/${tgt}/); \
      done;
    done

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/hppa/op_helper.c | 44 ++++++++++++++++++++---------------------
 1 file changed, 22 insertions(+), 22 deletions(-)

diff --git a/target/hppa/op_helper.c b/target/hppa/op_helper.c
index 0458378abb2..65faf03cd0a 100644
--- a/target/hppa/op_helper.c
+++ b/target/hppa/op_helper.c
@@ -107,7 +107,7 @@ static void do_stby_b(CPUHPPAState *env, target_ulong addr, target_ulong val,
         cpu_stb_data_ra(env, addr, val, ra);
         break;
     case 2:
-        cpu_stw_data_ra(env, addr, val, ra);
+        cpu_stw_be_data_ra(env, addr, val, ra);
         break;
     case 1:
         /* The 3 byte store must appear atomic.  */
@@ -115,11 +115,11 @@ static void do_stby_b(CPUHPPAState *env, target_ulong addr, target_ulong val,
             atomic_store_mask32(env, addr, val, 0x00ffffffu, ra);
         } else {
             cpu_stb_data_ra(env, addr, val >> 16, ra);
-            cpu_stw_data_ra(env, addr + 1, val, ra);
+            cpu_stw_be_data_ra(env, addr + 1, val, ra);
         }
         break;
     default:
-        cpu_stl_data_ra(env, addr, val, ra);
+        cpu_stl_be_data_ra(env, addr, val, ra);
         break;
     }
 }
@@ -132,7 +132,7 @@ static void do_stdby_b(CPUHPPAState *env, target_ulong addr, uint64_t val,
         cpu_stb_data_ra(env, addr, val, ra);
         break;
     case 6:
-        cpu_stw_data_ra(env, addr, val, ra);
+        cpu_stw_be_data_ra(env, addr, val, ra);
         break;
     case 5:
         /* The 3 byte store must appear atomic.  */
@@ -140,11 +140,11 @@ static void do_stdby_b(CPUHPPAState *env, target_ulong addr, uint64_t val,
             atomic_store_mask32(env, addr, val, 0x00ffffffu, ra);
         } else {
             cpu_stb_data_ra(env, addr, val >> 16, ra);
-            cpu_stw_data_ra(env, addr + 1, val, ra);
+            cpu_stw_be_data_ra(env, addr + 1, val, ra);
         }
         break;
     case 4:
-        cpu_stl_data_ra(env, addr, val, ra);
+        cpu_stl_be_data_ra(env, addr, val, ra);
         break;
     case 3:
         /* The 5 byte store must appear atomic.  */
@@ -152,7 +152,7 @@ static void do_stdby_b(CPUHPPAState *env, target_ulong addr, uint64_t val,
             atomic_store_mask64(env, addr, val, 0x000000ffffffffffull, 5, ra);
         } else {
             cpu_stb_data_ra(env, addr, val >> 32, ra);
-            cpu_stl_data_ra(env, addr + 1, val, ra);
+            cpu_stl_be_data_ra(env, addr + 1, val, ra);
         }
         break;
     case 2:
@@ -160,8 +160,8 @@ static void do_stdby_b(CPUHPPAState *env, target_ulong addr, uint64_t val,
         if (parallel) {
             atomic_store_mask64(env, addr, val, 0x0000ffffffffffffull, 6, ra);
         } else {
-            cpu_stw_data_ra(env, addr, val >> 32, ra);
-            cpu_stl_data_ra(env, addr + 2, val, ra);
+            cpu_stw_be_data_ra(env, addr, val >> 32, ra);
+            cpu_stl_be_data_ra(env, addr + 2, val, ra);
         }
         break;
     case 1:
@@ -170,12 +170,12 @@ static void do_stdby_b(CPUHPPAState *env, target_ulong addr, uint64_t val,
             atomic_store_mask64(env, addr, val, 0x00ffffffffffffffull, 7, ra);
         } else {
             cpu_stb_data_ra(env, addr, val >> 48, ra);
-            cpu_stw_data_ra(env, addr + 1, val >> 32, ra);
-            cpu_stl_data_ra(env, addr + 3, val, ra);
+            cpu_stw_be_data_ra(env, addr + 1, val >> 32, ra);
+            cpu_stl_be_data_ra(env, addr + 3, val, ra);
         }
         break;
     default:
-        cpu_stq_data_ra(env, addr, val, ra);
+        cpu_stq_be_data_ra(env, addr, val, ra);
         break;
     }
 }
@@ -211,12 +211,12 @@ static void do_stby_e(CPUHPPAState *env, target_ulong addr, target_ulong val,
         if (parallel) {
             atomic_store_mask32(env, addr - 3, val, 0xffffff00u, ra);
         } else {
-            cpu_stw_data_ra(env, addr - 3, val >> 16, ra);
+            cpu_stw_be_data_ra(env, addr - 3, val >> 16, ra);
             cpu_stb_data_ra(env, addr - 1, val >> 8, ra);
         }
         break;
     case 2:
-        cpu_stw_data_ra(env, addr - 2, val >> 16, ra);
+        cpu_stw_be_data_ra(env, addr - 2, val >> 16, ra);
         break;
     case 1:
         cpu_stb_data_ra(env, addr - 1, val >> 24, ra);
@@ -239,8 +239,8 @@ static void do_stdby_e(CPUHPPAState *env, target_ulong addr, uint64_t val,
             atomic_store_mask64(env, addr - 7, val,
                                 0xffffffffffffff00ull, 7, ra);
         } else {
-            cpu_stl_data_ra(env, addr - 7, val >> 32, ra);
-            cpu_stw_data_ra(env, addr - 3, val >> 16, ra);
+            cpu_stl_be_data_ra(env, addr - 7, val >> 32, ra);
+            cpu_stw_be_data_ra(env, addr - 3, val >> 16, ra);
             cpu_stb_data_ra(env, addr - 1, val >> 8, ra);
         }
         break;
@@ -250,8 +250,8 @@ static void do_stdby_e(CPUHPPAState *env, target_ulong addr, uint64_t val,
             atomic_store_mask64(env, addr - 6, val,
                                 0xffffffffffff0000ull, 6, ra);
         } else {
-            cpu_stl_data_ra(env, addr - 6, val >> 32, ra);
-            cpu_stw_data_ra(env, addr - 2, val >> 16, ra);
+            cpu_stl_be_data_ra(env, addr - 6, val >> 32, ra);
+            cpu_stw_be_data_ra(env, addr - 2, val >> 16, ra);
         }
         break;
     case 5:
@@ -260,24 +260,24 @@ static void do_stdby_e(CPUHPPAState *env, target_ulong addr, uint64_t val,
             atomic_store_mask64(env, addr - 5, val,
                                 0xffffffffff000000ull, 5, ra);
         } else {
-            cpu_stl_data_ra(env, addr - 5, val >> 32, ra);
+            cpu_stl_be_data_ra(env, addr - 5, val >> 32, ra);
             cpu_stb_data_ra(env, addr - 1, val >> 24, ra);
         }
         break;
     case 4:
-        cpu_stl_data_ra(env, addr - 4, val >> 32, ra);
+        cpu_stl_be_data_ra(env, addr - 4, val >> 32, ra);
         break;
     case 3:
         /* The 3 byte store must appear atomic.  */
         if (parallel) {
             atomic_store_mask32(env, addr - 3, val >> 32, 0xffffff00u, ra);
         } else {
-            cpu_stw_data_ra(env, addr - 3, val >> 48, ra);
+            cpu_stw_be_data_ra(env, addr - 3, val >> 48, ra);
             cpu_stb_data_ra(env, addr - 1, val >> 40, ra);
         }
         break;
     case 2:
-        cpu_stw_data_ra(env, addr - 2, val >> 48, ra);
+        cpu_stw_be_data_ra(env, addr - 2, val >> 48, ra);
         break;
     case 1:
         cpu_stb_data_ra(env, addr - 1, val >> 56, ra);
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 18/21] target/riscv: Use little-endian variant of cpu_ld/st_data*() for vector

[Philippe Mathieu-Daudé]

RISC-V vector "elements are simply packed in order from
the least-signicant to most-signicant bits of the vector
register" [*] which is little endianness, therefore the
cpu_ld/st_data*() definitions expand to the little endian
declarations. Use the explicit little-endian variants.

[*] RISC-V "V" Vector Extension v1.0

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/riscv/vector_helper.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
index 2de3358ee86..caa8dd9c125 100644
--- a/target/riscv/vector_helper.c
+++ b/target/riscv/vector_helper.c
@@ -193,9 +193,9 @@ void NAME##_host(void *vd, uint32_t idx, void *host)        \
 }
 
 GEN_VEXT_LD_ELEM(lde_b, uint8_t,  H1, ldub)
-GEN_VEXT_LD_ELEM(lde_h, uint16_t, H2, lduw)
-GEN_VEXT_LD_ELEM(lde_w, uint32_t, H4, ldl)
-GEN_VEXT_LD_ELEM(lde_d, uint64_t, H8, ldq)
+GEN_VEXT_LD_ELEM(lde_h, uint16_t, H2, lduw_le)
+GEN_VEXT_LD_ELEM(lde_w, uint32_t, H4, ldl_le)
+GEN_VEXT_LD_ELEM(lde_d, uint64_t, H8, ldq_le)
 
 #define GEN_VEXT_ST_ELEM(NAME, ETYPE, H, STSUF)             \
 static inline QEMU_ALWAYS_INLINE                            \
@@ -214,9 +214,9 @@ void NAME##_host(void *vd, uint32_t idx, void *host)        \
 }
 
 GEN_VEXT_ST_ELEM(ste_b, uint8_t,  H1, stb)
-GEN_VEXT_ST_ELEM(ste_h, uint16_t, H2, stw)
-GEN_VEXT_ST_ELEM(ste_w, uint32_t, H4, stl)
-GEN_VEXT_ST_ELEM(ste_d, uint64_t, H8, stq)
+GEN_VEXT_ST_ELEM(ste_h, uint16_t, H2, stw_le)
+GEN_VEXT_ST_ELEM(ste_w, uint32_t, H4, stl_le)
+GEN_VEXT_ST_ELEM(ste_d, uint64_t, H8, stq_le)
 
 static inline QEMU_ALWAYS_INLINE void
 vext_continuous_ldst_tlb(CPURISCVState *env, vext_ldst_elem_fn_tlb *ldst_tlb,
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 19/21] target/sh4: Replace cpu_stl_data() by explicit endianness variants

[Philippe Mathieu-Daudé]

See commit 852d481faf7 ("SH: Improve movca.l/ocbi emulation")
for more context on this code.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/sh4/op_helper.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/target/sh4/op_helper.c b/target/sh4/op_helper.c
index 557b1bf4972..63515cc5e6c 100644
--- a/target/sh4/op_helper.c
+++ b/target/sh4/op_helper.c
@@ -143,7 +143,11 @@ void helper_ocbi(CPUSH4State *env, uint32_t address)
         if ((a & ~0x1F) == (address & ~0x1F))
         {
             memory_content *next = (*current)->next;
-            cpu_stl_data(env, a, (*current)->value);
+            if (TARGET_BIG_ENDIAN) {
+                cpu_stl_be_data(env, a, (*current)->value);
+            } else {
+                cpu_stl_le_data(env, a, (*current)->value);
+            }
 
             if (next == NULL)
             {
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 20/21] target/mips: Use big-endian variant of cpu_ld/st_data*() for MSA

[Philippe Mathieu-Daudé]

MSA vectors are accessed in big endianness.

Per the "MIPS® SIMD Architecture" (MD00926 rev 1.03):

  3.1 Registers Layout

  MSA vectors are stored in memory starting from the 0th element at
  the lowest byte address. The byte order of each element follows the
  big- or little-endian convention of the system configuration.

Use the explicit big-endian variants of cpu_ld/st_data*().

Running files in tests/tcg/mips/user/ase/msa shows:

 NLOC.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.21 ms  |
 NLOC.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.22 ms  |
 NLOC.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.18 ms  |
 NLOC.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.18 ms  |
 NLZC.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.19 ms  |
 NLZC.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.18 ms  |
 NLZC.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.21 ms  |
 NLZC.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.18 ms  |
 PCNT.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.17 ms  |
 PCNT.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.19 ms  |
 PCNT.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.18 ms  |
 PCNT.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.19 ms  |
 BINSL.B           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 BINSL.H           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 BINSL.W           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 BINSL.D           | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 BINSR.B           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 BINSR.H           | PASS: 112  | FAIL:   0  | elapsed time:  0.58 ms  |
 BINSR.W           | PASS: 112  | FAIL:   0  | elapsed time:  0.58 ms  |
 BINSR.D           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 BMNZ.V            | PASS: 112  | FAIL:   0  | elapsed time:  0.65 ms  |
 BMZ.V             | PASS: 112  | FAIL:   0  | elapsed time:  0.59 ms  |
 BSEL.V            | PASS: 112  | FAIL:   0  | elapsed time:  0.60 ms  |
 BCLR.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BCLR.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BCLR.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BCLR.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 BNEG.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BNEG.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 BNEG.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BNEG.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BSET.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 BSET.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BSET.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 BSET.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.33 ms  |
 MADD_Q.H          | PASS: 112  | FAIL:   0  | elapsed time:  0.60 ms  |
 MADD_Q.W          | PASS: 112  | FAIL:   0  | elapsed time:  0.83 ms  |
 MADDR_Q.H         | PASS: 112  | FAIL:   0  | elapsed time:  0.62 ms  |
 MADDR_Q.W         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 MSUB_Q.H          | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 MSUB_Q.W          | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 MSUBR_Q.H         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 MSUBR_Q.W         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 MUL_Q.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MUL_Q.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MULR_Q.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MULR_Q.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 FMAX_A.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.36 ms  |
 FMAX_A.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 FMAX.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.33 ms  |
 FMAX.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 FMIN_A.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.36 ms  |
 FMIN_A.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 FMIN.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.35 ms  |
 FMIN.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 ADD_A.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADD_A.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADD_A.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADD_A.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 ADDS_A.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 ADDS_A.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.33 ms  |
 ADDS_A.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDS_A.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDS_S.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDS_S.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDS_S.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.80 ms  |
 ADDS_S.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 ADDS_U.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDS_U.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDS_U.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDS_U.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDV.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 ADDV.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ADDV.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 ADDV.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HADD_S.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HADD_S.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HADD_S.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HADD_U.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HADD_U.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.33 ms  |
 HADD_U.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVE_S.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVE_S.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVE_S.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVE_S.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVE_U.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVE_U.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVE_U.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 AVE_U.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVER_S.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVER_S.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 AVER_S.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVER_S.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVER_U.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 AVER_U.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVER_U.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AVER_U.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CEQ.B             | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 CEQ.H             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CEQ.W             | PASS:  80  | FAIL:   0  | elapsed time:  0.38 ms  |
 CEQ.D             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLE_S.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLE_S.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLE_S.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLE_S.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 CLE_U.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 CLE_U.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 CLE_U.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 CLE_U.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 CLT_S.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 CLT_S.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 CLT_S.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLT_S.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLT_U.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLT_U.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 CLT_U.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 CLT_U.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 DIV_S.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 DIV_S.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 DIV_S.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DIV_S.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DIV_U.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.39 ms  |
 DIV_U.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 DIV_U.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DIV_U.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DOTP_S.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DOTP_S.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.35 ms  |
 DOTP_S.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DOTP_U.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DOTP_U.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 DOTP_U.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 DPADD_S.H         | PASS: 112  | FAIL:   0  | elapsed time:  0.59 ms  |
 DPADD_S.W         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 DPADD_S.D         | PASS: 112  | FAIL:   0  | elapsed time:  0.58 ms  |
 DPADD_U.H         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 DPADD_U.W         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 DPADD_U.D         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 DPSUB_S.H         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 DPSUB_S.W         | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 DPSUB_S.D         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 DPSUB_U.H         | PASS: 112  | FAIL:   0  | elapsed time:  0.58 ms  |
 DPSUB_U.W         | PASS: 112  | FAIL:   0  | elapsed time:  0.59 ms  |
 DPSUB_U.D         | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 MAX_A.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MAX_A.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 MAX_A.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.33 ms  |
 MAX_A.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MAX_S.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MAX_S.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MAX_S.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MAX_S.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.35 ms  |
 MAX_U.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.35 ms  |
 MAX_U.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MAX_U.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 MAX_U.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_A.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 MIN_A.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_A.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_A.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_S.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_S.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.33 ms  |
 MIN_S.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_S.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_U.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 MIN_U.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_U.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MIN_U.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 MOD_S.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 MOD_S.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 MOD_S.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MOD_S.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MOD_U.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.37 ms  |
 MOD_U.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 MOD_U.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MOD_U.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MADDV.B           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 MADDV.H           | PASS: 112  | FAIL:   0  | elapsed time:  0.63 ms  |
 MADDV.W           | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 MADDV.D           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 MSUBV.B           | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 MSUBV.H           | PASS: 112  | FAIL:   0  | elapsed time:  0.65 ms  |
 MSUBV.W           | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 MSUBV.D           | PASS: 112  | FAIL:   0  | elapsed time:  0.65 ms  |
 MULV.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 MULV.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MULV.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 MULV.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ASUB_S.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ASUB_S.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ASUB_S.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ASUB_S.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ASUB_U.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.44 ms  |
 ASUB_U.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 ASUB_U.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ASUB_U.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HSUB_S.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HSUB_S.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 HSUB_S.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 HSUB_U.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HSUB_U.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 HSUB_U.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 SUBS_S.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 SUBS_S.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.38 ms  |
 SUBS_S.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 SUBS_S.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBS_U.B          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBS_U.H          | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 SUBS_U.W          | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 SUBS_U.D          | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBSUS_U.B        | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBSUS_U.H        | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 SUBSUS_U.W        | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBSUS_U.D        | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBSUU_S.B        | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBSUU_S.H        | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 SUBSUU_S.W        | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBSUU_S.D        | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBV.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBV.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBV.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SUBV.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVEV.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.35 ms  |
 ILVEV.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 ILVEV.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 ILVEV.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 ILVOD.B           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVOD.H           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVOD.W           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVOD.D           | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVL.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 ILVL.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVL.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVL.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVR.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVR.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 ILVR.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 ILVR.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 AND.V             | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 NOR.V             | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 OR.V              | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 XOR.V             | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 MOVE.V            | PASS:  80  | FAIL:   0  | elapsed time:  0.18 ms  |
 PCKEV.B           | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 PCKEV.H           | PASS: 112  | FAIL:   0  | elapsed time:  0.56 ms  |
 PCKEV.W           | PASS: 112  | FAIL:   0  | elapsed time:  0.68 ms  |
 PCKEV.D           | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 PCKOD.B           | PASS: 112  | FAIL:   0  | elapsed time:  0.61 ms  |
 PCKOD.H           | PASS: 112  | FAIL:   0  | elapsed time:  0.62 ms  |
 PCKOD.W           | PASS: 112  | FAIL:   0  | elapsed time:  0.58 ms  |
 PCKOD.D           | PASS: 112  | FAIL:   0  | elapsed time:  0.58 ms  |
 VSHF.B            | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 VSHF.H            | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 VSHF.W            | PASS: 112  | FAIL:   0  | elapsed time:  0.62 ms  |
 VSHF.D            | PASS: 112  | FAIL:   0  | elapsed time:  0.57 ms  |
 SLL.B             | PASS:  80  | FAIL:   0  | elapsed time:  0.34 ms  |
 SLL.H             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SLL.W             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SLL.D             | PASS:  80  | FAIL:   0  | elapsed time:  0.38 ms  |
 SRA.B             | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 SRA.H             | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 SRA.W             | PASS:  80  | FAIL:   0  | elapsed time:  0.37 ms  |
 SRA.D             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SRAR.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 SRAR.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 SRAR.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 SRAR.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SRL.B             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SRL.H             | PASS:  80  | FAIL:   0  | elapsed time:  0.32 ms  |
 SRL.W             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SRL.D             | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SRLR.B            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |
 SRLR.H            | PASS:  80  | FAIL:   0  | elapsed time:  0.29 ms  |
 SRLR.W            | PASS:  80  | FAIL:   0  | elapsed time:  0.31 ms  |
 SRLR.D            | PASS:  80  | FAIL:   0  | elapsed time:  0.30 ms  |

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
 target/mips/tcg/msa_helper.c | 51 ++++++++++++++++++++++--------------
 1 file changed, 31 insertions(+), 20 deletions(-)

diff --git a/target/mips/tcg/msa_helper.c b/target/mips/tcg/msa_helper.c
index f554b3d10ee..d6ce17abf9a 100644
--- a/target/mips/tcg/msa_helper.c
+++ b/target/mips/tcg/msa_helper.c
@@ -8231,8 +8231,8 @@ void helper_msa_ld_b(CPUMIPSState *env, uint32_t wd,
     uint64_t d0, d1;
 
     /* Load 8 bytes at a time.  Vector element ordering makes this LE.  */
-    d0 = cpu_ldq_le_data_ra(env, addr + 0, ra);
-    d1 = cpu_ldq_le_data_ra(env, addr + 8, ra);
+    d0 = cpu_ldq_be_data_ra(env, addr + 0, ra);
+    d1 = cpu_ldq_be_data_ra(env, addr + 8, ra);
     pwd->d[0] = d0;
     pwd->d[1] = d1;
 }
@@ -8248,9 +8248,9 @@ void helper_msa_ld_h(CPUMIPSState *env, uint32_t wd,
      * Load 8 bytes at a time.  Use little-endian load, then for
      * big-endian target, we must then swap the four halfwords.
      */
-    d0 = cpu_ldq_le_data_ra(env, addr + 0, ra);
-    d1 = cpu_ldq_le_data_ra(env, addr + 8, ra);
-    if (mips_env_is_bigendian(env)) {
+    d0 = cpu_ldq_be_data_ra(env, addr + 0, ra);
+    d1 = cpu_ldq_be_data_ra(env, addr + 8, ra);
+    if (!mips_env_is_bigendian(env)) {
         d0 = bswap16x4(d0);
         d1 = bswap16x4(d1);
     }
@@ -8269,9 +8269,9 @@ void helper_msa_ld_w(CPUMIPSState *env, uint32_t wd,
      * Load 8 bytes at a time.  Use little-endian load, then for
      * big-endian target, we must then bswap the two words.
      */
-    d0 = cpu_ldq_le_data_ra(env, addr + 0, ra);
-    d1 = cpu_ldq_le_data_ra(env, addr + 8, ra);
-    if (mips_env_is_bigendian(env)) {
+    d0 = cpu_ldq_be_data_ra(env, addr + 0, ra);
+    d1 = cpu_ldq_be_data_ra(env, addr + 8, ra);
+    if (!mips_env_is_bigendian(env)) {
         d0 = bswap32x2(d0);
         d1 = bswap32x2(d1);
     }
@@ -8286,8 +8286,12 @@ void helper_msa_ld_d(CPUMIPSState *env, uint32_t wd,
     uintptr_t ra = GETPC();
     uint64_t d0, d1;
 
-    d0 = cpu_ldq_data_ra(env, addr + 0, ra);
-    d1 = cpu_ldq_data_ra(env, addr + 8, ra);
+    d0 = cpu_ldq_be_data_ra(env, addr + 0, ra);
+    d1 = cpu_ldq_be_data_ra(env, addr + 8, ra);
+    if (!mips_env_is_bigendian(env)) {
+        d0 = bswap64(d0);
+        d1 = bswap64(d1);
+    }
     pwd->d[0] = d0;
     pwd->d[1] = d1;
 }
@@ -8320,8 +8324,8 @@ void helper_msa_st_b(CPUMIPSState *env, uint32_t wd,
     ensure_writable_pages(env, addr, mmu_idx, ra);
 
     /* Store 8 bytes at a time.  Vector element ordering makes this LE.  */
-    cpu_stq_le_data_ra(env, addr + 0, pwd->d[0], ra);
-    cpu_stq_le_data_ra(env, addr + 8, pwd->d[1], ra);
+    cpu_stq_be_data_ra(env, addr + 0, pwd->d[0], ra);
+    cpu_stq_be_data_ra(env, addr + 8, pwd->d[1], ra);
 }
 
 void helper_msa_st_h(CPUMIPSState *env, uint32_t wd,
@@ -8337,12 +8341,12 @@ void helper_msa_st_h(CPUMIPSState *env, uint32_t wd,
     /* Store 8 bytes at a time.  See helper_msa_ld_h. */
     d0 = pwd->d[0];
     d1 = pwd->d[1];
-    if (mips_env_is_bigendian(env)) {
+    if (!mips_env_is_bigendian(env)) {
         d0 = bswap16x4(d0);
         d1 = bswap16x4(d1);
     }
-    cpu_stq_le_data_ra(env, addr + 0, d0, ra);
-    cpu_stq_le_data_ra(env, addr + 8, d1, ra);
+    cpu_stq_be_data_ra(env, addr + 0, d0, ra);
+    cpu_stq_be_data_ra(env, addr + 8, d1, ra);
 }
 
 void helper_msa_st_w(CPUMIPSState *env, uint32_t wd,
@@ -8358,12 +8362,12 @@ void helper_msa_st_w(CPUMIPSState *env, uint32_t wd,
     /* Store 8 bytes at a time.  See helper_msa_ld_w. */
     d0 = pwd->d[0];
     d1 = pwd->d[1];
-    if (mips_env_is_bigendian(env)) {
+    if (!mips_env_is_bigendian(env)) {
         d0 = bswap32x2(d0);
         d1 = bswap32x2(d1);
     }
-    cpu_stq_le_data_ra(env, addr + 0, d0, ra);
-    cpu_stq_le_data_ra(env, addr + 8, d1, ra);
+    cpu_stq_be_data_ra(env, addr + 0, d0, ra);
+    cpu_stq_be_data_ra(env, addr + 8, d1, ra);
 }
 
 void helper_msa_st_d(CPUMIPSState *env, uint32_t wd,
@@ -8372,9 +8376,16 @@ void helper_msa_st_d(CPUMIPSState *env, uint32_t wd,
     wr_t *pwd = &(env->active_fpu.fpr[wd].wr);
     int mmu_idx = mips_env_mmu_index(env);
     uintptr_t ra = GETPC();
+    uint64_t d0, d1;
 
     ensure_writable_pages(env, addr, mmu_idx, GETPC());
 
-    cpu_stq_data_ra(env, addr + 0, pwd->d[0], ra);
-    cpu_stq_data_ra(env, addr + 8, pwd->d[1], ra);
+    d0 = pwd->d[0];
+    d1 = pwd->d[1];
+    if (!mips_env_is_bigendian(env)) {
+        d0 = bswap64(d0);
+        d1 = bswap64(d1);
+    }
+    cpu_stq_be_data_ra(env, addr + 0, d0, ra);
+    cpu_stq_be_data_ra(env, addr + 8, d1, ra);
 }
-- 
2.51.0

[RFC-PATCH-for-11.0 v3 21/21] accel/tcg: Remove non-explicit endian cpu_ld/st*_data*() helpers

[Philippe Mathieu-Daudé]

All uses were converted to the explicit cpu_ld/st*_{be,le}_data*()
helpers, no need for the non-explicit versions anymore.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
FIXME: Still some PPC/MIPS uses
---
 include/accel/tcg/cpu-ldst.h | 46 ------------------------------------
 1 file changed, 46 deletions(-)

diff --git a/include/accel/tcg/cpu-ldst.h b/include/accel/tcg/cpu-ldst.h
index a3125fc9026..70442bd08af 100644
--- a/include/accel/tcg/cpu-ldst.h
+++ b/include/accel/tcg/cpu-ldst.h
@@ -428,52 +428,6 @@ cpu_stq_le_data(CPUArchState *env, abi_ptr addr, uint64_t val)
     cpu_stq_le_data_ra(env, addr, val, 0);
 }
 
-#if TARGET_BIG_ENDIAN
-# define cpu_lduw_data        cpu_lduw_be_data
-# define cpu_ldsw_data        cpu_ldsw_be_data
-# define cpu_ldl_data         cpu_ldl_be_data
-# define cpu_ldq_data         cpu_ldq_be_data
-# define cpu_lduw_data_ra     cpu_lduw_be_data_ra
-# define cpu_ldsw_data_ra     cpu_ldsw_be_data_ra
-# define cpu_ldl_data_ra      cpu_ldl_be_data_ra
-# define cpu_ldq_data_ra      cpu_ldq_be_data_ra
-# define cpu_lduw_mmuidx_ra   cpu_lduw_be_mmuidx_ra
-# define cpu_ldsw_mmuidx_ra   cpu_ldsw_be_mmuidx_ra
-# define cpu_ldl_mmuidx_ra    cpu_ldl_be_mmuidx_ra
-# define cpu_ldq_mmuidx_ra    cpu_ldq_be_mmuidx_ra
-# define cpu_stw_data         cpu_stw_be_data
-# define cpu_stl_data         cpu_stl_be_data
-# define cpu_stq_data         cpu_stq_be_data
-# define cpu_stw_data_ra      cpu_stw_be_data_ra
-# define cpu_stl_data_ra      cpu_stl_be_data_ra
-# define cpu_stq_data_ra      cpu_stq_be_data_ra
-# define cpu_stw_mmuidx_ra    cpu_stw_be_mmuidx_ra
-# define cpu_stl_mmuidx_ra    cpu_stl_be_mmuidx_ra
-# define cpu_stq_mmuidx_ra    cpu_stq_be_mmuidx_ra
-#else
-# define cpu_lduw_data        cpu_lduw_le_data
-# define cpu_ldsw_data        cpu_ldsw_le_data
-# define cpu_ldl_data         cpu_ldl_le_data
-# define cpu_ldq_data         cpu_ldq_le_data
-# define cpu_lduw_data_ra     cpu_lduw_le_data_ra
-# define cpu_ldsw_data_ra     cpu_ldsw_le_data_ra
-# define cpu_ldl_data_ra      cpu_ldl_le_data_ra
-# define cpu_ldq_data_ra      cpu_ldq_le_data_ra
-# define cpu_lduw_mmuidx_ra   cpu_lduw_le_mmuidx_ra
-# define cpu_ldsw_mmuidx_ra   cpu_ldsw_le_mmuidx_ra
-# define cpu_ldl_mmuidx_ra    cpu_ldl_le_mmuidx_ra
-# define cpu_ldq_mmuidx_ra    cpu_ldq_le_mmuidx_ra
-# define cpu_stw_data         cpu_stw_le_data
-# define cpu_stl_data         cpu_stl_le_data
-# define cpu_stq_data         cpu_stq_le_data
-# define cpu_stw_data_ra      cpu_stw_le_data_ra
-# define cpu_stl_data_ra      cpu_stl_le_data_ra
-# define cpu_stq_data_ra      cpu_stq_le_data_ra
-# define cpu_stw_mmuidx_ra    cpu_stw_le_mmuidx_ra
-# define cpu_stl_mmuidx_ra    cpu_stl_le_mmuidx_ra
-# define cpu_stq_mmuidx_ra    cpu_stq_le_mmuidx_ra
-#endif
-
 static inline uint32_t cpu_ldub_code(CPUArchState *env, abi_ptr addr)
 {
     CPUState *cs = env_cpu(env);
-- 
2.51.0