Merge remote-tracking branch 'remotes/rth/tags/pull-tcg-20210921' into staging

Move cpu_signal_handler declaration.
Restrict cpu_handle_halt to sysemu.
Make do_unaligned_access noreturn.
Misc tcg/mips cleanup
Misc tcg/sparc cleanup
Misc tcg/riscv cleanup

# gpg: Signature made Tue 21 Sep 2021 10:47:29 PM EDT
# gpg:                using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F
# gpg:                issuer "richard.henderson@linaro.org"
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [ultimate]

* remotes/rth/tags/pull-tcg-20210921:
  tcg/riscv: Remove add with zero on user-only memory access
  hw/core: Make do_unaligned_access noreturn
  tcg/sparc: Introduce tcg_out_mov_delay
  tcg/sparc: Drop inline markers
  tcg/mips: Drop special alignment for code_gen_buffer
  tcg/mips: Unset TCG_TARGET_HAS_direct_jump
  tcg/mips: Allow JAL to be out of range in tcg_out_bswap_subr
  tcg/mips: Drop inline markers
  accel/tcg: Restrict cpu_handle_halt() to sysemu
  include/exec: Move cpu_signal_handler declaration

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

accel/tcg/cpu-exec.c

@@ -588,8 +588,9 @@ static inline void tb_add_jump(TranslationBlock *tb, int n,
 
 static inline bool cpu_handle_halt(CPUState *cpu)
 {
+#ifndef CONFIG_USER_ONLY
     if (cpu->halted) {
-#if defined(TARGET_I386) && !defined(CONFIG_USER_ONLY)
+#if defined(TARGET_I386)
         if (cpu->interrupt_request & CPU_INTERRUPT_POLL) {
             X86CPU *x86_cpu = X86_CPU(cpu);
             qemu_mutex_lock_iothread();
@@ -597,13 +598,14 @@ static inline bool cpu_handle_halt(CPUState *cpu)
             cpu_reset_interrupt(cpu, CPU_INTERRUPT_POLL);
             qemu_mutex_unlock_iothread();
         }
-#endif
+#endif /* TARGET_I386 */
         if (!cpu_has_work(cpu)) {
             return true;
         }
 
         cpu->halted = 0;
     }
+#endif /* !CONFIG_USER_ONLY */
 
     return false;
 }

include/exec/exec-all.h

@@ -662,6 +662,19 @@ static inline tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env,
     }
     return addr;
 }
+
+/**
+ * cpu_signal_handler
+ * @signum: host signal number
+ * @pinfo: host siginfo_t
+ * @puc: host ucontext_t
+ *
+ * To be called from the SIGBUS and SIGSEGV signal handler to inform the
+ * virtual cpu of exceptions.  Returns true if the signal was handled by
+ * the virtual CPU.
+ */
+int cpu_signal_handler(int signum, void *pinfo, void *puc);
+
 #else
 static inline void mmap_lock(void) {}
 static inline void mmap_unlock(void) {}

include/hw/core/tcg-cpu-ops.h

@@ -78,10 +78,11 @@ struct TCGCPUOps {
                                   MemTxResult response, uintptr_t retaddr);
     /**
      * @do_unaligned_access: Callback for unaligned access handling
+     * The callback must exit via raising an exception.
      */
     void (*do_unaligned_access)(CPUState *cpu, vaddr addr,
                                 MMUAccessType access_type,
-                                int mmu_idx, uintptr_t retaddr);
+                                int mmu_idx, uintptr_t retaddr) QEMU_NORETURN;
 
     /**
      * @adjust_watchpoint_address: hack for cpu_check_watchpoint used by ARM

target/alpha/cpu.h

@@ -283,11 +283,10 @@ hwaddr alpha_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr);
 int alpha_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
 int alpha_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
 void alpha_cpu_do_unaligned_access(CPUState *cpu, vaddr addr,
-                                   MMUAccessType access_type,
-                                   int mmu_idx, uintptr_t retaddr);
+                                   MMUAccessType access_type, int mmu_idx,
+                                   uintptr_t retaddr) QEMU_NORETURN;
 
 #define cpu_list alpha_cpu_list
-#define cpu_signal_handler cpu_alpha_signal_handler
 
 typedef CPUAlphaState CPUArchState;
 typedef AlphaCPU ArchCPU;
@@ -440,11 +439,6 @@ void alpha_translate_init(void);
 #define CPU_RESOLVING_TYPE TYPE_ALPHA_CPU
 
 void alpha_cpu_list(void);
-/* you can call this signal handler from your SIGBUS and SIGSEGV
-   signal handlers to inform the virtual CPU of exceptions. non zero
-   is returned if the signal was handled by the virtual CPU.  */
-int cpu_alpha_signal_handler(int host_signum, void *pinfo,
-                             void *puc);
 bool alpha_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
                         MMUAccessType access_type, int mmu_idx,
                         bool probe, uintptr_t retaddr);

target/arm/cpu.h

@@ -1121,12 +1121,6 @@ static inline bool is_a64(CPUARMState *env)
     return env->aarch64;
 }
 
-/* you can call this signal handler from your SIGBUS and SIGSEGV
-   signal handlers to inform the virtual CPU of exceptions. non zero
-   is returned if the signal was handled by the virtual CPU.  */
-int cpu_arm_signal_handler(int host_signum, void *pinfo,
-                           void *puc);
-
 /**
  * pmu_op_start/finish
  * @env: CPUARMState
@@ -3017,7 +3011,6 @@ bool write_cpustate_to_list(ARMCPU *cpu, bool kvm_sync);
 
 #define TYPE_ARM_HOST_CPU "host-" TYPE_ARM_CPU
 
-#define cpu_signal_handler cpu_arm_signal_handler
 #define cpu_list arm_cpu_list
 
 /* ARM has the following "translation regimes" (as the ARM ARM calls them):

target/arm/internals.h

@@ -594,7 +594,7 @@ bool arm_s1_regime_using_lpae_format(CPUARMState *env, ARMMMUIdx mmu_idx);
 /* Raise a data fault alignment exception for the specified virtual address */
 void arm_cpu_do_unaligned_access(CPUState *cs, vaddr vaddr,
                                  MMUAccessType access_type,
-                                 int mmu_idx, uintptr_t retaddr);
+                                 int mmu_idx, uintptr_t retaddr) QEMU_NORETURN;
 
 /* arm_cpu_do_transaction_failed: handle a memory system error response
  * (eg "no device/memory present at address") by raising an external abort

target/avr/cpu.h

@@ -175,7 +175,6 @@ static inline void set_avr_feature(CPUAVRState *env, int feature)
 }
 
 #define cpu_list avr_cpu_list
-#define cpu_signal_handler cpu_avr_signal_handler
 #define cpu_mmu_index avr_cpu_mmu_index
 
 static inline int avr_cpu_mmu_index(CPUAVRState *env, bool ifetch)
@@ -187,7 +186,6 @@ void avr_cpu_tcg_init(void);
 
 void avr_cpu_list(void);
 int cpu_avr_exec(CPUState *cpu);
-int cpu_avr_signal_handler(int host_signum, void *pinfo, void *puc);
 int avr_cpu_memory_rw_debug(CPUState *cs, vaddr address, uint8_t *buf,
                             int len, bool is_write);
 

target/cris/cpu.h

@@ -199,12 +199,6 @@ int crisv10_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
 int cris_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
 int cris_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
 
-/* you can call this signal handler from your SIGBUS and SIGSEGV
-   signal handlers to inform the virtual CPU of exceptions. non zero
-   is returned if the signal was handled by the virtual CPU.  */
-int cpu_cris_signal_handler(int host_signum, void *pinfo,
-                           void *puc);
-
 void cris_initialize_tcg(void);
 void cris_initialize_crisv10_tcg(void);
 
@@ -250,8 +244,6 @@ enum {
 #define CRIS_CPU_TYPE_NAME(name) (name CRIS_CPU_TYPE_SUFFIX)
 #define CPU_RESOLVING_TYPE TYPE_CRIS_CPU
 
-#define cpu_signal_handler cpu_cris_signal_handler
-
 /* MMU modes definitions */
 #define MMU_USER_IDX 1
 static inline int cpu_mmu_index (CPUCRISState *env, bool ifetch)

target/hexagon/cpu.h

@@ -129,9 +129,6 @@ typedef struct HexagonCPU {
 
 #include "cpu_bits.h"
 
-#define cpu_signal_handler cpu_hexagon_signal_handler
-int cpu_hexagon_signal_handler(int host_signum, void *pinfo, void *puc);
-
 static inline void cpu_get_tb_cpu_state(CPUHexagonState *env, target_ulong *pc,
                                         target_ulong *cs_base, uint32_t *flags)
 {

target/hppa/cpu.c

@@ -72,9 +72,10 @@ static void hppa_cpu_disas_set_info(CPUState *cs, disassemble_info *info)
 }
 
 #ifndef CONFIG_USER_ONLY
-static void hppa_cpu_do_unaligned_access(CPUState *cs, vaddr addr,
-                                         MMUAccessType access_type,
-                                         int mmu_idx, uintptr_t retaddr)
+static void QEMU_NORETURN
+hppa_cpu_do_unaligned_access(CPUState *cs, vaddr addr,
+                             MMUAccessType access_type, int mmu_idx,
+                             uintptr_t retaddr)
 {
     HPPACPU *cpu = HPPA_CPU(cs);
     CPUHPPAState *env = &cpu->env;

target/hppa/cpu.h

@@ -319,9 +319,6 @@ static inline void cpu_hppa_change_prot_id(CPUHPPAState *env) { }
 void cpu_hppa_change_prot_id(CPUHPPAState *env);
 #endif
 
-#define cpu_signal_handler cpu_hppa_signal_handler
-
-int cpu_hppa_signal_handler(int host_signum, void *pinfo, void *puc);
 hwaddr hppa_cpu_get_phys_page_debug(CPUState *cs, vaddr addr);
 int hppa_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
 int hppa_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);

target/i386/cpu.h

@@ -1947,12 +1947,6 @@ void cpu_x86_frstor(CPUX86State *s, target_ulong ptr, int data32);
 void cpu_x86_fxsave(CPUX86State *s, target_ulong ptr);
 void cpu_x86_fxrstor(CPUX86State *s, target_ulong ptr);
 
-/* you can call this signal handler from your SIGBUS and SIGSEGV
-   signal handlers to inform the virtual CPU of exceptions. non zero
-   is returned if the signal was handled by the virtual CPU.  */
-int cpu_x86_signal_handler(int host_signum, void *pinfo,
-                           void *puc);
-
 /* cpu.c */
 void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1,
                               uint32_t vendor2, uint32_t vendor3);
@@ -2020,7 +2014,6 @@ uint64_t cpu_get_tsc(CPUX86State *env);
 #define TARGET_DEFAULT_CPU_TYPE X86_CPU_TYPE_NAME("qemu32")
 #endif
 
-#define cpu_signal_handler cpu_x86_signal_handler
 #define cpu_list x86_cpu_list
 
 /* MMU modes definitions */

target/m68k/cpu.h

@@ -177,13 +177,6 @@ int m68k_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
 
 void m68k_tcg_init(void);
 void m68k_cpu_init_gdb(M68kCPU *cpu);
-/*
- * you can call this signal handler from your SIGBUS and SIGSEGV
- * signal handlers to inform the virtual CPU of exceptions. non zero
- * is returned if the signal was handled by the virtual CPU.
- */
-int cpu_m68k_signal_handler(int host_signum, void *pinfo,
-                           void *puc);
 uint32_t cpu_m68k_get_ccr(CPUM68KState *env);
 void cpu_m68k_set_ccr(CPUM68KState *env, uint32_t);
 void cpu_m68k_set_sr(CPUM68KState *env, uint32_t);
@@ -563,7 +556,6 @@ enum {
 #define M68K_CPU_TYPE_NAME(model) model M68K_CPU_TYPE_SUFFIX
 #define CPU_RESOLVING_TYPE TYPE_M68K_CPU
 
-#define cpu_signal_handler cpu_m68k_signal_handler
 #define cpu_list m68k_cpu_list
 
 /* MMU modes definitions */

target/microblaze/cpu.h

@@ -361,7 +361,7 @@ bool mb_cpu_exec_interrupt(CPUState *cs, int int_req);
 #endif /* !CONFIG_USER_ONLY */
 void mb_cpu_do_unaligned_access(CPUState *cs, vaddr vaddr,
                                 MMUAccessType access_type,
-                                int mmu_idx, uintptr_t retaddr);
+                                int mmu_idx, uintptr_t retaddr) QEMU_NORETURN;
 void mb_cpu_dump_state(CPUState *cpu, FILE *f, int flags);
 hwaddr mb_cpu_get_phys_page_attrs_debug(CPUState *cpu, vaddr addr,
                                         MemTxAttrs *attrs);
@@ -385,16 +385,9 @@ static inline void mb_cpu_write_msr(CPUMBState *env, uint32_t val)
 }
 
 void mb_tcg_init(void);
-/* you can call this signal handler from your SIGBUS and SIGSEGV
-   signal handlers to inform the virtual CPU of exceptions. non zero
-   is returned if the signal was handled by the virtual CPU.  */
-int cpu_mb_signal_handler(int host_signum, void *pinfo,
-                          void *puc);
 
 #define CPU_RESOLVING_TYPE TYPE_MICROBLAZE_CPU
 
-#define cpu_signal_handler cpu_mb_signal_handler
-
 /* MMU modes definitions */
 #define MMU_NOMMU_IDX   0
 #define MMU_KERNEL_IDX  1

target/mips/cpu.h

@@ -1193,7 +1193,6 @@ struct MIPSCPU {
 
 void mips_cpu_list(void);
 
-#define cpu_signal_handler cpu_mips_signal_handler
 #define cpu_list mips_cpu_list
 
 extern void cpu_wrdsp(uint32_t rs, uint32_t mask_num, CPUMIPSState *env);
@@ -1277,8 +1276,6 @@ enum {
  */
 #define CPU_INTERRUPT_WAKE CPU_INTERRUPT_TGT_INT_0
 
-int cpu_mips_signal_handler(int host_signum, void *pinfo, void *puc);
-
 #define MIPS_CPU_TYPE_SUFFIX "-" TYPE_MIPS_CPU
 #define MIPS_CPU_TYPE_NAME(model) model MIPS_CPU_TYPE_SUFFIX
 #define CPU_RESOLVING_TYPE TYPE_MIPS_CPU

target/mips/internal.h

@@ -156,8 +156,6 @@ extern const VMStateDescription vmstate_mips_cpu;
 
 #endif /* !CONFIG_USER_ONLY */
 
-#define cpu_signal_handler cpu_mips_signal_handler
-
 static inline bool cpu_mips_hw_interrupts_enabled(CPUMIPSState *env)
 {
     return (env->CP0_Status & (1 << CP0St_IE)) &&

target/mips/tcg/tcg-internal.h

@@ -22,8 +22,8 @@ bool mips_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
                        MMUAccessType access_type, int mmu_idx,
                        bool probe, uintptr_t retaddr);
 void mips_cpu_do_unaligned_access(CPUState *cpu, vaddr addr,
-                                  MMUAccessType access_type,
-                                  int mmu_idx, uintptr_t retaddr);
+                                  MMUAccessType access_type, int mmu_idx,
+                                  uintptr_t retaddr) QEMU_NORETURN;
 
 const char *mips_exception_name(int32_t exception);
 

target/nios2/cpu.h

@@ -193,20 +193,18 @@ struct Nios2CPU {
 
 void nios2_tcg_init(void);
 void nios2_cpu_do_interrupt(CPUState *cs);
-int cpu_nios2_signal_handler(int host_signum, void *pinfo, void *puc);
 void dump_mmu(CPUNios2State *env);
 void nios2_cpu_dump_state(CPUState *cpu, FILE *f, int flags);
 hwaddr nios2_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr);
 void nios2_cpu_do_unaligned_access(CPUState *cpu, vaddr addr,
-                                   MMUAccessType access_type,
-                                   int mmu_idx, uintptr_t retaddr);
+                                   MMUAccessType access_type, int mmu_idx,
+                                   uintptr_t retaddr) QEMU_NORETURN;
 
 void do_nios2_semihosting(CPUNios2State *env);
 
 #define CPU_RESOLVING_TYPE TYPE_NIOS2_CPU
 
 #define cpu_gen_code cpu_nios2_gen_code
-#define cpu_signal_handler cpu_nios2_signal_handler
 
 #define CPU_SAVE_VERSION 1
 

target/openrisc/cpu.h

@@ -320,11 +320,9 @@ void openrisc_translate_init(void);
 bool openrisc_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
                            MMUAccessType access_type, int mmu_idx,
                            bool probe, uintptr_t retaddr);
-int cpu_openrisc_signal_handler(int host_signum, void *pinfo, void *puc);
 int print_insn_or1k(bfd_vma addr, disassemble_info *info);
 
 #define cpu_list cpu_openrisc_list
-#define cpu_signal_handler cpu_openrisc_signal_handler
 
 #ifndef CONFIG_USER_ONLY
 extern const VMStateDescription vmstate_openrisc_cpu;

target/ppc/cpu.h

@@ -1278,12 +1278,6 @@ extern const VMStateDescription vmstate_ppc_cpu;
 
 /*****************************************************************************/
 void ppc_translate_init(void);
-/*
- * you can call this signal handler from your SIGBUS and SIGSEGV
- * signal handlers to inform the virtual CPU of exceptions. non zero
- * is returned if the signal was handled by the virtual CPU.
- */
-int cpu_ppc_signal_handler(int host_signum, void *pinfo, void *puc);
 bool ppc_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
                       MMUAccessType access_type, int mmu_idx,
                       bool probe, uintptr_t retaddr);
@@ -1371,7 +1365,6 @@ int ppc_dcr_write(ppc_dcr_t *dcr_env, int dcrn, uint32_t val);
 #define POWERPC_CPU_TYPE_NAME(model) model POWERPC_CPU_TYPE_SUFFIX
 #define CPU_RESOLVING_TYPE TYPE_POWERPC_CPU
 
-#define cpu_signal_handler cpu_ppc_signal_handler
 #define cpu_list ppc_cpu_list
 
 /* MMU modes definitions */

target/ppc/internal.h

@@ -213,8 +213,8 @@ void helper_compute_fprf_float128(CPUPPCState *env, float128 arg);
 
 /* Raise a data fault alignment exception for the specified virtual address */
 void ppc_cpu_do_unaligned_access(CPUState *cs, vaddr addr,
-                                 MMUAccessType access_type,
-                                 int mmu_idx, uintptr_t retaddr);
+                                 MMUAccessType access_type, int mmu_idx,
+                                 uintptr_t retaddr) QEMU_NORETURN;
 
 /* translate.c */
 

target/riscv/cpu.h

@@ -344,7 +344,7 @@ int riscv_cpu_mmu_index(CPURISCVState *env, bool ifetch);
 hwaddr riscv_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr);
 void  riscv_cpu_do_unaligned_access(CPUState *cs, vaddr addr,
                                     MMUAccessType access_type, int mmu_idx,
-                                    uintptr_t retaddr);
+                                    uintptr_t retaddr) QEMU_NORETURN;
 bool riscv_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
                         MMUAccessType access_type, int mmu_idx,
                         bool probe, uintptr_t retaddr);
@@ -356,7 +356,6 @@ void riscv_cpu_do_transaction_failed(CPUState *cs, hwaddr physaddr,
 char *riscv_isa_string(RISCVCPU *cpu);
 void riscv_cpu_list(void);
 
-#define cpu_signal_handler riscv_cpu_signal_handler
 #define cpu_list riscv_cpu_list
 #define cpu_mmu_index riscv_cpu_mmu_index
 
@@ -372,7 +371,6 @@ void riscv_cpu_set_rdtime_fn(CPURISCVState *env, uint64_t (*fn)(uint32_t),
 void riscv_cpu_set_mode(CPURISCVState *env, target_ulong newpriv);
 
 void riscv_translate_init(void);
-int riscv_cpu_signal_handler(int host_signum, void *pinfo, void *puc);
 void QEMU_NORETURN riscv_raise_exception(CPURISCVState *env,
                                          uint32_t exception, uintptr_t pc);
 

target/rx/cpu.h

@@ -134,13 +134,9 @@ int rx_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
 hwaddr rx_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr);
 
 void rx_translate_init(void);
-int cpu_rx_signal_handler(int host_signum, void *pinfo,
-                           void *puc);
-
 void rx_cpu_list(void);
 void rx_cpu_unpack_psw(CPURXState *env, uint32_t psw, int rte);
 
-#define cpu_signal_handler cpu_rx_signal_handler
 #define cpu_list rx_cpu_list
 
 #include "exec/cpu-all.h"

target/s390x/cpu.h

@@ -809,13 +809,6 @@ void s390_set_qemu_cpu_model(uint16_t type, uint8_t gen, uint8_t ec_ga,
 #define S390_CPU_TYPE_NAME(name) (name S390_CPU_TYPE_SUFFIX)
 #define CPU_RESOLVING_TYPE TYPE_S390_CPU
 
-/* you can call this signal handler from your SIGBUS and SIGSEGV
-   signal handlers to inform the virtual CPU of exceptions. non zero
-   is returned if the signal was handled by the virtual CPU.  */
-int cpu_s390x_signal_handler(int host_signum, void *pinfo, void *puc);
-#define cpu_signal_handler cpu_s390x_signal_handler
-
-
 /* interrupt.c */
 #define RA_IGNORED                  0
 void s390_program_interrupt(CPUS390XState *env, uint32_t code, uintptr_t ra);

target/s390x/s390x-internal.h

@@ -274,8 +274,8 @@ bool s390_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
                        MMUAccessType access_type, int mmu_idx,
                        bool probe, uintptr_t retaddr);
 void s390x_cpu_do_unaligned_access(CPUState *cs, vaddr addr,
-                                   MMUAccessType access_type,
-                                   int mmu_idx, uintptr_t retaddr);
+                                   MMUAccessType access_type, int mmu_idx,
+                                   uintptr_t retaddr) QEMU_NORETURN;
 
 
 /* fpu_helper.c */

target/sh4/cpu.h

@@ -209,12 +209,10 @@ hwaddr superh_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr);
 int superh_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
 int superh_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
 void superh_cpu_do_unaligned_access(CPUState *cpu, vaddr addr,
-                                    MMUAccessType access_type,
-                                    int mmu_idx, uintptr_t retaddr);
+                                    MMUAccessType access_type, int mmu_idx,
+                                    uintptr_t retaddr) QEMU_NORETURN;
 
 void sh4_translate_init(void);
-int cpu_sh4_signal_handler(int host_signum, void *pinfo,
-                           void *puc);
 bool superh_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
                          MMUAccessType access_type, int mmu_idx,
                          bool probe, uintptr_t retaddr);
@@ -250,7 +248,6 @@ void cpu_load_tlb(CPUSH4State * env);
 #define SUPERH_CPU_TYPE_NAME(model) model SUPERH_CPU_TYPE_SUFFIX
 #define CPU_RESOLVING_TYPE TYPE_SUPERH_CPU
 
-#define cpu_signal_handler cpu_sh4_signal_handler
 #define cpu_list sh4_cpu_list
 
 /* MMU modes definitions */

target/sparc/cpu.h

@@ -648,13 +648,11 @@ hwaddr cpu_get_phys_page_nofault(CPUSPARCState *env, target_ulong addr,
                                            int mmu_idx);
 #endif
 #endif
-int cpu_sparc_signal_handler(int host_signum, void *pinfo, void *puc);
 
 #define SPARC_CPU_TYPE_SUFFIX "-" TYPE_SPARC_CPU
 #define SPARC_CPU_TYPE_NAME(model) model SPARC_CPU_TYPE_SUFFIX
 #define CPU_RESOLVING_TYPE TYPE_SPARC_CPU
 
-#define cpu_signal_handler cpu_sparc_signal_handler
 #define cpu_list sparc_cpu_list
 
 /* MMU modes definitions */

target/tricore/cpu.h

@@ -362,7 +362,6 @@ void fpu_set_state(CPUTriCoreState *env);
 
 void tricore_cpu_list(void);
 
-#define cpu_signal_handler cpu_tricore_signal_handler
 #define cpu_list tricore_cpu_list
 
 static inline int cpu_mmu_index(CPUTriCoreState *env, bool ifetch)
@@ -377,7 +376,6 @@ typedef TriCoreCPU ArchCPU;
 
 void cpu_state_reset(CPUTriCoreState *s);
 void tricore_tcg_init(void);
-int cpu_tricore_signal_handler(int host_signum, void *pinfo, void *puc);
 
 static inline void cpu_get_tb_cpu_state(CPUTriCoreState *env, target_ulong *pc,
                                         target_ulong *cs_base, uint32_t *flags)

target/xtensa/cpu.h

@@ -581,10 +581,9 @@ void xtensa_count_regs(const XtensaConfig *config,
 int xtensa_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
 int xtensa_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
 void xtensa_cpu_do_unaligned_access(CPUState *cpu, vaddr addr,
-                                    MMUAccessType access_type,
-                                    int mmu_idx, uintptr_t retaddr);
+                                    MMUAccessType access_type, int mmu_idx,
+                                    uintptr_t retaddr) QEMU_NORETURN;
 
-#define cpu_signal_handler cpu_xtensa_signal_handler
 #define cpu_list xtensa_cpu_list
 
 #define XTENSA_CPU_TYPE_SUFFIX "-" TYPE_XTENSA_CPU
@@ -613,7 +612,6 @@ void check_interrupts(CPUXtensaState *s);
 void xtensa_irq_init(CPUXtensaState *env);
 qemu_irq *xtensa_get_extints(CPUXtensaState *env);
 qemu_irq xtensa_get_runstall(CPUXtensaState *env);
-int cpu_xtensa_signal_handler(int host_signum, void *pinfo, void *puc);
 void xtensa_cpu_list(void);
 void xtensa_sync_window_from_phys(CPUXtensaState *env);
 void xtensa_sync_phys_from_window(CPUXtensaState *env);

tcg/mips/tcg-target.c.inc

@@ -187,7 +187,7 @@ static bool patch_reloc(tcg_insn_unit *code_ptr, int type,
 #endif
 
 
-static inline bool is_p2m1(tcg_target_long val)
+static bool is_p2m1(tcg_target_long val)
 {
     return val && ((val + 1) & val) == 0;
 }
@@ -361,8 +361,8 @@ typedef enum {
 /*
  * Type reg
  */
-static inline void tcg_out_opc_reg(TCGContext *s, MIPSInsn opc,
-                                   TCGReg rd, TCGReg rs, TCGReg rt)
+static void tcg_out_opc_reg(TCGContext *s, MIPSInsn opc,
+                            TCGReg rd, TCGReg rs, TCGReg rt)
 {
     int32_t inst;
 
@@ -376,8 +376,8 @@ static inline void tcg_out_opc_reg(TCGContext *s, MIPSInsn opc,
 /*
  * Type immediate
  */
-static inline void tcg_out_opc_imm(TCGContext *s, MIPSInsn opc,
-                                   TCGReg rt, TCGReg rs, TCGArg imm)
+static void tcg_out_opc_imm(TCGContext *s, MIPSInsn opc,
+                            TCGReg rt, TCGReg rs, TCGArg imm)
 {
     int32_t inst;
 
@@ -391,8 +391,8 @@ static inline void tcg_out_opc_imm(TCGContext *s, MIPSInsn opc,
 /*
  * Type bitfield
  */
-static inline void tcg_out_opc_bf(TCGContext *s, MIPSInsn opc, TCGReg rt,
-                                  TCGReg rs, int msb, int lsb)
+static void tcg_out_opc_bf(TCGContext *s, MIPSInsn opc, TCGReg rt,
+                           TCGReg rs, int msb, int lsb)
 {
     int32_t inst;
 
@@ -404,8 +404,8 @@ static inline void tcg_out_opc_bf(TCGContext *s, MIPSInsn opc, TCGReg rt,
     tcg_out32(s, inst);
 }
 
-static inline void tcg_out_opc_bf64(TCGContext *s, MIPSInsn opc, MIPSInsn opm,
-                                    MIPSInsn oph, TCGReg rt, TCGReg rs,
+static void tcg_out_opc_bf64(TCGContext *s, MIPSInsn opc, MIPSInsn opm,
+                             MIPSInsn oph, TCGReg rt, TCGReg rs,
                                     int msb, int lsb)
 {
     if (lsb >= 32) {
@@ -422,8 +422,7 @@ static inline void tcg_out_opc_bf64(TCGContext *s, MIPSInsn opc, MIPSInsn opm,
 /*
  * Type branch
  */
-static inline void tcg_out_opc_br(TCGContext *s, MIPSInsn opc,
-                                  TCGReg rt, TCGReg rs)
+static void tcg_out_opc_br(TCGContext *s, MIPSInsn opc, TCGReg rt, TCGReg rs)
 {
     tcg_out_opc_imm(s, opc, rt, rs, 0);
 }
@@ -431,8 +430,8 @@ static inline void tcg_out_opc_br(TCGContext *s, MIPSInsn opc,
 /*
  * Type sa
  */
-static inline void tcg_out_opc_sa(TCGContext *s, MIPSInsn opc,
-                                  TCGReg rd, TCGReg rt, TCGArg sa)
+static void tcg_out_opc_sa(TCGContext *s, MIPSInsn opc,
+                           TCGReg rd, TCGReg rt, TCGArg sa)
 {
     int32_t inst;
 
@@ -479,28 +478,27 @@ static bool tcg_out_opc_jmp(TCGContext *s, MIPSInsn opc, const void *target)
     return true;
 }
 
-static inline void tcg_out_nop(TCGContext *s)
+static void tcg_out_nop(TCGContext *s)
 {
     tcg_out32(s, 0);
 }
 
-static inline void tcg_out_dsll(TCGContext *s, TCGReg rd, TCGReg rt, TCGArg sa)
+static void tcg_out_dsll(TCGContext *s, TCGReg rd, TCGReg rt, TCGArg sa)
 {
     tcg_out_opc_sa64(s, OPC_DSLL, OPC_DSLL32, rd, rt, sa);
 }
 
-static inline void tcg_out_dsrl(TCGContext *s, TCGReg rd, TCGReg rt, TCGArg sa)
+static void tcg_out_dsrl(TCGContext *s, TCGReg rd, TCGReg rt, TCGArg sa)
 {
     tcg_out_opc_sa64(s, OPC_DSRL, OPC_DSRL32, rd, rt, sa);
 }
 
-static inline void tcg_out_dsra(TCGContext *s, TCGReg rd, TCGReg rt, TCGArg sa)
+static void tcg_out_dsra(TCGContext *s, TCGReg rd, TCGReg rt, TCGArg sa)
 {
     tcg_out_opc_sa64(s, OPC_DSRA, OPC_DSRA32, rd, rt, sa);
 }
 
-static inline bool tcg_out_mov(TCGContext *s, TCGType type,
-                               TCGReg ret, TCGReg arg)
+static bool tcg_out_mov(TCGContext *s, TCGType type, TCGReg ret, TCGReg arg)
 {
     /* Simple reg-reg move, optimising out the 'do nothing' case */
     if (ret != arg) {
@@ -575,8 +573,10 @@ static void tcg_out_bswap16(TCGContext *s, TCGReg ret, TCGReg arg, int flags)
 
 static void tcg_out_bswap_subr(TCGContext *s, const tcg_insn_unit *sub)
 {
-    bool ok = tcg_out_opc_jmp(s, OPC_JAL, sub);
-    tcg_debug_assert(ok);
+    if (!tcg_out_opc_jmp(s, OPC_JAL, sub)) {
+        tcg_out_movi(s, TCG_TYPE_PTR, TCG_TMP1, (uintptr_t)sub);
+        tcg_out_opc_reg(s, OPC_JALR, TCG_REG_RA, TCG_TMP1, 0);
+    }
 }
 
 static void tcg_out_bswap32(TCGContext *s, TCGReg ret, TCGReg arg, int flags)
@@ -612,27 +612,7 @@ static void tcg_out_bswap64(TCGContext *s, TCGReg ret, TCGReg arg)
     }
 }
 
-static inline void tcg_out_ext8s(TCGContext *s, TCGReg ret, TCGReg arg)
-{
-    if (use_mips32r2_instructions) {
-        tcg_out_opc_reg(s, OPC_SEB, ret, 0, arg);
-    } else {
-        tcg_out_opc_sa(s, OPC_SLL, ret, arg, 24);
-        tcg_out_opc_sa(s, OPC_SRA, ret, ret, 24);
-    }
-}
-
-static inline void tcg_out_ext16s(TCGContext *s, TCGReg ret, TCGReg arg)
-{
-    if (use_mips32r2_instructions) {
-        tcg_out_opc_reg(s, OPC_SEH, ret, 0, arg);
-    } else {
-        tcg_out_opc_sa(s, OPC_SLL, ret, arg, 16);
-        tcg_out_opc_sa(s, OPC_SRA, ret, ret, 16);
-    }
-}
-
-static inline void tcg_out_ext32u(TCGContext *s, TCGReg ret, TCGReg arg)
+static void tcg_out_ext32u(TCGContext *s, TCGReg ret, TCGReg arg)
 {
     if (use_mips32r2_instructions) {
         tcg_out_opc_bf(s, OPC_DEXT, ret, arg, 31, 0);
@@ -656,8 +636,8 @@ static void tcg_out_ldst(TCGContext *s, MIPSInsn opc, TCGReg data,
     tcg_out_opc_imm(s, opc, data, addr, lo);
 }
 
-static inline void tcg_out_ld(TCGContext *s, TCGType type, TCGReg arg,
-                              TCGReg arg1, intptr_t arg2)
+static void tcg_out_ld(TCGContext *s, TCGType type, TCGReg arg,
+                       TCGReg arg1, intptr_t arg2)
 {
     MIPSInsn opc = OPC_LD;
     if (TCG_TARGET_REG_BITS == 32 || type == TCG_TYPE_I32) {
@@ -666,8 +646,8 @@ static inline void tcg_out_ld(TCGContext *s, TCGType type, TCGReg arg,
     tcg_out_ldst(s, opc, arg, arg1, arg2);
 }
 
-static inline void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg,
-                              TCGReg arg1, intptr_t arg2)
+static void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg,
+                       TCGReg arg1, intptr_t arg2)
 {
     MIPSInsn opc = OPC_SD;
     if (TCG_TARGET_REG_BITS == 32 || type == TCG_TYPE_I32) {
@@ -676,8 +656,8 @@ static inline void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg,
     tcg_out_ldst(s, opc, arg, arg1, arg2);
 }
 
-static inline bool tcg_out_sti(TCGContext *s, TCGType type, TCGArg val,
-                               TCGReg base, intptr_t ofs)
+static bool tcg_out_sti(TCGContext *s, TCGType type, TCGArg val,
+                        TCGReg base, intptr_t ofs)
 {
     if (val == 0) {
         tcg_out_st(s, type, TCG_REG_ZERO, base, ofs);
@@ -1637,9 +1617,9 @@ static void tcg_out_clz(TCGContext *s, MIPSInsn opcv2, MIPSInsn opcv6,
     }
 }
 
-static inline void tcg_out_op(TCGContext *s, TCGOpcode opc,
-                              const TCGArg args[TCG_MAX_OP_ARGS],
-                              const int const_args[TCG_MAX_OP_ARGS])
+static void tcg_out_op(TCGContext *s, TCGOpcode opc,
+                       const TCGArg args[TCG_MAX_OP_ARGS],
+                       const int const_args[TCG_MAX_OP_ARGS])
 {
     MIPSInsn i1, i2;
     TCGArg a0, a1, a2;
@@ -1674,17 +1654,11 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc,
         }
         break;
     case INDEX_op_goto_tb:
-        if (s->tb_jmp_insn_offset) {
-            /* direct jump method */
-            s->tb_jmp_insn_offset[a0] = tcg_current_code_size(s);
-            /* Avoid clobbering the address during retranslation.  */
-            tcg_out32(s, OPC_J | (*(uint32_t *)s->code_ptr & 0x3ffffff));
-        } else {
-            /* indirect jump method */
-            tcg_out_ld(s, TCG_TYPE_PTR, TCG_TMP0, TCG_REG_ZERO,
-                       (uintptr_t)(s->tb_jmp_target_addr + a0));
-            tcg_out_opc_reg(s, OPC_JR, 0, TCG_TMP0, 0);
-        }
+        /* indirect jump method */
+        tcg_debug_assert(s->tb_jmp_insn_offset == 0);
+        tcg_out_ld(s, TCG_TYPE_PTR, TCG_TMP0, TCG_REG_ZERO,
+                   (uintptr_t)(s->tb_jmp_target_addr + a0));
+        tcg_out_opc_reg(s, OPC_JR, 0, TCG_TMP0, 0);
         tcg_out_nop(s);
         set_jmp_reset_offset(s, a0);
         break;
@@ -2558,13 +2532,6 @@ static void tcg_target_init(TCGContext *s)
     tcg_regset_set_reg(s->reserved_regs, TCG_REG_GP);   /* global pointer */
 }
 
-void tb_target_set_jmp_target(uintptr_t tc_ptr, uintptr_t jmp_rx,
-                              uintptr_t jmp_rw, uintptr_t addr)
-{
-    qatomic_set((uint32_t *)jmp_rw, deposit32(OPC_J, 0, 26, addr >> 2));
-    flush_idcache_range(jmp_rx, jmp_rw, 4);
-}
-
 typedef struct {
     DebugFrameHeader h;
     uint8_t fde_def_cfa[4];

tcg/mips/tcg-target.h

@@ -39,11 +39,7 @@
 #define TCG_TARGET_TLB_DISPLACEMENT_BITS 16
 #define TCG_TARGET_NB_REGS 32
 
-/*
- * We have a 256MB branch region, but leave room to make sure the
- * main executable is also within that region.
- */
-#define MAX_CODE_GEN_BUFFER_SIZE  (128 * MiB)
+#define MAX_CODE_GEN_BUFFER_SIZE  ((size_t)-1)
 
 typedef enum {
     TCG_REG_ZERO = 0,
@@ -136,7 +132,7 @@ extern bool use_mips32r2_instructions;
 #define TCG_TARGET_HAS_muluh_i32        1
 #define TCG_TARGET_HAS_mulsh_i32        1
 #define TCG_TARGET_HAS_bswap32_i32      1
-#define TCG_TARGET_HAS_direct_jump      1
+#define TCG_TARGET_HAS_direct_jump      0
 
 #if TCG_TARGET_REG_BITS == 64
 #define TCG_TARGET_HAS_add2_i32         0
@@ -207,7 +203,9 @@ extern bool use_mips32r2_instructions;
 #define TCG_TARGET_DEFAULT_MO (0)
 #define TCG_TARGET_HAS_MEMORY_BSWAP     1
 
-void tb_target_set_jmp_target(uintptr_t, uintptr_t, uintptr_t, uintptr_t);
+/* not defined -- call should be eliminated at compile time */
+void tb_target_set_jmp_target(uintptr_t, uintptr_t, uintptr_t, uintptr_t)
+    QEMU_ERROR("code path is reachable");
 
 #ifdef CONFIG_SOFTMMU
 #define TCG_TARGET_NEED_LDST_LABELS

tcg/region.c

@@ -467,38 +467,6 @@ static size_t tcg_n_regions(size_t tb_size, unsigned max_cpus)
   (DEFAULT_CODE_GEN_BUFFER_SIZE_1 < MAX_CODE_GEN_BUFFER_SIZE \
    ? DEFAULT_CODE_GEN_BUFFER_SIZE_1 : MAX_CODE_GEN_BUFFER_SIZE)
 
-#ifdef __mips__
-/*
- * In order to use J and JAL within the code_gen_buffer, we require
- * that the buffer not cross a 256MB boundary.
- */
-static inline bool cross_256mb(void *addr, size_t size)
-{
-    return ((uintptr_t)addr ^ ((uintptr_t)addr + size)) & ~0x0ffffffful;
-}
-
-/*
- * We weren't able to allocate a buffer without crossing that boundary,
- * so make do with the larger portion of the buffer that doesn't cross.
- * Returns the new base and size of the buffer in *obuf and *osize.
- */
-static inline void split_cross_256mb(void **obuf, size_t *osize,
-                                     void *buf1, size_t size1)
-{
-    void *buf2 = (void *)(((uintptr_t)buf1 + size1) & ~0x0ffffffful);
-    size_t size2 = buf1 + size1 - buf2;
-
-    size1 = buf2 - buf1;
-    if (size1 < size2) {
-        size1 = size2;
-        buf1 = buf2;
-    }
-
-    *obuf = buf1;
-    *osize = size1;
-}
-#endif
-
 #ifdef USE_STATIC_CODE_GEN_BUFFER
 static uint8_t static_code_gen_buffer[DEFAULT_CODE_GEN_BUFFER_SIZE]
     __attribute__((aligned(CODE_GEN_ALIGN)));
@@ -526,12 +494,6 @@ static int alloc_code_gen_buffer(size_t tb_size, int splitwx, Error **errp)
         size = QEMU_ALIGN_DOWN(tb_size, qemu_real_host_page_size);
     }
 
-#ifdef __mips__
-    if (cross_256mb(buf, size)) {
-        split_cross_256mb(&buf, &size, buf, size);
-    }
-#endif
-
     region.start_aligned = buf;
     region.total_size = size;
 
@@ -573,39 +535,6 @@ static int alloc_code_gen_buffer_anon(size_t size, int prot,
         return -1;
     }
 
-#ifdef __mips__
-    if (cross_256mb(buf, size)) {
-        /*
-         * Try again, with the original still mapped, to avoid re-acquiring
-         * the same 256mb crossing.
-         */
-        size_t size2;
-        void *buf2 = mmap(NULL, size, prot, flags, -1, 0);
-        switch ((int)(buf2 != MAP_FAILED)) {
-        case 1:
-            if (!cross_256mb(buf2, size)) {
-                /* Success!  Use the new buffer.  */
-                munmap(buf, size);
-                break;
-            }
-            /* Failure.  Work with what we had.  */
-            munmap(buf2, size);
-            /* fallthru */
-        default:
-            /* Split the original buffer.  Free the smaller half.  */
-            split_cross_256mb(&buf2, &size2, buf, size);
-            if (buf == buf2) {
-                munmap(buf + size2, size - size2);
-            } else {
-                munmap(buf, size - size2);
-            }
-            size = size2;
-            break;
-        }
-        buf = buf2;
-    }
-#endif
-
     region.start_aligned = buf;
     region.total_size = size;
     return prot;
@@ -620,35 +549,15 @@ static bool alloc_code_gen_buffer_splitwx_memfd(size_t size, Error **errp)
     void *buf_rw = NULL, *buf_rx = MAP_FAILED;
     int fd = -1;
 
-#ifdef __mips__
-    /* Find space for the RX mapping, vs the 256MiB regions. */
-    if (alloc_code_gen_buffer_anon(size, PROT_NONE,
-                                   MAP_PRIVATE | MAP_ANONYMOUS |
-                                   MAP_NORESERVE, errp) < 0) {
-        return false;
-    }
-    /* The size of the mapping may have been adjusted. */
-    buf_rx = region.start_aligned;
-    size = region.total_size;
-#endif
-
     buf_rw = qemu_memfd_alloc("tcg-jit", size, 0, &fd, errp);
     if (buf_rw == NULL) {
         goto fail;
     }
 
-#ifdef __mips__
-    void *tmp = mmap(buf_rx, size, PROT_READ | PROT_EXEC,
-                     MAP_SHARED | MAP_FIXED, fd, 0);
-    if (tmp != buf_rx) {
-        goto fail_rx;
-    }
-#else
     buf_rx = mmap(NULL, size, PROT_READ | PROT_EXEC, MAP_SHARED, fd, 0);
     if (buf_rx == MAP_FAILED) {
         goto fail_rx;
     }
-#endif
 
     close(fd);
     region.start_aligned = buf_rw;

tcg/riscv/tcg-target.c.inc

@@ -1130,10 +1130,7 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, bool is_64)
         tcg_out_ext32u(s, base, addr_regl);
         addr_regl = base;
     }
-
-    if (guest_base == 0) {
-        tcg_out_opc_reg(s, OPC_ADD, base, addr_regl, TCG_REG_ZERO);
-    } else {
+    if (guest_base != 0) {
         tcg_out_opc_reg(s, OPC_ADD, base, TCG_GUEST_BASE_REG, addr_regl);
     }
     tcg_out_qemu_ld_direct(s, data_regl, data_regh, base, opc, is_64);
@@ -1199,10 +1196,7 @@ static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, bool is_64)
         tcg_out_ext32u(s, base, addr_regl);
         addr_regl = base;
     }
-
-    if (guest_base == 0) {
-        tcg_out_opc_reg(s, OPC_ADD, base, addr_regl, TCG_REG_ZERO);
-    } else {
+    if (guest_base != 0) {
         tcg_out_opc_reg(s, OPC_ADD, base, TCG_GUEST_BASE_REG, addr_regl);
     }
     tcg_out_qemu_st_direct(s, data_regl, data_regh, base, opc);

tcg/sparc/tcg-target.c.inc

@@ -294,12 +294,12 @@ static const int tcg_target_call_oarg_regs[] = {
 bool use_vis3_instructions;
 #endif
 
-static inline int check_fit_i64(int64_t val, unsigned int bits)
+static bool check_fit_i64(int64_t val, unsigned int bits)
 {
     return val == sextract64(val, 0, bits);
 }
 
-static inline int check_fit_i32(int32_t val, unsigned int bits)
+static bool check_fit_i32(int32_t val, unsigned int bits)
 {
     return val == sextract32(val, 0, bits);
 }
@@ -362,14 +362,19 @@ static bool tcg_target_const_match(int64_t val, TCGType type, int ct)
     }
 }
 
-static inline void tcg_out_arith(TCGContext *s, TCGReg rd, TCGReg rs1,
-                                 TCGReg rs2, int op)
+static void tcg_out_nop(TCGContext *s)
+{
+    tcg_out32(s, NOP);
+}
+
+static void tcg_out_arith(TCGContext *s, TCGReg rd, TCGReg rs1,
+                          TCGReg rs2, int op)
 {
     tcg_out32(s, op | INSN_RD(rd) | INSN_RS1(rs1) | INSN_RS2(rs2));
 }
 
-static inline void tcg_out_arithi(TCGContext *s, TCGReg rd, TCGReg rs1,
-                                  int32_t offset, int op)
+static void tcg_out_arithi(TCGContext *s, TCGReg rd, TCGReg rs1,
+                           int32_t offset, int op)
 {
     tcg_out32(s, op | INSN_RD(rd) | INSN_RS1(rs1) | INSN_IMM13(offset));
 }
@@ -381,8 +386,7 @@ static void tcg_out_arithc(TCGContext *s, TCGReg rd, TCGReg rs1,
               | (val2const ? INSN_IMM13(val2) : INSN_RS2(val2)));
 }
 
-static inline bool tcg_out_mov(TCGContext *s, TCGType type,
-                               TCGReg ret, TCGReg arg)
+static bool tcg_out_mov(TCGContext *s, TCGType type, TCGReg ret, TCGReg arg)
 {
     if (ret != arg) {
         tcg_out_arith(s, ret, arg, TCG_REG_G0, ARITH_OR);
@@ -390,12 +394,21 @@ static inline bool tcg_out_mov(TCGContext *s, TCGType type,
     return true;
 }
 
-static inline void tcg_out_sethi(TCGContext *s, TCGReg ret, uint32_t arg)
+static void tcg_out_mov_delay(TCGContext *s, TCGReg ret, TCGReg arg)
+{
+    if (ret != arg) {
+        tcg_out_arith(s, ret, arg, TCG_REG_G0, ARITH_OR);
+    } else {
+        tcg_out_nop(s);
+    }
+}
+
+static void tcg_out_sethi(TCGContext *s, TCGReg ret, uint32_t arg)
 {
     tcg_out32(s, SETHI | INSN_RD(ret) | ((arg & 0xfffffc00) >> 10));
 }
 
-static inline void tcg_out_movi_imm13(TCGContext *s, TCGReg ret, int32_t arg)
+static void tcg_out_movi_imm13(TCGContext *s, TCGReg ret, int32_t arg)
 {
     tcg_out_arithi(s, ret, TCG_REG_G0, arg, ARITH_OR);
 }
@@ -470,14 +483,14 @@ static void tcg_out_movi_int(TCGContext *s, TCGType type, TCGReg ret,
     }
 }
 
-static inline void tcg_out_movi(TCGContext *s, TCGType type,
-                                TCGReg ret, tcg_target_long arg)
+static void tcg_out_movi(TCGContext *s, TCGType type,
+                         TCGReg ret, tcg_target_long arg)
 {
     tcg_out_movi_int(s, type, ret, arg, false);
 }
 
-static inline void tcg_out_ldst_rr(TCGContext *s, TCGReg data, TCGReg a1,
-                                   TCGReg a2, int op)
+static void tcg_out_ldst_rr(TCGContext *s, TCGReg data, TCGReg a1,
+                            TCGReg a2, int op)
 {
     tcg_out32(s, op | INSN_RD(data) | INSN_RS1(a1) | INSN_RS2(a2));
 }
@@ -494,20 +507,20 @@ static void tcg_out_ldst(TCGContext *s, TCGReg ret, TCGReg addr,
     }
 }
 
-static inline void tcg_out_ld(TCGContext *s, TCGType type, TCGReg ret,
-                              TCGReg arg1, intptr_t arg2)
+static void tcg_out_ld(TCGContext *s, TCGType type, TCGReg ret,
+                       TCGReg arg1, intptr_t arg2)
 {
     tcg_out_ldst(s, ret, arg1, arg2, (type == TCG_TYPE_I32 ? LDUW : LDX));
 }
 
-static inline void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg,
-                              TCGReg arg1, intptr_t arg2)
+static void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg,
+                       TCGReg arg1, intptr_t arg2)
 {
     tcg_out_ldst(s, arg, arg1, arg2, (type == TCG_TYPE_I32 ? STW : STX));
 }
 
-static inline bool tcg_out_sti(TCGContext *s, TCGType type, TCGArg val,
-                               TCGReg base, intptr_t ofs)
+static bool tcg_out_sti(TCGContext *s, TCGType type, TCGArg val,
+                        TCGReg base, intptr_t ofs)
 {
     if (val == 0) {
         tcg_out_st(s, type, TCG_REG_G0, base, ofs);
@@ -527,12 +540,12 @@ static void tcg_out_ld_ptr(TCGContext *s, TCGReg ret, const void *arg)
     tcg_out_ld(s, TCG_TYPE_PTR, ret, ret, (uintptr_t)arg & 0x3ff);
 }
 
-static inline void tcg_out_sety(TCGContext *s, TCGReg rs)
+static void tcg_out_sety(TCGContext *s, TCGReg rs)
 {
     tcg_out32(s, WRY | INSN_RS1(TCG_REG_G0) | INSN_RS2(rs));
 }
 
-static inline void tcg_out_rdy(TCGContext *s, TCGReg rd)
+static void tcg_out_rdy(TCGContext *s, TCGReg rd)
 {
     tcg_out32(s, RDY | INSN_RD(rd));
 }
@@ -552,11 +565,6 @@ static void tcg_out_div32(TCGContext *s, TCGReg rd, TCGReg rs1,
                    uns ? ARITH_UDIV : ARITH_SDIV);
 }
 
-static inline void tcg_out_nop(TCGContext *s)
-{
-    tcg_out32(s, NOP);
-}
-
 static const uint8_t tcg_cond_to_bcond[] = {
     [TCG_COND_EQ] = COND_E,
     [TCG_COND_NE] = COND_NE,
@@ -1350,7 +1358,7 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc,
     case INDEX_op_goto_ptr:
         tcg_out_arithi(s, TCG_REG_G0, a0, 0, JMPL);
         if (USE_REG_TB) {
-            tcg_out_arith(s, TCG_REG_TB, a0, TCG_REG_G0, ARITH_OR);
+            tcg_out_mov_delay(s, TCG_REG_TB, a0);
         } else {
             tcg_out_nop(s);
         }