/* allow to see translation results - the slowdown should be negligible, so we leave it */
#define DEBUG_DISAS
-#ifndef glue
-#define xglue(x, y) x ## y
-#define glue(x, y) xglue(x, y)
-#define stringify(s) tostring(s)
-#define tostring(s) #s
-#endif
-
-#ifndef likely
-#if __GNUC__ < 3
-#define __builtin_expect(x, n) (x)
-#endif
-
-#define likely(x) __builtin_expect(!!(x), 1)
-#define unlikely(x) __builtin_expect(!!(x), 0)
-#endif
-
-#ifndef always_inline
-#if (__GNUC__ < 3) || defined(__APPLE__)
-#define always_inline inline
-#else
-#define always_inline __attribute__ (( always_inline )) inline
-#endif
-#endif
-
-#ifdef __i386__
-#define REGPARM(n) __attribute((regparm(n)))
-#else
-#define REGPARM(n)
-#endif
-
/* is_jmp field values */
#define DISAS_NEXT 0 /* next instruction can be analyzed */
#define DISAS_JUMP 1 /* only pc was modified dynamically */
/* XXX: make safe guess about sizes */
#define MAX_OP_PER_INSTR 32
+/* A Call op needs up to 6 + 2N parameters (N = number of arguments). */
+#define MAX_OPC_PARAM 10
#define OPC_BUF_SIZE 512
#define OPC_MAX_SIZE (OPC_BUF_SIZE - MAX_OP_PER_INSTR)
-#define OPPARAM_BUF_SIZE (OPC_BUF_SIZE * 3)
+/* Maximum size a TCG op can expand to. This is complicated because a
+ single op may require several host instructions and regirster reloads.
+ For now take a wild guess at 128 bytes, which should allow at least
+ a couple of fixup instructions per argument. */
+#define TCG_MAX_OP_SIZE 128
+
+#define OPPARAM_BUF_SIZE (OPC_BUF_SIZE * MAX_OPC_PARAM)
-extern uint16_t gen_opc_buf[OPC_BUF_SIZE];
-extern uint32_t gen_opparam_buf[OPPARAM_BUF_SIZE];
-extern long gen_labels[OPC_BUF_SIZE];
-extern int nb_gen_labels;
extern target_ulong gen_opc_pc[OPC_BUF_SIZE];
extern target_ulong gen_opc_npc[OPC_BUF_SIZE];
extern uint8_t gen_opc_cc_op[OPC_BUF_SIZE];
extern FILE *logfile;
extern int loglevel;
-void muls64(int64_t *phigh, int64_t *plow, int64_t a, int64_t b);
-void mulu64(uint64_t *phigh, uint64_t *plow, uint64_t a, uint64_t b);
-
int gen_intermediate_code(CPUState *env, struct TranslationBlock *tb);
int gen_intermediate_code_pc(CPUState *env, struct TranslationBlock *tb);
-void dump_ops(const uint16_t *opc_buf, const uint32_t *opparam_buf);
+unsigned long code_gen_max_block_size(void);
+void cpu_gen_init(void);
int cpu_gen_code(CPUState *env, struct TranslationBlock *tb,
- int max_code_size, int *gen_code_size_ptr);
+ int *gen_code_size_ptr);
int cpu_restore_state(struct TranslationBlock *tb,
CPUState *env, unsigned long searched_pc,
void *puc);
void cpu_resume_from_signal(CPUState *env1, void *puc);
void cpu_exec_init(CPUState *env);
int page_unprotect(target_ulong address, unsigned long pc, void *puc);
-void tb_invalidate_phys_page_range(target_ulong start, target_ulong end,
+void tb_invalidate_phys_page_range(target_phys_addr_t start, target_phys_addr_t end,
int is_cpu_write_access);
void tb_invalidate_page_range(target_ulong start, target_ulong end);
void tlb_flush_page(CPUState *env, target_ulong addr);
void tlb_flush(CPUState *env, int flush_global);
int tlb_set_page_exec(CPUState *env, target_ulong vaddr,
target_phys_addr_t paddr, int prot,
- int is_user, int is_softmmu);
+ int mmu_idx, int is_softmmu);
static inline int tlb_set_page(CPUState *env, target_ulong vaddr,
target_phys_addr_t paddr, int prot,
- int is_user, int is_softmmu)
+ int mmu_idx, int is_softmmu)
{
if (prot & PAGE_READ)
prot |= PAGE_EXEC;
- return tlb_set_page_exec(env, vaddr, paddr, prot, is_user, is_softmmu);
+ return tlb_set_page_exec(env, vaddr, paddr, prot, mmu_idx, is_softmmu);
}
-#define CODE_GEN_MAX_SIZE 65536
#define CODE_GEN_ALIGN 16 /* must be >= of the size of a icache line */
#define CODE_GEN_PHYS_HASH_BITS 15
#elif defined(__powerpc__)
#define CODE_GEN_BUFFER_SIZE (6 * 1024 * 1024)
#else
+/* XXX: make it dynamic on x86 */
#define CODE_GEN_BUFFER_SIZE (16 * 1024 * 1024)
#endif
#define CODE_GEN_MAX_BLOCKS (CODE_GEN_BUFFER_SIZE / CODE_GEN_AVG_BLOCK_SIZE)
-#if defined(__powerpc__)
+#if defined(__powerpc__) || defined(__x86_64__)
#define USE_DIRECT_JUMP
#endif
#if defined(__i386__) && !defined(_WIN32)
#ifdef USE_DIRECT_JUMP
uint16_t tb_jmp_offset[4]; /* offset of jump instruction */
#else
- uint32_t tb_next[2]; /* address of jump generated code */
+ unsigned long tb_next[2]; /* address of jump generated code */
#endif
/* list of TBs jumping to this one. This is a circular list using
the two least significant bits of the pointers to tell what is
asm volatile ("sync" : : : "memory");
asm volatile ("isync" : : : "memory");
}
-#elif defined(__i386__)
+#elif defined(__i386__) || defined(__x86_64__)
static inline void tb_set_jmp_target1(unsigned long jmp_addr, unsigned long addr)
{
/* patch the branch destination */
#define ASM_OP_LABEL_NAME(n, opname) \
ASM_NAME(__op_label) #n "." ASM_NAME(opname)
-#if defined(__powerpc__)
+extern CPUWriteMemoryFunc *io_mem_write[IO_MEM_NB_ENTRIES][4];
+extern CPUReadMemoryFunc *io_mem_read[IO_MEM_NB_ENTRIES][4];
+extern void *io_mem_opaque[IO_MEM_NB_ENTRIES];
-/* we patch the jump instruction directly */
-#define GOTO_TB(opname, tbparam, n)\
-do {\
- asm volatile (ASM_DATA_SECTION\
- ASM_OP_LABEL_NAME(n, opname) ":\n"\
- ".long 1f\n"\
- ASM_PREVIOUS_SECTION \
- "b " ASM_NAME(__op_jmp) #n "\n"\
- "1:\n");\
-} while (0)
-
-#elif defined(__i386__) && defined(USE_DIRECT_JUMP)
-
-/* we patch the jump instruction directly */
-#define GOTO_TB(opname, tbparam, n)\
-do {\
- asm volatile (".section .data\n"\
- ASM_OP_LABEL_NAME(n, opname) ":\n"\
- ".long 1f\n"\
- ASM_PREVIOUS_SECTION \
- "jmp " ASM_NAME(__op_jmp) #n "\n"\
- "1:\n");\
-} while (0)
+#if defined(__hppa__)
-#elif defined(__s390__)
-/* GCC spills R13, so we have to restore it before branching away */
-
-#define GOTO_TB(opname, tbparam, n)\
-do {\
- static void __attribute__((used)) *dummy ## n = &&dummy_label ## n;\
- static void __attribute__((used)) *__op_label ## n \
- __asm__(ASM_OP_LABEL_NAME(n, opname)) = &&label ## n;\
- __asm__ __volatile__ ( \
- "l %%r13,52(%%r15)\n" \
- "br %0\n" \
- : : "r" (((TranslationBlock*)tbparam)->tb_next[n]));\
- \
- for(;*((int*)0);); /* just to keep GCC busy */ \
-label ## n: ;\
-dummy_label ## n: ;\
-} while(0)
+typedef int spinlock_t[4];
+
+#define SPIN_LOCK_UNLOCKED { 1, 1, 1, 1 }
+
+static inline void resetlock (spinlock_t *p)
+{
+ (*p)[0] = (*p)[1] = (*p)[2] = (*p)[3] = 1;
+}
#else
-/* jump to next block operations (more portable code, does not need
- cache flushing, but slower because of indirect jump) */
-#define GOTO_TB(opname, tbparam, n)\
-do {\
- static void __attribute__((used)) *dummy ## n = &&dummy_label ## n;\
- static void __attribute__((used)) *__op_label ## n \
- __asm__(ASM_OP_LABEL_NAME(n, opname)) = &&label ## n;\
- goto *(void *)(((TranslationBlock *)tbparam)->tb_next[n]);\
-label ## n: ;\
-dummy_label ## n: ;\
-} while (0)
+typedef int spinlock_t;
-#endif
+#define SPIN_LOCK_UNLOCKED 0
-extern CPUWriteMemoryFunc *io_mem_write[IO_MEM_NB_ENTRIES][4];
-extern CPUReadMemoryFunc *io_mem_read[IO_MEM_NB_ENTRIES][4];
-extern void *io_mem_opaque[IO_MEM_NB_ENTRIES];
+static inline void resetlock (spinlock_t *p)
+{
+ *p = SPIN_LOCK_UNLOCKED;
+}
+
+#endif
#if defined(__powerpc__)
static inline int testandset (int *p)
: "cc","memory");
return ret;
}
+#elif defined(__hppa__)
+
+/* Because malloc only guarantees 8-byte alignment for malloc'd data,
+ and GCC only guarantees 8-byte alignment for stack locals, we can't
+ be assured of 16-byte alignment for atomic lock data even if we
+ specify "__attribute ((aligned(16)))" in the type declaration. So,
+ we use a struct containing an array of four ints for the atomic lock
+ type and dynamically select the 16-byte aligned int from the array
+ for the semaphore. */
+#define __PA_LDCW_ALIGNMENT 16
+static inline void *ldcw_align (void *p) {
+ unsigned long a = (unsigned long)p;
+ a = (a + __PA_LDCW_ALIGNMENT - 1) & ~(__PA_LDCW_ALIGNMENT - 1);
+ return (void *)a;
+}
+
+static inline int testandset (spinlock_t *p)
+{
+ unsigned int ret;
+ p = ldcw_align(p);
+ __asm__ __volatile__("ldcw 0(%1),%0"
+ : "=r" (ret)
+ : "r" (p)
+ : "memory" );
+ return !ret;
+}
+
#elif defined(__ia64)
#include <ia64intrin.h>
#error unimplemented CPU support
#endif
-typedef int spinlock_t;
-
-#define SPIN_LOCK_UNLOCKED 0
-
#if defined(CONFIG_USER_ONLY)
static inline void spin_lock(spinlock_t *lock)
{
static inline void spin_unlock(spinlock_t *lock)
{
- *lock = 0;
+ resetlock(lock);
}
static inline int spin_trylock(spinlock_t *lock)
#if !defined(CONFIG_USER_ONLY)
-void tlb_fill(target_ulong addr, int is_write, int is_user,
+void tlb_fill(target_ulong addr, int is_write, int mmu_idx,
void *retaddr);
-#define ACCESS_TYPE 3
+#define ACCESS_TYPE (NB_MMU_MODES + 1)
#define MEMSUFFIX _code
#define env cpu_single_env
is the offset relative to phys_ram_base */
static inline target_ulong get_phys_addr_code(CPUState *env, target_ulong addr)
{
- int is_user, index, pd;
+ int mmu_idx, index, pd;
index = (addr >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
-#if defined(TARGET_I386)
- is_user = ((env->hflags & HF_CPL_MASK) == 3);
-#elif defined (TARGET_PPC)
- is_user = msr_pr;
-#elif defined (TARGET_MIPS)
- is_user = ((env->hflags & MIPS_HFLAG_MODE) == MIPS_HFLAG_UM);
-#elif defined (TARGET_SPARC)
- is_user = (env->psrs == 0);
-#elif defined (TARGET_ARM)
- is_user = ((env->uncached_cpsr & CPSR_M) == ARM_CPU_MODE_USR);
-#elif defined (TARGET_SH4)
- is_user = ((env->sr & SR_MD) == 0);
-#elif defined (TARGET_ALPHA)
- is_user = ((env->ps >> 3) & 3);
-#elif defined (TARGET_M68K)
- is_user = ((env->sr & SR_S) == 0);
-#else
-#error unimplemented CPU
-#endif
- if (__builtin_expect(env->tlb_table[is_user][index].addr_code !=
+ mmu_idx = cpu_mmu_index(env);
+ if (__builtin_expect(env->tlb_table[mmu_idx][index].addr_code !=
(addr & TARGET_PAGE_MASK), 0)) {
ldub_code(addr);
}
- pd = env->tlb_table[is_user][index].addr_code & ~TARGET_PAGE_MASK;
+ pd = env->tlb_table[mmu_idx][index].addr_code & ~TARGET_PAGE_MASK;
if (pd > IO_MEM_ROM && !(pd & IO_MEM_ROMD)) {
-#ifdef TARGET_SPARC
+#if defined(TARGET_SPARC) || defined(TARGET_MIPS)
do_unassigned_access(addr, 0, 1, 0);
#else
cpu_abort(env, "Trying to execute code outside RAM or ROM at 0x" TARGET_FMT_lx "\n", addr);
#endif
}
- return addr + env->tlb_table[is_user][index].addend - (unsigned long)phys_ram_base;
+ return addr + env->tlb_table[mmu_idx][index].addend - (unsigned long)phys_ram_base;
}
#endif