diff options
Diffstat (limited to 'accel/tcg/translate-all.c')
-rw-r--r-- | accel/tcg/translate-all.c | 154 |
1 files changed, 61 insertions, 93 deletions
diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index fdf6d8a..d468667 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -21,48 +21,20 @@ #include "trace.h" #include "disas/disas.h" -#include "exec/exec-all.h" #include "tcg/tcg.h" -#if defined(CONFIG_USER_ONLY) -#include "qemu.h" -#if defined(__FreeBSD__) || defined(__FreeBSD_kernel__) -#include <sys/param.h> -#if __FreeBSD_version >= 700104 -#define HAVE_KINFO_GETVMMAP -#define sigqueue sigqueue_freebsd /* avoid redefinition */ -#include <sys/proc.h> -#include <machine/profile.h> -#define _KERNEL -#include <sys/user.h> -#undef _KERNEL -#undef sigqueue -#include <libutil.h> -#endif -#endif -#else -#include "exec/ram_addr.h" -#endif - -#include "exec/cputlb.h" -#include "exec/translate-all.h" -#include "exec/translator.h" +#include "exec/mmap-lock.h" +#include "tb-internal.h" #include "exec/tb-flush.h" -#include "qemu/bitmap.h" -#include "qemu/qemu-print.h" -#include "qemu/main-loop.h" #include "qemu/cacheinfo.h" -#include "qemu/timer.h" +#include "qemu/target-info.h" #include "exec/log.h" -#include "sysemu/cpus.h" -#include "sysemu/cpu-timers.h" -#include "sysemu/tcg.h" -#include "qapi/error.h" -#include "hw/core/tcg-cpu-ops.h" +#include "exec/icount.h" +#include "accel/tcg/cpu-ops.h" #include "tb-jmp-cache.h" #include "tb-hash.h" #include "tb-context.h" +#include "tb-internal.h" #include "internal-common.h" -#include "internal-target.h" #include "tcg/perf.h" #include "tcg/insn-start-words.h" @@ -105,7 +77,7 @@ static int64_t decode_sleb128(const uint8_t **pp) val |= (int64_t)(byte & 0x7f) << shift; shift += 7; } while (byte & 0x80); - if (shift < TARGET_LONG_BITS && (byte & 0x40)) { + if (shift < 64 && (byte & 0x40)) { val |= -(int64_t)1 << shift; } @@ -116,7 +88,7 @@ static int64_t decode_sleb128(const uint8_t **pp) /* Encode the data collected about the instructions while compiling TB. Place the data at BLOCK, and return the number of bytes consumed. - The logical table consists of TARGET_INSN_START_WORDS target_ulong's, + The logical table consists of INSN_START_WORDS uint64_t's, which come from the target's insn_start data, followed by a uintptr_t which comes from the host pc of the end of the code implementing the insn. @@ -136,13 +108,13 @@ static int encode_search(TranslationBlock *tb, uint8_t *block) for (i = 0, n = tb->icount; i < n; ++i) { uint64_t prev, curr; - for (j = 0; j < TARGET_INSN_START_WORDS; ++j) { + for (j = 0; j < INSN_START_WORDS; ++j) { if (i == 0) { prev = (!(tb_cflags(tb) & CF_PCREL) && j == 0 ? tb->pc : 0); } else { - prev = insn_data[(i - 1) * TARGET_INSN_START_WORDS + j]; + prev = insn_data[(i - 1) * INSN_START_WORDS + j]; } - curr = insn_data[i * TARGET_INSN_START_WORDS + j]; + curr = insn_data[i * INSN_START_WORDS + j]; p = encode_sleb128(p, curr - prev); } prev = (i == 0 ? 0 : insn_end_off[i - 1]); @@ -174,7 +146,7 @@ static int cpu_unwind_data_from_tb(TranslationBlock *tb, uintptr_t host_pc, return -1; } - memset(data, 0, sizeof(uint64_t) * TARGET_INSN_START_WORDS); + memset(data, 0, sizeof(uint64_t) * INSN_START_WORDS); if (!(tb_cflags(tb) & CF_PCREL)) { data[0] = tb->pc; } @@ -184,7 +156,7 @@ static int cpu_unwind_data_from_tb(TranslationBlock *tb, uintptr_t host_pc, * at which the end of the insn exceeds host_pc. */ for (i = 0; i < num_insns; ++i) { - for (j = 0; j < TARGET_INSN_START_WORDS; ++j) { + for (j = 0; j < INSN_START_WORDS; ++j) { data[j] += decode_sleb128(&p); } iter_pc += decode_sleb128(&p); @@ -202,7 +174,7 @@ static int cpu_unwind_data_from_tb(TranslationBlock *tb, uintptr_t host_pc, void cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb, uintptr_t host_pc) { - uint64_t data[TARGET_INSN_START_WORDS]; + uint64_t data[INSN_START_WORDS]; int insns_left = cpu_unwind_data_from_tb(tb, host_pc, data); if (insns_left < 0) { @@ -274,8 +246,10 @@ static int setjmp_gen_code(CPUArchState *env, TranslationBlock *tb, tcg_func_start(tcg_ctx); - tcg_ctx->cpu = env_cpu(env); - gen_intermediate_code(env_cpu(env), tb, max_insns, pc, host_pc); + CPUState *cs = env_cpu(env); + tcg_ctx->cpu = cs; + cs->cc->tcg_ops->translate_code(cs, tb, max_insns, pc, host_pc); + assert(tb->size != 0); tcg_ctx->cpu = NULL; *max_insns = tb->icount; @@ -284,9 +258,7 @@ static int setjmp_gen_code(CPUArchState *env, TranslationBlock *tb, } /* Called with mmap_lock held for user mode emulation. */ -TranslationBlock *tb_gen_code(CPUState *cpu, - vaddr pc, uint64_t cs_base, - uint32_t flags, int cflags) +TranslationBlock *tb_gen_code(CPUState *cpu, TCGTBCPUState s) { CPUArchState *env = cpu_env(cpu); TranslationBlock *tb, *existing_tb; @@ -299,14 +271,14 @@ TranslationBlock *tb_gen_code(CPUState *cpu, assert_memory_lock(); qemu_thread_jit_write(); - phys_pc = get_page_addr_code_hostp(env, pc, &host_pc); + phys_pc = get_page_addr_code_hostp(env, s.pc, &host_pc); if (phys_pc == -1) { /* Generate a one-shot TB with 1 insn in it */ - cflags = (cflags & ~CF_COUNT_MASK) | 1; + s.cflags = (s.cflags & ~CF_COUNT_MASK) | 1; } - max_insns = cflags & CF_COUNT_MASK; + max_insns = s.cflags & CF_COUNT_MASK; if (max_insns == 0) { max_insns = TCG_MAX_INSNS; } @@ -326,12 +298,12 @@ TranslationBlock *tb_gen_code(CPUState *cpu, gen_code_buf = tcg_ctx->code_gen_ptr; tb->tc.ptr = tcg_splitwx_to_rx(gen_code_buf); - if (!(cflags & CF_PCREL)) { - tb->pc = pc; + if (!(s.cflags & CF_PCREL)) { + tb->pc = s.pc; } - tb->cs_base = cs_base; - tb->flags = flags; - tb->cflags = cflags; + tb->cs_base = s.cs_base; + tb->flags = s.flags; + tb->cflags = s.cflags; tb_set_page_addr0(tb, phys_pc); tb_set_page_addr1(tb, -1); if (phys_pc != -1) { @@ -339,30 +311,20 @@ TranslationBlock *tb_gen_code(CPUState *cpu, } tcg_ctx->gen_tb = tb; - tcg_ctx->addr_type = TARGET_LONG_BITS == 32 ? TCG_TYPE_I32 : TCG_TYPE_I64; -#ifdef CONFIG_SOFTMMU - tcg_ctx->page_bits = TARGET_PAGE_BITS; - tcg_ctx->page_mask = TARGET_PAGE_MASK; - tcg_ctx->tlb_dyn_max_bits = CPU_TLB_DYN_MAX_BITS; -#endif - tcg_ctx->insn_start_words = TARGET_INSN_START_WORDS; -#ifdef TCG_GUEST_DEFAULT_MO - tcg_ctx->guest_mo = TCG_GUEST_DEFAULT_MO; -#else - tcg_ctx->guest_mo = TCG_MO_ALL; -#endif + tcg_ctx->addr_type = target_long_bits() == 32 ? TCG_TYPE_I32 : TCG_TYPE_I64; + tcg_ctx->guest_mo = cpu->cc->tcg_ops->guest_default_memory_order; restart_translate: - trace_translate_block(tb, pc, tb->tc.ptr); + trace_translate_block(tb, s.pc, tb->tc.ptr); - gen_code_size = setjmp_gen_code(env, tb, pc, host_pc, &max_insns, &ti); + gen_code_size = setjmp_gen_code(env, tb, s.pc, host_pc, &max_insns, &ti); if (unlikely(gen_code_size < 0)) { switch (gen_code_size) { case -1: /* * Overflow of code_gen_buffer, or the current slice of it. * - * TODO: We don't need to re-do gen_intermediate_code, nor + * TODO: We don't need to re-do tcg_ops->translate_code, nor * should we re-do the tcg optimization currently hidden * inside tcg_gen_code. All that should be required is to * flush the TBs, allocate a new TB, re-initialize it per @@ -432,10 +394,10 @@ TranslationBlock *tb_gen_code(CPUState *cpu, * For CF_PCREL, attribute all executions of the generated code * to its first mapping. */ - perf_report_code(pc, tb, tcg_splitwx_to_rx(gen_code_buf)); + perf_report_code(s.pc, tb, tcg_splitwx_to_rx(gen_code_buf)); if (qemu_loglevel_mask(CPU_LOG_TB_OUT_ASM) && - qemu_log_in_addr_range(pc)) { + qemu_log_in_addr_range(s.pc)) { FILE *logfile = qemu_log_trylock(); if (logfile) { int code_size, data_size; @@ -457,7 +419,7 @@ TranslationBlock *tb_gen_code(CPUState *cpu, fprintf(logfile, "OUT: [size=%d]\n", gen_code_size); fprintf(logfile, " -- guest addr 0x%016" PRIx64 " + tb prologue\n", - tcg_ctx->gen_insn_data[insn * TARGET_INSN_START_WORDS]); + tcg_ctx->gen_insn_data[insn * INSN_START_WORDS]); chunk_start = tcg_ctx->gen_insn_end_off[insn]; disas(logfile, tb->tc.ptr, chunk_start); @@ -470,7 +432,7 @@ TranslationBlock *tb_gen_code(CPUState *cpu, size_t chunk_end = tcg_ctx->gen_insn_end_off[insn]; if (chunk_end > chunk_start) { fprintf(logfile, " -- guest addr 0x%016" PRIx64 "\n", - tcg_ctx->gen_insn_data[insn * TARGET_INSN_START_WORDS]); + tcg_ctx->gen_insn_data[insn * INSN_START_WORDS]); disas(logfile, tb->tc.ptr + chunk_start, chunk_end - chunk_start); chunk_start = chunk_end; @@ -528,9 +490,25 @@ TranslationBlock *tb_gen_code(CPUState *cpu, } /* + * Insert TB into the corresponding region tree before publishing it + * through QHT. Otherwise rewinding happened in the TB might fail to + * lookup itself using host PC. + */ + tcg_tb_insert(tb); + + /* * If the TB is not associated with a physical RAM page then it must be - * a temporary one-insn TB, and we have nothing left to do. Return early - * before attempting to link to other TBs or add to the lookup table. + * a temporary one-insn TB. + * + * Such TBs must be added to region trees in order to make sure that + * restore_state_to_opc() - which on some architectures is not limited to + * rewinding, but also affects exception handling! - is called when such a + * TB causes an exception. + * + * At the same time, temporary one-insn TBs must be executed at most once, + * because subsequent reads from, e.g., I/O memory may return different + * values. So return early before attempting to link to other TBs or add + * to the QHT. */ if (tb_page_addr0(tb) == -1) { assert_no_pages_locked(); @@ -538,13 +516,6 @@ TranslationBlock *tb_gen_code(CPUState *cpu, } /* - * Insert TB into the corresponding region tree before publishing it - * through QHT. Otherwise rewinding happened in the TB might fail to - * lookup itself using host PC. - */ - tcg_tb_insert(tb); - - /* * No explicit memory barrier is required -- tb_link_page() makes the * TB visible in a consistent state. */ @@ -579,15 +550,11 @@ void tb_check_watchpoint(CPUState *cpu, uintptr_t retaddr) /* The exception probably happened in a helper. The CPU state should have been saved before calling it. Fetch the PC from there. */ CPUArchState *env = cpu_env(cpu); - vaddr pc; - uint64_t cs_base; - tb_page_addr_t addr; - uint32_t flags; + TCGTBCPUState s = cpu->cc->tcg_ops->get_tb_cpu_state(cpu); + tb_page_addr_t addr = get_page_addr_code(env, s.pc); - cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); - addr = get_page_addr_code(env, pc); if (addr != -1) { - tb_invalidate_phys_range(addr, addr); + tb_invalidate_phys_range(cpu, addr, addr); } } } @@ -618,7 +585,7 @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr) * to account for the re-execution of the branch. */ n = 1; - cc = CPU_GET_CLASS(cpu); + cc = cpu->cc; if (cc->tcg_ops->io_recompile_replay_branch && cc->tcg_ops->io_recompile_replay_branch(cpu, tb)) { cpu->neg.icount_decr.u16.low++; @@ -629,9 +596,10 @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr) * Exit the loop and potentially generate a new TB executing the * just the I/O insns. We also limit instrumentation to memory * operations only (which execute after completion) so we don't - * double instrument the instruction. + * double instrument the instruction. Also don't let an IRQ sneak + * in before we execute it. */ - cpu->cflags_next_tb = curr_cflags(cpu) | CF_MEMI_ONLY | n; + cpu->cflags_next_tb = curr_cflags(cpu) | CF_MEMI_ONLY | CF_NOIRQ | n; if (qemu_loglevel_mask(CPU_LOG_EXEC)) { vaddr pc = cpu->cc->get_pc(cpu); |