diff options
Diffstat (limited to 'accel')
-rw-r--r-- | accel/tcg/atomic_template.h | 8 | ||||
-rw-r--r-- | accel/tcg/cpu-exec.c | 23 | ||||
-rw-r--r-- | accel/tcg/cputlb.c | 226 | ||||
-rw-r--r-- | accel/tcg/tcg-all.c | 6 | ||||
-rw-r--r-- | accel/tcg/tcg-runtime.c | 4 | ||||
-rw-r--r-- | accel/tcg/translate-all.c | 10 | ||||
-rw-r--r-- | accel/tcg/user-exec.c | 2 |
7 files changed, 145 insertions, 134 deletions
diff --git a/accel/tcg/atomic_template.h b/accel/tcg/atomic_template.h index 685602b076..5aaf186253 100644 --- a/accel/tcg/atomic_template.h +++ b/accel/tcg/atomic_template.h @@ -62,21 +62,21 @@ #define ATOMIC_TRACE_RMW do { \ uint8_t info = glue(trace_mem_build_info_no_se, MEND)(SHIFT, false); \ \ - trace_guest_mem_before_exec(ENV_GET_CPU(env), addr, info); \ - trace_guest_mem_before_exec(ENV_GET_CPU(env), addr, \ + trace_guest_mem_before_exec(env_cpu(env), addr, info); \ + trace_guest_mem_before_exec(env_cpu(env), addr, \ info | TRACE_MEM_ST); \ } while (0) #define ATOMIC_TRACE_LD do { \ uint8_t info = glue(trace_mem_build_info_no_se, MEND)(SHIFT, false); \ \ - trace_guest_mem_before_exec(ENV_GET_CPU(env), addr, info); \ + trace_guest_mem_before_exec(env_cpu(env), addr, info); \ } while (0) # define ATOMIC_TRACE_ST do { \ uint8_t info = glue(trace_mem_build_info_no_se, MEND)(SHIFT, true); \ \ - trace_guest_mem_before_exec(ENV_GET_CPU(env), addr, info); \ + trace_guest_mem_before_exec(env_cpu(env), addr, info); \ } while (0) /* Define host-endian atomic operations. Note that END is used within diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index 45ef41ebb2..032a62672e 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -54,7 +54,7 @@ typedef struct SyncClocks { #define MAX_DELAY_PRINT_RATE 2000000000LL #define MAX_NB_PRINTS 100 -static void align_clocks(SyncClocks *sc, const CPUState *cpu) +static void align_clocks(SyncClocks *sc, CPUState *cpu) { int64_t cpu_icount; @@ -62,7 +62,7 @@ static void align_clocks(SyncClocks *sc, const CPUState *cpu) return; } - cpu_icount = cpu->icount_extra + cpu->icount_decr.u16.low; + cpu_icount = cpu->icount_extra + cpu_neg(cpu)->icount_decr.u16.low; sc->diff_clk += cpu_icount_to_ns(sc->last_cpu_icount - cpu_icount); sc->last_cpu_icount = cpu_icount; @@ -105,15 +105,15 @@ static void print_delay(const SyncClocks *sc) } } -static void init_delay_params(SyncClocks *sc, - const CPUState *cpu) +static void init_delay_params(SyncClocks *sc, CPUState *cpu) { if (!icount_align_option) { return; } sc->realtime_clock = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL_RT); sc->diff_clk = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) - sc->realtime_clock; - sc->last_cpu_icount = cpu->icount_extra + cpu->icount_decr.u16.low; + sc->last_cpu_icount + = cpu->icount_extra + cpu_neg(cpu)->icount_decr.u16.low; if (sc->diff_clk < max_delay) { max_delay = sc->diff_clk; } @@ -467,7 +467,7 @@ static inline bool cpu_handle_exception(CPUState *cpu, int *ret) if (cpu->exception_index < 0) { #ifndef CONFIG_USER_ONLY if (replay_has_exception() - && cpu->icount_decr.u16.low + cpu->icount_extra == 0) { + && cpu_neg(cpu)->icount_decr.u16.low + cpu->icount_extra == 0) { /* try to cause an exception pending in the log */ cpu_exec_nocache(cpu, 1, tb_find(cpu, NULL, 0, curr_cflags()), true); } @@ -525,7 +525,7 @@ static inline bool cpu_handle_interrupt(CPUState *cpu, * Ensure zeroing happens before reading cpu->exit_request or * cpu->interrupt_request (see also smp_wmb in cpu_exit()) */ - atomic_mb_set(&cpu->icount_decr.u16.high, 0); + atomic_mb_set(&cpu_neg(cpu)->icount_decr.u16.high, 0); if (unlikely(atomic_read(&cpu->interrupt_request))) { int interrupt_request; @@ -596,8 +596,9 @@ static inline bool cpu_handle_interrupt(CPUState *cpu, } /* Finally, check if we need to exit to the main loop. */ - if (unlikely(atomic_read(&cpu->exit_request) - || (use_icount && cpu->icount_decr.u16.low + cpu->icount_extra == 0))) { + if (unlikely(atomic_read(&cpu->exit_request)) + || (use_icount + && cpu_neg(cpu)->icount_decr.u16.low + cpu->icount_extra == 0)) { atomic_set(&cpu->exit_request, 0); if (cpu->exception_index == -1) { cpu->exception_index = EXCP_INTERRUPT; @@ -624,7 +625,7 @@ static inline void cpu_loop_exec_tb(CPUState *cpu, TranslationBlock *tb, } *last_tb = NULL; - insns_left = atomic_read(&cpu->icount_decr.u32); + insns_left = atomic_read(&cpu_neg(cpu)->icount_decr.u32); if (insns_left < 0) { /* Something asked us to stop executing chained TBs; just * continue round the main loop. Whatever requested the exit @@ -643,7 +644,7 @@ static inline void cpu_loop_exec_tb(CPUState *cpu, TranslationBlock *tb, cpu_update_icount(cpu); /* Refill decrementer and continue execution. */ insns_left = MIN(0xffff, cpu->icount_budget); - cpu->icount_decr.u16.low = insns_left; + cpu_neg(cpu)->icount_decr.u16.low = insns_left; cpu->icount_extra = cpu->icount_budget - insns_left; if (!cpu->icount_extra) { /* Execute any remaining instructions, then let the main loop diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index cdcc377102..baa3eb8f92 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -76,14 +76,14 @@ QEMU_BUILD_BUG_ON(NB_MMU_MODES > 16); static inline size_t sizeof_tlb(CPUArchState *env, uintptr_t mmu_idx) { - return env->tlb_mask[mmu_idx] + (1 << CPU_TLB_ENTRY_BITS); + return env_tlb(env)->f[mmu_idx].mask + (1 << CPU_TLB_ENTRY_BITS); } -static void tlb_window_reset(CPUTLBWindow *window, int64_t ns, +static void tlb_window_reset(CPUTLBDesc *desc, int64_t ns, size_t max_entries) { - window->begin_ns = ns; - window->max_entries = max_entries; + desc->window_begin_ns = ns; + desc->window_max_entries = max_entries; } static void tlb_dyn_init(CPUArchState *env) @@ -91,14 +91,14 @@ static void tlb_dyn_init(CPUArchState *env) int i; for (i = 0; i < NB_MMU_MODES; i++) { - CPUTLBDesc *desc = &env->tlb_d[i]; + CPUTLBDesc *desc = &env_tlb(env)->d[i]; size_t n_entries = 1 << CPU_TLB_DYN_DEFAULT_BITS; - tlb_window_reset(&desc->window, get_clock_realtime(), 0); + tlb_window_reset(desc, get_clock_realtime(), 0); desc->n_used_entries = 0; - env->tlb_mask[i] = (n_entries - 1) << CPU_TLB_ENTRY_BITS; - env->tlb_table[i] = g_new(CPUTLBEntry, n_entries); - env->iotlb[i] = g_new(CPUIOTLBEntry, n_entries); + env_tlb(env)->f[i].mask = (n_entries - 1) << CPU_TLB_ENTRY_BITS; + env_tlb(env)->f[i].table = g_new(CPUTLBEntry, n_entries); + env_tlb(env)->d[i].iotlb = g_new(CPUIOTLBEntry, n_entries); } } @@ -144,25 +144,25 @@ static void tlb_dyn_init(CPUArchState *env) */ static void tlb_mmu_resize_locked(CPUArchState *env, int mmu_idx) { - CPUTLBDesc *desc = &env->tlb_d[mmu_idx]; + CPUTLBDesc *desc = &env_tlb(env)->d[mmu_idx]; size_t old_size = tlb_n_entries(env, mmu_idx); size_t rate; size_t new_size = old_size; int64_t now = get_clock_realtime(); int64_t window_len_ms = 100; int64_t window_len_ns = window_len_ms * 1000 * 1000; - bool window_expired = now > desc->window.begin_ns + window_len_ns; + bool window_expired = now > desc->window_begin_ns + window_len_ns; - if (desc->n_used_entries > desc->window.max_entries) { - desc->window.max_entries = desc->n_used_entries; + if (desc->n_used_entries > desc->window_max_entries) { + desc->window_max_entries = desc->n_used_entries; } - rate = desc->window.max_entries * 100 / old_size; + rate = desc->window_max_entries * 100 / old_size; if (rate > 70) { new_size = MIN(old_size << 1, 1 << CPU_TLB_DYN_MAX_BITS); } else if (rate < 30 && window_expired) { - size_t ceil = pow2ceil(desc->window.max_entries); - size_t expected_rate = desc->window.max_entries * 100 / ceil; + size_t ceil = pow2ceil(desc->window_max_entries); + size_t expected_rate = desc->window_max_entries * 100 / ceil; /* * Avoid undersizing when the max number of entries seen is just below @@ -182,19 +182,19 @@ static void tlb_mmu_resize_locked(CPUArchState *env, int mmu_idx) if (new_size == old_size) { if (window_expired) { - tlb_window_reset(&desc->window, now, desc->n_used_entries); + tlb_window_reset(desc, now, desc->n_used_entries); } return; } - g_free(env->tlb_table[mmu_idx]); - g_free(env->iotlb[mmu_idx]); + g_free(env_tlb(env)->f[mmu_idx].table); + g_free(env_tlb(env)->d[mmu_idx].iotlb); - tlb_window_reset(&desc->window, now, 0); + tlb_window_reset(desc, now, 0); /* desc->n_used_entries is cleared by the caller */ - env->tlb_mask[mmu_idx] = (new_size - 1) << CPU_TLB_ENTRY_BITS; - env->tlb_table[mmu_idx] = g_try_new(CPUTLBEntry, new_size); - env->iotlb[mmu_idx] = g_try_new(CPUIOTLBEntry, new_size); + env_tlb(env)->f[mmu_idx].mask = (new_size - 1) << CPU_TLB_ENTRY_BITS; + env_tlb(env)->f[mmu_idx].table = g_try_new(CPUTLBEntry, new_size); + env_tlb(env)->d[mmu_idx].iotlb = g_try_new(CPUIOTLBEntry, new_size); /* * If the allocations fail, try smaller sizes. We just freed some * memory, so going back to half of new_size has a good chance of working. @@ -202,46 +202,47 @@ static void tlb_mmu_resize_locked(CPUArchState *env, int mmu_idx) * allocations to fail though, so we progressively reduce the allocation * size, aborting if we cannot even allocate the smallest TLB we support. */ - while (env->tlb_table[mmu_idx] == NULL || env->iotlb[mmu_idx] == NULL) { + while (env_tlb(env)->f[mmu_idx].table == NULL || + env_tlb(env)->d[mmu_idx].iotlb == NULL) { if (new_size == (1 << CPU_TLB_DYN_MIN_BITS)) { error_report("%s: %s", __func__, strerror(errno)); abort(); } new_size = MAX(new_size >> 1, 1 << CPU_TLB_DYN_MIN_BITS); - env->tlb_mask[mmu_idx] = (new_size - 1) << CPU_TLB_ENTRY_BITS; + env_tlb(env)->f[mmu_idx].mask = (new_size - 1) << CPU_TLB_ENTRY_BITS; - g_free(env->tlb_table[mmu_idx]); - g_free(env->iotlb[mmu_idx]); - env->tlb_table[mmu_idx] = g_try_new(CPUTLBEntry, new_size); - env->iotlb[mmu_idx] = g_try_new(CPUIOTLBEntry, new_size); + g_free(env_tlb(env)->f[mmu_idx].table); + g_free(env_tlb(env)->d[mmu_idx].iotlb); + env_tlb(env)->f[mmu_idx].table = g_try_new(CPUTLBEntry, new_size); + env_tlb(env)->d[mmu_idx].iotlb = g_try_new(CPUIOTLBEntry, new_size); } } static inline void tlb_table_flush_by_mmuidx(CPUArchState *env, int mmu_idx) { tlb_mmu_resize_locked(env, mmu_idx); - memset(env->tlb_table[mmu_idx], -1, sizeof_tlb(env, mmu_idx)); - env->tlb_d[mmu_idx].n_used_entries = 0; + memset(env_tlb(env)->f[mmu_idx].table, -1, sizeof_tlb(env, mmu_idx)); + env_tlb(env)->d[mmu_idx].n_used_entries = 0; } static inline void tlb_n_used_entries_inc(CPUArchState *env, uintptr_t mmu_idx) { - env->tlb_d[mmu_idx].n_used_entries++; + env_tlb(env)->d[mmu_idx].n_used_entries++; } static inline void tlb_n_used_entries_dec(CPUArchState *env, uintptr_t mmu_idx) { - env->tlb_d[mmu_idx].n_used_entries--; + env_tlb(env)->d[mmu_idx].n_used_entries--; } void tlb_init(CPUState *cpu) { CPUArchState *env = cpu->env_ptr; - qemu_spin_init(&env->tlb_c.lock); + qemu_spin_init(&env_tlb(env)->c.lock); /* Ensure that cpu_reset performs a full flush. */ - env->tlb_c.dirty = ALL_MMUIDX_BITS; + env_tlb(env)->c.dirty = ALL_MMUIDX_BITS; tlb_dyn_init(env); } @@ -273,9 +274,9 @@ void tlb_flush_counts(size_t *pfull, size_t *ppart, size_t *pelide) CPU_FOREACH(cpu) { CPUArchState *env = cpu->env_ptr; - full += atomic_read(&env->tlb_c.full_flush_count); - part += atomic_read(&env->tlb_c.part_flush_count); - elide += atomic_read(&env->tlb_c.elide_flush_count); + full += atomic_read(&env_tlb(env)->c.full_flush_count); + part += atomic_read(&env_tlb(env)->c.part_flush_count); + elide += atomic_read(&env_tlb(env)->c.elide_flush_count); } *pfull = full; *ppart = part; @@ -285,10 +286,11 @@ void tlb_flush_counts(size_t *pfull, size_t *ppart, size_t *pelide) static void tlb_flush_one_mmuidx_locked(CPUArchState *env, int mmu_idx) { tlb_table_flush_by_mmuidx(env, mmu_idx); - memset(env->tlb_v_table[mmu_idx], -1, sizeof(env->tlb_v_table[0])); - env->tlb_d[mmu_idx].large_page_addr = -1; - env->tlb_d[mmu_idx].large_page_mask = -1; - env->tlb_d[mmu_idx].vindex = 0; + env_tlb(env)->d[mmu_idx].large_page_addr = -1; + env_tlb(env)->d[mmu_idx].large_page_mask = -1; + env_tlb(env)->d[mmu_idx].vindex = 0; + memset(env_tlb(env)->d[mmu_idx].vtable, -1, + sizeof(env_tlb(env)->d[0].vtable)); } static void tlb_flush_by_mmuidx_async_work(CPUState *cpu, run_on_cpu_data data) @@ -301,31 +303,31 @@ static void tlb_flush_by_mmuidx_async_work(CPUState *cpu, run_on_cpu_data data) tlb_debug("mmu_idx:0x%04" PRIx16 "\n", asked); - qemu_spin_lock(&env->tlb_c.lock); + qemu_spin_lock(&env_tlb(env)->c.lock); - all_dirty = env->tlb_c.dirty; + all_dirty = env_tlb(env)->c.dirty; to_clean = asked & all_dirty; all_dirty &= ~to_clean; - env->tlb_c.dirty = all_dirty; + env_tlb(env)->c.dirty = all_dirty; for (work = to_clean; work != 0; work &= work - 1) { int mmu_idx = ctz32(work); tlb_flush_one_mmuidx_locked(env, mmu_idx); } - qemu_spin_unlock(&env->tlb_c.lock); + qemu_spin_unlock(&env_tlb(env)->c.lock); cpu_tb_jmp_cache_clear(cpu); if (to_clean == ALL_MMUIDX_BITS) { - atomic_set(&env->tlb_c.full_flush_count, - env->tlb_c.full_flush_count + 1); + atomic_set(&env_tlb(env)->c.full_flush_count, + env_tlb(env)->c.full_flush_count + 1); } else { - atomic_set(&env->tlb_c.part_flush_count, - env->tlb_c.part_flush_count + ctpop16(to_clean)); + atomic_set(&env_tlb(env)->c.part_flush_count, + env_tlb(env)->c.part_flush_count + ctpop16(to_clean)); if (to_clean != asked) { - atomic_set(&env->tlb_c.elide_flush_count, - env->tlb_c.elide_flush_count + + atomic_set(&env_tlb(env)->c.elide_flush_count, + env_tlb(env)->c.elide_flush_count + ctpop16(asked & ~to_clean)); } } @@ -410,11 +412,12 @@ static inline bool tlb_flush_entry_locked(CPUTLBEntry *tlb_entry, static inline void tlb_flush_vtlb_page_locked(CPUArchState *env, int mmu_idx, target_ulong page) { + CPUTLBDesc *d = &env_tlb(env)->d[mmu_idx]; int k; - assert_cpu_is_self(ENV_GET_CPU(env)); + assert_cpu_is_self(env_cpu(env)); for (k = 0; k < CPU_VTLB_SIZE; k++) { - if (tlb_flush_entry_locked(&env->tlb_v_table[mmu_idx][k], page)) { + if (tlb_flush_entry_locked(&d->vtable[k], page)) { tlb_n_used_entries_dec(env, mmu_idx); } } @@ -423,8 +426,8 @@ static inline void tlb_flush_vtlb_page_locked(CPUArchState *env, int mmu_idx, static void tlb_flush_page_locked(CPUArchState *env, int midx, target_ulong page) { - target_ulong lp_addr = env->tlb_d[midx].large_page_addr; - target_ulong lp_mask = env->tlb_d[midx].large_page_mask; + target_ulong lp_addr = env_tlb(env)->d[midx].large_page_addr; + target_ulong lp_mask = env_tlb(env)->d[midx].large_page_mask; /* Check if we need to flush due to large pages. */ if ((page & lp_mask) == lp_addr) { @@ -459,13 +462,13 @@ static void tlb_flush_page_by_mmuidx_async_work(CPUState *cpu, tlb_debug("page addr:" TARGET_FMT_lx " mmu_map:0x%lx\n", addr, mmu_idx_bitmap); - qemu_spin_lock(&env->tlb_c.lock); + qemu_spin_lock(&env_tlb(env)->c.lock); for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) { if (test_bit(mmu_idx, &mmu_idx_bitmap)) { tlb_flush_page_locked(env, mmu_idx, addr); } } - qemu_spin_unlock(&env->tlb_c.lock); + qemu_spin_unlock(&env_tlb(env)->c.lock); tb_flush_jmp_cache(cpu, addr); } @@ -609,22 +612,22 @@ void tlb_reset_dirty(CPUState *cpu, ram_addr_t start1, ram_addr_t length) int mmu_idx; env = cpu->env_ptr; - qemu_spin_lock(&env->tlb_c.lock); + qemu_spin_lock(&env_tlb(env)->c.lock); for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) { unsigned int i; unsigned int n = tlb_n_entries(env, mmu_idx); for (i = 0; i < n; i++) { - tlb_reset_dirty_range_locked(&env->tlb_table[mmu_idx][i], start1, - length); + tlb_reset_dirty_range_locked(&env_tlb(env)->f[mmu_idx].table[i], + start1, length); } for (i = 0; i < CPU_VTLB_SIZE; i++) { - tlb_reset_dirty_range_locked(&env->tlb_v_table[mmu_idx][i], start1, - length); + tlb_reset_dirty_range_locked(&env_tlb(env)->d[mmu_idx].vtable[i], + start1, length); } } - qemu_spin_unlock(&env->tlb_c.lock); + qemu_spin_unlock(&env_tlb(env)->c.lock); } /* Called with tlb_c.lock held */ @@ -646,7 +649,7 @@ void tlb_set_dirty(CPUState *cpu, target_ulong vaddr) assert_cpu_is_self(cpu); vaddr &= TARGET_PAGE_MASK; - qemu_spin_lock(&env->tlb_c.lock); + qemu_spin_lock(&env_tlb(env)->c.lock); for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) { tlb_set_dirty1_locked(tlb_entry(env, mmu_idx, vaddr), vaddr); } @@ -654,10 +657,10 @@ void tlb_set_dirty(CPUState *cpu, target_ulong vaddr) for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) { int k; for (k = 0; k < CPU_VTLB_SIZE; k++) { - tlb_set_dirty1_locked(&env->tlb_v_table[mmu_idx][k], vaddr); + tlb_set_dirty1_locked(&env_tlb(env)->d[mmu_idx].vtable[k], vaddr); } } - qemu_spin_unlock(&env->tlb_c.lock); + qemu_spin_unlock(&env_tlb(env)->c.lock); } /* Our TLB does not support large pages, so remember the area covered by @@ -665,7 +668,7 @@ void tlb_set_dirty(CPUState *cpu, target_ulong vaddr) static void tlb_add_large_page(CPUArchState *env, int mmu_idx, target_ulong vaddr, target_ulong size) { - target_ulong lp_addr = env->tlb_d[mmu_idx].large_page_addr; + target_ulong lp_addr = env_tlb(env)->d[mmu_idx].large_page_addr; target_ulong lp_mask = ~(size - 1); if (lp_addr == (target_ulong)-1) { @@ -675,13 +678,13 @@ static void tlb_add_large_page(CPUArchState *env, int mmu_idx, /* Extend the existing region to include the new page. This is a compromise between unnecessary flushes and the cost of maintaining a full variable size TLB. */ - lp_mask &= env->tlb_d[mmu_idx].large_page_mask; + lp_mask &= env_tlb(env)->d[mmu_idx].large_page_mask; while (((lp_addr ^ vaddr) & lp_mask) != 0) { lp_mask <<= 1; } } - env->tlb_d[mmu_idx].large_page_addr = lp_addr & lp_mask; - env->tlb_d[mmu_idx].large_page_mask = lp_mask; + env_tlb(env)->d[mmu_idx].large_page_addr = lp_addr & lp_mask; + env_tlb(env)->d[mmu_idx].large_page_mask = lp_mask; } /* Add a new TLB entry. At most one entry for a given virtual address @@ -696,6 +699,8 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, int mmu_idx, target_ulong size) { CPUArchState *env = cpu->env_ptr; + CPUTLB *tlb = env_tlb(env); + CPUTLBDesc *desc = &tlb->d[mmu_idx]; MemoryRegionSection *section; unsigned int index; target_ulong address; @@ -757,10 +762,10 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, * a longer critical section, but this is not a concern since the TLB lock * is unlikely to be contended. */ - qemu_spin_lock(&env->tlb_c.lock); + qemu_spin_lock(&tlb->c.lock); /* Note that the tlb is no longer clean. */ - env->tlb_c.dirty |= 1 << mmu_idx; + tlb->c.dirty |= 1 << mmu_idx; /* Make sure there's no cached translation for the new page. */ tlb_flush_vtlb_page_locked(env, mmu_idx, vaddr_page); @@ -770,12 +775,12 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, * different page; otherwise just overwrite the stale data. */ if (!tlb_hit_page_anyprot(te, vaddr_page) && !tlb_entry_is_empty(te)) { - unsigned vidx = env->tlb_d[mmu_idx].vindex++ % CPU_VTLB_SIZE; - CPUTLBEntry *tv = &env->tlb_v_table[mmu_idx][vidx]; + unsigned vidx = desc->vindex++ % CPU_VTLB_SIZE; + CPUTLBEntry *tv = &desc->vtable[vidx]; /* Evict the old entry into the victim tlb. */ copy_tlb_helper_locked(tv, te); - env->iotlb_v[mmu_idx][vidx] = env->iotlb[mmu_idx][index]; + desc->viotlb[vidx] = desc->iotlb[index]; tlb_n_used_entries_dec(env, mmu_idx); } @@ -792,8 +797,8 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, * subtract here is that of the page base, and not the same as the * vaddr we add back in io_readx()/io_writex()/get_page_addr_code(). */ - env->iotlb[mmu_idx][index].addr = iotlb - vaddr_page; - env->iotlb[mmu_idx][index].attrs = attrs; + desc->iotlb[index].addr = iotlb - vaddr_page; + desc->iotlb[index].attrs = attrs; /* Now calculate the new entry */ tn.addend = addend - vaddr_page; @@ -829,7 +834,7 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr, copy_tlb_helper_locked(te, &tn); tlb_n_used_entries_inc(env, mmu_idx); - qemu_spin_unlock(&env->tlb_c.lock); + qemu_spin_unlock(&tlb->c.lock); } /* Add a new TLB entry, but without specifying the memory @@ -878,7 +883,7 @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry, int mmu_idx, target_ulong addr, uintptr_t retaddr, MMUAccessType access_type, int size) { - CPUState *cpu = ENV_GET_CPU(env); + CPUState *cpu = env_cpu(env); hwaddr mr_offset; MemoryRegionSection *section; MemoryRegion *mr; @@ -922,7 +927,7 @@ static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry, int mmu_idx, uint64_t val, target_ulong addr, uintptr_t retaddr, int size) { - CPUState *cpu = ENV_GET_CPU(env); + CPUState *cpu = env_cpu(env); hwaddr mr_offset; MemoryRegionSection *section; MemoryRegion *mr; @@ -974,23 +979,30 @@ static bool victim_tlb_hit(CPUArchState *env, size_t mmu_idx, size_t index, { size_t vidx; - assert_cpu_is_self(ENV_GET_CPU(env)); + assert_cpu_is_self(env_cpu(env)); for (vidx = 0; vidx < CPU_VTLB_SIZE; ++vidx) { - CPUTLBEntry *vtlb = &env->tlb_v_table[mmu_idx][vidx]; - target_ulong cmp = tlb_read_ofs(vtlb, elt_ofs); + CPUTLBEntry *vtlb = &env_tlb(env)->d[mmu_idx].vtable[vidx]; + target_ulong cmp; + + /* elt_ofs might correspond to .addr_write, so use atomic_read */ +#if TCG_OVERSIZED_GUEST + cmp = *(target_ulong *)((uintptr_t)vtlb + elt_ofs); +#else + cmp = atomic_read((target_ulong *)((uintptr_t)vtlb + elt_ofs)); +#endif if (cmp == page) { /* Found entry in victim tlb, swap tlb and iotlb. */ - CPUTLBEntry tmptlb, *tlb = &env->tlb_table[mmu_idx][index]; + CPUTLBEntry tmptlb, *tlb = &env_tlb(env)->f[mmu_idx].table[index]; - qemu_spin_lock(&env->tlb_c.lock); + qemu_spin_lock(&env_tlb(env)->c.lock); copy_tlb_helper_locked(&tmptlb, tlb); copy_tlb_helper_locked(tlb, vtlb); copy_tlb_helper_locked(vtlb, &tmptlb); - qemu_spin_unlock(&env->tlb_c.lock); + qemu_spin_unlock(&env_tlb(env)->c.lock); - CPUIOTLBEntry tmpio, *io = &env->iotlb[mmu_idx][index]; - CPUIOTLBEntry *vio = &env->iotlb_v[mmu_idx][vidx]; + CPUIOTLBEntry tmpio, *io = &env_tlb(env)->d[mmu_idx].iotlb[index]; + CPUIOTLBEntry *vio = &env_tlb(env)->d[mmu_idx].viotlb[vidx]; tmpio = *io; *io = *vio; *vio = tmpio; return true; } @@ -1017,7 +1029,7 @@ tb_page_addr_t get_page_addr_code(CPUArchState *env, target_ulong addr) if (unlikely(!tlb_hit(entry->addr_code, addr))) { if (!VICTIM_TLB_HIT(addr_code, addr)) { - tlb_fill(ENV_GET_CPU(env), addr, 0, MMU_INST_FETCH, mmu_idx, 0); + tlb_fill(env_cpu(env), addr, 0, MMU_INST_FETCH, mmu_idx, 0); index = tlb_index(env, mmu_idx, addr); entry = tlb_entry(env, mmu_idx, addr); } @@ -1055,7 +1067,7 @@ void probe_write(CPUArchState *env, target_ulong addr, int size, int mmu_idx, if (!tlb_hit(tlb_addr_write(entry), addr)) { /* TLB entry is for a different page */ if (!VICTIM_TLB_HIT(addr_write, addr)) { - tlb_fill(ENV_GET_CPU(env), addr, size, MMU_DATA_STORE, + tlb_fill(env_cpu(env), addr, size, MMU_DATA_STORE, mmu_idx, retaddr); } } @@ -1089,7 +1101,7 @@ void *tlb_vaddr_to_host(CPUArchState *env, abi_ptr addr, uintptr_t index = tlb_index(env, mmu_idx, addr); if (!victim_tlb_hit(env, mmu_idx, index, elt_ofs, page)) { - CPUState *cs = ENV_GET_CPU(env); + CPUState *cs = env_cpu(env); CPUClass *cc = CPU_GET_CLASS(cs); if (!cc->tlb_fill(cs, addr, 0, access_type, mmu_idx, true, 0)) { @@ -1132,7 +1144,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr, /* Enforce guest required alignment. */ if (unlikely(a_bits > 0 && (addr & ((1 << a_bits) - 1)))) { /* ??? Maybe indicate atomic op to cpu_unaligned_access */ - cpu_unaligned_access(ENV_GET_CPU(env), addr, MMU_DATA_STORE, + cpu_unaligned_access(env_cpu(env), addr, MMU_DATA_STORE, mmu_idx, retaddr); } @@ -1148,7 +1160,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr, /* Check TLB entry and enforce page permissions. */ if (!tlb_hit(tlb_addr, addr)) { if (!VICTIM_TLB_HIT(addr_write, addr)) { - tlb_fill(ENV_GET_CPU(env), addr, 1 << s_bits, MMU_DATA_STORE, + tlb_fill(env_cpu(env), addr, 1 << s_bits, MMU_DATA_STORE, mmu_idx, retaddr); index = tlb_index(env, mmu_idx, addr); tlbe = tlb_entry(env, mmu_idx, addr); @@ -1165,7 +1177,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr, /* Let the guest notice RMW on a write-only page. */ if (unlikely(tlbe->addr_read != (tlb_addr & ~TLB_NOTDIRTY))) { - tlb_fill(ENV_GET_CPU(env), addr, 1 << s_bits, MMU_DATA_LOAD, + tlb_fill(env_cpu(env), addr, 1 << s_bits, MMU_DATA_LOAD, mmu_idx, retaddr); /* Since we don't support reads and writes to different addresses, and we do have the proper page loaded for write, this shouldn't @@ -1178,7 +1190,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr, ndi->active = false; if (unlikely(tlb_addr & TLB_NOTDIRTY)) { ndi->active = true; - memory_notdirty_write_prepare(ndi, ENV_GET_CPU(env), addr, + memory_notdirty_write_prepare(ndi, env_cpu(env), addr, qemu_ram_addr_from_host_nofail(hostaddr), 1 << s_bits); } @@ -1186,7 +1198,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr, return hostaddr; stop_the_world: - cpu_loop_exit_atomic(ENV_GET_CPU(env), retaddr); + cpu_loop_exit_atomic(env_cpu(env), retaddr); } #ifdef TARGET_WORDS_BIGENDIAN @@ -1251,7 +1263,7 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi, /* Handle CPU specific unaligned behaviour */ if (addr & ((1 << a_bits) - 1)) { - cpu_unaligned_access(ENV_GET_CPU(env), addr, access_type, + cpu_unaligned_access(env_cpu(env), addr, access_type, mmu_idx, retaddr); } @@ -1259,7 +1271,7 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi, if (!tlb_hit(tlb_addr, addr)) { if (!victim_tlb_hit(env, mmu_idx, index, tlb_off, addr & TARGET_PAGE_MASK)) { - tlb_fill(ENV_GET_CPU(env), addr, size, + tlb_fill(env_cpu(env), addr, size, access_type, mmu_idx, retaddr); index = tlb_index(env, mmu_idx, addr); entry = tlb_entry(env, mmu_idx, addr); @@ -1280,7 +1292,7 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi, * repeat the MMU check here. This tlb_fill() call might * longjump out if this access should cause a guest exception. */ - tlb_fill(ENV_GET_CPU(env), addr, size, + tlb_fill(env_cpu(env), addr, size, access_type, mmu_idx, retaddr); index = tlb_index(env, mmu_idx, addr); entry = tlb_entry(env, mmu_idx, addr); @@ -1293,8 +1305,8 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi, } } - res = io_readx(env, &env->iotlb[mmu_idx][index], mmu_idx, addr, - retaddr, access_type, size); + res = io_readx(env, &env_tlb(env)->d[mmu_idx].iotlb[index], + mmu_idx, addr, retaddr, access_type, size); return handle_bswap(res, size, big_endian); } @@ -1499,7 +1511,7 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, /* Handle CPU specific unaligned behaviour */ if (addr & ((1 << a_bits) - 1)) { - cpu_unaligned_access(ENV_GET_CPU(env), addr, MMU_DATA_STORE, + cpu_unaligned_access(env_cpu(env), addr, MMU_DATA_STORE, mmu_idx, retaddr); } @@ -1507,7 +1519,7 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, if (!tlb_hit(tlb_addr, addr)) { if (!victim_tlb_hit(env, mmu_idx, index, tlb_off, addr & TARGET_PAGE_MASK)) { - tlb_fill(ENV_GET_CPU(env), addr, size, MMU_DATA_STORE, + tlb_fill(env_cpu(env), addr, size, MMU_DATA_STORE, mmu_idx, retaddr); index = tlb_index(env, mmu_idx, addr); entry = tlb_entry(env, mmu_idx, addr); @@ -1528,7 +1540,7 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, * repeat the MMU check here. This tlb_fill() call might * longjump out if this access should cause a guest exception. */ - tlb_fill(ENV_GET_CPU(env), addr, size, MMU_DATA_STORE, + tlb_fill(env_cpu(env), addr, size, MMU_DATA_STORE, mmu_idx, retaddr); index = tlb_index(env, mmu_idx, addr); entry = tlb_entry(env, mmu_idx, addr); @@ -1541,7 +1553,7 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, } } - io_writex(env, &env->iotlb[mmu_idx][index], mmu_idx, + io_writex(env, &env_tlb(env)->d[mmu_idx].iotlb[index], mmu_idx, handle_bswap(val, size, big_endian), addr, retaddr, size); return; @@ -1568,7 +1580,7 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val, if (!tlb_hit_page(tlb_addr2, page2) && !victim_tlb_hit(env, mmu_idx, index2, tlb_off, page2 & TARGET_PAGE_MASK)) { - tlb_fill(ENV_GET_CPU(env), page2, size, MMU_DATA_STORE, + tlb_fill(env_cpu(env), page2, size, MMU_DATA_STORE, mmu_idx, retaddr); } diff --git a/accel/tcg/tcg-all.c b/accel/tcg/tcg-all.c index 3d25bdcc17..9b215dcc5a 100644 --- a/accel/tcg/tcg-all.c +++ b/accel/tcg/tcg-all.c @@ -28,13 +28,12 @@ #include "sysemu/sysemu.h" #include "qom/object.h" #include "qemu-common.h" -#include "qom/cpu.h" +#include "cpu.h" #include "sysemu/cpus.h" #include "qemu/main-loop.h" unsigned long tcg_tb_size; -#ifndef CONFIG_USER_ONLY /* mask must never be zero, except for A20 change call */ static void tcg_handle_interrupt(CPUState *cpu, int mask) { @@ -51,7 +50,7 @@ static void tcg_handle_interrupt(CPUState *cpu, int mask) if (!qemu_cpu_is_self(cpu)) { qemu_cpu_kick(cpu); } else { - atomic_set(&cpu->icount_decr.u16.high, -1); + atomic_set(&cpu_neg(cpu)->icount_decr.u16.high, -1); if (use_icount && !cpu->can_do_io && (mask & ~old_mask) != 0) { @@ -59,7 +58,6 @@ static void tcg_handle_interrupt(CPUState *cpu, int mask) } } } -#endif static int tcg_init(MachineState *ms) { diff --git a/accel/tcg/tcg-runtime.c b/accel/tcg/tcg-runtime.c index d0d4484406..8a1e408e31 100644 --- a/accel/tcg/tcg-runtime.c +++ b/accel/tcg/tcg-runtime.c @@ -146,7 +146,7 @@ uint64_t HELPER(ctpop_i64)(uint64_t arg) void *HELPER(lookup_tb_ptr)(CPUArchState *env) { - CPUState *cpu = ENV_GET_CPU(env); + CPUState *cpu = env_cpu(env); TranslationBlock *tb; target_ulong cs_base, pc; uint32_t flags; @@ -165,5 +165,5 @@ void *HELPER(lookup_tb_ptr)(CPUArchState *env) void HELPER(exit_atomic)(CPUArchState *env) { - cpu_loop_exit_atomic(ENV_GET_CPU(env), GETPC()); + cpu_loop_exit_atomic(env_cpu(env), GETPC()); } diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index 20b59f93f4..e24ee3a172 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -364,7 +364,7 @@ static int cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb, assert(use_icount); /* Reset the cycle counter to the start of the block and shift if to the number of actually executed instructions */ - cpu->icount_decr.u16.low += num_insns - i; + cpu_neg(cpu)->icount_decr.u16.low += num_insns - i; } restore_state_to_opc(env, tb, data); @@ -1732,7 +1732,7 @@ TranslationBlock *tb_gen_code(CPUState *cpu, tcg_func_start(tcg_ctx); - tcg_ctx->cpu = ENV_GET_CPU(env); + tcg_ctx->cpu = env_cpu(env); gen_intermediate_code(cpu, tb, max_insns); tcg_ctx->cpu = NULL; @@ -2200,7 +2200,7 @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr) if ((env->hflags & MIPS_HFLAG_BMASK) != 0 && env->active_tc.PC != tb->pc) { env->active_tc.PC -= (env->hflags & MIPS_HFLAG_B16 ? 2 : 4); - cpu->icount_decr.u16.low++; + cpu_neg(cpu)->icount_decr.u16.low++; env->hflags &= ~MIPS_HFLAG_BMASK; n = 2; } @@ -2208,7 +2208,7 @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr) if ((env->flags & ((DELAY_SLOT | DELAY_SLOT_CONDITIONAL))) != 0 && env->pc != tb->pc) { env->pc -= 2; - cpu->icount_decr.u16.low++; + cpu_neg(cpu)->icount_decr.u16.low++; env->flags &= ~(DELAY_SLOT | DELAY_SLOT_CONDITIONAL); n = 2; } @@ -2382,7 +2382,7 @@ void cpu_interrupt(CPUState *cpu, int mask) { g_assert(qemu_mutex_iothread_locked()); cpu->interrupt_request |= mask; - atomic_set(&cpu->icount_decr.u16.high, -1); + atomic_set(&cpu_neg(cpu)->icount_decr.u16.high, -1); } /* diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c index 8cfbeb1b56..cb5f4b19c5 100644 --- a/accel/tcg/user-exec.c +++ b/accel/tcg/user-exec.c @@ -680,7 +680,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr, { /* Enforce qemu required alignment. */ if (unlikely(addr & (size - 1))) { - cpu_loop_exit_atomic(ENV_GET_CPU(env), retaddr); + cpu_loop_exit_atomic(env_cpu(env), retaddr); } helper_retaddr = retaddr; return g2h(addr); |