Merge remote-tracking branch 'remotes/rth/tags/pull-tcg-20180702' into staging

Assorted tlb and tb caching fixes

# gpg: Signature made Mon 02 Jul 2018 17:03:07 BST
# gpg:                using RSA key 64DF38E8AF7E215F
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>"
# Primary key fingerprint: 7A48 1E78 868B 4DB6 A85A  05C0 64DF 38E8 AF7E 215F

* remotes/rth/tags/pull-tcg-20180702:
  cpu: Assert asidx_from_attrs return value in range
  accel/tcg: Avoid caching overwritten tlb entries
  accel/tcg: Don't treat invalid TLB entries as needing recheck
  accel/tcg: Correct "is this a TLB miss" check in get_page_addr_code()
  tcg: Define and use new tlb_hit() and tlb_hit_page() functions
  translate-all: fix locking of TBs whose two pages share the same physical page
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

accel/tcg/cputlb.c

@@ -235,20 +235,30 @@ void tlb_flush_by_mmuidx_all_cpus_synced(CPUState *src_cpu,
     async_safe_run_on_cpu(src_cpu, fn, RUN_ON_CPU_HOST_INT(idxmap));
 }
 
+static inline bool tlb_hit_page_anyprot(CPUTLBEntry *tlb_entry,
+                                        target_ulong page)
+{
+    return tlb_hit_page(tlb_entry->addr_read, page) ||
+           tlb_hit_page(tlb_entry->addr_write, page) ||
+           tlb_hit_page(tlb_entry->addr_code, page);
+}
 
-
-static inline void tlb_flush_entry(CPUTLBEntry *tlb_entry, target_ulong addr)
+static inline void tlb_flush_entry(CPUTLBEntry *tlb_entry, target_ulong page)
 {
-    if (addr == (tlb_entry->addr_read &
-                 (TARGET_PAGE_MASK | TLB_INVALID_MASK)) ||
-        addr == (tlb_entry->addr_write &
-                 (TARGET_PAGE_MASK | TLB_INVALID_MASK)) ||
-        addr == (tlb_entry->addr_code &
-                 (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    if (tlb_hit_page_anyprot(tlb_entry, page)) {
         memset(tlb_entry, -1, sizeof(*tlb_entry));
     }
 }
 
+static inline void tlb_flush_vtlb_page(CPUArchState *env, int mmu_idx,
+                                       target_ulong page)
+{
+    int k;
+    for (k = 0; k < CPU_VTLB_SIZE; k++) {
+        tlb_flush_entry(&env->tlb_v_table[mmu_idx][k], page);
+    }
+}
+
 static void tlb_flush_page_async_work(CPUState *cpu, run_on_cpu_data data)
 {
     CPUArchState *env = cpu->env_ptr;
@@ -274,14 +284,7 @@ static void tlb_flush_page_async_work(CPUState *cpu, run_on_cpu_data data)
     i = (addr >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
     for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
         tlb_flush_entry(&env->tlb_table[mmu_idx][i], addr);
-    }
-
-    /* check whether there are entries that need to be flushed in the vtlb */
-    for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
-        int k;
-        for (k = 0; k < CPU_VTLB_SIZE; k++) {
-            tlb_flush_entry(&env->tlb_v_table[mmu_idx][k], addr);
-        }
+        tlb_flush_vtlb_page(env, mmu_idx, addr);
     }
 
     tb_flush_jmp_cache(cpu, addr);
@@ -313,7 +316,6 @@ static void tlb_flush_page_by_mmuidx_async_work(CPUState *cpu,
     unsigned long mmu_idx_bitmap = addr_and_mmuidx & ALL_MMUIDX_BITS;
     int page = (addr >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
     int mmu_idx;
-    int i;
 
     assert_cpu_is_self(cpu);
 
@@ -323,11 +325,7 @@ static void tlb_flush_page_by_mmuidx_async_work(CPUState *cpu,
     for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
         if (test_bit(mmu_idx, &mmu_idx_bitmap)) {
             tlb_flush_entry(&env->tlb_table[mmu_idx][page], addr);
-
-            /* check whether there are vltb entries that need to be flushed */
-            for (i = 0; i < CPU_VTLB_SIZE; i++) {
-                tlb_flush_entry(&env->tlb_v_table[mmu_idx][i], addr);
-            }
+            tlb_flush_vtlb_page(env, mmu_idx, addr);
         }
     }
 
@@ -612,10 +610,9 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
     target_ulong address;
     target_ulong code_address;
     uintptr_t addend;
-    CPUTLBEntry *te, *tv, tn;
+    CPUTLBEntry *te, tn;
     hwaddr iotlb, xlat, sz, paddr_page;
     target_ulong vaddr_page;
-    unsigned vidx = env->vtlb_index++ % CPU_VTLB_SIZE;
     int asidx = cpu_asidx_from_attrs(cpu, attrs);
 
     assert_cpu_is_self(cpu);
@@ -657,19 +654,28 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
         addend = (uintptr_t)memory_region_get_ram_ptr(section->mr) + xlat;
     }
 
+    /* Make sure there's no cached translation for the new page.  */
+    tlb_flush_vtlb_page(env, mmu_idx, vaddr_page);
+
     code_address = address;
     iotlb = memory_region_section_get_iotlb(cpu, section, vaddr_page,
                                             paddr_page, xlat, prot, &address);
 
     index = (vaddr_page >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
     te = &env->tlb_table[mmu_idx][index];
-    /* do not discard the translation in te, evict it into a victim tlb */
-    tv = &env->tlb_v_table[mmu_idx][vidx];
 
-    /* addr_write can race with tlb_reset_dirty_range */
-    copy_tlb_helper(tv, te, true);
+    /*
+     * Only evict the old entry to the victim tlb if it's for a
+     * different page; otherwise just overwrite the stale data.
+     */
+    if (!tlb_hit_page_anyprot(te, vaddr_page)) {
+        unsigned vidx = env->vtlb_index++ % CPU_VTLB_SIZE;
+        CPUTLBEntry *tv = &env->tlb_v_table[mmu_idx][vidx];
 
-    env->iotlb_v[mmu_idx][vidx] = env->iotlb[mmu_idx][index];
+        /* Evict the old entry into the victim tlb.  */
+        copy_tlb_helper(tv, te, true);
+        env->iotlb_v[mmu_idx][vidx] = env->iotlb[mmu_idx][index];
+    }
 
     /* refill the tlb */
     /*
@@ -960,14 +966,14 @@ tb_page_addr_t get_page_addr_code(CPUArchState *env, target_ulong addr)
 
     index = (addr >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
     mmu_idx = cpu_mmu_index(env, true);
-    if (unlikely(env->tlb_table[mmu_idx][index].addr_code !=
-                 (addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK)))) {
+    if (unlikely(!tlb_hit(env->tlb_table[mmu_idx][index].addr_code, addr))) {
         if (!VICTIM_TLB_HIT(addr_read, addr)) {
             tlb_fill(ENV_GET_CPU(env), addr, 0, MMU_INST_FETCH, mmu_idx, 0);
         }
     }
 
-    if (unlikely(env->tlb_table[mmu_idx][index].addr_code & TLB_RECHECK)) {
+    if (unlikely((env->tlb_table[mmu_idx][index].addr_code &
+                  (TLB_RECHECK | TLB_INVALID_MASK)) == TLB_RECHECK)) {
         /*
          * This is a TLB_RECHECK access, where the MMU protection
          * covers a smaller range than a target page, and we must
@@ -1046,8 +1052,7 @@ void probe_write(CPUArchState *env, target_ulong addr, int size, int mmu_idx,
     int index = (addr >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
     target_ulong tlb_addr = env->tlb_table[mmu_idx][index].addr_write;
 
-    if ((addr & TARGET_PAGE_MASK)
-        != (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    if (!tlb_hit(tlb_addr, 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,
@@ -1091,8 +1096,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
     }
 
     /* Check TLB entry and enforce page permissions.  */
-    if ((addr & TARGET_PAGE_MASK)
-        != (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    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,
                      mmu_idx, retaddr);

accel/tcg/softmmu_template.h

@@ -123,8 +123,7 @@ WORD_TYPE helper_le_ld_name(CPUArchState *env, target_ulong addr,
     }
 
     /* If the TLB entry is for a different page, reload and try again.  */
-    if ((addr & TARGET_PAGE_MASK)
-         != (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    if (!tlb_hit(tlb_addr, addr)) {
         if (!VICTIM_TLB_HIT(ADDR_READ, addr)) {
             tlb_fill(ENV_GET_CPU(env), addr, DATA_SIZE, READ_ACCESS_TYPE,
                      mmu_idx, retaddr);
@@ -191,8 +190,7 @@ WORD_TYPE helper_be_ld_name(CPUArchState *env, target_ulong addr,
     }
 
     /* If the TLB entry is for a different page, reload and try again.  */
-    if ((addr & TARGET_PAGE_MASK)
-         != (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    if (!tlb_hit(tlb_addr, addr)) {
         if (!VICTIM_TLB_HIT(ADDR_READ, addr)) {
             tlb_fill(ENV_GET_CPU(env), addr, DATA_SIZE, READ_ACCESS_TYPE,
                      mmu_idx, retaddr);
@@ -286,8 +284,7 @@ void helper_le_st_name(CPUArchState *env, target_ulong addr, DATA_TYPE val,
     }
 
     /* If the TLB entry is for a different page, reload and try again.  */
-    if ((addr & TARGET_PAGE_MASK)
-        != (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    if (!tlb_hit(tlb_addr, addr)) {
         if (!VICTIM_TLB_HIT(addr_write, addr)) {
             tlb_fill(ENV_GET_CPU(env), addr, DATA_SIZE, MMU_DATA_STORE,
                      mmu_idx, retaddr);
@@ -322,7 +319,7 @@ void helper_le_st_name(CPUArchState *env, target_ulong addr, DATA_TYPE val,
         page2 = (addr + DATA_SIZE) & TARGET_PAGE_MASK;
         index2 = (page2 >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
         tlb_addr2 = env->tlb_table[mmu_idx][index2].addr_write;
-        if (page2 != (tlb_addr2 & (TARGET_PAGE_MASK | TLB_INVALID_MASK))
+        if (!tlb_hit_page(tlb_addr2, page2)
             && !VICTIM_TLB_HIT(addr_write, page2)) {
             tlb_fill(ENV_GET_CPU(env), page2, DATA_SIZE, MMU_DATA_STORE,
                      mmu_idx, retaddr);
@@ -364,8 +361,7 @@ void helper_be_st_name(CPUArchState *env, target_ulong addr, DATA_TYPE val,
     }
 
     /* If the TLB entry is for a different page, reload and try again.  */
-    if ((addr & TARGET_PAGE_MASK)
-        != (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    if (!tlb_hit(tlb_addr, addr)) {
         if (!VICTIM_TLB_HIT(addr_write, addr)) {
             tlb_fill(ENV_GET_CPU(env), addr, DATA_SIZE, MMU_DATA_STORE,
                      mmu_idx, retaddr);
@@ -400,7 +396,7 @@ void helper_be_st_name(CPUArchState *env, target_ulong addr, DATA_TYPE val,
         page2 = (addr + DATA_SIZE) & TARGET_PAGE_MASK;
         index2 = (page2 >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
         tlb_addr2 = env->tlb_table[mmu_idx][index2].addr_write;
-        if (page2 != (tlb_addr2 & (TARGET_PAGE_MASK | TLB_INVALID_MASK))
+        if (!tlb_hit_page(tlb_addr2, page2)
             && !VICTIM_TLB_HIT(addr_write, page2)) {
             tlb_fill(ENV_GET_CPU(env), page2, DATA_SIZE, MMU_DATA_STORE,
                      mmu_idx, retaddr);

accel/tcg/translate-all.c

@@ -669,9 +669,15 @@ static inline void page_lock_tb(const TranslationBlock *tb)
 
 static inline void page_unlock_tb(const TranslationBlock *tb)
 {
-    page_unlock(page_find(tb->page_addr[0] >> TARGET_PAGE_BITS));
+    PageDesc *p1 = page_find(tb->page_addr[0] >> TARGET_PAGE_BITS);
+
+    page_unlock(p1);
     if (unlikely(tb->page_addr[1] != -1)) {
-        page_unlock(page_find(tb->page_addr[1] >> TARGET_PAGE_BITS));
+        PageDesc *p2 = page_find(tb->page_addr[1] >> TARGET_PAGE_BITS);
+
+        if (p2 != p1) {
+            page_unlock(p2);
+        }
     }
 }
 
@@ -850,22 +856,34 @@ static void page_lock_pair(PageDesc **ret_p1, tb_page_addr_t phys1,
                            PageDesc **ret_p2, tb_page_addr_t phys2, int alloc)
 {
     PageDesc *p1, *p2;
+    tb_page_addr_t page1;
+    tb_page_addr_t page2;
 
     assert_memory_lock();
-    g_assert(phys1 != -1 && phys1 != phys2);
-    p1 = page_find_alloc(phys1 >> TARGET_PAGE_BITS, alloc);
+    g_assert(phys1 != -1);
+
+    page1 = phys1 >> TARGET_PAGE_BITS;
+    page2 = phys2 >> TARGET_PAGE_BITS;
+
+    p1 = page_find_alloc(page1, alloc);
     if (ret_p1) {
         *ret_p1 = p1;
     }
     if (likely(phys2 == -1)) {
         page_lock(p1);
         return;
+    } else if (page1 == page2) {
+        page_lock(p1);
+        if (ret_p2) {
+            *ret_p2 = p1;
+        }
+        return;
     }
-    p2 = page_find_alloc(phys2 >> TARGET_PAGE_BITS, alloc);
+    p2 = page_find_alloc(page2, alloc);
     if (ret_p2) {
         *ret_p2 = p2;
     }
-    if (phys1 < phys2) {
+    if (page1 < page2) {
         page_lock(p1);
         page_lock(p2);
     } else {
@@ -1623,7 +1641,7 @@ tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc,
         tb = existing_tb;
     }
 
-    if (p2) {
+    if (p2 && p2 != p) {
         page_unlock(p2);
     }
     page_unlock(p);

include/exec/cpu-all.h

@@ -339,6 +339,29 @@ CPUArchState *cpu_copy(CPUArchState *env);
 #define TLB_FLAGS_MASK  (TLB_INVALID_MASK | TLB_NOTDIRTY | TLB_MMIO \
                          | TLB_RECHECK)
 
+/**
+ * tlb_hit_page: return true if page aligned @addr is a hit against the
+ * TLB entry @tlb_addr
+ *
+ * @addr: virtual address to test (must be page aligned)
+ * @tlb_addr: TLB entry address (a CPUTLBEntry addr_read/write/code value)
+ */
+static inline bool tlb_hit_page(target_ulong tlb_addr, target_ulong addr)
+{
+    return addr == (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK));
+}
+
+/**
+ * tlb_hit: return true if @addr is a hit against the TLB entry @tlb_addr
+ *
+ * @addr: virtual address to test (need not be page aligned)
+ * @tlb_addr: TLB entry address (a CPUTLBEntry addr_read/write/code value)
+ */
+static inline bool tlb_hit(target_ulong tlb_addr, target_ulong addr)
+{
+    return tlb_hit_page(tlb_addr, addr & TARGET_PAGE_MASK);
+}
+
 void dump_exec_info(FILE *f, fprintf_function cpu_fprintf);
 void dump_opcount_info(FILE *f, fprintf_function cpu_fprintf);
 #endif /* !CONFIG_USER_ONLY */

include/exec/cpu_ldst.h

@@ -422,8 +422,7 @@ static inline void *tlb_vaddr_to_host(CPUArchState *env, target_ulong addr,
         g_assert_not_reached();
     }
 
-    if ((addr & TARGET_PAGE_MASK)
-        != (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
+    if (!tlb_hit(tlb_addr, addr)) {
         /* TLB entry is for a different page */
         return NULL;
     }

include/qom/cpu.h

@@ -620,11 +620,13 @@ static inline hwaddr cpu_get_phys_page_debug(CPUState *cpu, vaddr addr)
 static inline int cpu_asidx_from_attrs(CPUState *cpu, MemTxAttrs attrs)
 {
     CPUClass *cc = CPU_GET_CLASS(cpu);
+    int ret = 0;
 
     if (cc->asidx_from_attrs) {
-        return cc->asidx_from_attrs(cpu, attrs);
+        ret = cc->asidx_from_attrs(cpu, attrs);
+        assert(ret < cpu->num_ases && ret >= 0);
     }
-    return 0;
+    return ret;
 }
 #endif