Merge remote-tracking branch 'remotes/bonzini/softmmu-smap' into staging

* remotes/bonzini/softmmu-smap: (33 commits)
  target-i386: cleanup x86_cpu_get_phys_page_debug
  target-i386: fix protection bits in the TLB for SMEP
  target-i386: support long addresses for 4MB pages (PSE-36)
  target-i386: raise page fault for reserved bits in large pages
  target-i386: unify reserved bits and NX bit check
  target-i386: simplify pte/vaddr calculation
  target-i386: raise page fault for reserved physical address bits
  target-i386: test reserved PS bit on PML4Es
  target-i386: set correct error code for reserved bit access
  target-i386: introduce support for 1 GB pages
  target-i386: introduce do_check_protect label
  target-i386: tweak handling of PG_NX_MASK
  target-i386: commonize checks for PAE and non-PAE
  target-i386: commonize checks for 4MB and 4KB pages
  target-i386: commonize checks for 2MB and 4KB pages
  target-i386: fix coding standards in x86_cpu_handle_mmu_fault
  target-i386: simplify SMAP handling in MMU_KSMAP_IDX
  target-i386: fix kernel accesses with SMAP and CPL = 3
  target-i386: move check_io helpers to seg_helper.c
  target-i386: rename KSMAP to KNOSMAP
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

9 files changed, 242 insertions, 340 deletions

diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index cbf1d979cb..dde052cc42 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -552,8 +552,7 @@ struct X86CPUDefinition {
           CPUID_PSE36 | CPUID_CLFLUSH | CPUID_ACPI | CPUID_MMX | \
           CPUID_FXSR | CPUID_SSE | CPUID_SSE2 | CPUID_SS)
           /* partly implemented:
-          CPUID_MTRR, CPUID_MCA, CPUID_CLFLUSH (needed for Win64)
-          CPUID_PSE36 (needed for Solaris) */
+          CPUID_MTRR, CPUID_MCA, CPUID_CLFLUSH (needed for Win64) */
           /* missing:
           CPUID_VME, CPUID_DTS, CPUID_SS, CPUID_HT, CPUID_TM, CPUID_PBE */
 #define TCG_EXT_FEATURES (CPUID_EXT_SSE3 | CPUID_EXT_PCLMULQDQ | \
@@ -569,9 +568,7 @@ struct X86CPUDefinition {
           CPUID_EXT_RDRAND */
 #define TCG_EXT2_FEATURES ((TCG_FEATURES & CPUID_EXT2_AMD_ALIASES) | \
           CPUID_EXT2_NX | CPUID_EXT2_MMXEXT | CPUID_EXT2_RDTSCP | \
-          CPUID_EXT2_3DNOW | CPUID_EXT2_3DNOWEXT)
-          /* missing:
-          CPUID_EXT2_PDPE1GB */
+          CPUID_EXT2_3DNOW | CPUID_EXT2_3DNOWEXT | CPUID_EXT2_PDPE1GB)
 #define TCG_EXT3_FEATURES (CPUID_EXT3_LAHF_LM | CPUID_EXT3_SVM | \
           CPUID_EXT3_CR8LEG | CPUID_EXT3_ABM | CPUID_EXT3_SSE4A)
 #define TCG_SVM_FEATURES 0
diff --git a/target-i386/cpu.h b/target-i386/cpu.h
index ee410af7ae..b5e1b411fb 100644
--- a/target-i386/cpu.h
+++ b/target-i386/cpu.h
@@ -249,6 +249,7 @@
 #define PG_DIRTY_BIT    6
 #define PG_PSE_BIT      7
 #define PG_GLOBAL_BIT   8
+#define PG_PSE_PAT_BIT  12
 #define PG_NX_BIT       63
 
 #define PG_PRESENT_MASK  (1 << PG_PRESENT_BIT)
@@ -260,6 +261,9 @@
 #define PG_DIRTY_MASK    (1 << PG_DIRTY_BIT)
 #define PG_PSE_MASK      (1 << PG_PSE_BIT)
 #define PG_GLOBAL_MASK   (1 << PG_GLOBAL_BIT)
+#define PG_PSE_PAT_MASK  (1 << PG_PSE_PAT_BIT)
+#define PG_ADDRESS_MASK  0x000ffffffffff000LL
+#define PG_HI_RSVD_MASK  (PG_ADDRESS_MASK & ~PHYS_ADDR_MASK)
 #define PG_HI_USER_MASK  0x7ff0000000000000LL
 #define PG_NX_MASK       (1LL << PG_NX_BIT)
 
@@ -1135,6 +1139,14 @@ uint64_t cpu_get_tsc(CPUX86State *env);
 #define TARGET_VIRT_ADDR_SPACE_BITS 32
 #endif
 
+/* XXX: This value should match the one returned by CPUID
+ * and in exec.c */
+# if defined(TARGET_X86_64)
+# define PHYS_ADDR_MASK 0xffffffffffLL
+# else
+# define PHYS_ADDR_MASK 0xfffffffffLL
+# endif
+
 static inline CPUX86State *cpu_init(const char *cpu_model)
 {
     X86CPU *cpu = cpu_x86_init(cpu_model);
@@ -1151,17 +1163,24 @@ static inline CPUX86State *cpu_init(const char *cpu_model)
 #define cpudef_setup x86_cpudef_setup
 
 /* MMU modes definitions */
-#define MMU_MODE0_SUFFIX _kernel
+#define MMU_MODE0_SUFFIX _ksmap
 #define MMU_MODE1_SUFFIX _user
-#define MMU_MODE2_SUFFIX _ksmap /* Kernel with SMAP override */
-#define MMU_KERNEL_IDX  0
+#define MMU_MODE2_SUFFIX _knosmap /* SMAP disabled or CPL<3 && AC=1 */
+#define MMU_KSMAP_IDX   0
 #define MMU_USER_IDX    1
-#define MMU_KSMAP_IDX   2
-static inline int cpu_mmu_index (CPUX86State *env)
+#define MMU_KNOSMAP_IDX 2
+static inline int cpu_mmu_index(CPUX86State *env)
 {
     return (env->hflags & HF_CPL_MASK) == 3 ? MMU_USER_IDX :
-        ((env->hflags & HF_SMAP_MASK) && (env->eflags & AC_MASK))
-        ? MMU_KSMAP_IDX : MMU_KERNEL_IDX;
+        (!(env->hflags & HF_SMAP_MASK) || (env->eflags & AC_MASK))
+        ? MMU_KNOSMAP_IDX : MMU_KSMAP_IDX;
+}
+
+static inline int cpu_mmu_index_kernel(CPUX86State *env)
+{
+    return !(env->hflags & HF_SMAP_MASK) ? MMU_KNOSMAP_IDX :
+        ((env->hflags & HF_CPL_MASK) < 3 && (env->eflags & AC_MASK))
+        ? MMU_KNOSMAP_IDX : MMU_KSMAP_IDX;
 }
 
 #define CC_DST  (env->cc_dst)
diff --git a/target-i386/fpu_helper.c b/target-i386/fpu_helper.c
index a04e754e61..1b2900d5d2 100644
--- a/target-i386/fpu_helper.c
+++ b/target-i386/fpu_helper.c
@@ -22,10 +22,7 @@
 #include "exec/helper-proto.h"
 #include "qemu/aes.h"
 #include "qemu/host-utils.h"
-
-#if !defined(CONFIG_USER_ONLY)
-#include "exec/softmmu_exec.h"
-#endif /* !defined(CONFIG_USER_ONLY) */
+#include "exec/cpu_ldst.h"
 
 #define FPU_RC_MASK         0xc00
 #define FPU_RC_NEAR         0x000
diff --git a/target-i386/helper.c b/target-i386/helper.c
index 46d20e4b89..11ca8649b5 100644
--- a/target-i386/helper.c
+++ b/target-i386/helper.c
@@ -510,14 +510,6 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
 
 #else
 
-/* XXX: This value should match the one returned by CPUID
- * and in exec.c */
-# if defined(TARGET_X86_64)
-# define PHYS_ADDR_MASK 0xfffffff000LL
-# else
-# define PHYS_ADDR_MASK 0xffffff000LL
-# endif
-
 /* return value:
  * -1 = cannot handle fault
  * 0  = nothing more to do
@@ -530,10 +522,12 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
     CPUX86State *env = &cpu->env;
     uint64_t ptep, pte;
     target_ulong pde_addr, pte_addr;
-    int error_code, is_dirty, prot, page_size, is_write, is_user;
+    int error_code = 0;
+    int is_dirty, prot, page_size, is_write, is_user;
     hwaddr paddr;
+    uint64_t rsvd_mask = PG_HI_RSVD_MASK;
     uint32_t page_offset;
-    target_ulong vaddr, virt_addr;
+    target_ulong vaddr;
 
     is_user = mmu_idx == MMU_USER_IDX;
 #if defined(DEBUG_MMU)
@@ -550,12 +544,15 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
             pte = (uint32_t)pte;
         }
 #endif
-        virt_addr = addr & TARGET_PAGE_MASK;
         prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
         page_size = 4096;
         goto do_mapping;
     }
 
+    if (!(env->efer & MSR_EFER_NXE)) {
+        rsvd_mask |= PG_NX_MASK;
+    }
+
     if (env->cr[4] & CR4_PAE_MASK) {
         uint64_t pde, pdpe;
         target_ulong pdpe_addr;
@@ -577,34 +574,37 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
                 env->a20_mask;
             pml4e = ldq_phys(cs->as, pml4e_addr);
             if (!(pml4e & PG_PRESENT_MASK)) {
-                error_code = 0;
                 goto do_fault;
             }
-            if (!(env->efer & MSR_EFER_NXE) && (pml4e & PG_NX_MASK)) {
-                error_code = PG_ERROR_RSVD_MASK;
-                goto do_fault;
+            if (pml4e & (rsvd_mask | PG_PSE_MASK)) {
+                goto do_fault_rsvd;
             }
             if (!(pml4e & PG_ACCESSED_MASK)) {
                 pml4e |= PG_ACCESSED_MASK;
                 stl_phys_notdirty(cs->as, pml4e_addr, pml4e);
             }
             ptep = pml4e ^ PG_NX_MASK;
-            pdpe_addr = ((pml4e & PHYS_ADDR_MASK) + (((addr >> 30) & 0x1ff) << 3)) &
+            pdpe_addr = ((pml4e & PG_ADDRESS_MASK) + (((addr >> 30) & 0x1ff) << 3)) &
                 env->a20_mask;
             pdpe = ldq_phys(cs->as, pdpe_addr);
             if (!(pdpe & PG_PRESENT_MASK)) {
-                error_code = 0;
                 goto do_fault;
             }
-            if (!(env->efer & MSR_EFER_NXE) && (pdpe & PG_NX_MASK)) {
-                error_code = PG_ERROR_RSVD_MASK;
-                goto do_fault;
+            if (pdpe & rsvd_mask) {
+                goto do_fault_rsvd;
             }
             ptep &= pdpe ^ PG_NX_MASK;
             if (!(pdpe & PG_ACCESSED_MASK)) {
                 pdpe |= PG_ACCESSED_MASK;
                 stl_phys_notdirty(cs->as, pdpe_addr, pdpe);
             }
+            if (pdpe & PG_PSE_MASK) {
+                /* 1 GB page */
+                page_size = 1024 * 1024 * 1024;
+                pte_addr = pdpe_addr;
+                pte = pdpe;
+                goto do_check_protect;
+            }
         } else
 #endif
         {
@@ -613,134 +613,49 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
                 env->a20_mask;
             pdpe = ldq_phys(cs->as, pdpe_addr);
             if (!(pdpe & PG_PRESENT_MASK)) {
-                error_code = 0;
                 goto do_fault;
             }
+            rsvd_mask |= PG_HI_USER_MASK | PG_NX_MASK;
+            if (pdpe & rsvd_mask) {
+                goto do_fault_rsvd;
+            }
             ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
         }
 
-        pde_addr = ((pdpe & PHYS_ADDR_MASK) + (((addr >> 21) & 0x1ff) << 3)) &
+        pde_addr = ((pdpe & PG_ADDRESS_MASK) + (((addr >> 21) & 0x1ff) << 3)) &
             env->a20_mask;
         pde = ldq_phys(cs->as, pde_addr);
         if (!(pde & PG_PRESENT_MASK)) {
-            error_code = 0;
             goto do_fault;
         }
-        if (!(env->efer & MSR_EFER_NXE) && (pde & PG_NX_MASK)) {
-            error_code = PG_ERROR_RSVD_MASK;
-            goto do_fault;
+        if (pde & rsvd_mask) {
+            goto do_fault_rsvd;
         }
         ptep &= pde ^ PG_NX_MASK;
         if (pde & PG_PSE_MASK) {
             /* 2 MB page */
             page_size = 2048 * 1024;
-            ptep ^= PG_NX_MASK;
-            if ((ptep & PG_NX_MASK) && is_write1 == 2) {
-                goto do_fault_protect;
-            }
-            switch (mmu_idx) {
-            case MMU_USER_IDX:
-                if (!(ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if (is_write && !(ptep & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
-
-            case MMU_KERNEL_IDX:
-                if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
-                    (ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                /* fall through */
-            case MMU_KSMAP_IDX:
-                if (is_write1 == 2 && (env->cr[4] & CR4_SMEP_MASK) &&
-                    (ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if ((env->cr[0] & CR0_WP_MASK) &&
-                    is_write && !(ptep & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
-
-            default: /* cannot happen */
-                break;
-            }
-            is_dirty = is_write && !(pde & PG_DIRTY_MASK);
-            if (!(pde & PG_ACCESSED_MASK) || is_dirty) {
-                pde |= PG_ACCESSED_MASK;
-                if (is_dirty)
-                    pde |= PG_DIRTY_MASK;
-                stl_phys_notdirty(cs->as, pde_addr, pde);
-            }
-            /* align to page_size */
-            pte = pde & ((PHYS_ADDR_MASK & ~(page_size - 1)) | 0xfff);
-            virt_addr = addr & ~(page_size - 1);
-        } else {
-            /* 4 KB page */
-            if (!(pde & PG_ACCESSED_MASK)) {
-                pde |= PG_ACCESSED_MASK;
-                stl_phys_notdirty(cs->as, pde_addr, pde);
-            }
-            pte_addr = ((pde & PHYS_ADDR_MASK) + (((addr >> 12) & 0x1ff) << 3)) &
-                env->a20_mask;
-            pte = ldq_phys(cs->as, pte_addr);
-            if (!(pte & PG_PRESENT_MASK)) {
-                error_code = 0;
-                goto do_fault;
-            }
-            if (!(env->efer & MSR_EFER_NXE) && (pte & PG_NX_MASK)) {
-                error_code = PG_ERROR_RSVD_MASK;
-                goto do_fault;
-            }
-            /* combine pde and pte nx, user and rw protections */
-            ptep &= pte ^ PG_NX_MASK;
-            ptep ^= PG_NX_MASK;
-            if ((ptep & PG_NX_MASK) && is_write1 == 2)
-                goto do_fault_protect;
-            switch (mmu_idx) {
-            case MMU_USER_IDX:
-                if (!(ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if (is_write && !(ptep & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
-
-            case MMU_KERNEL_IDX:
-                if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
-                    (ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                /* fall through */
-            case MMU_KSMAP_IDX:
-                if (is_write1 == 2 && (env->cr[4] & CR4_SMEP_MASK) &&
-                    (ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if ((env->cr[0] & CR0_WP_MASK) &&
-                    is_write && !(ptep & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
-
-            default: /* cannot happen */
-                break;
-            }
-            is_dirty = is_write && !(pte & PG_DIRTY_MASK);
-            if (!(pte & PG_ACCESSED_MASK) || is_dirty) {
-                pte |= PG_ACCESSED_MASK;
-                if (is_dirty)
-                    pte |= PG_DIRTY_MASK;
-                stl_phys_notdirty(cs->as, pte_addr, pte);
-            }
-            page_size = 4096;
-            virt_addr = addr & ~0xfff;
-            pte = pte & (PHYS_ADDR_MASK | 0xfff);
+            pte_addr = pde_addr;
+            pte = pde;
+            goto do_check_protect;
+        }
+        /* 4 KB page */
+        if (!(pde & PG_ACCESSED_MASK)) {
+            pde |= PG_ACCESSED_MASK;
+            stl_phys_notdirty(cs->as, pde_addr, pde);
+        }
+        pte_addr = ((pde & PG_ADDRESS_MASK) + (((addr >> 12) & 0x1ff) << 3)) &
+            env->a20_mask;
+        pte = ldq_phys(cs->as, pte_addr);
+        if (!(pte & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        if (pte & rsvd_mask) {
+            goto do_fault_rsvd;
         }
+        /* combine pde and pte nx, user and rw protections */
+        ptep &= pte ^ PG_NX_MASK;
+        page_size = 4096;
     } else {
         uint32_t pde;
 
@@ -749,114 +664,95 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
             env->a20_mask;
         pde = ldl_phys(cs->as, pde_addr);
         if (!(pde & PG_PRESENT_MASK)) {
-            error_code = 0;
             goto do_fault;
         }
+        ptep = pde | PG_NX_MASK;
+
         /* if PSE bit is set, then we use a 4MB page */
         if ((pde & PG_PSE_MASK) && (env->cr[4] & CR4_PSE_MASK)) {
             page_size = 4096 * 1024;
-            switch (mmu_idx) {
-            case MMU_USER_IDX:
-                if (!(pde & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if (is_write && !(pde & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
-
-            case MMU_KERNEL_IDX:
-                if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
-                    (pde & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                /* fall through */
-            case MMU_KSMAP_IDX:
-                if (is_write1 == 2 && (env->cr[4] & CR4_SMEP_MASK) &&
-                    (pde & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if ((env->cr[0] & CR0_WP_MASK) &&
-                    is_write && !(pde & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
+            pte_addr = pde_addr;
+
+            /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved.
+             * Leave bits 20-13 in place for setting accessed/dirty bits below.
+             */
+            pte = pde | ((pde & 0x1fe000) << (32 - 13));
+            rsvd_mask = 0x200000;
+            goto do_check_protect_pse36;
+        }
 
-            default: /* cannot happen */
-                break;
-            }
-            is_dirty = is_write && !(pde & PG_DIRTY_MASK);
-            if (!(pde & PG_ACCESSED_MASK) || is_dirty) {
-                pde |= PG_ACCESSED_MASK;
-                if (is_dirty)
-                    pde |= PG_DIRTY_MASK;
-                stl_phys_notdirty(cs->as, pde_addr, pde);
-            }
+        if (!(pde & PG_ACCESSED_MASK)) {
+            pde |= PG_ACCESSED_MASK;
+            stl_phys_notdirty(cs->as, pde_addr, pde);
+        }
 
-            pte = pde & ~( (page_size - 1) & ~0xfff); /* align to page_size */
-            ptep = pte;
-            virt_addr = addr & ~(page_size - 1);
-        } else {
-            if (!(pde & PG_ACCESSED_MASK)) {
-                pde |= PG_ACCESSED_MASK;
-                stl_phys_notdirty(cs->as, pde_addr, pde);
-            }
+        /* page directory entry */
+        pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) &
+            env->a20_mask;
+        pte = ldl_phys(cs->as, pte_addr);
+        if (!(pte & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        /* combine pde and pte user and rw protections */
+        ptep &= pte | PG_NX_MASK;
+        page_size = 4096;
+        rsvd_mask = 0;
+    }
 
-            /* page directory entry */
-            pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) &
-                env->a20_mask;
-            pte = ldl_phys(cs->as, pte_addr);
-            if (!(pte & PG_PRESENT_MASK)) {
-                error_code = 0;
-                goto do_fault;
-            }
-            /* combine pde and pte user and rw protections */
-            ptep = pte & pde;
-            switch (mmu_idx) {
-            case MMU_USER_IDX:
-                if (!(ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if (is_write && !(ptep & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
+do_check_protect:
+    rsvd_mask |= (page_size - 1) & PG_ADDRESS_MASK & ~PG_PSE_PAT_MASK;
+do_check_protect_pse36:
+    if (pte & rsvd_mask) {
+        goto do_fault_rsvd;
+    }
+    ptep ^= PG_NX_MASK;
+    if ((ptep & PG_NX_MASK) && is_write1 == 2) {
+        goto do_fault_protect;
+    }
+    switch (mmu_idx) {
+    case MMU_USER_IDX:
+        if (!(ptep & PG_USER_MASK)) {
+            goto do_fault_protect;
+        }
+        if (is_write && !(ptep & PG_RW_MASK)) {
+            goto do_fault_protect;
+        }
+        break;
 
-            case MMU_KERNEL_IDX:
-                if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
-                    (ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                /* fall through */
-            case MMU_KSMAP_IDX:
-                if (is_write1 == 2 && (env->cr[4] & CR4_SMEP_MASK) &&
-                    (ptep & PG_USER_MASK)) {
-                    goto do_fault_protect;
-                }
-                if ((env->cr[0] & CR0_WP_MASK) &&
-                    is_write && !(ptep & PG_RW_MASK)) {
-                    goto do_fault_protect;
-                }
-                break;
+    case MMU_KSMAP_IDX:
+        if (is_write1 != 2 && (ptep & PG_USER_MASK)) {
+            goto do_fault_protect;
+        }
+        /* fall through */
+    case MMU_KNOSMAP_IDX:
+        if (is_write1 == 2 && (env->cr[4] & CR4_SMEP_MASK) &&
+            (ptep & PG_USER_MASK)) {
+            goto do_fault_protect;
+        }
+        if ((env->cr[0] & CR0_WP_MASK) &&
+            is_write && !(ptep & PG_RW_MASK)) {
+            goto do_fault_protect;
+        }
+        break;
 
-            default: /* cannot happen */
-                break;
-            }
-            is_dirty = is_write && !(pte & PG_DIRTY_MASK);
-            if (!(pte & PG_ACCESSED_MASK) || is_dirty) {
-                pte |= PG_ACCESSED_MASK;
-                if (is_dirty)
-                    pte |= PG_DIRTY_MASK;
-                stl_phys_notdirty(cs->as, pte_addr, pte);
-            }
-            page_size = 4096;
-            virt_addr = addr & ~0xfff;
+    default: /* cannot happen */
+        break;
+    }
+    is_dirty = is_write && !(pte & PG_DIRTY_MASK);
+    if (!(pte & PG_ACCESSED_MASK) || is_dirty) {
+        pte |= PG_ACCESSED_MASK;
+        if (is_dirty) {
+            pte |= PG_DIRTY_MASK;
         }
+        stl_phys_notdirty(cs->as, pte_addr, pte);
     }
+
     /* the page can be put in the TLB */
     prot = PAGE_READ;
-    if (!(ptep & PG_NX_MASK))
+    if (!(ptep & PG_NX_MASK) &&
+        !((env->cr[4] & CR4_SMEP_MASK) && (ptep & PG_USER_MASK))) {
         prot |= PAGE_EXEC;
+    }
     if (pte & PG_DIRTY_MASK) {
         /* only set write access if already dirty... otherwise wait
            for dirty access */
@@ -872,16 +768,21 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
  do_mapping:
     pte = pte & env->a20_mask;
 
+    /* align to page_size */
+    pte &= PG_ADDRESS_MASK & ~(page_size - 1);
+
     /* Even if 4MB pages, we map only one 4KB page in the cache to
        avoid filling it too fast */
-    page_offset = (addr & TARGET_PAGE_MASK) & (page_size - 1);
-    paddr = (pte & TARGET_PAGE_MASK) + page_offset;
-    vaddr = virt_addr + page_offset;
+    vaddr = addr & TARGET_PAGE_MASK;
+    page_offset = vaddr & (page_size - 1);
+    paddr = pte + page_offset;
 
     tlb_set_page(cs, vaddr, paddr, prot, mmu_idx, page_size);
     return 0;
+ do_fault_rsvd:
+    error_code |= PG_ERROR_RSVD_MASK;
  do_fault_protect:
-    error_code = PG_ERROR_P_MASK;
+    error_code |= PG_ERROR_P_MASK;
  do_fault:
     error_code |= (is_write << PG_ERROR_W_BIT);
     if (is_user)
@@ -910,7 +811,6 @@ hwaddr x86_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
     CPUX86State *env = &cpu->env;
     target_ulong pde_addr, pte_addr;
     uint64_t pte;
-    hwaddr paddr;
     uint32_t page_offset;
     int page_size;
 
@@ -928,25 +828,24 @@ hwaddr x86_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
 
             /* test virtual address sign extension */
             sext = (int64_t)addr >> 47;
-            if (sext != 0 && sext != -1)
+            if (sext != 0 && sext != -1) {
                 return -1;
-
+            }
             pml4e_addr = ((env->cr[3] & ~0xfff) + (((addr >> 39) & 0x1ff) << 3)) &
                 env->a20_mask;
             pml4e = ldq_phys(cs->as, pml4e_addr);
-            if (!(pml4e & PG_PRESENT_MASK))
+            if (!(pml4e & PG_PRESENT_MASK)) {
                 return -1;
-
-            pdpe_addr = ((pml4e & ~0xfff & ~(PG_NX_MASK | PG_HI_USER_MASK)) +
+            }
+            pdpe_addr = ((pml4e & PG_ADDRESS_MASK) +
                          (((addr >> 30) & 0x1ff) << 3)) & env->a20_mask;
             pdpe = ldq_phys(cs->as, pdpe_addr);
-            if (!(pdpe & PG_PRESENT_MASK))
+            if (!(pdpe & PG_PRESENT_MASK)) {
                 return -1;
-
+            }
             if (pdpe & PG_PSE_MASK) {
                 page_size = 1024 * 1024 * 1024;
-                pte = pdpe & ~( (page_size - 1) & ~0xfff);
-                pte &= ~(PG_NX_MASK | PG_HI_USER_MASK);
+                pte = pdpe;
                 goto out;
             }
 
@@ -960,7 +859,7 @@ hwaddr x86_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
                 return -1;
         }
 
-        pde_addr = ((pdpe & ~0xfff & ~(PG_NX_MASK | PG_HI_USER_MASK)) +
+        pde_addr = ((pdpe & PG_ADDRESS_MASK) +
                     (((addr >> 21) & 0x1ff) << 3)) & env->a20_mask;
         pde = ldq_phys(cs->as, pde_addr);
         if (!(pde & PG_PRESENT_MASK)) {
@@ -969,17 +868,17 @@ hwaddr x86_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
         if (pde & PG_PSE_MASK) {
             /* 2 MB page */
             page_size = 2048 * 1024;
-            pte = pde & ~( (page_size - 1) & ~0xfff); /* align to page_size */
+            pte = pde;
         } else {
             /* 4 KB page */
-            pte_addr = ((pde & ~0xfff & ~(PG_NX_MASK | PG_HI_USER_MASK)) +
+            pte_addr = ((pde & PG_ADDRESS_MASK) +
                         (((addr >> 12) & 0x1ff) << 3)) & env->a20_mask;
             page_size = 4096;
             pte = ldq_phys(cs->as, pte_addr);
         }
-        pte &= ~(PG_NX_MASK | PG_HI_USER_MASK);
-        if (!(pte & PG_PRESENT_MASK))
+        if (!(pte & PG_PRESENT_MASK)) {
             return -1;
+        }
     } else {
         uint32_t pde;
 
@@ -989,14 +888,15 @@ hwaddr x86_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
         if (!(pde & PG_PRESENT_MASK))
             return -1;
         if ((pde & PG_PSE_MASK) && (env->cr[4] & CR4_PSE_MASK)) {
-            pte = pde & ~0x003ff000; /* align to 4MB */
+            pte = pde | ((pde & 0x1fe000) << (32 - 13));
             page_size = 4096 * 1024;
         } else {
             /* page directory entry */
             pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) & env->a20_mask;
             pte = ldl_phys(cs->as, pte_addr);
-            if (!(pte & PG_PRESENT_MASK))
+            if (!(pte & PG_PRESENT_MASK)) {
                 return -1;
+            }
             page_size = 4096;
         }
         pte = pte & env->a20_mask;
@@ -1005,9 +905,9 @@ hwaddr x86_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
 #ifdef TARGET_X86_64
 out:
 #endif
+    pte &= PG_ADDRESS_MASK & ~(page_size - 1);
     page_offset = (addr & TARGET_PAGE_MASK) & (page_size - 1);
-    paddr = (pte & TARGET_PAGE_MASK) + page_offset;
-    return paddr;
+    return pte | page_offset;
 }
 
 void hw_breakpoint_insert(CPUX86State *env, int index)
diff --git a/target-i386/mem_helper.c b/target-i386/mem_helper.c
index 83aa1038d7..1aec8a5f19 100644
--- a/target-i386/mem_helper.c
+++ b/target-i386/mem_helper.c
@@ -19,10 +19,7 @@
 
 #include "cpu.h"
 #include "exec/helper-proto.h"
-
-#if !defined(CONFIG_USER_ONLY)
-#include "exec/softmmu_exec.h"
-#endif /* !defined(CONFIG_USER_ONLY) */
+#include "exec/cpu_ldst.h"
 
 /* broken thread support */
 
@@ -110,24 +107,6 @@ void helper_boundl(CPUX86State *env, target_ulong a0, int v)
 }
 
 #if !defined(CONFIG_USER_ONLY)
-
-#define MMUSUFFIX _mmu
-
-#define SHIFT 0
-#include "exec/softmmu_template.h"
-
-#define SHIFT 1
-#include "exec/softmmu_template.h"
-
-#define SHIFT 2
-#include "exec/softmmu_template.h"
-
-#define SHIFT 3
-#include "exec/softmmu_template.h"
-
-#endif
-
-#if !defined(CONFIG_USER_ONLY)
 /* try to fill the TLB and return an exception if error. If retaddr is
  * NULL, it means that the function was called in C code (i.e. not
  * from generated code or from helper.c)
diff --git a/target-i386/misc_helper.c b/target-i386/misc_helper.c
index 9cfa25f9ec..4aaf1e4d95 100644
--- a/target-i386/misc_helper.c
+++ b/target-i386/misc_helper.c
@@ -20,52 +20,7 @@
 #include "cpu.h"
 #include "exec/ioport.h"
 #include "exec/helper-proto.h"
-
-#if !defined(CONFIG_USER_ONLY)
-#include "exec/softmmu_exec.h"
-#endif /* !defined(CONFIG_USER_ONLY) */
-
-/* check if Port I/O is allowed in TSS */
-static inline void check_io(CPUX86State *env, int addr, int size)
-{
-    int io_offset, val, mask;
-
-    /* TSS must be a valid 32 bit one */
-    if (!(env->tr.flags & DESC_P_MASK) ||
-        ((env->tr.flags >> DESC_TYPE_SHIFT) & 0xf) != 9 ||
-        env->tr.limit < 103) {
-        goto fail;
-    }
-    io_offset = cpu_lduw_kernel(env, env->tr.base + 0x66);
-    io_offset += (addr >> 3);
-    /* Note: the check needs two bytes */
-    if ((io_offset + 1) > env->tr.limit) {
-        goto fail;
-    }
-    val = cpu_lduw_kernel(env, env->tr.base + io_offset);
-    val >>= (addr & 7);
-    mask = (1 << size) - 1;
-    /* all bits must be zero to allow the I/O */
-    if ((val & mask) != 0) {
-    fail:
-        raise_exception_err(env, EXCP0D_GPF, 0);
-    }
-}
-
-void helper_check_iob(CPUX86State *env, uint32_t t0)
-{
-    check_io(env, t0, 1);
-}
-
-void helper_check_iow(CPUX86State *env, uint32_t t0)
-{
-    check_io(env, t0, 2);
-}
-
-void helper_check_iol(CPUX86State *env, uint32_t t0)
-{
-    check_io(env, t0, 4);
-}
+#include "exec/cpu_ldst.h"
 
 void helper_outb(uint32_t port, uint32_t data)
 {
diff --git a/target-i386/seg_helper.c b/target-i386/seg_helper.c
index 51c2833ea5..2d970d0cb9 100644
--- a/target-i386/seg_helper.c
+++ b/target-i386/seg_helper.c
@@ -21,13 +21,10 @@
 #include "cpu.h"
 #include "qemu/log.h"
 #include "exec/helper-proto.h"
+#include "exec/cpu_ldst.h"
 
 //#define DEBUG_PCALL
 
-#if !defined(CONFIG_USER_ONLY)
-#include "exec/softmmu_exec.h"
-#endif /* !defined(CONFIG_USER_ONLY) */
-
 #ifdef DEBUG_PCALL
 # define LOG_PCALL(...) qemu_log_mask(CPU_LOG_PCALL, ## __VA_ARGS__)
 # define LOG_PCALL_STATE(cpu)                                  \
@@ -37,6 +34,24 @@
 # define LOG_PCALL_STATE(cpu) do { } while (0)
 #endif
 
+#ifndef CONFIG_USER_ONLY
+#define CPU_MMU_INDEX (cpu_mmu_index_kernel(env))
+#define MEMSUFFIX _kernel
+#define DATA_SIZE 1
+#include "exec/cpu_ldst_template.h"
+
+#define DATA_SIZE 2
+#include "exec/cpu_ldst_template.h"
+
+#define DATA_SIZE 4
+#include "exec/cpu_ldst_template.h"
+
+#define DATA_SIZE 8
+#include "exec/cpu_ldst_template.h"
+#undef CPU_MMU_INDEX
+#undef MEMSUFFIX
+#endif
+
 /* return non zero if error */
 static inline int load_segment(CPUX86State *env, uint32_t *e1_ptr,
                                uint32_t *e2_ptr, int selector)
@@ -2471,3 +2486,45 @@ void cpu_x86_load_seg(CPUX86State *env, int seg_reg, int selector)
     }
 }
 #endif
+
+/* check if Port I/O is allowed in TSS */
+static inline void check_io(CPUX86State *env, int addr, int size)
+{
+    int io_offset, val, mask;
+
+    /* TSS must be a valid 32 bit one */
+    if (!(env->tr.flags & DESC_P_MASK) ||
+        ((env->tr.flags >> DESC_TYPE_SHIFT) & 0xf) != 9 ||
+        env->tr.limit < 103) {
+        goto fail;
+    }
+    io_offset = cpu_lduw_kernel(env, env->tr.base + 0x66);
+    io_offset += (addr >> 3);
+    /* Note: the check needs two bytes */
+    if ((io_offset + 1) > env->tr.limit) {
+        goto fail;
+    }
+    val = cpu_lduw_kernel(env, env->tr.base + io_offset);
+    val >>= (addr & 7);
+    mask = (1 << size) - 1;
+    /* all bits must be zero to allow the I/O */
+    if ((val & mask) != 0) {
+    fail:
+        raise_exception_err(env, EXCP0D_GPF, 0);
+    }
+}
+
+void helper_check_iob(CPUX86State *env, uint32_t t0)
+{
+    check_io(env, t0, 1);
+}
+
+void helper_check_iow(CPUX86State *env, uint32_t t0)
+{
+    check_io(env, t0, 2);
+}
+
+void helper_check_iol(CPUX86State *env, uint32_t t0)
+{
+    check_io(env, t0, 4);
+}
diff --git a/target-i386/svm_helper.c b/target-i386/svm_helper.c
index ea5de6fa94..429d029a3d 100644
--- a/target-i386/svm_helper.c
+++ b/target-i386/svm_helper.c
@@ -20,10 +20,7 @@
 #include "cpu.h"
 #include "exec/cpu-all.h"
 #include "exec/helper-proto.h"
-
-#if !defined(CONFIG_USER_ONLY)
-#include "exec/softmmu_exec.h"
-#endif /* !defined(CONFIG_USER_ONLY) */
+#include "exec/cpu_ldst.h"
 
 /* Secure Virtual Machine helpers */
 
diff --git a/target-i386/translate.c b/target-i386/translate.c
index 3aa52eb795..2359787b42 100644
--- a/target-i386/translate.c
+++ b/target-i386/translate.c
@@ -27,6 +27,7 @@
 #include "cpu.h"
 #include "disas/disas.h"
 #include "tcg-op.h"
+#include "exec/cpu_ldst.h"
 
 #include "exec/helper-proto.h"
 #include "exec/helper-gen.h"