unicorn/qemu/target/i386/helper.c
Pavel Dovgalyuk b4bf3c776b
icount: fix cpu_restore_state_from_tb for non-tb-exit cases
In icount mode, instructions that access io memory spaces in the middle
of the translation block invoke TB recompilation. After recompilation,
such instructions become last in the TB and are allowed to access io
memory spaces.

When the code includes instruction like i386 'xchg eax, 0xffffd080'
which accesses APIC, QEMU goes into an infinite loop of the recompilation.

This instruction includes two memory accesses - one read and one write.
After the first access, APIC calls cpu_report_tpr_access, which restores
the CPU state to get the current eip. But cpu_restore_state_from_tb
resets the cpu->can_do_io flag which makes the second memory access invalid.
Therefore the second memory access causes a recompilation of the block.
Then these operations repeat again and again.

This patch moves resetting cpu->can_do_io flag from
cpu_restore_state_from_tb to cpu_loop_exit* functions.

It also adds a parameter for cpu_restore_state which controls restoring
icount. There is no need to restore icount when we only query CPU state
without breaking the TB. Restoring it in such cases leads to the
incorrect flow of the virtual time.

In most cases new parameter is true (icount should be recalculated).
But there are two cases in i386 and openrisc when the CPU state is only
queried without the need to break the TB. This patch fixes both of
these cases.

Backports commit afd46fcad2dceffda35c0586f5723c127b6e09d8 from qemu
2018-04-11 20:05:40 -04:00

1200 lines
36 KiB
C

/*
* i386 helpers (without register variable usage)
*
* Copyright (c) 2003 Fabrice Bellard
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, see <http://www.gnu.org/licenses/>.
*/
#include "qemu/osdep.h"
#include "cpu.h"
#include "exec/exec-all.h"
#ifndef CONFIG_USER_ONLY
#include "sysemu/sysemu.h"
#endif
void cpu_sync_bndcs_hflags(CPUX86State *env)
{
uint32_t hflags = env->hflags;
uint32_t hflags2 = env->hflags2;
uint32_t bndcsr;
if ((hflags & HF_CPL_MASK) == 3) {
bndcsr = env->bndcs_regs.cfgu;
} else {
bndcsr = env->msr_bndcfgs;
}
if ((env->cr[4] & CR4_OSXSAVE_MASK)
&& (env->xcr0 & XSTATE_BNDCSR_MASK)
&& (bndcsr & BNDCFG_ENABLE)) {
hflags |= HF_MPX_EN_MASK;
} else {
hflags &= ~HF_MPX_EN_MASK;
}
if (bndcsr & BNDCFG_BNDPRESERVE) {
hflags2 |= HF2_MPX_PR_MASK;
} else {
hflags2 &= ~HF2_MPX_PR_MASK;
}
env->hflags = hflags;
env->hflags2 = hflags2;
}
static void cpu_x86_version(CPUX86State *env, int *family, int *model)
{
int cpuver = env->cpuid_version;
if (family == NULL || model == NULL) {
return;
}
*family = (cpuver >> 8) & 0x0f;
*model = ((cpuver >> 12) & 0xf0) + ((cpuver >> 4) & 0x0f);
}
/* Broadcast MCA signal for processor version 06H_EH and above */
int cpu_x86_support_mca_broadcast(CPUX86State *env)
{
int family = 0;
int model = 0;
cpu_x86_version(env, &family, &model);
if ((family == 6 && model >= 14) || family > 6) {
return 1;
}
return 0;
}
/***********************************************************/
/* x86 debug */
static const char *cc_op_str[CC_OP_NB] = {
"DYNAMIC",
"EFLAGS",
"MULB",
"MULW",
"MULL",
"MULQ",
"ADDB",
"ADDW",
"ADDL",
"ADDQ",
"ADCB",
"ADCW",
"ADCL",
"ADCQ",
"SUBB",
"SUBW",
"SUBL",
"SUBQ",
"SBBB",
"SBBW",
"SBBL",
"SBBQ",
"LOGICB",
"LOGICW",
"LOGICL",
"LOGICQ",
"INCB",
"INCW",
"INCL",
"INCQ",
"DECB",
"DECW",
"DECL",
"DECQ",
"SHLB",
"SHLW",
"SHLL",
"SHLQ",
"SARB",
"SARW",
"SARL",
"SARQ",
"BMILGB",
"BMILGW",
"BMILGL",
"BMILGQ",
"ADCX",
"ADOX",
"ADCOX",
"CLR",
};
static void
cpu_x86_dump_seg_cache(CPUX86State *env, FILE *f, fprintf_function cpu_fprintf,
const char *name, struct SegmentCache *sc)
{
#ifdef TARGET_X86_64
if (env->hflags & HF_CS64_MASK) {
cpu_fprintf(f, "%-3s=%04x %016" PRIx64 " %08x %08x", name,
sc->selector, sc->base, sc->limit, sc->flags & 0x00ffff00);
} else
#endif
{
cpu_fprintf(f, "%-3s=%04x %08x %08x %08x", name, sc->selector,
(uint32_t)sc->base, sc->limit, sc->flags & 0x00ffff00);
}
if (!(env->hflags & HF_PE_MASK) || !(sc->flags & DESC_P_MASK))
goto done;
cpu_fprintf(f, " DPL=%d ", (sc->flags & DESC_DPL_MASK) >> DESC_DPL_SHIFT);
if (sc->flags & DESC_S_MASK) {
if (sc->flags & DESC_CS_MASK) {
cpu_fprintf(f, (sc->flags & DESC_L_MASK) ? "CS64" :
((sc->flags & DESC_B_MASK) ? "CS32" : "CS16"));
cpu_fprintf(f, " [%c%c", (sc->flags & DESC_C_MASK) ? 'C' : '-',
(sc->flags & DESC_R_MASK) ? 'R' : '-');
} else {
cpu_fprintf(f,
(sc->flags & DESC_B_MASK || env->hflags & HF_LMA_MASK)
? "DS " : "DS16");
cpu_fprintf(f, " [%c%c", (sc->flags & DESC_E_MASK) ? 'E' : '-',
(sc->flags & DESC_W_MASK) ? 'W' : '-');
}
cpu_fprintf(f, "%c]", (sc->flags & DESC_A_MASK) ? 'A' : '-');
} else {
static const char *sys_type_name[2][16] = {
{ /* 32 bit mode */
"Reserved", "TSS16-avl", "LDT", "TSS16-busy",
"CallGate16", "TaskGate", "IntGate16", "TrapGate16",
"Reserved", "TSS32-avl", "Reserved", "TSS32-busy",
"CallGate32", "Reserved", "IntGate32", "TrapGate32"
},
{ /* 64 bit mode */
"<hiword>", "Reserved", "LDT", "Reserved", "Reserved",
"Reserved", "Reserved", "Reserved", "Reserved",
"TSS64-avl", "Reserved", "TSS64-busy", "CallGate64",
"Reserved", "IntGate64", "TrapGate64"
}
};
cpu_fprintf(f, "%s",
sys_type_name[(env->hflags & HF_LMA_MASK) ? 1 : 0]
[(sc->flags & DESC_TYPE_MASK)
>> DESC_TYPE_SHIFT]);
}
done:
cpu_fprintf(f, "\n");
}
#define DUMP_CODE_BYTES_TOTAL 50
#define DUMP_CODE_BYTES_BACKWARD 20
void x86_cpu_dump_state(CPUState *cs, FILE *f, fprintf_function cpu_fprintf,
int flags)
{
X86CPU *cpu = X86_CPU(cs->uc, cs);
CPUX86State *env = &cpu->env;
int eflags, i, nb;
char cc_op_name[32];
static const char *seg_name[6] = { "ES", "CS", "SS", "DS", "FS", "GS" };
eflags = cpu_compute_eflags(env);
#ifdef TARGET_X86_64
if (env->hflags & HF_CS64_MASK) {
cpu_fprintf(f,
"RAX=%016" PRIx64 " RBX=%016" PRIx64 " RCX=%016" PRIx64 " RDX=%016" PRIx64 "\n"
"RSI=%016" PRIx64 " RDI=%016" PRIx64 " RBP=%016" PRIx64 " RSP=%016" PRIx64 "\n"
"R8 =%016" PRIx64 " R9 =%016" PRIx64 " R10=%016" PRIx64 " R11=%016" PRIx64 "\n"
"R12=%016" PRIx64 " R13=%016" PRIx64 " R14=%016" PRIx64 " R15=%016" PRIx64 "\n"
"RIP=%016" PRIx64 " RFL=%08x [%c%c%c%c%c%c%c] CPL=%d II=%d A20=%d SMM=%d HLT=%d\n",
env->regs[R_EAX],
env->regs[R_EBX],
env->regs[R_ECX],
env->regs[R_EDX],
env->regs[R_ESI],
env->regs[R_EDI],
env->regs[R_EBP],
env->regs[R_ESP],
env->regs[8],
env->regs[9],
env->regs[10],
env->regs[11],
env->regs[12],
env->regs[13],
env->regs[14],
env->regs[15],
env->eip, eflags,
eflags & DF_MASK ? 'D' : '-',
eflags & CC_O ? 'O' : '-',
eflags & CC_S ? 'S' : '-',
eflags & CC_Z ? 'Z' : '-',
eflags & CC_A ? 'A' : '-',
eflags & CC_P ? 'P' : '-',
eflags & CC_C ? 'C' : '-',
env->hflags & HF_CPL_MASK,
(env->hflags >> HF_INHIBIT_IRQ_SHIFT) & 1,
(env->a20_mask >> 20) & 1,
(env->hflags >> HF_SMM_SHIFT) & 1,
cs->halted);
} else
#endif
{
cpu_fprintf(f, "EAX=%08x EBX=%08x ECX=%08x EDX=%08x\n"
"ESI=%08x EDI=%08x EBP=%08x ESP=%08x\n"
"EIP=%08x EFL=%08x [%c%c%c%c%c%c%c] CPL=%d II=%d A20=%d SMM=%d HLT=%d\n",
(uint32_t)env->regs[R_EAX],
(uint32_t)env->regs[R_EBX],
(uint32_t)env->regs[R_ECX],
(uint32_t)env->regs[R_EDX],
(uint32_t)env->regs[R_ESI],
(uint32_t)env->regs[R_EDI],
(uint32_t)env->regs[R_EBP],
(uint32_t)env->regs[R_ESP],
(uint32_t)env->eip, eflags,
eflags & DF_MASK ? 'D' : '-',
eflags & CC_O ? 'O' : '-',
eflags & CC_S ? 'S' : '-',
eflags & CC_Z ? 'Z' : '-',
eflags & CC_A ? 'A' : '-',
eflags & CC_P ? 'P' : '-',
eflags & CC_C ? 'C' : '-',
env->hflags & HF_CPL_MASK,
(env->hflags >> HF_INHIBIT_IRQ_SHIFT) & 1,
(env->a20_mask >> 20) & 1,
(env->hflags >> HF_SMM_SHIFT) & 1,
cs->halted);
}
for(i = 0; i < 6; i++) {
cpu_x86_dump_seg_cache(env, f, cpu_fprintf, seg_name[i],
&env->segs[i]);
}
cpu_x86_dump_seg_cache(env, f, cpu_fprintf, "LDT", &env->ldt);
cpu_x86_dump_seg_cache(env, f, cpu_fprintf, "TR", &env->tr);
#ifdef TARGET_X86_64
if (env->hflags & HF_LMA_MASK) {
cpu_fprintf(f, "GDT= %016" PRIx64 " %08x\n",
env->gdt.base, env->gdt.limit);
cpu_fprintf(f, "IDT= %016" PRIx64 " %08x\n",
env->idt.base, env->idt.limit);
cpu_fprintf(f, "CR0=%08x CR2=%016" PRIx64 " CR3=%016" PRIx64 " CR4=%08x\n",
(uint32_t)env->cr[0],
env->cr[2],
env->cr[3],
(uint32_t)env->cr[4]);
for(i = 0; i < 4; i++)
cpu_fprintf(f, "DR%d=%016" PRIx64 " ", i, env->dr[i]);
cpu_fprintf(f, "\nDR6=%016" PRIx64 " DR7=%016" PRIx64 "\n",
env->dr[6], env->dr[7]);
} else
#endif
{
cpu_fprintf(f, "GDT= %08x %08x\n",
(uint32_t)env->gdt.base, env->gdt.limit);
cpu_fprintf(f, "IDT= %08x %08x\n",
(uint32_t)env->idt.base, env->idt.limit);
cpu_fprintf(f, "CR0=%08x CR2=%08x CR3=%08x CR4=%08x\n",
(uint32_t)env->cr[0],
(uint32_t)env->cr[2],
(uint32_t)env->cr[3],
(uint32_t)env->cr[4]);
for(i = 0; i < 4; i++) {
cpu_fprintf(f, "DR%d=" TARGET_FMT_lx " ", i, env->dr[i]);
}
cpu_fprintf(f, "\nDR6=" TARGET_FMT_lx " DR7=" TARGET_FMT_lx "\n",
env->dr[6], env->dr[7]);
}
if (flags & CPU_DUMP_CCOP) {
if ((unsigned)env->cc_op < CC_OP_NB)
snprintf(cc_op_name, sizeof(cc_op_name), "%s", cc_op_str[env->cc_op]);
else
snprintf(cc_op_name, sizeof(cc_op_name), "[%d]", env->cc_op);
#ifdef TARGET_X86_64
if (env->hflags & HF_CS64_MASK) {
cpu_fprintf(f, "CCS=%016" PRIx64 " CCD=%016" PRIx64 " CCO=%-8s\n",
env->cc_src, env->cc_dst,
cc_op_name);
} else
#endif
{
cpu_fprintf(f, "CCS=%08x CCD=%08x CCO=%-8s\n",
(uint32_t)env->cc_src, (uint32_t)env->cc_dst,
cc_op_name);
}
}
cpu_fprintf(f, "EFER=%016" PRIx64 "\n", env->efer);
if (flags & CPU_DUMP_FPU) {
int fptag;
fptag = 0;
for(i = 0; i < 8; i++) {
fptag |= ((!env->fptags[i]) << i);
}
cpu_fprintf(f, "FCW=%04x FSW=%04x [ST=%d] FTW=%02x MXCSR=%08x\n",
env->fpuc,
(env->fpus & ~0x3800) | (env->fpstt & 0x7) << 11,
env->fpstt,
fptag,
env->mxcsr);
for(i=0;i<8;i++) {
CPU_LDoubleU u;
u.d = env->fpregs[i].d;
cpu_fprintf(f, "FPR%d=%016" PRIx64 " %04x",
i, u.l.lower, u.l.upper);
if ((i & 1) == 1)
cpu_fprintf(f, "\n");
else
cpu_fprintf(f, " ");
}
if (env->hflags & HF_CS64_MASK)
nb = 16;
else
nb = 8;
for(i=0;i<nb;i++) {
cpu_fprintf(f, "XMM%02d=%08x%08x%08x%08x",
i,
env->xmm_regs[i].ZMM_L(3),
env->xmm_regs[i].ZMM_L(2),
env->xmm_regs[i].ZMM_L(1),
env->xmm_regs[i].ZMM_L(0));
if ((i & 1) == 1)
cpu_fprintf(f, "\n");
else
cpu_fprintf(f, " ");
}
}
if (flags & CPU_DUMP_CODE) {
target_ulong base = env->segs[R_CS].base + env->eip;
target_ulong offs = MIN(env->eip, DUMP_CODE_BYTES_BACKWARD);
uint8_t code;
char codestr[3];
cpu_fprintf(f, "Code=");
for (i = 0; i < DUMP_CODE_BYTES_TOTAL; i++) {
if (cpu_memory_rw_debug(cs, base - offs + i, &code, 1, 0) == 0) {
snprintf(codestr, sizeof(codestr), "%02x", code);
} else {
snprintf(codestr, sizeof(codestr), "??");
}
cpu_fprintf(f, "%s%s%s%s", i > 0 ? " " : "",
i == offs ? "<" : "", codestr, i == offs ? ">" : "");
}
cpu_fprintf(f, "\n");
}
}
/***********************************************************/
/* x86 mmu */
/* XXX: add PGE support */
void x86_cpu_set_a20(X86CPU *cpu, int a20_state)
{
CPUX86State *env = &cpu->env;
a20_state = (a20_state != 0);
if (a20_state != ((env->a20_mask >> 20) & 1)) {
CPUState *cs = CPU(cpu);
#if defined(DEBUG_MMU)
printf("A20 update: a20=%d\n", a20_state);
#endif
/* if the cpu is currently executing code, we must unlink it and
all the potentially executing TB */
cpu_interrupt(cs, CPU_INTERRUPT_EXITTB);
/* when a20 is changed, all the MMU mappings are invalid, so
we must flush everything */
tlb_flush(cs);
env->a20_mask = ~(1 << 20) | (a20_state << 20);
}
}
void cpu_x86_update_cr0(CPUX86State *env, uint32_t new_cr0)
{
X86CPU *cpu = x86_env_get_cpu(env);
int pe_state;
#if defined(DEBUG_MMU)
printf("CR0 update: CR0=0x%08x\n", new_cr0);
#endif
if ((new_cr0 & (CR0_PG_MASK | CR0_WP_MASK | CR0_PE_MASK)) !=
(env->cr[0] & (CR0_PG_MASK | CR0_WP_MASK | CR0_PE_MASK))) {
tlb_flush(CPU(cpu));
}
#ifdef TARGET_X86_64
if (!(env->cr[0] & CR0_PG_MASK) && (new_cr0 & CR0_PG_MASK) &&
(env->efer & MSR_EFER_LME)) {
/* enter in long mode */
/* XXX: generate an exception */
if (!(env->cr[4] & CR4_PAE_MASK))
return;
env->efer |= MSR_EFER_LMA;
env->hflags |= HF_LMA_MASK;
} else if ((env->cr[0] & CR0_PG_MASK) && !(new_cr0 & CR0_PG_MASK) &&
(env->efer & MSR_EFER_LMA)) {
/* exit long mode */
env->efer &= ~MSR_EFER_LMA;
env->hflags &= ~(HF_LMA_MASK | HF_CS64_MASK);
env->eip &= 0xffffffff;
}
#endif
env->cr[0] = new_cr0 | CR0_ET_MASK;
/* update PE flag in hidden flags */
pe_state = (env->cr[0] & CR0_PE_MASK);
env->hflags = (env->hflags & ~HF_PE_MASK) | (pe_state << HF_PE_SHIFT);
/* ensure that ADDSEG is always set in real mode */
env->hflags |= ((pe_state ^ 1) << HF_ADDSEG_SHIFT);
/* update FPU flags */
env->hflags = (env->hflags & ~(HF_MP_MASK | HF_EM_MASK | HF_TS_MASK)) |
((new_cr0 << (HF_MP_SHIFT - 1)) & (HF_MP_MASK | HF_EM_MASK | HF_TS_MASK));
}
/* XXX: in legacy PAE mode, generate a GPF if reserved bits are set in
the PDPT */
void cpu_x86_update_cr3(CPUX86State *env, target_ulong new_cr3)
{
X86CPU *cpu = x86_env_get_cpu(env);
env->cr[3] = new_cr3;
if (env->cr[0] & CR0_PG_MASK) {
#if defined(DEBUG_MMU)
printf("CR3 update: CR3=" TARGET_FMT_lx "\n", new_cr3);
#endif
tlb_flush(CPU(cpu));
}
}
void cpu_x86_update_cr4(CPUX86State *env, uint32_t new_cr4)
{
X86CPU *cpu = x86_env_get_cpu(env);
uint32_t hflags;
#if defined(DEBUG_MMU)
printf("CR4 update: %08x -> %08x\n", (uint32_t)env->cr[4], new_cr4);
#endif
if ((new_cr4 ^ env->cr[4]) &
(CR4_PGE_MASK | CR4_PAE_MASK | CR4_PSE_MASK |
CR4_SMEP_MASK | CR4_SMAP_MASK | CR4_LA57_MASK)) {
tlb_flush(CPU(cpu));
}
/* Clear bits we're going to recompute. */
hflags = env->hflags & ~(HF_OSFXSR_MASK | HF_SMAP_MASK);
/* SSE handling */
if (!(env->features[FEAT_1_EDX] & CPUID_SSE)) {
new_cr4 &= ~CR4_OSFXSR_MASK;
}
if (new_cr4 & CR4_OSFXSR_MASK) {
hflags |= HF_OSFXSR_MASK;
}
if (!(env->features[FEAT_7_0_EBX] & CPUID_7_0_EBX_SMAP)) {
new_cr4 &= ~CR4_SMAP_MASK;
}
if (new_cr4 & CR4_SMAP_MASK) {
hflags |= HF_SMAP_MASK;
}
if (!(env->features[FEAT_7_0_ECX] & CPUID_7_0_ECX_PKU)) {
new_cr4 &= ~CR4_PKE_MASK;
}
env->cr[4] = new_cr4;
env->hflags = hflags;
cpu_sync_bndcs_hflags(env);
}
#if defined(CONFIG_USER_ONLY)
int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size,
int is_write, int mmu_idx)
{
X86CPU *cpu = X86_CPU(cs->uc, cs);
CPUX86State *env = &cpu->env;
/* user mode only emulation */
is_write &= 1;
env->cr[2] = addr;
env->error_code = (is_write << PG_ERROR_W_BIT);
env->error_code |= PG_ERROR_U_MASK;
cs->exception_index = EXCP0E_PAGE;
env->exception_is_int = 0;
env->exception_next_eip = -1;
return 1;
}
#else
/* return value:
* -1 = cannot handle fault
* 0 = nothing more to do
* 1 = generate PF fault
*/
int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size,
int is_write1, int mmu_idx)
{
X86CPU *cpu = X86_CPU(cs->uc, cs);
CPUX86State *env = &cpu->env;
uint64_t ptep, pte;
int32_t a20_mask;
target_ulong pde_addr, pte_addr;
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;
is_user = mmu_idx == MMU_USER_IDX;
#if defined(DEBUG_MMU)
printf("MMU fault: addr=%" VADDR_PRIx " w=%d u=%d eip=" TARGET_FMT_lx "\n",
addr, is_write1, is_user, env->eip);
#endif
is_write = is_write1 & 1;
a20_mask = x86_get_a20_mask(env);
if (!(env->cr[0] & CR0_PG_MASK)) {
pte = addr;
#ifdef TARGET_X86_64
if (!(env->hflags & HF_LMA_MASK)) {
/* Without long mode we can only address 32bits in real mode */
pte = (uint32_t)pte;
}
#endif
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;
#ifdef TARGET_X86_64
if (env->hflags & HF_LMA_MASK) {
bool la57 = env->cr[4] & CR4_LA57_MASK;
uint64_t pml5e_addr, pml5e;
uint64_t pml4e_addr, pml4e;
int32_t sext;
/* test virtual address sign extension */
sext = la57 ? (int64_t)addr >> 56 : (int64_t)addr >> 47;
if (sext != 0 && sext != -1) {
env->error_code = 0;
cs->exception_index = EXCP0D_GPF;
return 1;
}
if (la57) {
pml5e_addr = ((env->cr[3] & ~0xfff) +
(((addr >> 48) & 0x1ff) << 3)) & a20_mask;
pml5e = x86_ldq_phys(cs, pml5e_addr);
if (!(pml5e & PG_PRESENT_MASK)) {
goto do_fault;
}
if (pml5e & (rsvd_mask | PG_PSE_MASK)) {
goto do_fault_rsvd;
}
if (!(pml5e & PG_ACCESSED_MASK)) {
pml5e |= PG_ACCESSED_MASK;
x86_stl_phys_notdirty(cs, pml5e_addr, pml5e);
}
ptep = pml5e ^ PG_NX_MASK;
} else {
pml5e = env->cr[3];
ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
}
pml4e_addr = ((pml5e & PG_ADDRESS_MASK) +
(((addr >> 39) & 0x1ff) << 3)) & a20_mask;
pml4e = x86_ldq_phys(cs, pml4e_addr);
if (!(pml4e & PG_PRESENT_MASK)) {
goto do_fault;
}
if (pml4e & (rsvd_mask | PG_PSE_MASK)) {
goto do_fault_rsvd;
}
if (!(pml4e & PG_ACCESSED_MASK)) {
pml4e |= PG_ACCESSED_MASK;
x86_stl_phys_notdirty(cs, pml4e_addr, pml4e);
}
ptep &= pml4e ^ PG_NX_MASK;
pdpe_addr = ((pml4e & PG_ADDRESS_MASK) + (((addr >> 30) & 0x1ff) << 3)) &
a20_mask;
pdpe = x86_ldq_phys(cs, pdpe_addr);
if (!(pdpe & PG_PRESENT_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;
x86_stl_phys_notdirty(cs, 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
{
/* XXX: load them when cr3 is loaded ? */
pdpe_addr = ((env->cr[3] & ~0x1f) + ((addr >> 27) & 0x18)) &
env->a20_mask;
pdpe = x86_ldq_phys(cs, pdpe_addr);
if (!(pdpe & PG_PRESENT_MASK)) {
goto do_fault;
}
rsvd_mask |= PG_HI_USER_MASK;
if (pdpe & (rsvd_mask | PG_NX_MASK)) {
goto do_fault_rsvd;
}
ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
}
pde_addr = ((pdpe & PG_ADDRESS_MASK) + (((addr >> 21) & 0x1ff) << 3)) &
a20_mask;
pde = x86_ldq_phys(cs, pde_addr);
if (!(pde & PG_PRESENT_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;
pte_addr = pde_addr;
pte = pde;
goto do_check_protect;
}
/* 4 KB page */
if (!(pde & PG_ACCESSED_MASK)) {
pde |= PG_ACCESSED_MASK;
x86_stl_phys_notdirty(cs, pde_addr, pde);
}
pte_addr = ((pde & PG_ADDRESS_MASK) + (((addr >> 12) & 0x1ff) << 3)) &
a20_mask;
pte = x86_ldq_phys(cs, 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;
/* page directory entry */
pde_addr = ((env->cr[3] & ~0xfff) + ((addr >> 20) & 0xffc)) &
a20_mask;
pde = x86_ldl_phys(cs, pde_addr);
if (!(pde & PG_PRESENT_MASK)) {
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;
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 & 0x1fe000LL) << (32 - 13));
rsvd_mask = 0x200000;
goto do_check_protect_pse36;
}
if (!(pde & PG_ACCESSED_MASK)) {
pde |= PG_ACCESSED_MASK;
x86_stl_phys_notdirty(cs, pde_addr, pde);
}
/* page directory entry */
pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) &
env->a20_mask;
pte = x86_ldl_phys(cs, 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;
}
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;
/* can the page can be put in the TLB? prot will tell us */
if (is_user && !(ptep & PG_USER_MASK)) {
goto do_fault_protect;
}
prot = 0;
if (mmu_idx != MMU_KSMAP_IDX || !(ptep & PG_USER_MASK)) {
prot |= PAGE_READ;
if ((ptep & PG_RW_MASK) || (!is_user && !(env->cr[0] & CR0_WP_MASK))) {
prot |= PAGE_WRITE;
}
}
if (!(ptep & PG_NX_MASK) &&
(mmu_idx == MMU_USER_IDX ||
!((env->cr[4] & CR4_SMEP_MASK) && (ptep & PG_USER_MASK)))) {
prot |= PAGE_EXEC;
}
if ((env->cr[4] & CR4_PKE_MASK) && (env->hflags & HF_LMA_MASK) &&
(ptep & PG_USER_MASK) && env->pkru) {
uint32_t pk = (pte & PG_PKRU_MASK) >> PG_PKRU_BIT;
uint32_t pkru_ad = (env->pkru >> pk * 2) & 1;
uint32_t pkru_wd = (env->pkru >> pk * 2) & 2;
uint32_t pkru_prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
if (pkru_ad) {
pkru_prot &= ~(PAGE_READ | PAGE_WRITE);
} else if (pkru_wd && (is_user || env->cr[0] & CR0_WP_MASK)) {
pkru_prot &= ~PAGE_WRITE;
}
prot &= pkru_prot;
if ((pkru_prot & (1 << is_write1)) == 0) {
assert(is_write1 != 2);
error_code |= PG_ERROR_PK_MASK;
goto do_fault_protect;
}
}
if ((prot & (1 << is_write1)) == 0) {
goto do_fault_protect;
}
/* yes, it can! */
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;
}
x86_stl_phys_notdirty(cs, pte_addr, pte);
}
if (!(pte & PG_DIRTY_MASK)) {
/* only set write access if already dirty... otherwise wait
for dirty access */
assert(!is_write);
prot &= ~PAGE_WRITE;
}
do_mapping:
#if 0
pte = pte & 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 */
vaddr = addr & TARGET_PAGE_MASK;
page_offset = vaddr & (page_size - 1);
paddr = pte + page_offset;
#endif
// Unicorn: indentity map guest virtual address to host virtual address
vaddr = addr & TARGET_PAGE_MASK;
paddr = vaddr;
//printf(">>> map address %"PRIx64" to %"PRIx64"\n", vaddr, paddr);
assert(prot & (1 << is_write1));
tlb_set_page_with_attrs(cs, vaddr, paddr, cpu_get_mem_attrs(env),
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;
do_fault:
error_code |= (is_write << PG_ERROR_W_BIT);
if (is_user)
error_code |= PG_ERROR_U_MASK;
if (is_write1 == 2 &&
(((env->efer & MSR_EFER_NXE) &&
(env->cr[4] & CR4_PAE_MASK)) ||
(env->cr[4] & CR4_SMEP_MASK)))
error_code |= PG_ERROR_I_D_MASK;
if (env->intercept_exceptions & (1 << EXCP0E_PAGE)) {
/* cr2 is not modified in case of exceptions */
x86_stq_phys(cs,
env->vm_vmcb + offsetof(struct vmcb, control.exit_info_2),
addr);
} else {
env->cr[2] = addr;
}
env->error_code = error_code;
cs->exception_index = EXCP0E_PAGE;
return 1;
}
hwaddr x86_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
{
X86CPU *cpu = X86_CPU(cs->uc, cs);
CPUX86State *env = &cpu->env;
target_ulong pde_addr, pte_addr;
uint64_t pte;
int32_t a20_mask;
uint32_t page_offset;
int page_size;
a20_mask = x86_get_a20_mask(env);
if (!(env->cr[0] & CR0_PG_MASK)) {
pte = addr & a20_mask;
page_size = 4096;
} else if (env->cr[4] & CR4_PAE_MASK) {
target_ulong pdpe_addr;
uint64_t pde, pdpe;
#ifdef TARGET_X86_64
if (env->hflags & HF_LMA_MASK) {
bool la57 = env->cr[4] & CR4_LA57_MASK;
uint64_t pml5e_addr, pml5e;
uint64_t pml4e_addr, pml4e;
int32_t sext;
/* test virtual address sign extension */
sext = la57 ? (int64_t)addr >> 56 : (int64_t)addr >> 47;
if (sext != 0 && sext != -1) {
return -1;
}
if (la57) {
pml5e_addr = ((env->cr[3] & ~0xfff) +
(((addr >> 48) & 0x1ff) << 3)) & a20_mask;
pml5e = x86_ldq_phys(cs, pml5e_addr);
if (!(pml5e & PG_PRESENT_MASK)) {
return -1;
}
} else {
pml5e = env->cr[3];
}
pml4e_addr = ((pml5e & PG_ADDRESS_MASK) +
(((addr >> 39) & 0x1ff) << 3)) & a20_mask;
pml4e = x86_ldq_phys(cs, pml4e_addr);
if (!(pml4e & PG_PRESENT_MASK)) {
return -1;
}
pdpe_addr = ((pml4e & PG_ADDRESS_MASK) +
(((addr >> 30) & 0x1ff) << 3)) & a20_mask;
pdpe = x86_ldq_phys(cs, pdpe_addr);
if (!(pdpe & PG_PRESENT_MASK)) {
return -1;
}
if (pdpe & PG_PSE_MASK) {
page_size = 1024 * 1024 * 1024;
pte = pdpe;
goto out;
}
} else
#endif
{
pdpe_addr = ((env->cr[3] & ~0x1f) + ((addr >> 27) & 0x18)) &
a20_mask;
pdpe = x86_ldq_phys(cs, pdpe_addr);
if (!(pdpe & PG_PRESENT_MASK))
return -1;
}
pde_addr = ((pdpe & PG_ADDRESS_MASK) +
(((addr >> 21) & 0x1ff) << 3)) & a20_mask;
pde = x86_ldq_phys(cs, pde_addr);
if (!(pde & PG_PRESENT_MASK)) {
return -1;
}
if (pde & PG_PSE_MASK) {
/* 2 MB page */
page_size = 2048 * 1024;
pte = pde;
} else {
/* 4 KB page */
pte_addr = ((pde & PG_ADDRESS_MASK) +
(((addr >> 12) & 0x1ff) << 3)) & a20_mask;
page_size = 4096;
pte = x86_ldq_phys(cs, pte_addr);
}
if (!(pte & PG_PRESENT_MASK)) {
return -1;
}
} else {
uint32_t pde;
/* page directory entry */
pde_addr = ((env->cr[3] & ~0xfff) + ((addr >> 20) & 0xffc)) & a20_mask;
pde = x86_ldl_phys(cs, pde_addr);
if (!(pde & PG_PRESENT_MASK))
return -1;
if ((pde & PG_PSE_MASK) && (env->cr[4] & CR4_PSE_MASK)) {
pte = pde | ((pde & 0x1fe000LL) << (32 - 13));
page_size = 4096 * 1024;
} else {
/* page directory entry */
pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) & a20_mask;
pte = x86_ldl_phys(cs, pte_addr);
if (!(pte & PG_PRESENT_MASK)) {
return -1;
}
page_size = 4096;
}
pte = pte & a20_mask;
}
#ifdef TARGET_X86_64
out:
#endif
pte &= PG_ADDRESS_MASK & ~(page_size - 1);
page_offset = (addr & TARGET_PAGE_MASK) & (page_size - 1);
return pte | page_offset;
}
typedef struct MCEInjectionParams {
int bank;
uint64_t status;
uint64_t mcg_status;
uint64_t addr;
uint64_t misc;
int flags;
} MCEInjectionParams;
void cpu_report_tpr_access(CPUX86State *env, TPRAccess access)
{
X86CPU *cpu = x86_env_get_cpu(env);
CPUState *cs = CPU(cpu);
if (tcg_enabled(env->uc)) {
cpu_restore_state(cs, cs->mem_io_pc, false);
apic_handle_tpr_access_report(cpu->apic_state, env->eip, access);
}
}
#endif /* !CONFIG_USER_ONLY */
int cpu_x86_get_descr_debug(CPUX86State *env, unsigned int selector,
target_ulong *base, unsigned int *limit,
unsigned int *flags)
{
X86CPU *cpu = x86_env_get_cpu(env);
CPUState *cs = CPU(cpu);
SegmentCache *dt;
target_ulong ptr;
uint32_t e1, e2;
int index;
if (selector & 0x4)
dt = &env->ldt;
else
dt = &env->gdt;
index = selector & ~7;
ptr = dt->base + index;
if ((uint32_t)(index + 7) > dt->limit
|| cpu_memory_rw_debug(cs, ptr, (uint8_t *)&e1, sizeof(e1), 0) != 0
|| cpu_memory_rw_debug(cs, ptr+4, (uint8_t *)&e2, sizeof(e2), 0) != 0)
return 0;
*base = ((e1 >> 16) | ((e2 & 0xff) << 16) | (e2 & 0xff000000));
*limit = (e1 & 0xffff) | (e2 & 0x000f0000);
if (e2 & DESC_G_MASK)
*limit = (*limit << 12) | 0xfff;
*flags = e2;
return 1;
}
#if !defined(CONFIG_USER_ONLY)
void do_cpu_init(X86CPU *cpu)
{
CPUState *cs = CPU(cpu);
CPUX86State *env = &cpu->env;
CPUX86State *save = g_new(CPUX86State, 1);
int sipi = cs->interrupt_request & CPU_INTERRUPT_SIPI;
*save = *env;
cpu_reset(cs);
cs->interrupt_request = sipi;
memcpy(&env->start_init_save, &save->start_init_save,
offsetof(CPUX86State, end_init_save) -
offsetof(CPUX86State, start_init_save));
g_free(save);
apic_init_reset(env->uc, cpu->apic_state);
}
void do_cpu_sipi(X86CPU *cpu)
{
apic_sipi(cpu->apic_state);
}
#else
void do_cpu_init(X86CPU *cpu)
{
}
void do_cpu_sipi(X86CPU *cpu)
{
}
#endif
/* Frob eflags into and out of the CPU temporary format. */
void x86_cpu_exec_enter(CPUState *cs)
{
X86CPU *cpu = X86_CPU(cs->uc, cs);
CPUX86State *env = &cpu->env;
CC_SRC = env->eflags & (CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C);
env->df = 1 - (2 * ((env->eflags >> 10) & 1));
CC_OP = CC_OP_EFLAGS;
env->eflags &= ~(DF_MASK | CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C);
}
void x86_cpu_exec_exit(CPUState *cs)
{
X86CPU *cpu = X86_CPU(cs->uc, cs);
CPUX86State *env = &cpu->env;
env->eflags = cpu_compute_eflags(env);
env->eflags0 = env->eflags;
}
#ifndef CONFIG_USER_ONLY
uint8_t x86_ldub_phys(CPUState *cs, hwaddr addr)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
return address_space_ldub(as, addr, attrs, NULL);
}
uint32_t x86_lduw_phys(CPUState *cs, hwaddr addr)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
return address_space_lduw(as, addr, attrs, NULL);
}
uint32_t x86_ldl_phys(CPUState *cs, hwaddr addr)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
return address_space_ldl(as, addr, attrs, NULL);
}
uint64_t x86_ldq_phys(CPUState *cs, hwaddr addr)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
return address_space_ldq(as, addr, attrs, NULL);
}
void x86_stb_phys(CPUState *cs, hwaddr addr, uint8_t val)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
address_space_stb(as, addr, val, attrs, NULL);
}
void x86_stl_phys_notdirty(CPUState *cs, hwaddr addr, uint32_t val)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
address_space_stl_notdirty(as, addr, val, attrs, NULL);
}
void x86_stw_phys(CPUState *cs, hwaddr addr, uint32_t val)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
address_space_stw(as, addr, val, attrs, NULL);
}
void x86_stl_phys(CPUState *cs, hwaddr addr, uint32_t val)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
address_space_stl(as, addr, val, attrs, NULL);
}
void x86_stq_phys(CPUState *cs, hwaddr addr, uint64_t val)
{
X86CPU *cpu = X86_CPU(NULL, cs);
CPUX86State *env = &cpu->env;
MemTxAttrs attrs = cpu_get_mem_attrs(env);
AddressSpace *as = cpu_addressspace(cs, attrs);
address_space_stq(as, addr, val, attrs, NULL);
}
#endif