[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] [PATCH 2/5] x86emul: limit-check branch targets



All branches need to #GP when their target violates the segment limit
(in 16- and 32-bit modes) or is non-canonical (in 64-bit mode). For
near branches facilitate this via a zero-byte instruction fetch from
the target address (resulting in address translation and validation
without an actual read from memory), while far branches get dealt with
by breaking up the segment register loading into a read-and-validate
part and a write one. The latter at once allows correcting some
ordering issues in how the individual emulation steps get carried out:
Before updating machine state, all exceptions unrelated to that state
updating should have got raised (i.e. the only ones possibly resulting
in partly updated state are faulting memory writes [pushes]).

Note that while not immediately needed here, write and distinct read
emulation routines get updated to deal with zero byte accesses too, for
overall consistency.

Reported-by: ÃËËÃÂÂ <liuling-it@xxxxxx>
Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>

--- a/tools/tests/x86_emulator/x86_emulate.c
+++ b/tools/tests/x86_emulator/x86_emulate.c
@@ -8,6 +8,8 @@

 typedef bool bool_t;

+#define is_canonical_address(x) (((int64_t)(x) >> 47) == ((int64_t)(x) >> 63))
+
 #define BUG() abort()
 #define ASSERT assert

--- a/xen/arch/x86/hvm/emulate.c
+++ b/xen/arch/x86/hvm/emulate.c
@@ -745,7 +745,7 @@ static int __hvmemul_read(

     rc = hvmemul_virtual_to_linear(
         seg, offset, bytes, &reps, access_type, hvmemul_ctxt, &addr);
-    if ( rc != X86EMUL_OKAY )
+    if ( rc != X86EMUL_OKAY || !bytes )
         return rc;
     if ( ((access_type != hvm_access_insn_fetch
            ? vio->mmio_access.read_access
@@ -811,13 +811,17 @@ static int hvmemul_insn_fetch(
         container_of(ctxt, struct hvm_emulate_ctxt, ctxt);
     unsigned int insn_off = offset - hvmemul_ctxt->insn_buf_eip;

-    /* Fall back if requested bytes are not in the prefetch cache. */
-    if ( unlikely((insn_off + bytes) > hvmemul_ctxt->insn_buf_bytes) )
+    /*
+     * Fall back if requested bytes are not in the prefetch cache.
+     * But always perform the (fake) read when bytes == 0.
+     */
+    if ( !bytes ||
+         unlikely((insn_off + bytes) > hvmemul_ctxt->insn_buf_bytes) )
     {
         int rc = __hvmemul_read(seg, offset, p_data, bytes,
                                 hvm_access_insn_fetch, hvmemul_ctxt);

-        if ( rc == X86EMUL_OKAY )
+        if ( rc == X86EMUL_OKAY && bytes )
         {
             ASSERT(insn_off + bytes <= sizeof(hvmemul_ctxt->insn_buf));
             memcpy(&hvmemul_ctxt->insn_buf[insn_off], p_data, bytes);
@@ -849,7 +853,7 @@ static int hvmemul_write(

     rc = hvmemul_virtual_to_linear(
         seg, offset, bytes, &reps, hvm_access_write, hvmemul_ctxt, &addr);
-    if ( rc != X86EMUL_OKAY )
+    if ( rc != X86EMUL_OKAY || !bytes )
         return rc;

     if ( vio->mmio_access.write_access &&
--- a/xen/arch/x86/hvm/hvm.c
+++ b/xen/arch/x86/hvm/hvm.c
@@ -3685,7 +3685,7 @@ int hvm_virtual_to_linear_addr(
          * Certain of them are not done in native real mode anyway.
          */
         addr = (uint32_t)(addr + reg->base);
-        last_byte = (uint32_t)addr + bytes - 1;
+        last_byte = (uint32_t)addr + bytes - !!bytes;
         if ( last_byte < addr )
             return 0;
     }
@@ -3709,7 +3709,7 @@ int hvm_virtual_to_linear_addr(
             break;
         }

-        last_byte = (uint32_t)offset + bytes - 1;
+        last_byte = (uint32_t)offset + bytes - !!bytes;

         /* Is this a grows-down data segment? Special limit check if so. */
         if ( (reg->attr.fields.type & 0xc) == 0x4 )
@@ -3740,7 +3740,7 @@ int hvm_virtual_to_linear_addr(
         if ( (seg == x86_seg_fs) || (seg == x86_seg_gs) )
             addr += reg->base;

-        last_byte = addr + bytes - 1;
+        last_byte = addr + bytes - !!bytes;
         if ( !is_canonical_address(addr) || last_byte < addr ||
              !is_canonical_address(last_byte) )
             return 0;
--- a/xen/arch/x86/mm/shadow/common.c
+++ b/xen/arch/x86/mm/shadow/common.c
@@ -157,7 +157,7 @@ hvm_read(enum x86_segment seg,

     rc = hvm_translate_linear_addr(
         seg, offset, bytes, access_type, sh_ctxt, &addr);
-    if ( rc )
+    if ( rc || !bytes )
         return rc;

     if ( access_type == hvm_access_insn_fetch )
@@ -241,7 +241,7 @@ hvm_emulate_write(enum x86_segment seg,

     rc = hvm_translate_linear_addr(
         seg, offset, bytes, hvm_access_write, sh_ctxt, &addr);
-    if ( rc )
+    if ( rc || !bytes )
         return rc;

     return v->arch.paging.mode->shadow.x86_emulate_write(
--- a/xen/arch/x86/mm.c
+++ b/xen/arch/x86/mm.c
@@ -5126,10 +5126,11 @@ static int ptwr_emulated_read(
     unsigned int bytes,
     struct x86_emulate_ctxt *ctxt)
 {
-    unsigned int rc;
+    unsigned int rc = bytes;
     unsigned long addr = offset;

-    if ( (rc = copy_from_user(p_data, (void *)addr, bytes)) != 0 )
+    if ( !__addr_ok(addr) ||
+         (rc = __copy_from_user(p_data, (void *)addr, bytes)) )
     {
         propagate_page_fault(addr + bytes - rc, 0); /* read fault */
         return X86EMUL_EXCEPTION;
@@ -5278,7 +5279,7 @@ static int ptwr_emulated_write(
 {
     paddr_t val = 0;

-    if ( (bytes > sizeof(paddr_t)) || (bytes & (bytes -1)) )
+    if ( (bytes > sizeof(paddr_t)) || (bytes & (bytes - 1)) || !bytes )
     {
         MEM_LOG("ptwr_emulate: bad write size (addr=%lx, bytes=%u)",
                 offset, bytes);
@@ -5394,7 +5395,8 @@ int mmio_ro_emulated_write(
     struct mmio_ro_emulate_ctxt *mmio_ro_ctxt = ctxt->data;

     /* Only allow naturally-aligned stores at the original %cr2 address. */
-    if ( ((bytes | offset) & (bytes - 1)) || offset != mmio_ro_ctxt->cr2 )
+    if ( ((bytes | offset) & (bytes - 1)) || !bytes ||
+         offset != mmio_ro_ctxt->cr2 )
     {
         MEM_LOG("mmio_ro_emulate: bad access (cr2=%lx, addr=%lx, bytes=%u)",
                 mmio_ro_ctxt->cr2, offset, bytes);
@@ -5423,7 +5425,7 @@ int mmcfg_intercept_write(
      * Only allow naturally-aligned stores no wider than 4 bytes to the
      * original %cr2 address.
      */
-    if ( ((bytes | offset) & (bytes - 1)) || bytes > 4 ||
+    if ( ((bytes | offset) & (bytes - 1)) || bytes > 4 || !bytes ||
          offset != mmio_ctxt->cr2 )
     {
         MEM_LOG("mmcfg_intercept: bad write (cr2=%lx, addr=%lx, bytes=%u)",
--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -642,14 +642,26 @@ do {

 #define jmp_rel(rel)                                                    \
 do {                                                                    \
-    int _rel = (int)(rel);                                              \
-    _regs.eip += _rel;                                                  \
+    unsigned long ip = _regs.eip + (int)(rel);                          \
     if ( op_bytes == 2 )                                                \
-        _regs.eip = (uint16_t)_regs.eip;                                \
+        ip = (uint16_t)ip;                                              \
     else if ( !mode_64bit() )                                           \
-        _regs.eip = (uint32_t)_regs.eip;                                \
+        ip = (uint32_t)ip;                                              \
+    rc = ops->insn_fetch(x86_seg_cs, ip, NULL, 0, ctxt);                \
+    if ( rc ) goto done;                                                \
+    _regs.eip = ip;                                                     \
 } while (0)

+#define validate_far_branch(cs, ip)                                     \
+    generate_exception_if(in_longmode(ctxt, ops) && (cs)->attr.fields.l \
+                          ? !is_canonical_address(ip)                   \
+                          : (ip) > (cs)->limit, EXC_GP, 0)
+
+#define commit_far_branch(cs, ip) ({                                    \
+    validate_far_branch(cs, ip);                                        \
+    ops->write_segment(x86_seg_cs, cs, ctxt);                           \
+})
+
 struct fpu_insn_ctxt {
     uint8_t insn_bytes;
     uint8_t exn_raised;
@@ -1099,29 +1111,30 @@ static int
 realmode_load_seg(
     enum x86_segment seg,
     uint16_t sel,
+    struct segment_register *sreg,
     struct x86_emulate_ctxt *ctxt,
     const struct x86_emulate_ops *ops)
 {
-    struct segment_register reg;
-    int rc;
+    int rc = ops->read_segment(seg, sreg, ctxt);

-    if ( (rc = ops->read_segment(seg, &reg, ctxt)) != 0 )
-        return rc;
-
-    reg.sel  = sel;
-    reg.base = (uint32_t)sel << 4;
+    if ( !rc )
+    {
+        sreg->sel  = sel;
+        sreg->base = (uint32_t)sel << 4;
+    }

-    return ops->write_segment(seg, &reg, ctxt);
+    return rc;
 }

 static int
 protmode_load_seg(
     enum x86_segment seg,
     uint16_t sel, bool_t is_ret,
+    struct segment_register *sreg,
     struct x86_emulate_ctxt *ctxt,
     const struct x86_emulate_ops *ops)
 {
-    struct segment_register desctab, ss, segr;
+    struct segment_register desctab, ss;
     struct { uint32_t a, b; } desc;
     uint8_t dpl, rpl, cpl;
     uint32_t new_desc_b, a_flag = 0x100;
@@ -1132,8 +1145,8 @@ protmode_load_seg(
     {
         if ( (seg == x86_seg_cs) || (seg == x86_seg_ss) )
             goto raise_exn;
-        memset(&segr, 0, sizeof(segr));
-        return ops->write_segment(seg, &segr, ctxt);
+        memset(sreg, 0, sizeof(*sreg));
+        return X86EMUL_OKAY;
     }

     /* System segment descriptors must reside in the GDT. */
@@ -1242,16 +1255,16 @@ protmode_load_seg(
     desc.b |= a_flag;

  skip_accessed_flag:
-    segr.base = (((desc.b <<  0) & 0xff000000u) |
-                 ((desc.b << 16) & 0x00ff0000u) |
-                 ((desc.a >> 16) & 0x0000ffffu));
-    segr.attr.bytes = (((desc.b >>  8) & 0x00ffu) |
-                       ((desc.b >> 12) & 0x0f00u));
-    segr.limit = (desc.b & 0x000f0000u) | (desc.a & 0x0000ffffu);
-    if ( segr.attr.fields.g )
-        segr.limit = (segr.limit << 12) | 0xfffu;
-    segr.sel = sel;
-    return ops->write_segment(seg, &segr, ctxt);
+    sreg->base = (((desc.b <<  0) & 0xff000000u) |
+                  ((desc.b << 16) & 0x00ff0000u) |
+                  ((desc.a >> 16) & 0x0000ffffu));
+    sreg->attr.bytes = (((desc.b >>  8) & 0x00ffu) |
+                        ((desc.b >> 12) & 0x0f00u));
+    sreg->limit = (desc.b & 0x000f0000u) | (desc.a & 0x0000ffffu);
+    if ( sreg->attr.fields.g )
+        sreg->limit = (sreg->limit << 12) | 0xfffu;
+    sreg->sel = sel;
+    return X86EMUL_OKAY;

  raise_exn:
     if ( ops->inject_hw_exception == NULL )
@@ -1265,17 +1278,29 @@ static int
 load_seg(
     enum x86_segment seg,
     uint16_t sel, bool_t is_ret,
+    struct segment_register *sreg,
     struct x86_emulate_ctxt *ctxt,
     const struct x86_emulate_ops *ops)
 {
+    struct segment_register reg;
+    int rc;
+
     if ( (ops->read_segment == NULL) ||
          (ops->write_segment == NULL) )
         return X86EMUL_UNHANDLEABLE;

+    if ( !sreg )
+        sreg = &reg;
+
     if ( in_protmode(ctxt, ops) )
-        return protmode_load_seg(seg, sel, is_ret, ctxt, ops);
+        rc = protmode_load_seg(seg, sel, is_ret, sreg, ctxt, ops);
+    else
+        rc = realmode_load_seg(seg, sel, sreg, ctxt, ops);
+
+    if ( !rc && sreg == &reg )
+        rc = ops->write_segment(seg, sreg, ctxt);

-    return realmode_load_seg(seg, sel, ctxt, ops);
+    return rc;
 }

 void *
@@ -1970,6 +1995,8 @@ x86_emulate(

     switch ( b )
     {
+        struct segment_register cs;
+
     case 0x00 ... 0x05: add: /* add */
         emulate_2op_SrcV("add", src, dst, _regs.eflags);
         break;
@@ -2031,7 +2058,7 @@ x86_emulate(
         if ( (rc = read_ulong(x86_seg_ss, sp_post_inc(op_bytes),
                               &dst.val, op_bytes, ctxt, ops)) != 0 )
             goto done;
-        if ( (rc = load_seg(src.val, dst.val, 0, ctxt, ops)) != 0 )
+        if ( (rc = load_seg(src.val, dst.val, 0, NULL, ctxt, ops)) != 0 )
             return rc;
         break;

@@ -2364,7 +2391,7 @@ x86_emulate(
         enum x86_segment seg = decode_segment(modrm_reg);
         generate_exception_if(seg == decode_segment_failed, EXC_UD, -1);
         generate_exception_if(seg == x86_seg_cs, EXC_UD, -1);
-        if ( (rc = load_seg(seg, src.val, 0, ctxt, ops)) != 0 )
+        if ( (rc = load_seg(seg, src.val, 0, NULL, ctxt, ops)) != 0 )
             goto done;
         if ( seg == x86_seg_ss )
             ctxt->retire.flags.mov_ss = 1;
@@ -2439,14 +2466,15 @@ x86_emulate(
         sel = insn_fetch_type(uint16_t);

         if ( (rc = ops->read_segment(x86_seg_cs, &reg, ctxt)) ||
-             (rc = ops->write(x86_seg_ss, sp_pre_dec(op_bytes),
+             (rc = load_seg(x86_seg_cs, sel, 0, &cs, ctxt, ops)) ||
+             (validate_far_branch(&cs, eip),
+              rc = ops->write(x86_seg_ss, sp_pre_dec(op_bytes),
                               &reg.sel, op_bytes, ctxt)) ||
              (rc = ops->write(x86_seg_ss, sp_pre_dec(op_bytes),
-                              &_regs.eip, op_bytes, ctxt)) )
+                              &_regs.eip, op_bytes, ctxt)) ||
+             (rc = ops->write_segment(x86_seg_cs, &cs, ctxt)) )
             goto done;

-        if ( (rc = load_seg(x86_seg_cs, sel, 0, ctxt, ops)) != 0 )
-            goto done;
         _regs.eip = eip;
         break;
     }
@@ -2664,7 +2692,8 @@ x86_emulate(
         int offset = (b == 0xc2) ? insn_fetch_type(uint16_t) : 0;
         op_bytes = ((op_bytes == 4) && mode_64bit()) ? 8 : op_bytes;
         if ( (rc = read_ulong(x86_seg_ss, sp_post_inc(op_bytes + offset),
-                              &dst.val, op_bytes, ctxt, ops)) != 0 )
+                              &dst.val, op_bytes, ctxt, ops)) != 0 ||
+             (rc = ops->insn_fetch(x86_seg_cs, dst.val, NULL, 0, ctxt)) )
             goto done;
         _regs.eip = dst.val;
         break;
@@ -2679,7 +2708,7 @@ x86_emulate(
         if ( (rc = read_ulong(src.mem.seg, src.mem.off + src.bytes,
                               &sel, 2, ctxt, ops)) != 0 )
             goto done;
-        if ( (rc = load_seg(dst.val, sel, 0, ctxt, ops)) != 0 )
+        if ( (rc = load_seg(dst.val, sel, 0, NULL, ctxt, ops)) != 0 )
             goto done;
         dst.val = src.val;
         break;
@@ -2753,7 +2782,8 @@ x86_emulate(
                               &dst.val, op_bytes, ctxt, ops)) ||
              (rc = read_ulong(x86_seg_ss, sp_post_inc(op_bytes + offset),
                               &src.val, op_bytes, ctxt, ops)) ||
-             (rc = load_seg(x86_seg_cs, src.val, 1, ctxt, ops)) )
+             (rc = load_seg(x86_seg_cs, src.val, 1, &cs, ctxt, ops)) ||
+             (rc = commit_far_branch(&cs, dst.val)) )
             goto done;
         _regs.eip = dst.val;
         break;
@@ -2782,7 +2812,7 @@ x86_emulate(
         goto swint;

     case 0xcf: /* iret */ {
-        unsigned long cs, eip, eflags;
+        unsigned long sel, eip, eflags;
         uint32_t mask = EFLG_VIP | EFLG_VIF | EFLG_VM;
         if ( !mode_ring0() )
             mask |= EFLG_IOPL;
@@ -2792,7 +2822,7 @@ x86_emulate(
         if ( (rc = read_ulong(x86_seg_ss, sp_post_inc(op_bytes),
                               &eip, op_bytes, ctxt, ops)) ||
              (rc = read_ulong(x86_seg_ss, sp_post_inc(op_bytes),
-                              &cs, op_bytes, ctxt, ops)) ||
+                              &sel, op_bytes, ctxt, ops)) ||
              (rc = read_ulong(x86_seg_ss, sp_post_inc(op_bytes),
                               &eflags, op_bytes, ctxt, ops)) )
             goto done;
@@ -2802,7 +2832,8 @@ x86_emulate(
         _regs.eflags &= mask;
         _regs.eflags |= (uint32_t)(eflags & ~mask) | 0x02;
         _regs.eip = eip;
-        if ( (rc = load_seg(x86_seg_cs, cs, 1, ctxt, ops)) != 0 )
+        if ( (rc = load_seg(x86_seg_cs, sel, 1, &cs, ctxt, ops)) ||
+             (rc = commit_far_branch(&cs, eip)) )
             goto done;
         break;
     }
@@ -3432,7 +3463,8 @@ x86_emulate(
         generate_exception_if(mode_64bit(), EXC_UD, -1);
         eip = insn_fetch_bytes(op_bytes);
         sel = insn_fetch_type(uint16_t);
-        if ( (rc = load_seg(x86_seg_cs, sel, 0, ctxt, ops)) != 0 )
+        if ( (rc = load_seg(x86_seg_cs, sel, 0, &cs, ctxt, ops)) ||
+             (rc = commit_far_branch(&cs, eip)) )
             goto done;
         _regs.eip = eip;
         break;
@@ -3702,10 +3734,14 @@ x86_emulate(
             break;
         case 2: /* call (near) */
             dst.val = _regs.eip;
+            if ( (rc = ops->insn_fetch(x86_seg_cs, src.val, NULL, 0, ctxt)) )
+                goto done;
             _regs.eip = src.val;
             src.val = dst.val;
             goto push;
         case 4: /* jmp (near) */
+            if ( (rc = ops->insn_fetch(x86_seg_cs, src.val, NULL, 0, ctxt)) )
+                goto done;
             _regs.eip = src.val;
             dst.type = OP_NONE;
             break;
@@ -3724,14 +3760,17 @@ x86_emulate(
                 struct segment_register reg;
                 fail_if(ops->read_segment == NULL);
                 if ( (rc = ops->read_segment(x86_seg_cs, &reg, ctxt)) ||
-                     (rc = ops->write(x86_seg_ss, sp_pre_dec(op_bytes),
+                     (rc = load_seg(x86_seg_cs, sel, 0, &cs, ctxt, ops)) ||
+                     (validate_far_branch(&cs, src.val),
+                      rc = ops->write(x86_seg_ss, sp_pre_dec(op_bytes),
                                       &reg.sel, op_bytes, ctxt)) ||
                      (rc = ops->write(x86_seg_ss, sp_pre_dec(op_bytes),
-                                      &_regs.eip, op_bytes, ctxt)) )
+                                      &_regs.eip, op_bytes, ctxt)) ||
+                     (rc = ops->write_segment(x86_seg_cs, &cs, ctxt)) )
                     goto done;
             }
-
-            if ( (rc = load_seg(x86_seg_cs, sel, 0, ctxt, ops)) != 0 )
+            else if ( (rc = load_seg(x86_seg_cs, sel, 0, &cs, ctxt, ops)) ||
+                      (rc = commit_far_branch(&cs, src.val)) )
                 goto done;
             _regs.eip = src.val;

@@ -3816,7 +3855,7 @@ x86_emulate(
         generate_exception_if(!in_protmode(ctxt, ops), EXC_UD, -1);
         generate_exception_if(!mode_ring0(), EXC_GP, 0);
         if ( (rc = load_seg((modrm_reg & 1) ? x86_seg_tr : x86_seg_ldtr,
-                            src.val, 0, ctxt, ops)) != 0 )
+                            src.val, 0, NULL, ctxt, ops)) != 0 )
             goto done;
         break;

@@ -4269,6 +4308,9 @@ x86_emulate(
             goto done;

         generate_exception_if(!(msr_content & 0xfffc), EXC_GP, 0);
+        generate_exception_if(user64 && (!is_canonical_address(_regs.edx) ||
+                                         !is_canonical_address(_regs.ecx)),
+                              EXC_GP, 0);

         cs.sel = (msr_content | 3) + /* SELECTOR_RPL_MASK */
                  (user64 ? 32 : 16);


Attachment: x86emul-branch-limit-checks.patch
Description: Text document

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.