1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
|
/*
* pv/iret.c
*
* iret hypercall handling code
*
* This program is free software; you can redistribute it and/or
* modify it under the terms and conditions of the GNU General Public
* License, version 2, as published by the Free Software Foundation.
*
* This program 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
* General Public License for more details.
*
* You should have received a copy of the GNU General Public
* License along with this program; If not, see
* <http://www.gnu.org/licenses/>.
*/
#include <xen/guest_access.h>
#include <xen/hypercall.h>
#include <xen/lib.h>
#include <xen/sched.h>
#include <asm/current.h>
static void async_exception_cleanup(struct vcpu *curr)
{
unsigned int trap;
if ( !curr->arch.async_exception_mask )
return;
if ( !(curr->arch.async_exception_mask & (curr->arch.async_exception_mask - 1)) )
trap = __scanbit(curr->arch.async_exception_mask, VCPU_TRAP_NONE);
else
for ( trap = VCPU_TRAP_NONE + 1; trap <= VCPU_TRAP_LAST; ++trap )
if ( (curr->arch.async_exception_mask ^
curr->arch.async_exception_state(trap).old_mask) == (1u << trap) )
break;
if ( unlikely(trap > VCPU_TRAP_LAST) )
{
ASSERT_UNREACHABLE();
return;
}
/* Restore previous asynchronous exception mask. */
curr->arch.async_exception_mask =
curr->arch.async_exception_state(trap).old_mask;
}
long do_iret(void)
{
struct cpu_user_regs *regs = guest_cpu_user_regs();
struct iret_context iret_saved;
struct vcpu *v = current;
if ( unlikely(copy_from_guest_pv(&iret_saved, (void __user *)regs->rsp,
sizeof(iret_saved))) )
{
gprintk(XENLOG_ERR,
"Fault while reading IRET context from guest stack\n");
goto exit_and_crash;
}
/* Returning to user mode? */
if ( (iret_saved.cs & 3) == 3 )
{
if ( unlikely(pagetable_is_null(v->arch.guest_table_user)) )
{
gprintk(XENLOG_ERR,
"Guest switching to user mode with no user page tables\n");
goto exit_and_crash;
}
toggle_guest_mode(v);
}
if ( VM_ASSIST(v->domain, architectural_iopl) )
v->arch.pv.iopl = iret_saved.rflags & X86_EFLAGS_IOPL;
regs->rip = iret_saved.rip;
regs->cs = iret_saved.cs | 3; /* force guest privilege */
regs->rflags = ((iret_saved.rflags & ~(X86_EFLAGS_IOPL|X86_EFLAGS_VM))
| X86_EFLAGS_IF);
regs->rsp = iret_saved.rsp;
regs->ss = iret_saved.ss | 3; /* force guest privilege */
if ( !(iret_saved.flags & VGCF_in_syscall) )
{
regs->entry_vector &= ~TRAP_syscall;
regs->r11 = iret_saved.r11;
regs->rcx = iret_saved.rcx;
}
/* Restore upcall mask from supplied EFLAGS.IF. */
vcpu_info(v, evtchn_upcall_mask) = !(iret_saved.rflags & X86_EFLAGS_IF);
async_exception_cleanup(v);
/* Saved %rax gets written back to regs->rax in entry.S. */
return iret_saved.rax;
exit_and_crash:
domain_crash(v->domain);
return 0;
}
#ifdef CONFIG_PV32
int compat_iret(void)
{
struct cpu_user_regs *regs = guest_cpu_user_regs();
struct vcpu *v = current;
u32 eflags;
/* Trim stack pointer to 32 bits. */
regs->rsp = (u32)regs->rsp;
/* Restore EAX (clobbered by hypercall). */
if ( unlikely(__get_guest(regs->eax, (u32 *)regs->rsp)) )
{
domain_crash(v->domain);
return 0;
}
/* Restore CS and EIP. */
if ( unlikely(__get_guest(regs->eip, (u32 *)regs->rsp + 1)) ||
unlikely(__get_guest(regs->cs, (u32 *)regs->rsp + 2)) )
{
domain_crash(v->domain);
return 0;
}
/*
* Fix up and restore EFLAGS. We fix up in a local staging area
* to avoid firing the BUG_ON(IOPL) check in arch_get_info_guest.
*/
if ( unlikely(__get_guest(eflags, (u32 *)regs->rsp + 3)) )
{
domain_crash(v->domain);
return 0;
}
if ( VM_ASSIST(v->domain, architectural_iopl) )
v->arch.pv.iopl = eflags & X86_EFLAGS_IOPL;
regs->eflags = (eflags & ~X86_EFLAGS_IOPL) | X86_EFLAGS_IF;
if ( unlikely(eflags & X86_EFLAGS_VM) )
{
/*
* Cannot return to VM86 mode: inject a GP fault instead. Note that
* the GP fault is reported on the first VM86 mode instruction, not on
* the IRET (which is why we can simply leave the stack frame as-is
* (except for perhaps having to copy it), which in turn seems better
* than teaching create_bounce_frame() to needlessly deal with vm86
* mode frames).
*/
const struct trap_info *ti;
u32 x, ksp = v->arch.pv.kernel_sp - 40;
unsigned int i;
int rc = 0;
gdprintk(XENLOG_ERR, "VM86 mode unavailable (ksp:%08X->%08X)\n",
regs->esp, ksp);
if ( ksp < regs->esp )
{
for (i = 1; i < 10; ++i)
{
rc |= __get_guest(x, (u32 *)regs->rsp + i);
rc |= __put_guest(x, (u32 *)(unsigned long)ksp + i);
}
}
else if ( ksp > regs->esp )
{
for ( i = 9; i > 0; --i )
{
rc |= __get_guest(x, (u32 *)regs->rsp + i);
rc |= __put_guest(x, (u32 *)(unsigned long)ksp + i);
}
}
if ( rc )
{
domain_crash(v->domain);
return 0;
}
regs->esp = ksp;
regs->ss = v->arch.pv.kernel_ss;
ti = &v->arch.pv.trap_ctxt[X86_EXC_GP];
if ( TI_GET_IF(ti) )
eflags &= ~X86_EFLAGS_IF;
regs->eflags &= ~(X86_EFLAGS_VM|X86_EFLAGS_RF|
X86_EFLAGS_NT|X86_EFLAGS_TF);
if ( unlikely(__put_guest(0, (u32 *)regs->rsp)) )
{
domain_crash(v->domain);
return 0;
}
regs->eip = ti->address;
regs->cs = ti->cs;
}
else if ( unlikely(ring_0(regs)) )
{
domain_crash(v->domain);
return 0;
}
else if ( ring_1(regs) )
regs->esp += 16;
/* Return to ring 2/3: restore ESP and SS. */
else if ( __get_guest(regs->ss, (u32 *)regs->rsp + 5) ||
__get_guest(regs->esp, (u32 *)regs->rsp + 4) )
{
domain_crash(v->domain);
return 0;
}
/* Restore upcall mask from supplied EFLAGS.IF. */
vcpu_info(v, evtchn_upcall_mask) = !(eflags & X86_EFLAGS_IF);
async_exception_cleanup(v);
/*
* The hypercall exit path will overwrite EAX with this return
* value.
*/
return regs->eax;
}
#endif
/*
* Local variables:
* mode: C
* c-file-style: "BSD"
* c-basic-offset: 4
* tab-width: 4
* indent-tabs-mode: nil
* End:
*/
|
