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
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
|
From 7ad38a39f08aadc1578bdb46ccabaad79ed0faee Mon Sep 17 00:00:00 2001
From: Henry Wang <Henry.Wang@arm.com>
Date: Tue, 11 Oct 2022 15:10:34 +0200
Subject: [PATCH 054/126] xen/arm: Allocate and free P2M pages from the P2M
pool
This commit sets/tearsdown of p2m pages pool for non-privileged Arm
guests by calling `p2m_set_allocation` and `p2m_teardown_allocation`.
- For dom0, P2M pages should come from heap directly instead of p2m
pool, so that the kernel may take advantage of the extended regions.
- For xl guests, the setting of the p2m pool is called in
`XEN_DOMCTL_shadow_op` and the p2m pool is destroyed in
`domain_relinquish_resources`. Note that domctl->u.shadow_op.mb is
updated with the new size when setting the p2m pool.
- For dom0less domUs, the setting of the p2m pool is called before
allocating memory during domain creation. Users can specify the p2m
pool size by `xen,domain-p2m-mem-mb` dts property.
To actually allocate/free pages from the p2m pool, this commit adds
two helper functions namely `p2m_alloc_page` and `p2m_free_page` to
`struct p2m_domain`. By replacing the `alloc_domheap_page` and
`free_domheap_page` with these two helper functions, p2m pages can
be added/removed from the list of p2m pool rather than from the heap.
Since page from `p2m_alloc_page` is cleaned, take the opportunity
to remove the redundant `clean_page` in `p2m_create_table`.
This is part of CVE-2022-33747 / XSA-409.
Signed-off-by: Henry Wang <Henry.Wang@arm.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
master commit: cbea5a1149ca7fd4b7cdbfa3ec2e4f109b601ff7
master date: 2022-10-11 14:28:44 +0200
---
docs/misc/arm/device-tree/booting.txt | 8 ++++
xen/arch/arm/domain.c | 6 +++
xen/arch/arm/domain_build.c | 29 ++++++++++++++
xen/arch/arm/domctl.c | 23 ++++++++++-
xen/arch/arm/p2m.c | 57 +++++++++++++++++++++++++--
5 files changed, 118 insertions(+), 5 deletions(-)
diff --git a/docs/misc/arm/device-tree/booting.txt b/docs/misc/arm/device-tree/booting.txt
index 5243bc7fd344..470c9491a781 100644
--- a/docs/misc/arm/device-tree/booting.txt
+++ b/docs/misc/arm/device-tree/booting.txt
@@ -164,6 +164,14 @@ with the following properties:
Both #address-cells and #size-cells need to be specified because
both sub-nodes (described shortly) have reg properties.
+- xen,domain-p2m-mem-mb
+
+ Optional. A 32-bit integer specifying the amount of megabytes of RAM
+ used for the domain P2M pool. This is in-sync with the shadow_memory
+ option in xl.cfg. Leaving this field empty in device tree will lead to
+ the default size of domain P2M pool, i.e. 1MB per guest vCPU plus 4KB
+ per MB of guest RAM plus 512KB for guest extended regions.
+
Under the "xen,domain" compatible node, one or more sub-nodes are present
for the DomU kernel and ramdisk.
diff --git a/xen/arch/arm/domain.c b/xen/arch/arm/domain.c
index 223ec9694df1..a5ffd952ecd0 100644
--- a/xen/arch/arm/domain.c
+++ b/xen/arch/arm/domain.c
@@ -985,6 +985,7 @@ enum {
PROG_page,
PROG_mapping,
PROG_p2m,
+ PROG_p2m_pool,
PROG_done,
};
@@ -1044,6 +1045,11 @@ int domain_relinquish_resources(struct domain *d)
if ( ret )
return ret;
+ PROGRESS(p2m_pool):
+ ret = p2m_teardown_allocation(d);
+ if( ret )
+ return ret;
+
PROGRESS(done):
break;
diff --git a/xen/arch/arm/domain_build.c b/xen/arch/arm/domain_build.c
index 26c13429488d..df0ec84f034c 100644
--- a/xen/arch/arm/domain_build.c
+++ b/xen/arch/arm/domain_build.c
@@ -2333,6 +2333,21 @@ static void __init find_gnttab_region(struct domain *d,
kinfo->gnttab_start, kinfo->gnttab_start + kinfo->gnttab_size);
}
+static unsigned long __init domain_p2m_pages(unsigned long maxmem_kb,
+ unsigned int smp_cpus)
+{
+ /*
+ * Keep in sync with libxl__get_required_paging_memory().
+ * 256 pages (1MB) per vcpu, plus 1 page per MiB of RAM for the P2M map,
+ * plus 128 pages to cover extended regions.
+ */
+ unsigned long memkb = 4 * (256 * smp_cpus + (maxmem_kb / 1024) + 128);
+
+ BUILD_BUG_ON(PAGE_SIZE != SZ_4K);
+
+ return DIV_ROUND_UP(memkb, 1024) << (20 - PAGE_SHIFT);
+}
+
static int __init construct_domain(struct domain *d, struct kernel_info *kinfo)
{
unsigned int i;
@@ -2424,6 +2439,8 @@ static int __init construct_domU(struct domain *d,
struct kernel_info kinfo = {};
int rc;
u64 mem;
+ u32 p2m_mem_mb;
+ unsigned long p2m_pages;
rc = dt_property_read_u64(node, "memory", &mem);
if ( !rc )
@@ -2433,6 +2450,18 @@ static int __init construct_domU(struct domain *d,
}
kinfo.unassigned_mem = (paddr_t)mem * SZ_1K;
+ rc = dt_property_read_u32(node, "xen,domain-p2m-mem-mb", &p2m_mem_mb);
+ /* If xen,domain-p2m-mem-mb is not specified, use the default value. */
+ p2m_pages = rc ?
+ p2m_mem_mb << (20 - PAGE_SHIFT) :
+ domain_p2m_pages(mem, d->max_vcpus);
+
+ spin_lock(&d->arch.paging.lock);
+ rc = p2m_set_allocation(d, p2m_pages, NULL);
+ spin_unlock(&d->arch.paging.lock);
+ if ( rc != 0 )
+ return rc;
+
printk("*** LOADING DOMU cpus=%u memory=%"PRIx64"KB ***\n", d->max_vcpus, mem);
kinfo.vpl011 = dt_property_read_bool(node, "vpl011");
diff --git a/xen/arch/arm/domctl.c b/xen/arch/arm/domctl.c
index a049bc7f3e52..4ab5ed4ab24d 100644
--- a/xen/arch/arm/domctl.c
+++ b/xen/arch/arm/domctl.c
@@ -48,6 +48,9 @@ static int handle_vuart_init(struct domain *d,
static long p2m_domctl(struct domain *d, struct xen_domctl_shadow_op *sc,
XEN_GUEST_HANDLE_PARAM(xen_domctl_t) u_domctl)
{
+ long rc;
+ bool preempted = false;
+
if ( unlikely(d == current->domain) )
{
printk(XENLOG_ERR "Tried to do a p2m domctl op on itself.\n");
@@ -64,9 +67,27 @@ static long p2m_domctl(struct domain *d, struct xen_domctl_shadow_op *sc,
switch ( sc->op )
{
case XEN_DOMCTL_SHADOW_OP_SET_ALLOCATION:
- return 0;
+ {
+ /* Allow and handle preemption */
+ spin_lock(&d->arch.paging.lock);
+ rc = p2m_set_allocation(d, sc->mb << (20 - PAGE_SHIFT), &preempted);
+ spin_unlock(&d->arch.paging.lock);
+
+ if ( preempted )
+ /* Not finished. Set up to re-run the call. */
+ rc = hypercall_create_continuation(__HYPERVISOR_domctl, "h",
+ u_domctl);
+ else
+ /* Finished. Return the new allocation. */
+ sc->mb = p2m_get_allocation(d);
+
+ return rc;
+ }
case XEN_DOMCTL_SHADOW_OP_GET_ALLOCATION:
+ {
+ sc->mb = p2m_get_allocation(d);
return 0;
+ }
default:
{
printk(XENLOG_ERR "Bad p2m domctl op %u\n", sc->op);
diff --git a/xen/arch/arm/p2m.c b/xen/arch/arm/p2m.c
index 6883d8627702..c1055ff2a745 100644
--- a/xen/arch/arm/p2m.c
+++ b/xen/arch/arm/p2m.c
@@ -50,6 +50,54 @@ static uint64_t generate_vttbr(uint16_t vmid, mfn_t root_mfn)
return (mfn_to_maddr(root_mfn) | ((uint64_t)vmid << 48));
}
+static struct page_info *p2m_alloc_page(struct domain *d)
+{
+ struct page_info *pg;
+
+ spin_lock(&d->arch.paging.lock);
+ /*
+ * For hardware domain, there should be no limit in the number of pages that
+ * can be allocated, so that the kernel may take advantage of the extended
+ * regions. Hence, allocate p2m pages for hardware domains from heap.
+ */
+ if ( is_hardware_domain(d) )
+ {
+ pg = alloc_domheap_page(NULL, 0);
+ if ( pg == NULL )
+ {
+ printk(XENLOG_G_ERR "Failed to allocate P2M pages for hwdom.\n");
+ spin_unlock(&d->arch.paging.lock);
+ return NULL;
+ }
+ }
+ else
+ {
+ pg = page_list_remove_head(&d->arch.paging.p2m_freelist);
+ if ( unlikely(!pg) )
+ {
+ spin_unlock(&d->arch.paging.lock);
+ return NULL;
+ }
+ d->arch.paging.p2m_total_pages--;
+ }
+ spin_unlock(&d->arch.paging.lock);
+
+ return pg;
+}
+
+static void p2m_free_page(struct domain *d, struct page_info *pg)
+{
+ spin_lock(&d->arch.paging.lock);
+ if ( is_hardware_domain(d) )
+ free_domheap_page(pg);
+ else
+ {
+ d->arch.paging.p2m_total_pages++;
+ page_list_add_tail(pg, &d->arch.paging.p2m_freelist);
+ }
+ spin_unlock(&d->arch.paging.lock);
+}
+
/* Return the size of the pool, rounded up to the nearest MB */
unsigned int p2m_get_allocation(struct domain *d)
{
@@ -751,7 +799,7 @@ static int p2m_create_table(struct p2m_domain *p2m, lpae_t *entry)
ASSERT(!p2m_is_valid(*entry));
- page = alloc_domheap_page(NULL, 0);
+ page = p2m_alloc_page(p2m->domain);
if ( page == NULL )
return -ENOMEM;
@@ -878,7 +926,7 @@ static void p2m_free_entry(struct p2m_domain *p2m,
pg = mfn_to_page(mfn);
page_list_del(pg, &p2m->pages);
- free_domheap_page(pg);
+ p2m_free_page(p2m->domain, pg);
}
static bool p2m_split_superpage(struct p2m_domain *p2m, lpae_t *entry,
@@ -902,7 +950,7 @@ static bool p2m_split_superpage(struct p2m_domain *p2m, lpae_t *entry,
ASSERT(level < target);
ASSERT(p2m_is_superpage(*entry, level));
- page = alloc_domheap_page(NULL, 0);
+ page = p2m_alloc_page(p2m->domain);
if ( !page )
return false;
@@ -1644,7 +1692,7 @@ int p2m_teardown(struct domain *d)
while ( (pg = page_list_remove_head(&p2m->pages)) )
{
- free_domheap_page(pg);
+ p2m_free_page(p2m->domain, pg);
count++;
/* Arbitrarily preempt every 512 iterations */
if ( !(count % 512) && hypercall_preempt_check() )
@@ -1668,6 +1716,7 @@ void p2m_final_teardown(struct domain *d)
return;
ASSERT(page_list_empty(&p2m->pages));
+ ASSERT(page_list_empty(&d->arch.paging.p2m_freelist));
if ( p2m->root )
free_domheap_pages(p2m->root, P2M_ROOT_ORDER);
--
2.37.4
|