-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathghost.c
1257 lines (983 loc) · 30 KB
/
ghost.c
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
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
* Copyright (C) 2012 gonzoj
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* 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 <generated/autoconf.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/moduleparam.h>
#include <linux/slab.h>
#include <linux/pid.h>
#include <linux/mm.h>
#include <linux/semaphore.h>
#include <linux/list.h>
/* what the fuck? defines kunmap_atomic */
#include <linux/hugetlb.h>
#include <asm/desc.h>
#include <asm/segment.h>
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
#include <asm/page.h>
#include <asm/processor.h>
#include <asm/traps.h>
#include <asm/highmem.h>
#include <net/netlink.h>
#include <net/genetlink.h>
#define MODULE_NAME "ghost"
/* un-/comment to toggle debug output */
#define _DEBUG_BUILD
#ifdef _DEBUG_BUILD
#define __WARNING KERN_ALERT
#define __ERROR KERN_ALERT
#define __DEBUG KERN_ALERT
#define __INFO KERN_ALERT
#define debugk printk
#else
#define __WARNING KERN_WARNING
#define __ERROR KERN_ERR
#define __DEBUG KERN_DEBUG
#define __INFO KERN_INFO
#define debugk if (_debug) printk
#endif
#define _WARNING __WARNING "["MODULE_NAME"] warning: "
#define _ERROR __ERROR "["MODULE_NAME"] error: "
#define _DEBUG __DEBUG "["MODULE_NAME"] debug: "
#define _INFO __INFO "["MODULE_NAME"] info: "
enum {
SUCCESS,
ERROR,
ERROR_MISSING_PARAMETERS,
ERROR_INVALID_PARAMETERS,
ERROR_PAGE_IN_LIST,
ERROR_OUT_OF_MEMORY,
ERROR_INVALID_ADDRESS,
ERROR_PAGE_NOT_IN_LIST,
ERROR_TASK_NOT_IN_LIST,
};
/* module parameters */
static int _debug = 0;
module_param(_debug, bool, 0);
static int _log = 1;
module_param(_log, bool, 0);
/* netlink family definitions */
enum {
_NLA_UNSPEC,
NLA_PID,
NLA_I_ADDR,
NLA_D_PAGE,
NLA_Z_PAGE,
NLA_RET,
__NLA_MAX,
};
#define NLA_MAX (__NLA_MAX - 1)
enum {
_NLO_UNSPEC,
NLO_REGISTER_PAGE,
NLO_RELEASE_PAGE,
__NLO_MAX,
};
#define NLO_MAX (__NLO_MAX - 1)
#define NL_INTERFACE_VERSION 1
struct _page {
unsigned long i_page;
unsigned long d_page;
unsigned long i_frame;
unsigned long d_frame;
int (*ret)(void);
unsigned long z_frame;
struct list_head lpages;
};
struct _task {
pid_t pid;
struct timespec start_time;
struct list_head managed_pages;
struct list_head ltasks;
};
enum X86_PF_ERROR_CODE {
PF_PROT = 1 << 0, // 0: no page found 1: protection fault
PF_WRITE = 1 << 1, // 0: read access 1: write access
PF_USER = 1 << 2, // 0: kernel-mode access 1: user-mode access
PF_RSVD = 1 << 3, // 1: use of reserved bit detected
PF_INSTR = 1 << 4, // 1: fault was an instruction fetch
};
/* netlink family declarations */
static struct genl_family genl_family_mod = {
.id = GENL_ID_GENERATE,
.hdrsize = 0,
.name = MODULE_NAME,
.version = NL_INTERFACE_VERSION,
.maxattr = NLA_MAX
};
static struct nla_policy nla_policy_mod[__NLA_MAX] = {
[NLA_PID] = { .type = NLA_U32 },
[NLA_I_ADDR] = { .type = NLA_U64 },
[NLA_D_PAGE] = { .type = NLA_U64 },
[NLA_Z_PAGE] = { .type = NLA_U64 },
[NLA_RET] = { .type = NLA_U64 }
};
static int nlo_register_page(struct sk_buff *, struct genl_info *);
static int nlo_release_page(struct sk_buff *, struct genl_info *);
static struct genl_ops genl_ops_register_page = {
.cmd = NLO_REGISTER_PAGE,
.flags = 0,
.policy = nla_policy_mod,
.doit = nlo_register_page,
.dumpit = NULL
};
static struct genl_ops genl_ops_release_page = {
.cmd = NLO_RELEASE_PAGE,
.flags = 0,
.policy = nla_policy_mod,
.doit = nlo_release_page,
.dumpit = NULL
};
/* pattern declaratins for resolving unexported kernel symbols */
#ifdef CONFIG_X86_64
/*
* sub $0x78, %rsp
* callq ...
*/
static unsigned char call_error_entry[] = { 0x48, 0x83, 0xEC, 0x78, 0xE8 };
/*
* mov %rsp, %rdi
* mov 0x78(%rsp), %rsi
* movq $0xffffffffffffffff, 0x78(%rsp)
* callq ...
*/
static unsigned char call_do_page_fault[] = { 0x48, 0x89, 0xE7, 0x48, 0x8B, 0x74, 0x24, 0x78, 0x48, 0xC7, 0x44, 0x24, 0x78, 0xFF, 0xFF, 0xFF, 0xFF, 0xE8 };
/*
* movq $0xffffffffffffffff, 0x78(%rsp)
* callq ?? ?? ?? ??
* jmpq ...
*/
static unsigned char jmp_error_exit[] = { 0x48, 0xC7, 0x44, 0x24, 0x78, 0xFF, 0xFF, 0xFF, 0xFF, 0xE8, 0x00, 0x00, 0x00, 0x00, 0xE9 };
static unsigned int jmp_error_exit_mask[] = { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1 };
#else
/*
* mov %esp, %eax
* call *%edi
* jmp ...
*/
static unsigned char jmp_ret_from_exception[] = { 0x89, 0xE0, 0xFF, 0xD7, 0xE9 };
#endif
LIST_HEAD(managed_tasks);
static struct rw_semaphore manage_s;
static void *kernel_isr0x0E_stub = NULL;
#ifdef CONFIG_X86_64
static void *kernel64_error_entry;
static void *kernel64_do_page_fault;
static void *kernel64_error_exit;
#else
static void *kernel32_ret_from_exception;
#endif
static int register_managed_page(pid_t, unsigned long, unsigned long, int(*)(void), unsigned long);
static int release_managed_page(pid_t, unsigned long);
static int nlo_register_page(struct sk_buff *skb, struct genl_info *info) {
pid_t pid;
unsigned long i_addr, d_page, z_page;
int (*ret)(void);
printk(_INFO "netlink: received register request\n");
if (!info) {
printk(_ERROR "no netlink info available\n");
return ERROR;
}
if (!info->attrs[NLA_PID] || !info->attrs[NLA_I_ADDR] || !info->attrs[NLA_D_PAGE] || !info->attrs[NLA_Z_PAGE] || !info->attrs[NLA_RET]) {
printk(_ERROR "critical parameters are missing\n");
return ERROR_MISSING_PARAMETERS;
}
pid = (pid_t) nla_get_u32(info->attrs[NLA_PID]);
i_addr = (unsigned long) nla_get_u64(info->attrs[NLA_I_ADDR]);
d_page = (unsigned long) nla_get_u64(info->attrs[NLA_D_PAGE]);
ret = (int(*)(void)) (unsigned long) nla_get_u64(info->attrs[NLA_RET]);
z_page = (unsigned long) nla_get_u64(info->attrs[NLA_Z_PAGE]);
printk(_INFO "[%i] registering %lX (read/write %lX) using return at %lX\n", pid, i_addr, d_page, (unsigned long) ret);
if (z_page) printk(_INFO "using zero page %lX\n", z_page);
if (!i_addr || !d_page || !ret) {
printk(_ERROR "one or more critical parameters are not set\n");
return ERROR_INVALID_PARAMETERS;
}
return register_managed_page(pid, i_addr, d_page, ret, z_page);
}
static int nlo_release_page(struct sk_buff *skb, struct genl_info *info) {
pid_t pid;
unsigned long i_addr;
printk(_INFO "netlink: received release request\n");
if (!info) {
printk(_ERROR "no netlink info availabe\n");
return ERROR;
}
if (!info->attrs[NLA_PID] || !info->attrs[NLA_I_ADDR]) {
printk(_ERROR "critical parameters are missing\n");
return ERROR_MISSING_PARAMETERS;
}
pid = (pid_t) nla_get_u32(info->attrs[NLA_PID]);
i_addr = (unsigned long) nla_get_u64(info->attrs[NLA_I_ADDR]);
printk(_INFO "[%i] releasing %lX\n", pid, i_addr);
if (!i_addr) {
printk(_ERROR "one or more critical parameters are not set\n");
return ERROR_INVALID_PARAMETERS;
}
return release_managed_page(pid, i_addr);
}
/* fix for 2.6.36 */
#include <linux/rcupdate.h>
static struct task_struct * _get_pid_task(struct pid *pid, enum pid_type type) {
struct task_struct *result;
rcu_read_lock();
result = pid_task(pid, type);
if (result) get_task_struct(result);
rcu_read_unlock();
return result;
}
/* end fix */
static struct task_struct * get_task_by_pid(pid_t pid) {
struct task_struct *tsk;
struct pid *_pid;
_pid = find_get_pid(pid);
if (!_pid) return NULL;
/* fix for 2.6.36 */
tsk = _get_pid_task(_pid, PIDTYPE_PID);
return tsk;
}
/*
* NOTE: we can't handle hugetlb pages at the moment
*/
static pte_t * resolve_address_to_pte(struct mm_struct *mm, unsigned long address, pmd_t **pmd) {
pgd_t *pgd;
pud_t *pud;
/* 2.6.36: no transparent hugepage support yet */
//pmd_t orig_pmd;
pte_t *pte = NULL;
/* we might wanna hold page_table_lock while doing a page table walk */
pgd = pgd_offset(mm, address);
/*
* we require clients to lock managed pages in memory, so the corresponding
* page table should be in memory at any time.
* if it's not, we assume the kernel page fault handler will deal with it
*/
if (!pgd_present(*pgd)) {
printk(_WARNING "when retrievig PTE: PGD not present\n");
}
pud = pud_offset(pgd, address);
if (!pud_present(*pud)) {
printk(_WARNING "when retrieving PTE: PUD not presend\n");
} else if (pud_large(*pud)) {
printk(_WARNING "when retrieving PTE: large page (PUD) detected\n");
goto out;
} else if (pud_none(*pud)) {
printk(_WARNING "when retrieving PTE: PUD invalid\n");
goto out;
}
*pmd = pmd_offset(pud, address);
//orig_pmd = **pmd;
if (!pmd_present(**pmd)) {
printk(_WARNING "when retrieving PTE: PMD not present\n");
} else if (pmd_large(**pmd)) {
printk(_WARNING "when retrieving PTE: large page (PMD) detected\n");
goto out;
} else if (pmd_none(**pmd)) {
printk(_WARNING "when retrieving PTE: PMD invalid\n");
goto out;
}/* else if (pmd_trans_huge(orig_pmd)) {
printk(_WARNING "when retrieving PTE: transparent hugepage detected\n");
goto out;
}*/
pte = pte_offset_map(*pmd, address);
debugk(_DEBUG "resolving %lX: PTE (%lX): %lX\n", address, (unsigned long) pte, (unsigned long) pte->pte);
out:
return pte;
}
static int resolve_linear_address(struct mm_struct *mm, unsigned long laddr, unsigned long *paddr) {
pmd_t *pmd;
pte_t *pte;
/* just in case ;-) */
spin_lock(&mm->page_table_lock);
pte = resolve_address_to_pte(mm, laddr, &pmd);
if (!pte) {
spin_unlock(&mm->page_table_lock);
return ERROR;
}
if (pte_none(*pte)) {
pte_unmap(pte);
spin_unlock(&mm->page_table_lock);
return ERROR;
}
*paddr = pte_pfn(*pte);
pte_unmap(pte);
spin_unlock(&mm->page_table_lock);
return SUCCESS;
}
static void _flush_tlb(void *laddr) {
__flush_tlb_one((unsigned long) laddr);
}
static void _flush_tlb_page(unsigned long laddr) {
/* do we have to flush the cpu's TLB we're running on? -- I don't think so */
_flush_tlb((void *) laddr);
on_each_cpu(_flush_tlb, (void *) laddr, 1);
}
static int prepare_page_table(struct mm_struct *mm, struct _page *p) {
pmd_t *pmd;
pte_t *pte;
spinlock_t *ptl;
debugk(_DEBUG "locking mm->mmap_sem\n");
down_read(&mm->mmap_sem);
pte = resolve_address_to_pte(mm, p->i_page, &pmd);
if (!pte) {
up_read(&mm->mmap_sem);
return ERROR;
}
ptl = pte_lockptr(mm, pmd);
debugk(_DEBUG "locking PTE spinlock %lX\n", (unsigned long) ptl);
spin_lock(ptl);
printk(_INFO "marking page non present\n");
pte->pte &= ~_PAGE_PRESENT;
//pte->pte |= _PAGE_GLOBAL;
printk(_INFO "flushing TLBs (%lX)\n", p->i_page);
_flush_tlb_page(p->i_page);
pte_unmap(pte);
spin_unlock(ptl);
if (p->z_frame) {
pte = resolve_address_to_pte(mm, p->d_page, &pmd);
if (!pte) {
up_read(&mm->mmap_sem);
return ERROR;
}
ptl = pte_lockptr(mm, pmd);
debugk(_DEBUG "locking PTE spinlock %lX\n", (unsigned long) ptl);
spin_lock(ptl);
printk(_INFO "replacing read/write frame number\n");
pte->pte |= PTE_PFN_MASK;
pte->pte &= ((p->z_frame << PAGE_SHIFT) | ~PTE_PFN_MASK);
printk(_INFO "flushing TLBs (%lX)\n", p->d_page);
_flush_tlb_page(p->d_page);
pte_unmap(pte);
spin_unlock(ptl);
}
up_read(&mm->mmap_sem);
return SUCCESS;
}
static int fixup_page_table(struct mm_struct *mm, struct _page *p) {
pmd_t *pmd;
pte_t *pte;
spinlock_t *ptl;
debugk(_DEBUG "locking mm->mmap_sem\n");
down_read(&mm->mmap_sem);
pte = resolve_address_to_pte(mm, p->i_page, &pmd);
if (!pte) {
up_read(&mm->mmap_sem);
return ERROR;
}
ptl = pte_lockptr(mm, pmd);
debugk(_DEBUG "locking PTE spinlock %lX\n", (unsigned long) ptl);
spin_lock(ptl);
printk(_INFO "adjusting frame number and marking page present\n");
pte->pte |= PTE_PFN_MASK;
pte->pte &= ((p->i_frame << PAGE_SHIFT) | ~PTE_PFN_MASK);
pte->pte |= _PAGE_PRESENT;
printk(_INFO "flushing TLBs (%lX)\n", p->i_page);
_flush_tlb_page(p->i_page);
pte_unmap(pte);
spin_unlock(ptl);
if (p->z_frame) {
pte = resolve_address_to_pte(mm, p->d_page, &pmd);
if (!pte) {
up_read(&mm->mmap_sem);
return ERROR;
}
ptl = pte_lockptr(mm, pmd);
debugk(_DEBUG "locking PTE spinlock %lX\n", (unsigned long) ptl);
spin_lock(ptl);
printk(_INFO "restoring original read/write frame number\n");
pte->pte |= PTE_PFN_MASK;
pte->pte &= ((p->d_frame << PAGE_SHIFT) | ~PTE_PFN_MASK);
printk(_INFO "flushing TLBs (%lX)\n", p->d_page);
_flush_tlb_page(p->d_page);
pte_unmap(pte);
spin_unlock(ptl);
}
up_read(&mm->mmap_sem);
return SUCCESS;
}
static int is_task_equal_to(struct _task *t, struct task_struct *tsk) {
return ((t->pid == tsk->pid) && (t->start_time.tv_sec == tsk->start_time.tv_sec) && (t->start_time.tv_nsec == tsk->start_time.tv_nsec));
}
static struct _task * is_task_in_list(struct task_struct *tsk) {
struct _task *t;
t = NULL;
list_for_each_entry(t, &managed_tasks, ltasks) {
if (is_task_equal_to(t, tsk)) return t;
}
return NULL;
}
static int register_managed_page(pid_t pid, unsigned long i_addr, unsigned long d_addr, int (*ret)(void), unsigned long z_page) {
struct _task *t, *t_new;
struct _page *p, *p_new;
struct task_struct *tsk;
tsk = get_task_by_pid(pid);
if (!tsk || !tsk->mm) return ERROR;
down_read(&manage_s);
t = NULL;
list_for_each_entry(t, &managed_tasks, ltasks) {
if (is_task_equal_to(t, tsk)) goto register_page;
}
t_new = (struct _task *) kmalloc(sizeof(struct _task), GFP_KERNEL);
if (!t_new) {
printk(_ERROR "out of memory\n");
return ERROR_OUT_OF_MEMORY;
}
t_new->pid = pid;
t_new->start_time = tsk->start_time;
INIT_LIST_HEAD(&t_new->managed_pages);
up_read(&manage_s);
down_write(&manage_s);
list_add(&t_new->ltasks, &managed_tasks);
up_write(&manage_s);
down_read(&manage_s);
t = t_new;
register_page:
p = NULL;
list_for_each_entry(p, &t->managed_pages, lpages) {
if ((i_addr & PAGE_MASK) == p->i_page) {
up_read(&manage_s);
return ERROR_PAGE_IN_LIST;
}
}
p_new = (struct _page *) kmalloc(sizeof(struct _page), GFP_KERNEL);
if (!p_new) {
printk(_ERROR "out of memory\n");
return ERROR_OUT_OF_MEMORY;
}
p_new->i_page = i_addr & PAGE_MASK;
p_new->d_page = d_addr & PAGE_MASK;
down_read(&tsk->mm->mmap_sem);
if (resolve_linear_address(tsk->mm, p_new->i_page, &p_new->i_frame) || resolve_linear_address(tsk->mm, p_new->d_page, &p_new->d_frame)) {
kfree(p_new);
up_read(&tsk->mm->mmap_sem);
up_read(&manage_s);
return ERROR_INVALID_ADDRESS;
}
p_new->z_frame = 0;
if (z_page && resolve_linear_address(tsk->mm, z_page, &p_new->z_frame)) {
kfree(p_new);
up_read(&tsk->mm->mmap_sem);
up_read(&manage_s);
return ERROR_INVALID_ADDRESS;;
}
up_read(&tsk->mm->mmap_sem);
p_new->ret = ret;
INIT_LIST_HEAD(&p_new->lpages);
/* moved */
up_read(&manage_s);
down_write(&manage_s);
if (prepare_page_table(tsk->mm, p_new)) {
up_read(&tsk->mm->mmap_sem);
up_read(&manage_s);
printk(_ERROR "failed to prepare page table entries\n");
return ERROR;
}
list_add(&p_new->lpages, &t->managed_pages);
up_write(&manage_s);
printk(_INFO "[%i] managing page %lX (%lX)", t->pid, p_new->i_page, p_new->i_frame);
printk(_INFO "read/write page: %lX (%lX)\n", p_new->d_page, p_new->d_frame);
if (z_page) {
printk(_INFO "using zero page %lX (%lX)\n", z_page, p_new->z_frame);
}
return SUCCESS;
}
static int release_managed_page(pid_t pid, unsigned long i_addr) {
struct _task *t;
struct _page *p;
struct task_struct *tsk;
int task_in_list = 0;
tsk = get_task_by_pid(pid);
if (!tsk || !tsk->mm) {
down_read(&manage_s);
if ((t = is_task_in_list(tsk))) {
up_read(&manage_s);
down_write(&manage_s);
p = NULL;
list_for_each_entry(p, &t->managed_pages, lpages) {
list_del(&p->lpages);
kfree(p);
}
list_del(&t->ltasks);
kfree(t);
up_write(&manage_s);
return SUCCESS;
} else {
up_read(&manage_s);
return ERROR_TASK_NOT_IN_LIST;
}
}
down_read(&manage_s);
t = NULL;
list_for_each_entry(t, &managed_tasks, ltasks) {
if (is_task_equal_to(t, tsk)) {
task_in_list = 1;
p = NULL;
list_for_each_entry(p, &t->managed_pages, lpages) {
if ((i_addr & PAGE_MASK) == p->i_page) {
/* moved */
up_read(&manage_s);
down_write(&manage_s);
if (fixup_page_table(tsk->mm, p)) {
printk(_WARNING "failed to restore valid page table state\n");
}
list_del(&p->lpages);
kfree(p);
if (list_empty(&t->managed_pages)) {
list_del(&t->ltasks);
kfree(t);
}
up_write(&manage_s);
return SUCCESS;
}
}
}
}
up_read(&manage_s);
return task_in_list ? ERROR_PAGE_NOT_IN_LIST : ERROR_TASK_NOT_IN_LIST;
}
static void release_managed_tasks(void) {
struct _task *t;
struct _page *p;
struct task_struct *tsk;
down_write(&manage_s);
t = NULL;
list_for_each_entry(t, &managed_tasks, ltasks) {
tsk = get_task_by_pid(t->pid);
p = NULL;
if (!tsk || !tsk->mm) {
list_for_each_entry(p, &t->managed_pages, lpages) {
list_del(&p->lpages);
kfree(p);
}
list_del(&t->ltasks);
kfree(t);
} else {
list_for_each_entry(p, &t->managed_pages, lpages) {
if (fixup_page_table(tsk->mm, p)) {
printk(_WARNING "failed to restore valid page table state\n");
}
list_del(&p->lpages);
kfree(p);
}
list_del(&t->ltasks);
kfree(t);
}
}
up_write(&manage_s);
}
static void * get_registered_isr(int gate) {
struct desc_ptr __idtr;
gate_desc *__idt, entry;
store_idt(&__idtr);
__idt = (gate_desc *) __idtr.address;
memcpy(&entry, &__idt[gate], sizeof(entry));
return (void *) gate_offset(entry);
}
/* fix for 2.6.36 */
typedef void (*smp_call_func_t)(void *info);
static inline void _load_idt(const struct desc_ptr *dtr) {
asm volatile("lidt %0"::"m" (*dtr));
}
/* end fix */
static void * register_isr(int gate, void *new, void **old) {
struct desc_ptr __idtr;
gate_desc *__idt, entry;
void *isr;
store_idt(&__idtr);
__idt = (gate_desc *) __idtr.address;
memcpy(&entry, &__idt[gate], sizeof(entry));
isr = (void *) gate_offset(entry);
if (old) {
*old = isr;
printk(_INFO "saving ISR %lX\n", (unsigned long) *old);
}
#ifdef CONFIG_X86_64
pack_gate(&entry, __idt[gate].type, (unsigned long) new, __idt[gate].dpl, __idt[gate].ist, gate_segment(__idt[gate]));
#else
pack_gate(&entry, __idt[gate].type, (unsigned long) new, __idt[gate].dpl, 0, gate_segment(__idt[gate]));
#endif
printk(_INFO "registering ISR %lX\n", (unsigned long) new);
/* not necessary, smp_call_function should do the trick */
asm("cli");
write_idt_entry(__idt, gate, &entry);
asm("sti");
/* fix for 2.6.36 */
smp_call_function((smp_call_func_t) _load_idt, &__idtr, 1);
return isr;
}
static void * find_pattern(unsigned char *base, int range, unsigned char *pattern, int size, int *mask) {
int i, j;
for (i = 0; i < range - size; i++) {
if (!mask) {
if (!memcmp(&base[i], pattern, size)) {
return (void *) (base + i + size);
}
} else {
for (j = 0; j < size; j++) {
if (mask[j] && (&base[i])[j] != pattern[j]) break;
}
if (j == size) return (void *) (base + i + size);
}
}
return NULL;
}
static int resolve_kernel_symbol(unsigned char *base, int range, unsigned char *pattern, int length, int *mask, void **symbol) {
unsigned long *label = find_pattern(base, range, pattern, length, mask);
if (label) {
*(unsigned long *)symbol = (*(int32_t *)label + (unsigned long) label + sizeof(int32_t));
if (*(unsigned long *)symbol % 4) {
printk(_WARNING "resolved kernel symbol is not aligned\n");
}
return SUCCESS;
} else {
return ERROR;
}
}
static void noinline load_dtlb(unsigned long address) {
__asm__ __volatile__ (
#ifdef CONFIG_X86_64
"movq (%0), %%rax\n"
:
: "r"(address)
: "%rax"
#else
"movl (%0), %%eax\n"
:
: "r"(address)
: "%eax"
#endif
);
}
/*static void noinline load_itlb(void *address) {
__asm__ __volatile__ (
"call *%0\n"
:
: "r"(address)
);
}*/
static int fault_in_kernel_space(unsigned long address) {
return address >= TASK_SIZE_MAX;
}
int dotraplinkage isr0x0E(unsigned long error_code, unsigned long ip) {
struct task_struct *tsk;
unsigned long address;
unsigned long frame;
struct _task *t;
struct _page *p;
pmd_t *pmd;
pte_t *pte;
spinlock_t *ptl;
address = read_cr2();
/* we don't handle page faults in kernel space, pass it down */
if (fault_in_kernel_space(address)) {
debugk(_DEBUG "page fault in kernelspace (%i)\n", current->pid);
return 0;
}
/* pass the page fault down if there is no valid user context */
tsk = current;
if (!tsk->mm) {
printk(_ERROR "no context for process %i\n", tsk->pid);
return 0;
}
down_read(&manage_s);
//debugk(_DEBUG "locking mm->mmap_sem\n");
down_read(&tsk->mm->mmap_sem);
/* we identify access to managed pages by the faulting linear address and the corresponding physical frame - meh, this is error-prone; maybe we should use the PTE's address instead? */
t = NULL;
list_for_each_entry(t, &managed_tasks, ltasks) {
if (is_task_equal_to(t, tsk)) {
debugk(_DEBUG "%s: page fault in managed process (%lX: %lX)\n", error_code & PF_USER ? "userspace" : "kernelspace", ip, address);
}
p = NULL;
list_for_each_entry(p, &t->managed_pages, lpages) {
/* expect odd behaviour if a client failed to release a managed page and we hit a matching linear address resp. physical frame */
if ((address & PAGE_MASK) != p->i_page) continue;
/* we ignore page table present flags when resolving addresses */
if (!resolve_linear_address(tsk->mm, address, &frame)) {
if (frame == p->i_frame) {
debugk(_DEBUG "page fault on managed page (%lX: %lX)\n", ip, address);
/* is this necessary? */
__set_current_state(TASK_RUNNING);
pte = resolve_address_to_pte(tsk->mm, address, &pmd);
ptl = pte_lockptr(tsk->mm, pmd);
debugk(_DEBUG "locking PTE spinlock %lX\n", (unsigned long) ptl);
spin_lock(ptl);
if (address == ip && !(error_code & PF_WRITE)) {
/* execute access */
debugk(_DEBUG "loading ITLB\n");
/* mark the page present temporarily and load ITLB */
pte->pte |= _PAGE_PRESENT;
/* call into managed page */
p->ret();
pte->pte &= ~_PAGE_PRESENT;
} else {
/* read/write access */
debugk(_DEBUG "loading DTLB\n");
/* replace the page frame number with the fake */
pte->pte |= PTE_PFN_MASK;
pte->pte &= ((p->d_frame << PAGE_SHIFT) | ~PTE_PFN_MASK);
/* mark the page present temporarily and load DTLB */
pte->pte |= _PAGE_PRESENT;
//printk(KERN_ALERT "pte when loading %lX: %lX\n", (unsigned long) pte, (unsigned long) pte->pte);
load_dtlb(address);
pte->pte &= ~_PAGE_PRESENT;
/* restore the original page frame number */
pte->pte |= PTE_PFN_MASK;
pte->pte &= ((p->i_frame << PAGE_SHIFT) | ~PTE_PFN_MASK);
if (_log) {
/* we might want to notify our client that a managed page has been accessed via netlink */
printk(_INFO "%lX accessing (%s) %lX\n", ip, (error_code & PF_WRITE) ? "write" : "read", address);
}
}
pte_unmap(pte);
spin_unlock(ptl);
up_read(&tsk->mm->mmap_sem);
//local_irq_enable();
up_read(&manage_s);
debugk(_DEBUG "successfully handled page fault\n");
/* we handled the page fault, return from the interrupt */
return 1;
}
} else {
/*
* actually we don't have to worry about that particular situation; in case we miss a page fault that
* we should handle, the kernel page fault handler will fail to proccess it and just kill the offending process
*/