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
|
/*
* Linux/PA-RISC Project (http://www.parisc-linux.org/)
*
* System call entry code Copyright (c) Matthew Wilcox 1999 <willy@bofh.ai>
* Licensed under the GNU GPL.
* thanks to Philipp Rumpf, Mike Shaver and various others
* sorry about the wall, puffin..
*/
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/errno.h>
#include <asm/page.h>
#include <asm/psw.h>
#include <asm/thread_info.h>
#include <asm/assembly.h>
#include <asm/processor.h>
#include <linux/linkage.h>
/* We fill the empty parts of the gateway page with
* something that will kill the kernel or a
* userspace application.
*/
#define KILL_INSN break 0,0
.level LEVEL
.text
.import syscall_exit,code
.import syscall_exit_rfi,code
/* Linux gateway page is aliased to virtual page 0 in the kernel
* address space. Since it is a gateway page it cannot be
* dereferenced, so null pointers will still fault. We start
* the actual entry point at 0x100. We put break instructions
* at the beginning of the page to trap null indirect function
* pointers.
*/
.align PAGE_SIZE
ENTRY(linux_gateway_page)
/* ADDRESS 0x00 to 0xb0 = 176 bytes / 4 bytes per insn = 44 insns */
.rept 44
KILL_INSN
.endr
/* ADDRESS 0xb0 to 0xb4, lws uses 1 insns for entry */
/* Light-weight-syscall entry must always be located at 0xb0 */
/* WARNING: Keep this number updated with table size changes */
#define __NR_lws_entries (2)
lws_entry:
/* Unconditional branch to lws_start, located on the
same gateway page */
b,n lws_start
/* Fill from 0xb4 to 0xe0 */
.rept 11
KILL_INSN
.endr
/* This function MUST be located at 0xe0 for glibc's threading
mechanism to work. DO NOT MOVE THIS CODE EVER! */
set_thread_pointer:
gate .+8, %r0 /* increase privilege */
depi 3, 31, 2, %r31 /* Ensure we return into user mode. */
be 0(%sr7,%r31) /* return to user space */
mtctl %r26, %cr27 /* move arg0 to the control register */
/* Increase the chance of trapping if random jumps occur to this
address, fill from 0xf0 to 0x100 */
.rept 4
KILL_INSN
.endr
/* This address must remain fixed at 0x100 for glibc's syscalls to work */
.align 256
linux_gateway_entry:
gate .+8, %r0 /* become privileged */
mtsp %r0,%sr4 /* get kernel space into sr4 */
mtsp %r0,%sr5 /* get kernel space into sr5 */
mtsp %r0,%sr6 /* get kernel space into sr6 */
mfsp %sr7,%r1 /* save user sr7 */
mtsp %r1,%sr3 /* and store it in sr3 */
#ifdef CONFIG_64BIT
/* for now we can *always* set the W bit on entry to the syscall
* since we don't support wide userland processes. We could
* also save the current SM other than in r0 and restore it on
* exit from the syscall, and also use that value to know
* whether to do narrow or wide syscalls. -PB
*/
ssm PSW_SM_W, %r1
extrd,u %r1,PSW_W_BIT,1,%r1
/* sp must be aligned on 4, so deposit the W bit setting into
* the bottom of sp temporarily */
or,ev %r1,%r30,%r30
b,n 1f
/* The top halves of argument registers must be cleared on syscall
* entry from narrow executable.
*/
depdi 0, 31, 32, %r26
depdi 0, 31, 32, %r25
depdi 0, 31, 32, %r24
depdi 0, 31, 32, %r23
depdi 0, 31, 32, %r22
depdi 0, 31, 32, %r21
1:
#endif
mfctl %cr30,%r1
xor %r1,%r30,%r30 /* ye olde xor trick */
xor %r1,%r30,%r1
xor %r1,%r30,%r30
ldo THREAD_SZ_ALGN+FRAME_SIZE(%r30),%r30 /* set up kernel stack */
/* N.B.: It is critical that we don't set sr7 to 0 until r30
* contains a valid kernel stack pointer. It is also
* critical that we don't start using the kernel stack
* until after sr7 has been set to 0.
*/
mtsp %r0,%sr7 /* get kernel space into sr7 */
STREGM %r1,FRAME_SIZE(%r30) /* save r1 (usp) here for now */
mfctl %cr30,%r1 /* get task ptr in %r1 */
LDREG TI_TASK(%r1),%r1
/* Save some registers for sigcontext and potential task
switch (see entry.S for the details of which ones are
saved/restored). TASK_PT_PSW is zeroed so we can see whether
a process is on a syscall or not. For an interrupt the real
PSW value is stored. This is needed for gdb and sys_ptrace. */
STREG %r0, TASK_PT_PSW(%r1)
STREG %r2, TASK_PT_GR2(%r1) /* preserve rp */
STREG %r19, TASK_PT_GR19(%r1)
LDREGM -FRAME_SIZE(%r30), %r2 /* get users sp back */
#ifdef CONFIG_64BIT
extrd,u %r2,63,1,%r19 /* W hidden in bottom bit */
#if 0
xor %r19,%r2,%r2 /* clear bottom bit */
depd,z %r19,1,1,%r19
std %r19,TASK_PT_PSW(%r1)
#endif
#endif
STREG %r2, TASK_PT_GR30(%r1) /* ... and save it */
STREG %r20, TASK_PT_GR20(%r1) /* Syscall number */
STREG %r21, TASK_PT_GR21(%r1)
STREG %r22, TASK_PT_GR22(%r1)
STREG %r23, TASK_PT_GR23(%r1) /* 4th argument */
STREG %r24, TASK_PT_GR24(%r1) /* 3rd argument */
STREG %r25, TASK_PT_GR25(%r1) /* 2nd argument */
STREG %r26, TASK_PT_GR26(%r1) /* 1st argument */
STREG %r27, TASK_PT_GR27(%r1) /* user dp */
STREG %r28, TASK_PT_GR28(%r1) /* return value 0 */
STREG %r28, TASK_PT_ORIG_R28(%r1) /* return value 0 (saved for signals) */
STREG %r29, TASK_PT_GR29(%r1) /* return value 1 */
STREG %r31, TASK_PT_GR31(%r1) /* preserve syscall return ptr */
ldo TASK_PT_FR0(%r1), %r27 /* save fpregs from the kernel */
save_fp %r27 /* or potential task switch */
mfctl %cr11, %r27 /* i.e. SAR */
STREG %r27, TASK_PT_SAR(%r1)
loadgp
#ifdef CONFIG_64BIT
ldo -16(%r30),%r29 /* Reference param save area */
copy %r19,%r2 /* W bit back to r2 */
#else
/* no need to save these on stack in wide mode because the first 8
* args are passed in registers */
stw %r22, -52(%r30) /* 5th argument */
stw %r21, -56(%r30) /* 6th argument */
#endif
/* Are we being ptraced? */
mfctl %cr30, %r1
LDREG TI_TASK(%r1),%r1
ldw TASK_PTRACE(%r1), %r1
bb,<,n %r1,31,.Ltracesys
/* Note! We cannot use the syscall table that is mapped
nearby since the gateway page is mapped execute-only. */
#ifdef CONFIG_64BIT
ldil L%sys_call_table, %r1
or,= %r2,%r2,%r2
addil L%(sys_call_table64-sys_call_table), %r1
ldo R%sys_call_table(%r1), %r19
or,= %r2,%r2,%r2
ldo R%sys_call_table64(%r1), %r19
#else
ldil L%sys_call_table, %r1
ldo R%sys_call_table(%r1), %r19
#endif
comiclr,>> __NR_Linux_syscalls, %r20, %r0
b,n .Lsyscall_nosys
LDREGX %r20(%r19), %r19
/* If this is a sys_rt_sigreturn call, and the signal was received
* when not in_syscall, then we want to return via syscall_exit_rfi,
* not syscall_exit. Signal no. in r20, in_syscall in r25 (see
* trampoline code in signal.c).
*/
ldi __NR_rt_sigreturn,%r2
comb,= %r2,%r20,.Lrt_sigreturn
.Lin_syscall:
ldil L%syscall_exit,%r2
be 0(%sr7,%r19)
ldo R%syscall_exit(%r2),%r2
.Lrt_sigreturn:
comib,<> 0,%r25,.Lin_syscall
ldil L%syscall_exit_rfi,%r2
be 0(%sr7,%r19)
ldo R%syscall_exit_rfi(%r2),%r2
/* Note! Because we are not running where we were linked, any
calls to functions external to this file must be indirect. To
be safe, we apply the opposite rule to functions within this
file, with local labels given to them to ensure correctness. */
.Lsyscall_nosys:
syscall_nosys:
ldil L%syscall_exit,%r1
be R%syscall_exit(%sr7,%r1)
ldo -ENOSYS(%r0),%r28 /* set errno */
/* Warning! This trace code is a virtual duplicate of the code above so be
* sure to maintain both! */
.Ltracesys:
tracesys:
/* Need to save more registers so the debugger can see where we
* are. This saves only the lower 8 bits of PSW, so that the C
* bit is still clear on syscalls, and the D bit is set if this
* full register save path has been executed. We check the D
* bit on syscall_return_rfi to determine which registers to
* restore. An interrupt results in a full PSW saved with the
* C bit set, a non-straced syscall entry results in C and D clear
* in the saved PSW.
*/
ldo -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1
|