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
|
/*
* drivers/s390/s390mach.c
* S/390 machine check handler
*
* S390 version
* Copyright (C) 2000 IBM Deutschland Entwicklung GmbH, IBM Corporation
* Author(s): Ingo Adlung (adlung@de.ibm.com)
* Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <linux/config.h>
#include <linux/init.h>
#include <linux/sched.h>
#include <linux/errno.h>
#include <linux/workqueue.h>
#include <asm/lowcore.h>
#include "s390mach.h"
#define DBG printk
// #define DBG(args,...) do {} while (0);
static struct semaphore m_sem;
extern int css_process_crw(int);
extern int chsc_process_crw(void);
extern int chp_process_crw(int, int);
extern void css_reiterate_subchannels(void);
extern struct workqueue_struct *slow_path_wq;
extern struct work_struct slow_path_work;
static NORET_TYPE void
s390_handle_damage(char *msg)
{
#ifdef CONFIG_SMP
smp_send_stop();
#endif
disabled_wait((unsigned long) __builtin_return_address(0));
for(;;);
}
/*
* Retrieve CRWs and call function to handle event.
*
* Note : we currently process CRWs for io and chsc subchannels only
*/
static int
s390_collect_crw_info(void *param)
{
struct crw crw;
int ccode, ret, slow;
struct semaphore *sem;
sem = (struct semaphore *)param;
/* Set a nice name. */
daemonize("kmcheck");
repeat:
down_interruptible(sem);
slow = 0;
while (1) {
ccode = stcrw(&crw);
if (ccode != 0)
break;
DBG(KERN_DEBUG "crw_info : CRW reports slct=%d, oflw=%d, "
"chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
crw.slct, crw.oflw, crw.chn, crw.rsc, crw.anc,
crw.erc, crw.rsid);
/* Check for overflows. */
if (crw.oflw) {
pr_debug("%s: crw overflow detected!\n", __FUNCTION__);
css_reiterate_subchannels();
slow = 1;
continue;
}
switch (crw.rsc) {
case CRW_RSC_SCH:
pr_debug("source is subchannel %04X\n", crw.rsid);
ret = css_process_crw (crw.rsid);
if (ret == -EAGAIN)
slow = 1;
break;
case CRW_RSC_MONITOR:
pr_debug("source is monitoring facility\n");
break;
case CRW_RSC_CPATH:
pr_debug("source is channel path %02X\n", crw.rsid);
switch (crw.erc) {
case CRW_ERC_IPARM: /* Path has come. */
ret = chp_process_crw(crw.rsid, 1);
break;
case CRW_ERC_PERRI: /* Path has gone. */
case CRW_ERC_PERRN:
ret = chp_process_crw(crw.rsid, 0);
break;
default:
pr_debug("Don't know how to handle erc=%x\n",
crw.erc);
ret = 0;
}
if (ret == -EAGAIN)
slow = 1;
break;
case CRW_RSC_CONFIG:
pr_debug("source is configuration-alert facility\n");
break;
case CRW_RSC_CSS:
pr_debug("source is channel subsystem\n");
ret = chsc_process_crw();
if (ret == -EAGAIN)
slow = 1;
break;
default:
pr_debug("unknown source\n");
break;
}
}
if (slow)
queue_work(slow_path_wq, &slow_path_work);
goto repeat;
return 0;
}
struct mcck_struct {
int kill_task;
int channel_report;
int warning;
unsigned long long mcck_code;
};
static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck);
/*
* Main machine check handler function. Will be called with interrupts enabled
* or disabled and machine checks enabled or disabled.
*/
void
s390_handle_mcck(void)
{
unsigned long flags;
struct mcck_struct mcck;
/*
* Disable machine checks and get the current state of accumulated
* machine checks. Afterwards delete the old state and enable machine
* checks again.
*/
local_irq_save(flags);
local_mcck_disable();
mcck = __get_cpu_var(cpu_mcck);
memset(&__get_cpu_var(cpu_mcck), 0, sizeof(struct mcck_struct));
clear_thread_flag(TIF_MCCK_PENDING);
local_mcck_enable();
local_irq_restore(flags);
if (mcck.channel_report)
up(&m_sem);
#ifdef CONFIG_MACHCHK_WARNING
/*
* The warning may remain for a prolonged period on the bare iron.
* (actually till the machine is powered off, or until the problem is gone)
* So we just stop listening for the WARNING MCH and prevent continuously
* being interrupted. One caveat is however, that we must do this per
* processor and cannot use the smp version of ctl_clear_bit().
* On VM we only get one interrupt per virtally presented machinecheck.
* Though one suffices, we may get one interrupt per (virtual) processor.
*/
if (mcck.warning) { /* WARNING pending ? */
static int mchchk_wng_posted = 0;
/*
* Use single machine clear, as we cannot handle smp right now
*/
__ctl_clear_bit(14, 24); /* Disable WARNING MCH */
if (xchg(&mchchk_wng_posted, 1) == 0)
kill_proc(1, SIGPWR, 1);
}
#endif
if (mcck.kill_task) {
local_irq_enable();
printk(KERN_EMERG "mcck: Terminating task because of machine "
"malfunction (code 0x%016llx).\n", mcck.mcck_code);
printk(KERN_EMERG "mcck: task: %s, pid: %d.\n",
current->comm, current->pid);
do_exit(SIGSEGV);
}
}
/*
* returns 0 if all registers could be validated
* returns 1 otherwise
*/
static int
s390_revalidate_registers(struct mci *mci)
{
int kill_task;
u64 tmpclock;
u64 zero;
void *fpt_save_area, *fpt_creg_save_area;
kill_task = 0;
zero = 0;
/* General purpose registers */
if (!mci->gr)
/*
* General purpose registers couldn't be restored and have
* unknown contents. Process needs to be terminated.
*/
kill_task = 1;
/* Revalidate floating point registers */
if (!mci->fp)
/*
* Floating point registers can't be restored and
* therefore the process needs to be terminated.
*/
kill_task = 1;
#ifndef __s390x__
asm volatile("ld 0,0(%0)\n"
"ld 2,8(%0)\n"
"ld 4,16(%0)\n"
"ld 6,24(%0)"
: : "a" (&S390_lowcore.floating_pt_save_area));
#endif
if (MACHINE_HAS_IEEE) {
#ifdef __s390x__
fpt_save_area = &S390_lowcore.floating_pt_save_area;
fpt_creg_save_area = &S390_lowcore.fpt_creg_save_area;
#else
fpt_save_area = (void *) S390_lowcore.extended_save_area_addr;
fpt_creg_save_area = fpt_save_area+128;
#endif
/* Floating point control register */
if (!mci->fc) {
/*
* Floating point control register can't be restored.
* Task will be terminated.
*/
asm volatile ("lfpc 0(%0)" : : "a" (&zero), "m" (zero));
kill_task = 1;
}
else
asm volatile (
"lfpc 0(%0)"
: : "a" (fpt_creg_save_area));
asm volatile("ld 0,0(%0)\n"
"ld 1,8(%0)\n"
"ld 2,16(%0)\n"
"ld 3,24(%0)\n"
"ld 4,32(%0)\n"
"ld 5,40(%0)\n"
"ld 6,48(%0)\n"
"ld 7,56(%0)\n"
"ld 8,64(%0)\n"
"ld 9,72(%0)\n"
"ld 10,80(%0)\n"
"ld 11,88(%0)\n"
"ld 12,96(%0)\n"
"ld 13,104(%0)\n"
"ld 14,112(%0)\n"
"ld 15,120(%0)\n"
: : "a" (fpt_save_area));
}
/* Revalidate access registers */
asm volatile("lam 0,15,0(%0)"
: : "a" (&S390_lowcore.access_regs_save_area));
if (!mci->ar)
/*
* Access registers have unknown contents.
* Terminating task.
*/
kill_task = 1;
/* Revalidate control registers */
if (!mci->cr)
/*
* Control registers have unknown contents.
* Can't recover and therefore stopping machine.
*/
s390_handle_damage("invalid control registers.");
else
#ifdef __s390x__
asm volatile("lctlg 0,15,0(%0)"
: : "a" (&S390_lowcore.cregs_save_area));
#else
asm volatile("lctl 0,15,0(%0)"
: : "a" (&S390_lowcore.cregs_save_area));
#endif
/*
* We don't even try to revalidate the TOD register, since we simply
* can't write something sensible into that register.
*/
#ifdef __s390x__
/*
* See if we can revalidate the TOD programmable register with its
* old contents (should be zero) otherwise set it to zero.
*/
if (!mci->pr)
asm volatile("sr 0,0\n"
"sckpf"
: : : "0", "cc");
else
asm volatile(
"l 0,0(%0)\n"
"sckpf"
: : "a" (&S390_lowcore.tod_progreg_save_area) : "0", "cc");
|