summaryrefslogtreecommitdiff
path: root/arch/sh64/mm/fault.c
blob: 7aea586fc3d0962a6c1bf9e22f29863d26454030 (plain)
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
/*
 * This file is subject to the terms and conditions of the GNU General Public
 * License.  See the file "COPYING" in the main directory of this archive
 * for more details.
 *
 * arch/sh64/mm/fault.c
 *
 * Copyright (C) 2000, 2001  Paolo Alberelli
 * Copyright (C) 2003  Richard Curnow (/proc/tlb, bug fixes)
 * Copyright (C) 2003  Paul Mundt
 *
 */

#include <linux/signal.h>
#include <linux/rwsem.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/types.h>
#include <linux/ptrace.h>
#include <linux/mman.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/interrupt.h>

#include <asm/system.h>
#include <asm/io.h>
#include <asm/tlb.h>
#include <asm/uaccess.h>
#include <asm/pgalloc.h>
#include <asm/mmu_context.h>
#include <asm/registers.h>		/* required by inline asm statements */

#if defined(CONFIG_SH64_PROC_TLB)
#include <linux/init.h>
#include <linux/proc_fs.h>
/* Count numbers of tlb refills in each region */
static unsigned long long calls_to_update_mmu_cache = 0ULL;
static unsigned long long calls_to_flush_tlb_page   = 0ULL;
static unsigned long long calls_to_flush_tlb_range  = 0ULL;
static unsigned long long calls_to_flush_tlb_mm     = 0ULL;
static unsigned long long calls_to_flush_tlb_all    = 0ULL;
unsigned long long calls_to_do_slow_page_fault = 0ULL;
unsigned long long calls_to_do_fast_page_fault = 0ULL;

/* Count size of ranges for flush_tlb_range */
static unsigned long long flush_tlb_range_1         = 0ULL;
static unsigned long long flush_tlb_range_2         = 0ULL;
static unsigned long long flush_tlb_range_3_4       = 0ULL;
static unsigned long long flush_tlb_range_5_7       = 0ULL;
static unsigned long long flush_tlb_range_8_11      = 0ULL;
static unsigned long long flush_tlb_range_12_15     = 0ULL;
static unsigned long long flush_tlb_range_16_up     = 0ULL;

static unsigned long long page_not_present          = 0ULL;

#endif

extern void die(const char *,struct pt_regs *,long);

#define PFLAG(val,flag)   (( (val) & (flag) ) ? #flag : "" )
#define PPROT(flag) PFLAG(pgprot_val(prot),flag)

static inline void print_prots(pgprot_t prot)
{
	printk("prot is 0x%08lx\n",pgprot_val(prot));

	printk("%s %s %s %s %s\n",PPROT(_PAGE_SHARED),PPROT(_PAGE_READ),
	       PPROT(_PAGE_EXECUTE),PPROT(_PAGE_WRITE),PPROT(_PAGE_USER));
}

static inline void print_vma(struct vm_area_struct *vma)
{
	printk("vma start 0x%08lx\n", vma->vm_start);
	printk("vma end   0x%08lx\n", vma->vm_end);

	print_prots(vma->vm_page_prot);
	printk("vm_flags 0x%08lx\n", vma->vm_flags);
}

static inline void print_task(struct task_struct *tsk)
{
	printk("Task pid %d\n", tsk->pid);
}

static pte_t *lookup_pte(struct mm_struct *mm, unsigned long address)
{
	pgd_t *dir;
	pmd_t *pmd;
	pte_t *pte;
	pte_t entry;

	dir = pgd_offset(mm, address);
	if (pgd_none(*dir)) {
		return NULL;
	}

	pmd = pmd_offset(dir, address);
	if (pmd_none(*pmd)) {
		return NULL;
	}

	pte = pte_offset_kernel(pmd, address);
	entry = *pte;

	if (pte_none(entry)) {
		return NULL;
	}
	if (!pte_present(entry)) {
		return NULL;
	}

	return pte;
}

/*
 * This routine handles page faults.  It determines the address,
 * and the problem, and then passes it off to one of the appropriate
 * routines.
 */
asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
			      unsigned long textaccess, unsigned long address)
{
	struct task_struct *tsk;
	struct mm_struct *mm;
	struct vm_area_struct * vma;
	const struct exception_table_entry *fixup;
	pte_t *pte;
	int fault;

#if defined(CONFIG_SH64_PROC_TLB)
        ++calls_to_do_slow_page_fault;
#endif

	/* SIM
	 * Note this is now called with interrupts still disabled
	 * This is to cope with being called for a missing IO port
	 * address with interrupts disabled. This should be fixed as
	 * soon as we have a better 'fast path' miss handler.
	 *
	 * Plus take care how you try and debug this stuff.
	 * For example, writing debug data to a port which you
	 * have just faulted on is not going to work.
	 */

	tsk = current;
	mm = tsk->mm;

	/* Not an IO address, so reenable interrupts */
	local_irq_enable();

	/*
	 * If we're in an interrupt or have no user
	 * context, we must not take the fault..
	 */
	if (in_atomic() || !mm)
		goto no_context;

	/* TLB misses upon some cache flushes get done under cli() */
	down_read(&mm->mmap_sem);

	vma = find_vma(mm, address);

	if (!vma) {
#ifdef DEBUG_FAULT
		print_task(tsk);
		printk("%s:%d fault, address is 0x%08x PC %016Lx textaccess %d writeaccess %d\n",
		       __FUNCTION__,__LINE__,
		       address,regs->pc,textaccess,writeaccess);
		show_regs(regs);
#endif
		goto bad_area;
	}
	if (vma->vm_start <= address) {
		goto good_area;
	}

	if (!(vma->vm_flags & VM_GROWSDOWN)) {
#ifdef DEBUG_FAULT
		print_task(tsk);
		printk("%s:%d fault, address is 0x%08x PC %016Lx textaccess %d writeaccess %d\n",
		       __FUNCTION__,__LINE__,
		       address,regs->pc,textaccess,writeaccess);
		show_regs(regs);

		print_vma(vma);
#endif
		goto bad_area;
	}
	if (expand_stack(vma, address)) {
#ifdef DEBUG_FAULT
		print_task(tsk);
		printk("%s:%d fault, address is 0x%08x PC %016Lx textaccess %d writeaccess %d\n",
		       __FUNCTION__,__LINE__,
		       address,regs->pc,textaccess,writeaccess);
		show_regs(regs);
#endif
		goto bad_area;
	}
/*
 * Ok, we have a good vm_area for this memory access, so
 * we can handle it..
 */
good_area:
	if (textaccess) {
		if (!(vma->vm_flags & VM_EXEC))
			goto bad_area;
	} else {
		if (writeaccess) {
			if (!(vma->vm_flags & VM_WRITE))
				goto bad_area;
		} else {
			if (!(vma->vm_flags & VM_READ))
				goto bad_area;
		}
	}

	/*
	 * If for any reason at all we couldn't handle the fault,
	 * make sure we exit gracefully rather than endlessly redo
	 * the fault.
	 */
survive:
	fault = handle_mm_fault(mm, vma, address, writeaccess);
	if (unlikely(fault & VM_FAULT_ERROR)) {
		if (fault & VM_FAULT_OOM)
			goto out_of_memory;
		else if (fault & VM_FAULT_SIGBUS)
			goto do_sigbus;
		BUG();
	}
	if (fault & VM_FAULT_MAJOR)
		tsk->maj_flt++;
	else
		tsk->min_flt++;

	/* If we get here, the page fault has been handled.  Do the TLB refill
	   now from the newly-setup PTE, to avoid having to fault again right
	   away on the same instruction. */
	pte = lookup_pte (mm, address);
	if (!pte) {
		/* From empirical evidence, we can get here, due to
		   !pte_present(pte).  (e.g. if a swap-in occurs, and the page
		   is swapped back out again before the process that wanted it
		   gets rescheduled?) */
		goto no_pte;
	}

	__do_tlb_refill(address, textaccess, pte);

no_pte:

	up_read(&mm->mmap_sem);
	return;

/*
 * Something tried to access memory that isn't in our memory map..
 * Fix it, but check if it's kernel or user first..
 */
bad_area:
#ifdef DEBUG_FAULT
	printk("fault:bad area\n");
#endif
	up_read(&mm->mmap_sem);

	if (user_mode(regs)) {
		static int count=0;
		siginfo_t info;
		if (count < 4) {
			/* This is really to help debug faults when starting
			 * usermode, so only need a few */
			count++;
			printk("user mode bad_area address=%08lx pid=%d (%s) pc=%08lx\n",
				address, current->pid, current->comm,
				(unsigned long) regs->pc);
#if 0
			show_regs(regs);
#endif
		}
		if (is_global_init(tsk)) {
			panic("INIT had user mode bad_area\n");
		}
		tsk->thread.address = address;
		tsk->thread.error_code = writeaccess;
		info.si_signo = SIGSEGV;
		info.si_errno = 0;
		info.si_addr = (void *) address;
		force_sig_info(SIGSEGV, &info, tsk);
		return;
	}

no_context:
#ifdef DEBUG_FAULT
	printk("fault:No context\n");
#endif
	/* Are we prepared to handle this kernel fault?  */
	fixup = search_exception_tables(regs->pc);
	if (fixup) {
		regs->pc = fixup->fixup;
		return;
	}

/*
 * Oops. The kernel tried to access some bad page. We'll have to
 * terminate things with extreme prejudice.
 *
 */
	if (address < PAGE_SIZE)
		printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
	else
		printk(KERN_ALERT "Unable to handle kernel paging request");
	printk(" at virtual address %08lx\n", address);
	printk(KERN_ALERT "pc = %08Lx%08Lx\n", regs->pc >> 32, regs->pc & 0xffffffff);
	die("Oops", regs, writeaccess);
	do_exit(SIGKILL);

/*
 * We ran out of memory, or some other thing happened to us that made
 * us unable to handle the page fault gracefully.
 */
out_of_memory:
	if (is_global_init(current)) {
		panic("INIT out of memory\n");
		yield();
		goto survive;
	}
	printk("fault:Out of memory\n");
	up_read(&mm->mmap_sem);
	if (is_global_init(current)) {
		yield();
		down_read(&mm->mmap_sem);
		goto survive;
	}
	printk("VM: killing process %s\n", tsk->comm);
	if (user_mode(regs))
		do_group_exit(SIGKILL);
	goto no_context;

do_sigbus:
	printk("fault:Do sigbus\n");
	up_read(&mm->mmap_sem);

	/*
	 * Send a sigbus, regardless of whether we were in kernel
	 * or user mode.
	 */
	tsk->thread.address = address;
	tsk->thread.error_code = writeaccess;
	tsk->thread.trap_no = 14;
	force_sig(SIGBUS, tsk);

	/* Kernel mode? Handle exceptions or die */
	if (!user_mode(regs))
		goto no_context;
}


void flush_tlb_all(void);

void update_mmu_cache(struct vm_area_struct * vma,
			unsigned long address, pte_t pte)
{
#if defined(CONFIG_SH64_PROC_TLB)
	++calls_to_update_mmu_cache;
#endif

	/*
	 * This appears to get called once for every pte entry that gets
	 * established => I don't think it's efficient to try refilling the
	 * TLBs with the pages - some may not get accessed even.  Also, for
	 * executable pages, it is impossible to determine reliably here which
	 * TLB they should be mapped into (or both even).
	 *
	 * So, just do nothing here and handle faults on demand.  In the
	 * TLBMISS handling case, the refill is now done anyway after the pte
	 * has been fixed up, so that deals with most useful cases.
	 */
}

static void __flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
{
	unsigned long long match, pteh=0, lpage;
	unsigned long tlb;
	struct mm_struct *mm;

	mm = vma->vm_mm;

	if (mm->context == NO_CONTEXT)
		return;

	/*
	 * Sign-extend based on neff.
	 */
	lpage = (page & NEFF_SIGN) ? (page | NEFF_MASK) : page;
	match = ((mm->context & MMU_CONTEXT_ASID_MASK) << PTEH_ASID_SHIFT) | PTEH_VALID;
	match |= lpage;

        /* Do ITLB : don't bother for pages in non-exectutable VMAs */
	if (vma->vm_flags & VM_EXEC) {
		for_each_itlb_entry(tlb) {
			asm volatile ("getcfg	%1, 0, %0"
				      : "=r" (pteh)
				      : "r" (tlb) );

			if (pteh == match) {
				__flush_tlb_slot(tlb);
				break;
			}

		}
	}

        /* Do DTLB : any page could potentially be in here. */
	for_each_dtlb_entry(tlb) {
		asm volatile ("getcfg	%1, 0, %0"
			      : "=r" (pteh)
			      : "r" (tlb) );

		if (pteh == match) {
			__flush_tlb_slot(tlb);
			break;
		}

	}
}

void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
{
	unsigned long flags;

#if defined(CONFIG_SH64_PROC_TLB)
        ++calls_to_flush_tlb_page;
#endif

	if (vma->vm_mm) {
		page &= PAGE_MASK;
		local_irq_save(flags);
		__flush_tlb_page(vma, page);
		local_irq_restore(flags);
	}
}

void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
		     unsigned long end)
{
	unsigned long flags;
	unsigned long long match, pteh=0, pteh_epn, pteh_low;
	unsigned long tlb;
	struct mm_struct *mm;

	mm = vma->vm_mm;

#if defined(CONFIG_SH64_PROC_TLB)
	++calls_to_flush_tlb_range;

	{
		unsigned long size = (end - 1) - start;
		size >>= 12; /* divide by PAGE_SIZE */
		size++; /* end=start+4096 => 1 page */
		switch (size) {
		  case  1        : flush_tlb_range_1++;     break;
		  case  2        : flush_tlb_range_2++;     break;
		  case  3 ...  4 : flush_tlb_range_3_4++;   break;
		  case  5 ...  7 : flush_tlb_range_5_7++;   break;
		  case  8 ... 11 : flush_tlb_range_8_11++;  break;
		  case 12 ... 15 : flush_tlb_range_12_15++; break;
		  default        : flush_tlb_range_16_up++; break;
		}
	}
#endif

	if (mm->context == NO_CONTEXT)
		return;

	local_irq_save(flags);

	start &= PAGE_MASK;
	end &= PAGE_MASK;

	match = ((mm->context & MMU_CONTEXT_ASID_MASK) << PTEH_ASID_SHIFT) | PTEH_VALID;

	/* Flush ITLB */
	for_each_itlb_entry(tlb) {
		asm volatile ("getcfg	%1, 0, %0"
			      : "=r" (pteh)
			      : "r" (tlb) );

		pteh_epn = pteh & PAGE_MASK;
		pteh_low = pteh & ~PAGE_MASK;

		if (pteh_low == match && pteh_epn >= start && pteh_epn <= end)
			__flush_tlb_slot(tlb);
	}

	/* Flush DTLB */
	for_each_dtlb_entry(tlb) {
		asm volatile ("getcfg	%1, 0, %0"
			      : "=r" (pteh)
			      : "r" (tlb) );

		pteh_epn = pteh & PAGE_MASK;
		pteh_low = pteh & ~PAGE_MASK;

		if (pteh_low == match && pteh_epn >= start && pteh_epn <= end)
			__flush_tlb_slot(tlb);
	}

	local_irq_restore(flags);
}

void flush_tlb_mm(struct mm_struct *mm)
{
	unsigned long flags;

#if defined(CONFIG_SH64_PROC_TLB)
	++calls_to_flush_tlb_mm;
#endif

	if (mm->context == NO_CONTEXT)
		return;

	local_irq_save(flags);

	mm->context=NO_CONTEXT;
	if(mm==current->mm)
		activate_context(mm);

	local_irq_restore(flags);

}

void flush_tlb_all(void)
{
	/* Invalidate all, including shared pages, excluding fixed TLBs */

	unsigned long flags, tlb;

#if defined(CONFIG_SH64_PROC_TLB)
	++calls_to_flush_tlb_all;
#endif

	local_irq_save(flags);

	/* Flush each ITLB entry */
	for_each_itlb_entry(tlb) {
		__flush_tlb_slot(tlb);
	}

	/* Flush each DTLB entry */
	for_each_dtlb_entry(tlb) {
		__flush_tlb_slot(tlb);
	}

	local_irq_restore(flags);
}

void flush_tlb_kernel_range(unsigned long start, unsigned long end)
{
        /* FIXME: Optimize this later.. */
        flush_tlb_all();
}

#if defined(CONFIG_SH64_PROC_TLB)
/* Procfs interface to read the performance information */

static int
tlb_proc_info(char *buf, char **start, off_t fpos, int length, int *eof, void *data)
{
  int len=0;
  len += sprintf(buf+len, "do_fast_page_fault   called %12lld times\n", calls_to_do_fast_page_fault);
  len += sprintf(buf+len, "do_slow_page_fault   called %12lld times\n", calls_to_do_slow_page_fault);
  len += sprintf(buf+len, "update_mmu_cache     called %12lld times\n", calls_to_update_mmu_cache);
  len += sprintf(buf+len, "flush_tlb_page       called %12lld times\n", calls_to_flush_tlb_page);
  len += sprintf(buf+len, "flush_tlb_range      called %12lld times\n", calls_to_flush_tlb_range);
  len += sprintf(buf+len, "flush_tlb_mm         called %12lld times\n", calls_to_flush_tlb_mm);
  len += sprintf(buf+len, "flush_tlb_all        called %12lld times\n", calls_to_flush_tlb_all);
  len += sprintf(buf+len, "flush_tlb_range_sizes\n"
                          " 1      : %12lld\n"
                          " 2      : %12lld\n"
                          " 3 -  4 : %12lld\n"
                          " 5 -  7 : %12lld\n"
                          " 8 - 11 : %12lld\n"
                          "12 - 15 : %12lld\n"
                          "16+     : %12lld\n",
                          flush_tlb_range_1, flush_tlb_range_2, flush_tlb_range_3_4,
                          flush_tlb_range_5_7, flush_tlb_range_8_11, flush_tlb_range_12_15,
                          flush_tlb_range_16_up);
  len += sprintf(buf+len, "page not present           %12lld times\n", page_not_present);
  *eof = 1;
  return len;
}

static int __init register_proc_tlb(void)
{
  create_proc_read_entry("tlb", 0, NULL, tlb_proc_info, NULL);
  return 0;
}

__initcall(register_proc_tlb);

#endif