process.c 32.0 KB
Newer Older
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
/*
 *  Derived from "arch/i386/kernel/process.c"
 *    Copyright (C) 1995  Linus Torvalds
 *
 *  Updated and modified by Cort Dougan (cort@cs.nmt.edu) and
 *  Paul Mackerras (paulus@cs.anu.edu.au)
 *
 *  PowerPC version
 *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
 *
 *  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
 *  2 of the License, or (at your option) any later version.
 */

#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/stddef.h>
#include <linux/unistd.h>
#include <linux/ptrace.h>
#include <linux/slab.h>
#include <linux/user.h>
#include <linux/elf.h>
#include <linux/init.h>
#include <linux/prctl.h>
#include <linux/init_task.h>
31
#include <linux/export.h>
32 33 34
#include <linux/kallsyms.h>
#include <linux/mqueue.h>
#include <linux/hardirq.h>
35
#include <linux/utsname.h>
36
#include <linux/ftrace.h>
37
#include <linux/kernel_stat.h>
38 39
#include <linux/personality.h>
#include <linux/random.h>
40
#include <linux/hw_breakpoint.h>
41 42 43 44 45 46 47

#include <asm/pgtable.h>
#include <asm/uaccess.h>
#include <asm/io.h>
#include <asm/processor.h>
#include <asm/mmu.h>
#include <asm/prom.h>
48
#include <asm/machdep.h>
49
#include <asm/time.h>
50
#include <asm/runlatch.h>
51
#include <asm/syscalls.h>
52 53
#include <asm/switch_to.h>
#include <asm/debug.h>
54 55 56
#ifdef CONFIG_PPC64
#include <asm/firmware.h>
#endif
57 58
#include <linux/kprobes.h>
#include <linux/kdebug.h>
59 60 61 62 63 64

extern unsigned long _get_SP(void);

#ifndef CONFIG_SMP
struct task_struct *last_task_used_math = NULL;
struct task_struct *last_task_used_altivec = NULL;
65
struct task_struct *last_task_used_vsx = NULL;
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
struct task_struct *last_task_used_spe = NULL;
#endif

/*
 * Make sure the floating-point register state in the
 * the thread_struct is up to date for task tsk.
 */
void flush_fp_to_thread(struct task_struct *tsk)
{
	if (tsk->thread.regs) {
		/*
		 * We need to disable preemption here because if we didn't,
		 * another process could get scheduled after the regs->msr
		 * test but before we have finished saving the FP registers
		 * to the thread_struct.  That process could take over the
		 * FPU, and then when we get scheduled again we would store
		 * bogus values for the remaining FP registers.
		 */
		preempt_disable();
		if (tsk->thread.regs->msr & MSR_FP) {
#ifdef CONFIG_SMP
			/*
			 * This should only ever be called for current or
			 * for a stopped child process.  Since we save away
			 * the FP register state on context switch on SMP,
			 * there is something wrong if a stopped child appears
			 * to still have its FP state in the CPU registers.
			 */
			BUG_ON(tsk != current);
#endif
96
			giveup_fpu(tsk);
97 98 99 100
		}
		preempt_enable();
	}
}
101
EXPORT_SYMBOL_GPL(flush_fp_to_thread);
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

void enable_kernel_fp(void)
{
	WARN_ON(preemptible());

#ifdef CONFIG_SMP
	if (current->thread.regs && (current->thread.regs->msr & MSR_FP))
		giveup_fpu(current);
	else
		giveup_fpu(NULL);	/* just enables FP for kernel */
#else
	giveup_fpu(last_task_used_math);
#endif /* CONFIG_SMP */
}
EXPORT_SYMBOL(enable_kernel_fp);

#ifdef CONFIG_ALTIVEC
void enable_kernel_altivec(void)
{
	WARN_ON(preemptible());

#ifdef CONFIG_SMP
	if (current->thread.regs && (current->thread.regs->msr & MSR_VEC))
		giveup_altivec(current);
	else
127
		giveup_altivec_notask();
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145
#else
	giveup_altivec(last_task_used_altivec);
#endif /* CONFIG_SMP */
}
EXPORT_SYMBOL(enable_kernel_altivec);

/*
 * Make sure the VMX/Altivec register state in the
 * the thread_struct is up to date for task tsk.
 */
void flush_altivec_to_thread(struct task_struct *tsk)
{
	if (tsk->thread.regs) {
		preempt_disable();
		if (tsk->thread.regs->msr & MSR_VEC) {
#ifdef CONFIG_SMP
			BUG_ON(tsk != current);
#endif
146
			giveup_altivec(tsk);
147 148 149 150
		}
		preempt_enable();
	}
}
151
EXPORT_SYMBOL_GPL(flush_altivec_to_thread);
152 153
#endif /* CONFIG_ALTIVEC */

154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
#ifdef CONFIG_VSX
#if 0
/* not currently used, but some crazy RAID module might want to later */
void enable_kernel_vsx(void)
{
	WARN_ON(preemptible());

#ifdef CONFIG_SMP
	if (current->thread.regs && (current->thread.regs->msr & MSR_VSX))
		giveup_vsx(current);
	else
		giveup_vsx(NULL);	/* just enable vsx for kernel - force */
#else
	giveup_vsx(last_task_used_vsx);
#endif /* CONFIG_SMP */
}
EXPORT_SYMBOL(enable_kernel_vsx);
#endif

173 174 175 176 177 178 179
void giveup_vsx(struct task_struct *tsk)
{
	giveup_fpu(tsk);
	giveup_altivec(tsk);
	__giveup_vsx(tsk);
}

180 181 182 183 184 185 186 187 188 189 190 191 192
void flush_vsx_to_thread(struct task_struct *tsk)
{
	if (tsk->thread.regs) {
		preempt_disable();
		if (tsk->thread.regs->msr & MSR_VSX) {
#ifdef CONFIG_SMP
			BUG_ON(tsk != current);
#endif
			giveup_vsx(tsk);
		}
		preempt_enable();
	}
}
193
EXPORT_SYMBOL_GPL(flush_vsx_to_thread);
194 195
#endif /* CONFIG_VSX */

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
#ifdef CONFIG_SPE

void enable_kernel_spe(void)
{
	WARN_ON(preemptible());

#ifdef CONFIG_SMP
	if (current->thread.regs && (current->thread.regs->msr & MSR_SPE))
		giveup_spe(current);
	else
		giveup_spe(NULL);	/* just enable SPE for kernel - force */
#else
	giveup_spe(last_task_used_spe);
#endif /* __SMP __ */
}
EXPORT_SYMBOL(enable_kernel_spe);

void flush_spe_to_thread(struct task_struct *tsk)
{
	if (tsk->thread.regs) {
		preempt_disable();
		if (tsk->thread.regs->msr & MSR_SPE) {
#ifdef CONFIG_SMP
			BUG_ON(tsk != current);
#endif
221
			tsk->thread.spefscr = mfspr(SPRN_SPEFSCR);
222
			giveup_spe(tsk);
223 224 225 226 227 228
		}
		preempt_enable();
	}
}
#endif /* CONFIG_SPE */

229
#ifndef CONFIG_SMP
230 231 232 233
/*
 * If we are doing lazy switching of CPU state (FP, altivec or SPE),
 * and the current task has some state, discard it.
 */
234
void discard_lazy_cpu_state(void)
235 236 237 238 239 240 241 242
{
	preempt_disable();
	if (last_task_used_math == current)
		last_task_used_math = NULL;
#ifdef CONFIG_ALTIVEC
	if (last_task_used_altivec == current)
		last_task_used_altivec = NULL;
#endif /* CONFIG_ALTIVEC */
243 244 245 246
#ifdef CONFIG_VSX
	if (last_task_used_vsx == current)
		last_task_used_vsx = NULL;
#endif /* CONFIG_VSX */
247 248 249 250 251 252
#ifdef CONFIG_SPE
	if (last_task_used_spe == current)
		last_task_used_spe = NULL;
#endif
	preempt_enable();
}
253
#endif /* CONFIG_SMP */
254

255 256 257 258 259 260
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
void do_send_trap(struct pt_regs *regs, unsigned long address,
		  unsigned long error_code, int signal_code, int breakpt)
{
	siginfo_t info;

261
	current->thread.trap_nr = signal_code;
262 263 264 265 266 267 268 269 270 271 272 273
	if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code,
			11, SIGSEGV) == NOTIFY_STOP)
		return;

	/* Deliver the signal to userspace */
	info.si_signo = SIGTRAP;
	info.si_errno = breakpt;	/* breakpoint or watchpoint id */
	info.si_code = signal_code;
	info.si_addr = (void __user *)address;
	force_sig_info(SIGTRAP, &info, current);
}
#else	/* !CONFIG_PPC_ADV_DEBUG_REGS */
274 275 276 277 278
void do_dabr(struct pt_regs *regs, unsigned long address,
		    unsigned long error_code)
{
	siginfo_t info;

279
	current->thread.trap_nr = TRAP_HWBKPT;
280 281 282 283 284 285 286 287
	if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code,
			11, SIGSEGV) == NOTIFY_STOP)
		return;

	if (debugger_dabr_match(regs))
		return;

	/* Clear the DABR */
288
	set_dabr(0, 0);
289 290 291 292 293 294 295 296

	/* Deliver the signal to userspace */
	info.si_signo = SIGTRAP;
	info.si_errno = 0;
	info.si_code = TRAP_HWBKPT;
	info.si_addr = (void __user *)address;
	force_sig_info(SIGTRAP, &info, current);
}
297
#endif	/* CONFIG_PPC_ADV_DEBUG_REGS */
298

299 300
static DEFINE_PER_CPU(unsigned long, current_dabr);

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
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
/*
 * Set the debug registers back to their default "safe" values.
 */
static void set_debug_reg_defaults(struct thread_struct *thread)
{
	thread->iac1 = thread->iac2 = 0;
#if CONFIG_PPC_ADV_DEBUG_IACS > 2
	thread->iac3 = thread->iac4 = 0;
#endif
	thread->dac1 = thread->dac2 = 0;
#if CONFIG_PPC_ADV_DEBUG_DVCS > 0
	thread->dvc1 = thread->dvc2 = 0;
#endif
	thread->dbcr0 = 0;
#ifdef CONFIG_BOOKE
	/*
	 * Force User/Supervisor bits to b11 (user-only MSR[PR]=1)
	 */
	thread->dbcr1 = DBCR1_IAC1US | DBCR1_IAC2US |	\
			DBCR1_IAC3US | DBCR1_IAC4US;
	/*
	 * Force Data Address Compare User/Supervisor bits to be User-only
	 * (0b11 MSR[PR]=1) and set all other bits in DBCR2 register to be 0.
	 */
	thread->dbcr2 = DBCR2_DAC1US | DBCR2_DAC2US;
#else
	thread->dbcr1 = 0;
#endif
}

static void prime_debug_regs(struct thread_struct *thread)
{
	mtspr(SPRN_IAC1, thread->iac1);
	mtspr(SPRN_IAC2, thread->iac2);
#if CONFIG_PPC_ADV_DEBUG_IACS > 2
	mtspr(SPRN_IAC3, thread->iac3);
	mtspr(SPRN_IAC4, thread->iac4);
#endif
	mtspr(SPRN_DAC1, thread->dac1);
	mtspr(SPRN_DAC2, thread->dac2);
#if CONFIG_PPC_ADV_DEBUG_DVCS > 0
	mtspr(SPRN_DVC1, thread->dvc1);
	mtspr(SPRN_DVC2, thread->dvc2);
#endif
	mtspr(SPRN_DBCR0, thread->dbcr0);
	mtspr(SPRN_DBCR1, thread->dbcr1);
#ifdef CONFIG_BOOKE
	mtspr(SPRN_DBCR2, thread->dbcr2);
#endif
}
/*
 * Unless neither the old or new thread are making use of the
 * debug registers, set the debug registers from the values
 * stored in the new thread.
 */
static void switch_booke_debug_regs(struct thread_struct *new_thread)
{
	if ((current->thread.dbcr0 & DBCR0_IDM)
		|| (new_thread->dbcr0 & DBCR0_IDM))
			prime_debug_regs(new_thread);
}
#else	/* !CONFIG_PPC_ADV_DEBUG_REGS */
364
#ifndef CONFIG_HAVE_HW_BREAKPOINT
365 366 367 368
static void set_debug_reg_defaults(struct thread_struct *thread)
{
	if (thread->dabr) {
		thread->dabr = 0;
369 370
		thread->dabrx = 0;
		set_dabr(0, 0);
371 372
	}
}
373
#endif /* !CONFIG_HAVE_HW_BREAKPOINT */
374 375
#endif	/* CONFIG_PPC_ADV_DEBUG_REGS */

376
int set_dabr(unsigned long dabr, unsigned long dabrx)
377
{
378 379
	__get_cpu_var(current_dabr) = dabr;

380
	if (ppc_md.set_dabr)
381
		return ppc_md.set_dabr(dabr, dabrx);
382

383
	/* XXX should we have a CPU_FTR_HAS_DABR ? */
384
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
385
	mtspr(SPRN_DAC1, dabr);
386 387 388
#ifdef CONFIG_PPC_47x
	isync();
#endif
389 390
#elif defined(CONFIG_PPC_BOOK3S)
	mtspr(SPRN_DABR, dabr);
391
	mtspr(SPRN_DABRX, dabrx);
392
#endif
393
	return 0;
394 395
}

396 397 398
#ifdef CONFIG_PPC64
DEFINE_PER_CPU(struct cpu_usage, cpu_usage_array);
#endif
399 400 401 402 403 404 405

struct task_struct *__switch_to(struct task_struct *prev,
	struct task_struct *new)
{
	struct thread_struct *new_thread, *old_thread;
	unsigned long flags;
	struct task_struct *last;
P
Peter Zijlstra 已提交
406 407 408
#ifdef CONFIG_PPC_BOOK3S_64
	struct ppc64_tlb_batch *batch;
#endif
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

#ifdef CONFIG_SMP
	/* avoid complexity of lazy save/restore of fpu
	 * by just saving it every time we switch out if
	 * this task used the fpu during the last quantum.
	 *
	 * If it tries to use the fpu again, it'll trap and
	 * reload its fp regs.  So we don't have to do a restore
	 * every switch, just a save.
	 *  -- Cort
	 */
	if (prev->thread.regs && (prev->thread.regs->msr & MSR_FP))
		giveup_fpu(prev);
#ifdef CONFIG_ALTIVEC
	/*
	 * If the previous thread used altivec in the last quantum
	 * (thus changing altivec regs) then save them.
	 * We used to check the VRSAVE register but not all apps
	 * set it, so we don't rely on it now (and in fact we need
	 * to save & restore VSCR even if VRSAVE == 0).  -- paulus
	 *
	 * On SMP we always save/restore altivec regs just to avoid the
	 * complexity of changing processors.
	 *  -- Cort
	 */
	if (prev->thread.regs && (prev->thread.regs->msr & MSR_VEC))
		giveup_altivec(prev);
#endif /* CONFIG_ALTIVEC */
437 438
#ifdef CONFIG_VSX
	if (prev->thread.regs && (prev->thread.regs->msr & MSR_VSX))
439 440
		/* VMX and FPU registers are already save here */
		__giveup_vsx(prev);
441
#endif /* CONFIG_VSX */
442 443 444 445 446 447 448 449 450 451
#ifdef CONFIG_SPE
	/*
	 * If the previous thread used spe in the last quantum
	 * (thus changing spe regs) then save them.
	 *
	 * On SMP we always save/restore spe regs just to avoid the
	 * complexity of changing processors.
	 */
	if ((prev->thread.regs && (prev->thread.regs->msr & MSR_SPE)))
		giveup_spe(prev);
452 453 454 455 456 457 458 459 460 461
#endif /* CONFIG_SPE */

#else  /* CONFIG_SMP */
#ifdef CONFIG_ALTIVEC
	/* Avoid the trap.  On smp this this never happens since
	 * we don't set last_task_used_altivec -- Cort
	 */
	if (new->thread.regs && last_task_used_altivec == new)
		new->thread.regs->msr |= MSR_VEC;
#endif /* CONFIG_ALTIVEC */
462 463 464 465
#ifdef CONFIG_VSX
	if (new->thread.regs && last_task_used_vsx == new)
		new->thread.regs->msr |= MSR_VSX;
#endif /* CONFIG_VSX */
466
#ifdef CONFIG_SPE
467 468 469 470 471 472
	/* Avoid the trap.  On smp this this never happens since
	 * we don't set last_task_used_spe
	 */
	if (new->thread.regs && last_task_used_spe == new)
		new->thread.regs->msr |= MSR_SPE;
#endif /* CONFIG_SPE */
473

474 475
#endif /* CONFIG_SMP */

476
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
477
	switch_booke_debug_regs(&new->thread);
478
#else
479 480 481 482 483
/*
 * For PPC_BOOK3S_64, we use the hw-breakpoint interfaces that would
 * schedule DABR
 */
#ifndef CONFIG_HAVE_HW_BREAKPOINT
484
	if (unlikely(__get_cpu_var(current_dabr) != new->thread.dabr))
485
		set_dabr(new->thread.dabr, new->thread.dabrx);
486
#endif /* CONFIG_HAVE_HW_BREAKPOINT */
487 488
#endif

489

490 491
	new_thread = &new->thread;
	old_thread = &current->thread;
492 493 494 495 496 497 498 499 500 501 502 503 504

#ifdef CONFIG_PPC64
	/*
	 * Collect processor utilization data per process
	 */
	if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
		struct cpu_usage *cu = &__get_cpu_var(cpu_usage_array);
		long unsigned start_tb, current_tb;
		start_tb = old_thread->start_tb;
		cu->current_tb = current_tb = mfspr(SPRN_PURR);
		old_thread->accum_tb += (current_tb - start_tb);
		new_thread->start_tb = current_tb;
	}
P
Peter Zijlstra 已提交
505 506 507 508 509 510 511 512 513 514 515
#endif /* CONFIG_PPC64 */

#ifdef CONFIG_PPC_BOOK3S_64
	batch = &__get_cpu_var(ppc64_tlb_batch);
	if (batch->active) {
		current_thread_info()->local_flags |= _TLF_LAZY_MMU;
		if (batch->index)
			__flush_tlb_pending(batch);
		batch->active = 0;
	}
#endif /* CONFIG_PPC_BOOK3S_64 */
516

517
	local_irq_save(flags);
518

519 520 521 522 523 524
	/*
	 * We can't take a PMU exception inside _switch() since there is a
	 * window where the kernel stack SLB and the kernel stack are out
	 * of sync. Hard disable here.
	 */
	hard_irq_disable();
525 526
	last = _switch(old_thread, new_thread);

P
Peter Zijlstra 已提交
527 528 529 530 531 532 533 534
#ifdef CONFIG_PPC_BOOK3S_64
	if (current_thread_info()->local_flags & _TLF_LAZY_MMU) {
		current_thread_info()->local_flags &= ~_TLF_LAZY_MMU;
		batch = &__get_cpu_var(ppc64_tlb_batch);
		batch->active = 1;
	}
#endif /* CONFIG_PPC_BOOK3S_64 */

535 536 537 538 539
	local_irq_restore(flags);

	return last;
}

540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555
static int instructions_to_print = 16;

static void show_instructions(struct pt_regs *regs)
{
	int i;
	unsigned long pc = regs->nip - (instructions_to_print * 3 / 4 *
			sizeof(int));

	printk("Instruction dump:");

	for (i = 0; i < instructions_to_print; i++) {
		int instr;

		if (!(i % 8))
			printk("\n");

556 557 558 559 560 561 562 563
#if !defined(CONFIG_BOOKE)
		/* If executing with the IMMU off, adjust pc rather
		 * than print XXXXXXXX.
		 */
		if (!(regs->msr & MSR_IR))
			pc = (unsigned long)phys_to_virt(pc);
#endif

564 565 566 567
		/* We use __get_user here *only* to avoid an OOPS on a
		 * bad address because the pc *should* only be a
		 * kernel address.
		 */
568 569
		if (!__kernel_text_address(pc) ||
		     __get_user(instr, (unsigned int __user *)pc)) {
570
			printk(KERN_CONT "XXXXXXXX ");
571 572
		} else {
			if (regs->nip == pc)
573
				printk(KERN_CONT "<%08x> ", instr);
574
			else
575
				printk(KERN_CONT "%08x ", instr);
576 577 578 579 580 581 582 583 584 585 586 587
		}

		pc += sizeof(int);
	}

	printk("\n");
}

static struct regbit {
	unsigned long bit;
	const char *name;
} msr_bits[] = {
588 589 590 591 592 593 594 595 596
#if defined(CONFIG_PPC64) && !defined(CONFIG_BOOKE)
	{MSR_SF,	"SF"},
	{MSR_HV,	"HV"},
#endif
	{MSR_VEC,	"VEC"},
	{MSR_VSX,	"VSX"},
#ifdef CONFIG_BOOKE
	{MSR_CE,	"CE"},
#endif
597 598 599 600
	{MSR_EE,	"EE"},
	{MSR_PR,	"PR"},
	{MSR_FP,	"FP"},
	{MSR_ME,	"ME"},
601
#ifdef CONFIG_BOOKE
602
	{MSR_DE,	"DE"},
603 604 605 606
#else
	{MSR_SE,	"SE"},
	{MSR_BE,	"BE"},
#endif
607 608
	{MSR_IR,	"IR"},
	{MSR_DR,	"DR"},
609 610 611 612 613
	{MSR_PMM,	"PMM"},
#ifndef CONFIG_BOOKE
	{MSR_RI,	"RI"},
	{MSR_LE,	"LE"},
#endif
614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630
	{0,		NULL}
};

static void printbits(unsigned long val, struct regbit *bits)
{
	const char *sep = "";

	printk("<");
	for (; bits->bit; ++bits)
		if (val & bits->bit) {
			printk("%s%s", sep, bits->name);
			sep = ",";
		}
	printk(">");
}

#ifdef CONFIG_PPC64
631
#define REG		"%016lx"
632 633 634
#define REGS_PER_LINE	4
#define LAST_VOLATILE	13
#else
635
#define REG		"%08lx"
636 637 638 639
#define REGS_PER_LINE	8
#define LAST_VOLATILE	12
#endif

640 641 642 643
void show_regs(struct pt_regs * regs)
{
	int i, trap;

644 645 646
	printk("NIP: "REG" LR: "REG" CTR: "REG"\n",
	       regs->nip, regs->link, regs->ctr);
	printk("REGS: %p TRAP: %04lx   %s  (%s)\n",
647
	       regs, regs->trap, print_tainted(), init_utsname()->release);
648 649
	printk("MSR: "REG" ", regs->msr);
	printbits(regs->msr, msr_bits);
650
	printk("  CR: %08lx  XER: %08lx\n", regs->ccr, regs->xer);
651 652 653
#ifdef CONFIG_PPC64
	printk("SOFTE: %ld\n", regs->softe);
#endif
654
	trap = TRAP(regs);
655 656
	if ((regs->trap != 0xc00) && cpu_has_feature(CPU_FTR_CFAR))
		printk("CFAR: "REG"\n", regs->orig_gpr3);
657
	if (trap == 0x300 || trap == 0x600)
658
#if defined(CONFIG_4xx) || defined(CONFIG_BOOKE)
659 660
		printk("DEAR: "REG", ESR: "REG"\n", regs->dar, regs->dsisr);
#else
661
		printk("DAR: "REG", DSISR: %08lx\n", regs->dar, regs->dsisr);
662
#endif
663
	printk("TASK = %p[%d] '%s' THREAD: %p",
664
	       current, task_pid_nr(current), current->comm, task_thread_info(current));
665 666

#ifdef CONFIG_SMP
667
	printk(" CPU: %d", raw_smp_processor_id());
668 669 670
#endif /* CONFIG_SMP */

	for (i = 0;  i < 32;  i++) {
671
		if ((i % REGS_PER_LINE) == 0)
K
Kumar Gala 已提交
672
			printk("\nGPR%02d: ", i);
673 674
		printk(REG " ", regs->gpr[i]);
		if (i == LAST_VOLATILE && !FULL_REGS(regs))
675 676 677 678 679 680 681 682
			break;
	}
	printk("\n");
#ifdef CONFIG_KALLSYMS
	/*
	 * Lookup NIP late so we have the best change of getting the
	 * above info out without failing
	 */
683 684
	printk("NIP ["REG"] %pS\n", regs->nip, (void *)regs->nip);
	printk("LR ["REG"] %pS\n", regs->link, (void *)regs->link);
685 686
#endif
	show_stack(current, (unsigned long *) regs->gpr[1]);
687 688
	if (!user_mode(regs))
		show_instructions(regs);
689 690 691 692
}

void exit_thread(void)
{
693
	discard_lazy_cpu_state();
694 695 696 697
}

void flush_thread(void)
{
698
	discard_lazy_cpu_state();
699

700
#ifdef CONFIG_HAVE_HW_BREAKPOINT
701
	flush_ptrace_hw_breakpoint(current);
702
#else /* CONFIG_HAVE_HW_BREAKPOINT */
703
	set_debug_reg_defaults(&current->thread);
704
#endif /* CONFIG_HAVE_HW_BREAKPOINT */
705 706 707 708 709 710 711 712
}

void
release_thread(struct task_struct *t)
{
}

/*
713 714
 * this gets called so that we can store coprocessor state into memory and
 * copy the current task into the new thread.
715
 */
716
int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
717
{
718 719 720 721
	flush_fp_to_thread(src);
	flush_altivec_to_thread(src);
	flush_vsx_to_thread(src);
	flush_spe_to_thread(src);
722
#ifdef CONFIG_HAVE_HW_BREAKPOINT
723
	flush_ptrace_hw_breakpoint(src);
724
#endif /* CONFIG_HAVE_HW_BREAKPOINT */
725 726 727

	*dst = *src;
	return 0;
728 729 730 731 732
}

/*
 * Copy a thread..
 */
733 734
extern unsigned long dscr_default; /* defined in arch/powerpc/kernel/sysfs.c */

A
Alexey Dobriyan 已提交
735
int copy_thread(unsigned long clone_flags, unsigned long usp,
A
Al Viro 已提交
736
		unsigned long arg, struct task_struct *p,
737
		struct pt_regs *regs)
738 739 740
{
	struct pt_regs *childregs, *kregs;
	extern void ret_from_fork(void);
A
Al Viro 已提交
741 742
	extern void ret_from_kernel_thread(void);
	void (*f)(void);
A
Al Viro 已提交
743
	unsigned long sp = (unsigned long)task_stack_page(p) + THREAD_SIZE;
744 745 746 747

	/* Copy registers */
	sp -= sizeof(struct pt_regs);
	childregs = (struct pt_regs *) sp;
A
Al Viro 已提交
748
	if (!regs) {
749
		/* for kernel thread, set `current' and stackptr in new task */
A
Al Viro 已提交
750
		memset(childregs, 0, sizeof(struct pt_regs));
751
		childregs->gpr[1] = sp + sizeof(struct pt_regs);
A
Al Viro 已提交
752 753 754
#ifdef CONFIG_PPC64
		childregs->gpr[14] = *(unsigned long *)usp;
		childregs->gpr[2] = ((unsigned long *)usp)[1],
A
Al Viro 已提交
755
		clear_tsk_thread_flag(p, TIF_32BIT);
A
Al Viro 已提交
756 757 758
#else
		childregs->gpr[14] = usp;	/* function */
		childregs->gpr[2] = (unsigned long) p;
759
#endif
A
Al Viro 已提交
760
		childregs->gpr[15] = arg;
761
		p->thread.regs = NULL;	/* no user register state */
A
Al Viro 已提交
762
		f = ret_from_kernel_thread;
763
	} else {
A
Al Viro 已提交
764 765
		CHECK_FULL_REGS(regs);
		*childregs = *regs;
766 767
		childregs->gpr[1] = usp;
		p->thread.regs = childregs;
A
Al Viro 已提交
768
		childregs->gpr[3] = 0;  /* Result from fork() */
769 770
		if (clone_flags & CLONE_SETTLS) {
#ifdef CONFIG_PPC64
771
			if (!is_32bit_task())
772 773 774 775 776
				childregs->gpr[13] = childregs->gpr[6];
			else
#endif
				childregs->gpr[2] = childregs->gpr[6];
		}
A
Al Viro 已提交
777 778

		f = ret_from_fork;
779 780 781 782 783 784 785 786 787 788 789 790 791 792 793
	}
	sp -= STACK_FRAME_OVERHEAD;

	/*
	 * The way this works is that at some point in the future
	 * some task will call _switch to switch to the new task.
	 * That will pop off the stack frame created below and start
	 * the new task running at ret_from_fork.  The new task will
	 * do some house keeping and then return from the fork or clone
	 * system call, using the stack frame created above.
	 */
	sp -= sizeof(struct pt_regs);
	kregs = (struct pt_regs *) sp;
	sp -= STACK_FRAME_OVERHEAD;
	p->thread.ksp = sp;
794 795
	p->thread.ksp_limit = (unsigned long)task_stack_page(p) +
				_ALIGN_UP(sizeof(struct thread_info), 16);
796

797
#ifdef CONFIG_PPC_STD_MMU_64
798
	if (mmu_has_feature(MMU_FTR_SLB)) {
P
Paul Mackerras 已提交
799
		unsigned long sp_vsid;
800
		unsigned long llp = mmu_psize_defs[mmu_linear_psize].sllp;
801

802
		if (mmu_has_feature(MMU_FTR_1T_SEGMENT))
P
Paul Mackerras 已提交
803 804 805 806 807
			sp_vsid = get_kernel_vsid(sp, MMU_SEGSIZE_1T)
				<< SLB_VSID_SHIFT_1T;
		else
			sp_vsid = get_kernel_vsid(sp, MMU_SEGSIZE_256M)
				<< SLB_VSID_SHIFT;
808
		sp_vsid |= SLB_VSID_KERNEL | llp;
809 810
		p->thread.ksp_vsid = sp_vsid;
	}
811
#endif /* CONFIG_PPC_STD_MMU_64 */
812 813
#ifdef CONFIG_PPC64 
	if (cpu_has_feature(CPU_FTR_DSCR)) {
814 815
		p->thread.dscr_inherit = current->thread.dscr_inherit;
		p->thread.dscr = current->thread.dscr;
816 817
	}
#endif
818 819 820 821 822
	/*
	 * The PPC64 ABI makes use of a TOC to contain function 
	 * pointers.  The function (ret_from_except) is actually a pointer
	 * to the TOC entry.  The first entry is a pointer to the actual
	 * function.
A
Al Viro 已提交
823
	 */
824
#ifdef CONFIG_PPC64
A
Al Viro 已提交
825
	kregs->nip = *((unsigned long *)f);
826
#else
A
Al Viro 已提交
827
	kregs->nip = (unsigned long)f;
828
#endif
829 830 831 832 833 834
	return 0;
}

/*
 * Set up a thread for executing a new program
 */
835
void start_thread(struct pt_regs *regs, unsigned long start, unsigned long sp)
836
{
837 838 839 840
#ifdef CONFIG_PPC64
	unsigned long load_addr = regs->gpr[2];	/* saved by ELF_PLAT_INIT */
#endif

841 842 843 844 845
	/*
	 * If we exec out of a kernel thread then thread.regs will not be
	 * set.  Do it now.
	 */
	if (!current->thread.regs) {
A
Al Viro 已提交
846 847
		struct pt_regs *regs = task_stack_page(current) + THREAD_SIZE;
		current->thread.regs = regs - 1;
848 849
	}

850 851 852 853 854 855
	memset(regs->gpr, 0, sizeof(regs->gpr));
	regs->ctr = 0;
	regs->link = 0;
	regs->xer = 0;
	regs->ccr = 0;
	regs->gpr[1] = sp;
856

857 858 859 860 861 862 863
	/*
	 * We have just cleared all the nonvolatile GPRs, so make
	 * FULL_REGS(regs) return true.  This is necessary to allow
	 * ptrace to examine the thread immediately after exec.
	 */
	regs->trap &= ~1UL;

864 865 866
#ifdef CONFIG_PPC32
	regs->mq = 0;
	regs->nip = start;
867
	regs->msr = MSR_USER;
868
#else
869
	if (!is_32bit_task()) {
870
		unsigned long entry, toc;
871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889

		/* start is a relocated pointer to the function descriptor for
		 * the elf _start routine.  The first entry in the function
		 * descriptor is the entry address of _start and the second
		 * entry is the TOC value we need to use.
		 */
		__get_user(entry, (unsigned long __user *)start);
		__get_user(toc, (unsigned long __user *)start+1);

		/* Check whether the e_entry function descriptor entries
		 * need to be relocated before we can use them.
		 */
		if (load_addr != 0) {
			entry += load_addr;
			toc   += load_addr;
		}
		regs->nip = entry;
		regs->gpr[2] = toc;
		regs->msr = MSR_USER64;
S
Stephen Rothwell 已提交
890 891 892 893
	} else {
		regs->nip = start;
		regs->gpr[2] = 0;
		regs->msr = MSR_USER32;
894 895 896
	}
#endif

897
	discard_lazy_cpu_state();
898 899 900
#ifdef CONFIG_VSX
	current->thread.used_vsr = 0;
#endif
901
	memset(current->thread.fpr, 0, sizeof(current->thread.fpr));
902
	current->thread.fpscr.val = 0;
903 904 905
#ifdef CONFIG_ALTIVEC
	memset(current->thread.vr, 0, sizeof(current->thread.vr));
	memset(&current->thread.vscr, 0, sizeof(current->thread.vscr));
906
	current->thread.vscr.u[3] = 0x00010000; /* Java mode disabled */
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930
	current->thread.vrsave = 0;
	current->thread.used_vr = 0;
#endif /* CONFIG_ALTIVEC */
#ifdef CONFIG_SPE
	memset(current->thread.evr, 0, sizeof(current->thread.evr));
	current->thread.acc = 0;
	current->thread.spefscr = 0;
	current->thread.used_spe = 0;
#endif /* CONFIG_SPE */
}

#define PR_FP_ALL_EXCEPT (PR_FP_EXC_DIV | PR_FP_EXC_OVF | PR_FP_EXC_UND \
		| PR_FP_EXC_RES | PR_FP_EXC_INV)

int set_fpexc_mode(struct task_struct *tsk, unsigned int val)
{
	struct pt_regs *regs = tsk->thread.regs;

	/* This is a bit hairy.  If we are an SPE enabled  processor
	 * (have embedded fp) we store the IEEE exception enable flags in
	 * fpexc_mode.  fpexc_mode is also used for setting FP exception
	 * mode (asyn, precise, disabled) for 'Classic' FP. */
	if (val & PR_FP_EXC_SW_ENABLE) {
#ifdef CONFIG_SPE
931 932 933 934 935 936 937
		if (cpu_has_feature(CPU_FTR_SPE)) {
			tsk->thread.fpexc_mode = val &
				(PR_FP_EXC_SW_ENABLE | PR_FP_ALL_EXCEPT);
			return 0;
		} else {
			return -EINVAL;
		}
938 939 940 941
#else
		return -EINVAL;
#endif
	}
942 943 944 945 946 947 948 949 950 951 952 953

	/* on a CONFIG_SPE this does not hurt us.  The bits that
	 * __pack_fe01 use do not overlap with bits used for
	 * PR_FP_EXC_SW_ENABLE.  Additionally, the MSR[FE0,FE1] bits
	 * on CONFIG_SPE implementations are reserved so writing to
	 * them does not change anything */
	if (val > PR_FP_EXC_PRECISE)
		return -EINVAL;
	tsk->thread.fpexc_mode = __pack_fe01(val);
	if (regs != NULL && (regs->msr & MSR_FP) != 0)
		regs->msr = (regs->msr & ~(MSR_FE0|MSR_FE1))
			| tsk->thread.fpexc_mode;
954 955 956 957 958 959 960 961 962
	return 0;
}

int get_fpexc_mode(struct task_struct *tsk, unsigned long adr)
{
	unsigned int val;

	if (tsk->thread.fpexc_mode & PR_FP_EXC_SW_ENABLE)
#ifdef CONFIG_SPE
963 964 965 966
		if (cpu_has_feature(CPU_FTR_SPE))
			val = tsk->thread.fpexc_mode;
		else
			return -EINVAL;
967 968 969 970 971 972 973 974
#else
		return -EINVAL;
#endif
	else
		val = __unpack_fe01(tsk->thread.fpexc_mode);
	return put_user(val, (unsigned int __user *) adr);
}

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 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018
int set_endian(struct task_struct *tsk, unsigned int val)
{
	struct pt_regs *regs = tsk->thread.regs;

	if ((val == PR_ENDIAN_LITTLE && !cpu_has_feature(CPU_FTR_REAL_LE)) ||
	    (val == PR_ENDIAN_PPC_LITTLE && !cpu_has_feature(CPU_FTR_PPC_LE)))
		return -EINVAL;

	if (regs == NULL)
		return -EINVAL;

	if (val == PR_ENDIAN_BIG)
		regs->msr &= ~MSR_LE;
	else if (val == PR_ENDIAN_LITTLE || val == PR_ENDIAN_PPC_LITTLE)
		regs->msr |= MSR_LE;
	else
		return -EINVAL;

	return 0;
}

int get_endian(struct task_struct *tsk, unsigned long adr)
{
	struct pt_regs *regs = tsk->thread.regs;
	unsigned int val;

	if (!cpu_has_feature(CPU_FTR_PPC_LE) &&
	    !cpu_has_feature(CPU_FTR_REAL_LE))
		return -EINVAL;

	if (regs == NULL)
		return -EINVAL;

	if (regs->msr & MSR_LE) {
		if (cpu_has_feature(CPU_FTR_REAL_LE))
			val = PR_ENDIAN_LITTLE;
		else
			val = PR_ENDIAN_PPC_LITTLE;
	} else
		val = PR_ENDIAN_BIG;

	return put_user(val, (unsigned int __user *)adr);
}

1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029
int set_unalign_ctl(struct task_struct *tsk, unsigned int val)
{
	tsk->thread.align_ctl = val;
	return 0;
}

int get_unalign_ctl(struct task_struct *tsk, unsigned long adr)
{
	return put_user(tsk->thread.align_ctl, (unsigned int __user *)adr);
}

1030 1031
#define TRUNC_PTR(x)	((typeof(x))(((unsigned long)(x)) & 0xffffffff))

1032 1033 1034 1035 1036 1037 1038 1039
int sys_clone(unsigned long clone_flags, unsigned long usp,
	      int __user *parent_tidp, void __user *child_threadptr,
	      int __user *child_tidp, int p6,
	      struct pt_regs *regs)
{
	CHECK_FULL_REGS(regs);
	if (usp == 0)
		usp = regs->gpr[1];	/* stack pointer for child */
1040
#ifdef CONFIG_PPC64
1041
	if (is_32bit_task()) {
1042 1043 1044 1045
		parent_tidp = TRUNC_PTR(parent_tidp);
		child_tidp = TRUNC_PTR(child_tidp);
	}
#endif
1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065
 	return do_fork(clone_flags, usp, regs, 0, parent_tidp, child_tidp);
}

int sys_fork(unsigned long p1, unsigned long p2, unsigned long p3,
	     unsigned long p4, unsigned long p5, unsigned long p6,
	     struct pt_regs *regs)
{
	CHECK_FULL_REGS(regs);
	return do_fork(SIGCHLD, regs->gpr[1], regs, 0, NULL, NULL);
}

int sys_vfork(unsigned long p1, unsigned long p2, unsigned long p3,
	      unsigned long p4, unsigned long p5, unsigned long p6,
	      struct pt_regs *regs)
{
	CHECK_FULL_REGS(regs);
	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->gpr[1],
			regs, 0, NULL, NULL);
}

1066 1067 1068 1069
void __ret_from_kernel_execve(struct pt_regs *normal)
__noreturn;

void ret_from_kernel_execve(struct pt_regs *normal)
1070
{
1071 1072
	set_thread_flag(TIF_RESTOREALL);
	__ret_from_kernel_execve(normal);
1073 1074
}

1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098
static inline int valid_irq_stack(unsigned long sp, struct task_struct *p,
				  unsigned long nbytes)
{
	unsigned long stack_page;
	unsigned long cpu = task_cpu(p);

	/*
	 * Avoid crashing if the stack has overflowed and corrupted
	 * task_cpu(p), which is in the thread_info struct.
	 */
	if (cpu < NR_CPUS && cpu_possible(cpu)) {
		stack_page = (unsigned long) hardirq_ctx[cpu];
		if (sp >= stack_page + sizeof(struct thread_struct)
		    && sp <= stack_page + THREAD_SIZE - nbytes)
			return 1;

		stack_page = (unsigned long) softirq_ctx[cpu];
		if (sp >= stack_page + sizeof(struct thread_struct)
		    && sp <= stack_page + THREAD_SIZE - nbytes)
			return 1;
	}
	return 0;
}

1099
int validate_sp(unsigned long sp, struct task_struct *p,
1100 1101
		       unsigned long nbytes)
{
A
Al Viro 已提交
1102
	unsigned long stack_page = (unsigned long)task_stack_page(p);
1103 1104 1105 1106 1107

	if (sp >= stack_page + sizeof(struct thread_struct)
	    && sp <= stack_page + THREAD_SIZE - nbytes)
		return 1;

1108
	return valid_irq_stack(sp, p, nbytes);
1109 1110
}

1111 1112
EXPORT_SYMBOL(validate_sp);

1113 1114 1115 1116 1117 1118 1119 1120 1121
unsigned long get_wchan(struct task_struct *p)
{
	unsigned long ip, sp;
	int count = 0;

	if (!p || p == current || p->state == TASK_RUNNING)
		return 0;

	sp = p->thread.ksp;
1122
	if (!validate_sp(sp, p, STACK_FRAME_OVERHEAD))
1123 1124 1125 1126
		return 0;

	do {
		sp = *(unsigned long *)sp;
1127
		if (!validate_sp(sp, p, STACK_FRAME_OVERHEAD))
1128 1129
			return 0;
		if (count > 0) {
1130
			ip = ((unsigned long *)sp)[STACK_FRAME_LR_SAVE];
1131 1132 1133 1134 1135 1136
			if (!in_sched_functions(ip))
				return ip;
		}
	} while (count++ < 16);
	return 0;
}
1137

1138
static int kstack_depth_to_print = CONFIG_PRINT_STACK_DEPTH;
1139 1140 1141 1142 1143 1144

void show_stack(struct task_struct *tsk, unsigned long *stack)
{
	unsigned long sp, ip, lr, newsp;
	int count = 0;
	int firstframe = 1;
1145 1146 1147
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
	int curr_frame = current->curr_ret_stack;
	extern void return_to_handler(void);
1148 1149
	unsigned long rth = (unsigned long)return_to_handler;
	unsigned long mrth = -1;
1150
#ifdef CONFIG_PPC64
1151 1152 1153 1154
	extern void mod_return_to_handler(void);
	rth = *(unsigned long *)rth;
	mrth = (unsigned long)mod_return_to_handler;
	mrth = *(unsigned long *)mrth;
1155 1156
#endif
#endif
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170

	sp = (unsigned long) stack;
	if (tsk == NULL)
		tsk = current;
	if (sp == 0) {
		if (tsk == current)
			asm("mr %0,1" : "=r" (sp));
		else
			sp = tsk->thread.ksp;
	}

	lr = 0;
	printk("Call Trace:\n");
	do {
1171
		if (!validate_sp(sp, tsk, STACK_FRAME_OVERHEAD))
1172 1173 1174 1175
			return;

		stack = (unsigned long *) sp;
		newsp = stack[0];
1176
		ip = stack[STACK_FRAME_LR_SAVE];
1177
		if (!firstframe || ip != lr) {
1178
			printk("["REG"] ["REG"] %pS", sp, ip, (void *)ip);
1179
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
1180
			if ((ip == rth || ip == mrth) && curr_frame >= 0) {
1181 1182 1183 1184 1185
				printk(" (%pS)",
				       (void *)current->ret_stack[curr_frame].ret);
				curr_frame--;
			}
#endif
1186 1187 1188 1189 1190 1191 1192 1193 1194 1195
			if (firstframe)
				printk(" (unreliable)");
			printk("\n");
		}
		firstframe = 0;

		/*
		 * See if this is an exception frame.
		 * We look for the "regshere" marker in the current frame.
		 */
1196 1197
		if (validate_sp(sp, tsk, STACK_INT_FRAME_SIZE)
		    && stack[STACK_FRAME_MARKER] == STACK_FRAME_REGS_MARKER) {
1198 1199 1200
			struct pt_regs *regs = (struct pt_regs *)
				(sp + STACK_FRAME_OVERHEAD);
			lr = regs->link;
1201 1202
			printk("--- Exception: %lx at %pS\n    LR = %pS\n",
			       regs->trap, (void *)regs->nip, (void *)lr);
1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214
			firstframe = 1;
		}

		sp = newsp;
	} while (count++ < kstack_depth_to_print);
}

void dump_stack(void)
{
	show_stack(current, NULL);
}
EXPORT_SYMBOL(dump_stack);
1215 1216

#ifdef CONFIG_PPC64
1217 1218
/* Called with hard IRQs off */
void __ppc64_runlatch_on(void)
1219
{
1220
	struct thread_info *ti = current_thread_info();
1221 1222
	unsigned long ctrl;

1223 1224 1225
	ctrl = mfspr(SPRN_CTRLF);
	ctrl |= CTRL_RUNLATCH;
	mtspr(SPRN_CTRLT, ctrl);
1226

1227
	ti->local_flags |= _TLF_RUNLATCH;
1228 1229
}

1230
/* Called with hard IRQs off */
1231
void __ppc64_runlatch_off(void)
1232
{
1233
	struct thread_info *ti = current_thread_info();
1234 1235
	unsigned long ctrl;

1236
	ti->local_flags &= ~_TLF_RUNLATCH;
1237

1238 1239 1240
	ctrl = mfspr(SPRN_CTRLF);
	ctrl &= ~CTRL_RUNLATCH;
	mtspr(SPRN_CTRLT, ctrl);
1241
}
1242
#endif /* CONFIG_PPC64 */
1243

1244 1245 1246 1247 1248 1249
unsigned long arch_align_stack(unsigned long sp)
{
	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
		sp -= get_random_int() & ~PAGE_MASK;
	return sp & ~0xf;
}
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265

static inline unsigned long brk_rnd(void)
{
        unsigned long rnd = 0;

	/* 8MB for 32bit, 1GB for 64bit */
	if (is_32bit_task())
		rnd = (long)(get_random_int() % (1<<(23-PAGE_SHIFT)));
	else
		rnd = (long)(get_random_int() % (1<<(30-PAGE_SHIFT)));

	return rnd << PAGE_SHIFT;
}

unsigned long arch_randomize_brk(struct mm_struct *mm)
{
1266 1267 1268
	unsigned long base = mm->brk;
	unsigned long ret;

1269
#ifdef CONFIG_PPC_STD_MMU_64
1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281
	/*
	 * If we are using 1TB segments and we are allowed to randomise
	 * the heap, we can put it above 1TB so it is backed by a 1TB
	 * segment. Otherwise the heap will be in the bottom 1TB
	 * which always uses 256MB segments and this may result in a
	 * performance penalty.
	 */
	if (!is_32bit_task() && (mmu_highuser_ssize == MMU_SEGSIZE_1T))
		base = max_t(unsigned long, mm->brk, 1UL << SID_SHIFT_1T);
#endif

	ret = PAGE_ALIGN(base + brk_rnd());
1282 1283 1284 1285 1286 1287

	if (ret < mm->brk)
		return mm->brk;

	return ret;
}
A
Anton Blanchard 已提交
1288 1289 1290 1291 1292 1293 1294 1295 1296 1297

unsigned long randomize_et_dyn(unsigned long base)
{
	unsigned long ret = PAGE_ALIGN(base + brk_rnd());

	if (ret < base)
		return base;

	return ret;
}