process.c 37.0 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
#include <asm/switch_to.h>
53
#include <asm/tm.h>
54
#include <asm/debug.h>
55 56 57
#ifdef CONFIG_PPC64
#include <asm/firmware.h>
#endif
58 59
#include <linux/kprobes.h>
#include <linux/kdebug.h>
60

61 62 63 64 65 66 67
/* Transactional Memory debug */
#ifdef TM_DEBUG_SW
#define TM_DEBUG(x...) printk(KERN_INFO x)
#else
#define TM_DEBUG(x...) do { } while(0)
#endif

68 69 70 71 72
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;
73
struct task_struct *last_task_used_vsx = NULL;
74 75 76
struct task_struct *last_task_used_spe = NULL;
#endif

77
#ifdef CONFIG_PPC_FPU
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
/*
 * 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
105
			giveup_fpu(tsk);
106 107 108 109
		}
		preempt_enable();
	}
}
110
EXPORT_SYMBOL_GPL(flush_fp_to_thread);
111
#endif
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

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
137
		giveup_altivec_notask();
138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155
#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
156
			giveup_altivec(tsk);
157 158 159 160
		}
		preempt_enable();
	}
}
161
EXPORT_SYMBOL_GPL(flush_altivec_to_thread);
162 163
#endif /* CONFIG_ALTIVEC */

164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182
#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

183 184 185 186 187 188 189
void giveup_vsx(struct task_struct *tsk)
{
	giveup_fpu(tsk);
	giveup_altivec(tsk);
	__giveup_vsx(tsk);
}

190 191 192 193 194 195 196 197 198 199 200 201 202
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();
	}
}
203
EXPORT_SYMBOL_GPL(flush_vsx_to_thread);
204 205
#endif /* CONFIG_VSX */

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
#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
231
			tsk->thread.spefscr = mfspr(SPRN_SPEFSCR);
232
			giveup_spe(tsk);
233 234 235 236 237 238
		}
		preempt_enable();
	}
}
#endif /* CONFIG_SPE */

239
#ifndef CONFIG_SMP
240 241 242 243
/*
 * If we are doing lazy switching of CPU state (FP, altivec or SPE),
 * and the current task has some state, discard it.
 */
244
void discard_lazy_cpu_state(void)
245 246 247 248 249 250 251 252
{
	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 */
253 254 255 256
#ifdef CONFIG_VSX
	if (last_task_used_vsx == current)
		last_task_used_vsx = NULL;
#endif /* CONFIG_VSX */
257 258 259 260 261 262
#ifdef CONFIG_SPE
	if (last_task_used_spe == current)
		last_task_used_spe = NULL;
#endif
	preempt_enable();
}
263
#endif /* CONFIG_SMP */
264

265 266 267 268 269 270
#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;

271
	current->thread.trap_nr = signal_code;
272 273 274 275 276 277 278 279 280 281 282 283
	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 */
284
void do_break (struct pt_regs *regs, unsigned long address,
285 286 287 288
		    unsigned long error_code)
{
	siginfo_t info;

289
	current->thread.trap_nr = TRAP_HWBKPT;
290 291 292 293
	if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code,
			11, SIGSEGV) == NOTIFY_STOP)
		return;

294
	if (debugger_break_match(regs))
295 296
		return;

297 298
	/* Clear the breakpoint */
	hw_breakpoint_disable();
299 300 301 302 303 304 305 306

	/* 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);
}
307
#endif	/* CONFIG_PPC_ADV_DEBUG_REGS */
308

309
static DEFINE_PER_CPU(struct arch_hw_breakpoint, current_brk);
310

311 312 313 314 315 316
#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)
{
317
	thread->debug.iac1 = thread->debug.iac2 = 0;
318
#if CONFIG_PPC_ADV_DEBUG_IACS > 2
319
	thread->debug.iac3 = thread->debug.iac4 = 0;
320
#endif
321
	thread->debug.dac1 = thread->debug.dac2 = 0;
322
#if CONFIG_PPC_ADV_DEBUG_DVCS > 0
323
	thread->debug.dvc1 = thread->debug.dvc2 = 0;
324
#endif
325
	thread->debug.dbcr0 = 0;
326 327 328 329
#ifdef CONFIG_BOOKE
	/*
	 * Force User/Supervisor bits to b11 (user-only MSR[PR]=1)
	 */
330
	thread->debug.dbcr1 = DBCR1_IAC1US | DBCR1_IAC2US |
331 332 333 334 335
			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.
	 */
336
	thread->debug.dbcr2 = DBCR2_DAC1US | DBCR2_DAC2US;
337
#else
338
	thread->debug.dbcr1 = 0;
339 340 341 342 343
#endif
}

static void prime_debug_regs(struct thread_struct *thread)
{
344 345 346 347 348 349 350
	/*
	 * We could have inherited MSR_DE from userspace, since
	 * it doesn't get cleared on exception entry.  Make sure
	 * MSR_DE is clear before we enable any debug events.
	 */
	mtmsr(mfmsr() & ~MSR_DE);

351 352
	mtspr(SPRN_IAC1, thread->debug.iac1);
	mtspr(SPRN_IAC2, thread->debug.iac2);
353
#if CONFIG_PPC_ADV_DEBUG_IACS > 2
354 355
	mtspr(SPRN_IAC3, thread->debug.iac3);
	mtspr(SPRN_IAC4, thread->debug.iac4);
356
#endif
357 358
	mtspr(SPRN_DAC1, thread->debug.dac1);
	mtspr(SPRN_DAC2, thread->debug.dac2);
359
#if CONFIG_PPC_ADV_DEBUG_DVCS > 0
360 361
	mtspr(SPRN_DVC1, thread->debug.dvc1);
	mtspr(SPRN_DVC2, thread->debug.dvc2);
362
#endif
363 364
	mtspr(SPRN_DBCR0, thread->debug.dbcr0);
	mtspr(SPRN_DBCR1, thread->debug.dbcr1);
365
#ifdef CONFIG_BOOKE
366
	mtspr(SPRN_DBCR2, thread->debug.dbcr2);
367 368 369 370 371 372 373 374 375
#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)
{
376 377
	if ((current->thread.debug.dbcr0 & DBCR0_IDM)
		|| (new_thread->debug.dbcr0 & DBCR0_IDM))
378 379 380
			prime_debug_regs(new_thread);
}
#else	/* !CONFIG_PPC_ADV_DEBUG_REGS */
381
#ifndef CONFIG_HAVE_HW_BREAKPOINT
382 383
static void set_debug_reg_defaults(struct thread_struct *thread)
{
384 385
	thread->hw_brk.address = 0;
	thread->hw_brk.type = 0;
386
	set_breakpoint(&thread->hw_brk);
387
}
388
#endif /* !CONFIG_HAVE_HW_BREAKPOINT */
389 390
#endif	/* CONFIG_PPC_ADV_DEBUG_REGS */

391
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
392 393
static inline int __set_dabr(unsigned long dabr, unsigned long dabrx)
{
394
	mtspr(SPRN_DAC1, dabr);
395 396 397
#ifdef CONFIG_PPC_47x
	isync();
#endif
398 399
	return 0;
}
400
#elif defined(CONFIG_PPC_BOOK3S)
401 402
static inline int __set_dabr(unsigned long dabr, unsigned long dabrx)
{
403
	mtspr(SPRN_DABR, dabr);
404 405
	if (cpu_has_feature(CPU_FTR_DABRX))
		mtspr(SPRN_DABRX, dabrx);
406
	return 0;
407
}
408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427
#else
static inline int __set_dabr(unsigned long dabr, unsigned long dabrx)
{
	return -EINVAL;
}
#endif

static inline int set_dabr(struct arch_hw_breakpoint *brk)
{
	unsigned long dabr, dabrx;

	dabr = brk->address | (brk->type & HW_BRK_TYPE_DABR);
	dabrx = ((brk->type >> 3) & 0x7);

	if (ppc_md.set_dabr)
		return ppc_md.set_dabr(dabr, dabrx);

	return __set_dabr(dabr, dabrx);
}

428 429
static inline int set_dawr(struct arch_hw_breakpoint *brk)
{
430
	unsigned long dawr, dawrx, mrd;
431 432 433 434 435 436 437 438 439

	dawr = brk->address;

	dawrx  = (brk->type & (HW_BRK_TYPE_READ | HW_BRK_TYPE_WRITE)) \
		                   << (63 - 58); //* read/write bits */
	dawrx |= ((brk->type & (HW_BRK_TYPE_TRANSLATE)) >> 2) \
		                   << (63 - 59); //* translate */
	dawrx |= (brk->type & (HW_BRK_TYPE_PRIV_ALL)) \
		                   >> 3; //* PRIM bits */
440 441 442 443 444 445 446 447
	/* dawr length is stored in field MDR bits 48:53.  Matches range in
	   doublewords (64 bits) baised by -1 eg. 0b000000=1DW and
	   0b111111=64DW.
	   brk->len is in bytes.
	   This aligns up to double word size, shifts and does the bias.
	*/
	mrd = ((brk->len + 7) >> 3) - 1;
	dawrx |= (mrd & 0x3f) << (63 - 53);
448 449 450 451 452 453 454 455

	if (ppc_md.set_dawr)
		return ppc_md.set_dawr(dawr, dawrx);
	mtspr(SPRN_DAWR, dawr);
	mtspr(SPRN_DAWRX, dawrx);
	return 0;
}

456
int set_breakpoint(struct arch_hw_breakpoint *brk)
457 458 459
{
	__get_cpu_var(current_brk) = *brk;

460 461 462
	if (cpu_has_feature(CPU_FTR_DAWR))
		return set_dawr(brk);

463 464
	return set_dabr(brk);
}
465

466 467 468
#ifdef CONFIG_PPC64
DEFINE_PER_CPU(struct cpu_usage, cpu_usage_array);
#endif
469

470 471 472 473 474 475 476 477 478 479 480
static inline bool hw_brk_match(struct arch_hw_breakpoint *a,
			      struct arch_hw_breakpoint *b)
{
	if (a->address != b->address)
		return false;
	if (a->type != b->type)
		return false;
	if (a->len != b->len)
		return false;
	return true;
}
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
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
static inline void tm_reclaim_task(struct task_struct *tsk)
{
	/* We have to work out if we're switching from/to a task that's in the
	 * middle of a transaction.
	 *
	 * In switching we need to maintain a 2nd register state as
	 * oldtask->thread.ckpt_regs.  We tm_reclaim(oldproc); this saves the
	 * checkpointed (tbegin) state in ckpt_regs and saves the transactional
	 * (current) FPRs into oldtask->thread.transact_fpr[].
	 *
	 * We also context switch (save) TFHAR/TEXASR/TFIAR in here.
	 */
	struct thread_struct *thr = &tsk->thread;

	if (!thr->regs)
		return;

	if (!MSR_TM_ACTIVE(thr->regs->msr))
		goto out_and_saveregs;

	/* Stash the original thread MSR, as giveup_fpu et al will
	 * modify it.  We hold onto it to see whether the task used
	 * FP & vector regs.
	 */
	thr->tm_orig_msr = thr->regs->msr;

	TM_DEBUG("--- tm_reclaim on pid %d (NIP=%lx, "
		 "ccr=%lx, msr=%lx, trap=%lx)\n",
		 tsk->pid, thr->regs->nip,
		 thr->regs->ccr, thr->regs->msr,
		 thr->regs->trap);

	tm_reclaim(thr, thr->regs->msr, TM_CAUSE_RESCHED);

	TM_DEBUG("--- tm_reclaim on pid %d complete\n",
		 tsk->pid);

out_and_saveregs:
	/* Always save the regs here, even if a transaction's not active.
	 * This context-switches a thread's TM info SPRs.  We do it here to
	 * be consistent with the restore path (in recheckpoint) which
	 * cannot happen later in _switch().
	 */
	tm_save_sprs(thr);
}

528
static inline void tm_recheckpoint_new_task(struct task_struct *new)
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
{
	unsigned long msr;

	if (!cpu_has_feature(CPU_FTR_TM))
		return;

	/* Recheckpoint the registers of the thread we're about to switch to.
	 *
	 * If the task was using FP, we non-lazily reload both the original and
	 * the speculative FP register states.  This is because the kernel
	 * doesn't see if/when a TM rollback occurs, so if we take an FP
	 * unavoidable later, we are unable to determine which set of FP regs
	 * need to be restored.
	 */
	if (!new->thread.regs)
		return;

	/* The TM SPRs are restored here, so that TEXASR.FS can be set
	 * before the trecheckpoint and no explosion occurs.
	 */
	tm_restore_sprs(&new->thread);

	if (!MSR_TM_ACTIVE(new->thread.regs->msr))
		return;
	msr = new->thread.tm_orig_msr;
	/* Recheckpoint to restore original checkpointed register state. */
	TM_DEBUG("*** tm_recheckpoint of pid %d "
		 "(new->msr 0x%lx, new->origmsr 0x%lx)\n",
		 new->pid, new->thread.regs->msr, msr);

	/* This loads the checkpointed FP/VEC state, if used */
	tm_recheckpoint(&new->thread, msr);

	/* This loads the speculative FP/VEC state, if used */
	if (msr & MSR_FP) {
		do_load_up_transact_fpu(&new->thread);
		new->thread.regs->msr |=
			(MSR_FP | new->thread.fpexc_mode);
	}
568
#ifdef CONFIG_ALTIVEC
569 570 571 572
	if (msr & MSR_VEC) {
		do_load_up_transact_altivec(&new->thread);
		new->thread.regs->msr |= MSR_VEC;
	}
573
#endif
574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593
	/* We may as well turn on VSX too since all the state is restored now */
	if (msr & MSR_VSX)
		new->thread.regs->msr |= MSR_VSX;

	TM_DEBUG("*** tm_recheckpoint of pid %d complete "
		 "(kernel msr 0x%lx)\n",
		 new->pid, mfmsr());
}

static inline void __switch_to_tm(struct task_struct *prev)
{
	if (cpu_has_feature(CPU_FTR_TM)) {
		tm_enable();
		tm_reclaim_task(prev);
	}
}
#else
#define tm_recheckpoint_new_task(new)
#define __switch_to_tm(prev)
#endif /* CONFIG_PPC_TRANSACTIONAL_MEM */
594

595 596 597 598 599 600
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
601 602 603
#ifdef CONFIG_PPC_BOOK3S_64
	struct ppc64_tlb_batch *batch;
#endif
604

605 606 607 608 609 610 611 612 613 614
	/* Back up the TAR across context switches.
	 * Note that the TAR is not available for use in the kernel.  (To
	 * provide this, the TAR should be backed up/restored on exception
	 * entry/exit instead, and be in pt_regs.  FIXME, this should be in
	 * pt_regs anyway (for debug).)
	 * Save the TAR here before we do treclaim/trecheckpoint as these
	 * will change the TAR.
	 */
	save_tar(&prev->thread);

615 616
	__switch_to_tm(prev);

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
#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 */
644 645
#ifdef CONFIG_VSX
	if (prev->thread.regs && (prev->thread.regs->msr & MSR_VSX))
646 647
		/* VMX and FPU registers are already save here */
		__giveup_vsx(prev);
648
#endif /* CONFIG_VSX */
649 650 651 652 653 654 655 656 657 658
#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);
659 660 661 662 663 664 665 666 667 668
#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 */
669 670 671 672
#ifdef CONFIG_VSX
	if (new->thread.regs && last_task_used_vsx == new)
		new->thread.regs->msr |= MSR_VSX;
#endif /* CONFIG_VSX */
673
#ifdef CONFIG_SPE
674 675 676 677 678 679
	/* 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 */
680

681 682
#endif /* CONFIG_SMP */

683
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
684
	switch_booke_debug_regs(&new->thread);
685
#else
686 687 688 689 690
/*
 * For PPC_BOOK3S_64, we use the hw-breakpoint interfaces that would
 * schedule DABR
 */
#ifndef CONFIG_HAVE_HW_BREAKPOINT
691
	if (unlikely(hw_brk_match(&__get_cpu_var(current_brk), &new->thread.hw_brk)))
692
		set_breakpoint(&new->thread.hw_brk);
693
#endif /* CONFIG_HAVE_HW_BREAKPOINT */
694 695
#endif

696

697 698
	new_thread = &new->thread;
	old_thread = &current->thread;
699 700 701 702 703 704 705 706 707 708 709 710 711

#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
712 713 714 715 716 717 718 719 720 721 722
#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 */
723

724
	local_irq_save(flags);
725

726 727 728 729 730 731
	/*
	 * 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();
732 733 734

	tm_recheckpoint_new_task(new);

735 736
	last = _switch(old_thread, new_thread);

P
Peter Zijlstra 已提交
737 738 739 740 741 742 743 744
#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 */

745 746 747 748 749
	local_irq_restore(flags);

	return last;
}

750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765
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");

766 767 768 769 770 771 772 773
#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

774 775 776 777
		/* We use __get_user here *only* to avoid an OOPS on a
		 * bad address because the pc *should* only be a
		 * kernel address.
		 */
778 779
		if (!__kernel_text_address(pc) ||
		     __get_user(instr, (unsigned int __user *)pc)) {
780
			printk(KERN_CONT "XXXXXXXX ");
781 782
		} else {
			if (regs->nip == pc)
783
				printk(KERN_CONT "<%08x> ", instr);
784
			else
785
				printk(KERN_CONT "%08x ", instr);
786 787 788 789 790 791 792 793 794 795 796 797
		}

		pc += sizeof(int);
	}

	printk("\n");
}

static struct regbit {
	unsigned long bit;
	const char *name;
} msr_bits[] = {
798 799 800 801 802 803 804 805 806
#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
807 808 809 810
	{MSR_EE,	"EE"},
	{MSR_PR,	"PR"},
	{MSR_FP,	"FP"},
	{MSR_ME,	"ME"},
811
#ifdef CONFIG_BOOKE
812
	{MSR_DE,	"DE"},
813 814 815 816
#else
	{MSR_SE,	"SE"},
	{MSR_BE,	"BE"},
#endif
817 818
	{MSR_IR,	"IR"},
	{MSR_DR,	"DR"},
819 820 821 822 823
	{MSR_PMM,	"PMM"},
#ifndef CONFIG_BOOKE
	{MSR_RI,	"RI"},
	{MSR_LE,	"LE"},
#endif
824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840
	{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
841
#define REG		"%016lx"
842 843 844
#define REGS_PER_LINE	4
#define LAST_VOLATILE	13
#else
845
#define REG		"%08lx"
846 847 848 849
#define REGS_PER_LINE	8
#define LAST_VOLATILE	12
#endif

850 851 852 853
void show_regs(struct pt_regs * regs)
{
	int i, trap;

854 855
	show_regs_print_info(KERN_DEFAULT);

856 857 858
	printk("NIP: "REG" LR: "REG" CTR: "REG"\n",
	       regs->nip, regs->link, regs->ctr);
	printk("REGS: %p TRAP: %04lx   %s  (%s)\n",
859
	       regs, regs->trap, print_tainted(), init_utsname()->release);
860 861
	printk("MSR: "REG" ", regs->msr);
	printbits(regs->msr, msr_bits);
862
	printk("  CR: %08lx  XER: %08lx\n", regs->ccr, regs->xer);
863 864 865
#ifdef CONFIG_PPC64
	printk("SOFTE: %ld\n", regs->softe);
#endif
866
	trap = TRAP(regs);
867 868
	if ((regs->trap != 0xc00) && cpu_has_feature(CPU_FTR_CFAR))
		printk("CFAR: "REG"\n", regs->orig_gpr3);
869
	if (trap == 0x300 || trap == 0x600)
870
#if defined(CONFIG_4xx) || defined(CONFIG_BOOKE)
871 872
		printk("DEAR: "REG", ESR: "REG"\n", regs->dar, regs->dsisr);
#else
873
		printk("DAR: "REG", DSISR: %08lx\n", regs->dar, regs->dsisr);
874
#endif
875 876

	for (i = 0;  i < 32;  i++) {
877
		if ((i % REGS_PER_LINE) == 0)
K
Kumar Gala 已提交
878
			printk("\nGPR%02d: ", i);
879 880
		printk(REG " ", regs->gpr[i]);
		if (i == LAST_VOLATILE && !FULL_REGS(regs))
881 882 883 884 885 886 887 888
			break;
	}
	printk("\n");
#ifdef CONFIG_KALLSYMS
	/*
	 * Lookup NIP late so we have the best change of getting the
	 * above info out without failing
	 */
889 890
	printk("NIP ["REG"] %pS\n", regs->nip, (void *)regs->nip);
	printk("LR ["REG"] %pS\n", regs->link, (void *)regs->link);
891
#endif
892 893 894
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
	printk("PACATMSCRATCH [%llx]\n", get_paca()->tm_scratch);
#endif
895
	show_stack(current, (unsigned long *) regs->gpr[1]);
896 897
	if (!user_mode(regs))
		show_instructions(regs);
898 899 900 901
}

void exit_thread(void)
{
902
	discard_lazy_cpu_state();
903 904 905 906
}

void flush_thread(void)
{
907
	discard_lazy_cpu_state();
908

909
#ifdef CONFIG_HAVE_HW_BREAKPOINT
910
	flush_ptrace_hw_breakpoint(current);
911
#else /* CONFIG_HAVE_HW_BREAKPOINT */
912
	set_debug_reg_defaults(&current->thread);
913
#endif /* CONFIG_HAVE_HW_BREAKPOINT */
914 915 916 917 918 919 920 921
}

void
release_thread(struct task_struct *t)
{
}

/*
922 923
 * this gets called so that we can store coprocessor state into memory and
 * copy the current task into the new thread.
924
 */
925
int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
926
{
927 928 929 930
	flush_fp_to_thread(src);
	flush_altivec_to_thread(src);
	flush_vsx_to_thread(src);
	flush_spe_to_thread(src);
931

932
	*dst = *src;
933 934 935

	clear_task_ebb(dst);

936
	return 0;
937 938 939 940 941
}

/*
 * Copy a thread..
 */
942 943
extern unsigned long dscr_default; /* defined in arch/powerpc/kernel/sysfs.c */

A
Alexey Dobriyan 已提交
944
int copy_thread(unsigned long clone_flags, unsigned long usp,
945
		unsigned long arg, struct task_struct *p)
946 947 948
{
	struct pt_regs *childregs, *kregs;
	extern void ret_from_fork(void);
A
Al Viro 已提交
949 950
	extern void ret_from_kernel_thread(void);
	void (*f)(void);
A
Al Viro 已提交
951
	unsigned long sp = (unsigned long)task_stack_page(p) + THREAD_SIZE;
952 953 954 955

	/* Copy registers */
	sp -= sizeof(struct pt_regs);
	childregs = (struct pt_regs *) sp;
956
	if (unlikely(p->flags & PF_KTHREAD)) {
957
		struct thread_info *ti = (void *)task_stack_page(p);
A
Al Viro 已提交
958
		memset(childregs, 0, sizeof(struct pt_regs));
959
		childregs->gpr[1] = sp + sizeof(struct pt_regs);
960
		childregs->gpr[14] = usp;	/* function */
A
Al Viro 已提交
961
#ifdef CONFIG_PPC64
A
Al Viro 已提交
962
		clear_tsk_thread_flag(p, TIF_32BIT);
963
		childregs->softe = 1;
964
#endif
A
Al Viro 已提交
965
		childregs->gpr[15] = arg;
966
		p->thread.regs = NULL;	/* no user register state */
967
		ti->flags |= _TIF_RESTOREALL;
A
Al Viro 已提交
968
		f = ret_from_kernel_thread;
969
	} else {
970
		struct pt_regs *regs = current_pt_regs();
A
Al Viro 已提交
971 972
		CHECK_FULL_REGS(regs);
		*childregs = *regs;
973 974
		if (usp)
			childregs->gpr[1] = usp;
975
		p->thread.regs = childregs;
A
Al Viro 已提交
976
		childregs->gpr[3] = 0;  /* Result from fork() */
977 978
		if (clone_flags & CLONE_SETTLS) {
#ifdef CONFIG_PPC64
979
			if (!is_32bit_task())
980 981 982 983 984
				childregs->gpr[13] = childregs->gpr[6];
			else
#endif
				childregs->gpr[2] = childregs->gpr[6];
		}
A
Al Viro 已提交
985 986

		f = ret_from_fork;
987 988 989 990 991 992 993 994 995 996 997
	}
	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.
	 */
998
	((unsigned long *)sp)[0] = 0;
999 1000 1001 1002
	sp -= sizeof(struct pt_regs);
	kregs = (struct pt_regs *) sp;
	sp -= STACK_FRAME_OVERHEAD;
	p->thread.ksp = sp;
1003
#ifdef CONFIG_PPC32
1004 1005
	p->thread.ksp_limit = (unsigned long)task_stack_page(p) +
				_ALIGN_UP(sizeof(struct thread_info), 16);
1006
#endif
1007 1008 1009 1010
#ifdef CONFIG_HAVE_HW_BREAKPOINT
	p->thread.ptrace_bps[0] = NULL;
#endif

1011 1012 1013 1014 1015
	p->thread.fp_save_area = NULL;
#ifdef CONFIG_ALTIVEC
	p->thread.vr_save_area = NULL;
#endif

1016
#ifdef CONFIG_PPC_STD_MMU_64
1017
	if (mmu_has_feature(MMU_FTR_SLB)) {
P
Paul Mackerras 已提交
1018
		unsigned long sp_vsid;
1019
		unsigned long llp = mmu_psize_defs[mmu_linear_psize].sllp;
1020

1021
		if (mmu_has_feature(MMU_FTR_1T_SEGMENT))
P
Paul Mackerras 已提交
1022 1023 1024 1025 1026
			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;
1027
		sp_vsid |= SLB_VSID_KERNEL | llp;
1028 1029
		p->thread.ksp_vsid = sp_vsid;
	}
1030
#endif /* CONFIG_PPC_STD_MMU_64 */
1031 1032
#ifdef CONFIG_PPC64 
	if (cpu_has_feature(CPU_FTR_DSCR)) {
1033 1034
		p->thread.dscr_inherit = current->thread.dscr_inherit;
		p->thread.dscr = current->thread.dscr;
1035
	}
1036 1037
	if (cpu_has_feature(CPU_FTR_HAS_PPR))
		p->thread.ppr = INIT_PPR;
1038
#endif
1039 1040 1041 1042 1043
	/*
	 * 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
1044
	 */
1045
#ifdef CONFIG_PPC64
A
Al Viro 已提交
1046
	kregs->nip = *((unsigned long *)f);
1047
#else
A
Al Viro 已提交
1048
	kregs->nip = (unsigned long)f;
1049
#endif
1050 1051 1052 1053 1054 1055
	return 0;
}

/*
 * Set up a thread for executing a new program
 */
1056
void start_thread(struct pt_regs *regs, unsigned long start, unsigned long sp)
1057
{
1058 1059 1060 1061
#ifdef CONFIG_PPC64
	unsigned long load_addr = regs->gpr[2];	/* saved by ELF_PLAT_INIT */
#endif

1062 1063 1064 1065 1066
	/*
	 * 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
1067 1068
		struct pt_regs *regs = task_stack_page(current) + THREAD_SIZE;
		current->thread.regs = regs - 1;
1069 1070
	}

1071 1072 1073 1074 1075 1076
	memset(regs->gpr, 0, sizeof(regs->gpr));
	regs->ctr = 0;
	regs->link = 0;
	regs->xer = 0;
	regs->ccr = 0;
	regs->gpr[1] = sp;
1077

1078 1079 1080 1081 1082 1083 1084
	/*
	 * 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;

1085 1086 1087
#ifdef CONFIG_PPC32
	regs->mq = 0;
	regs->nip = start;
1088
	regs->msr = MSR_USER;
1089
#else
1090
	if (!is_32bit_task()) {
1091
		unsigned long entry, toc;
1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110

		/* 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
1111 1112 1113 1114
	} else {
		regs->nip = start;
		regs->gpr[2] = 0;
		regs->msr = MSR_USER32;
1115 1116
	}
#endif
1117
	discard_lazy_cpu_state();
1118 1119 1120
#ifdef CONFIG_VSX
	current->thread.used_vsr = 0;
#endif
1121
	memset(&current->thread.fp_state, 0, sizeof(current->thread.fp_state));
1122
	current->thread.fp_save_area = NULL;
1123
#ifdef CONFIG_ALTIVEC
1124 1125
	memset(&current->thread.vr_state, 0, sizeof(current->thread.vr_state));
	current->thread.vr_state.vscr.u[3] = 0x00010000; /* Java mode disabled */
1126
	current->thread.vr_save_area = NULL;
1127 1128 1129 1130 1131 1132 1133 1134 1135
	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 */
1136 1137 1138 1139 1140 1141 1142
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
	if (cpu_has_feature(CPU_FTR_TM))
		regs->msr |= MSR_TM;
	current->thread.tm_tfhar = 0;
	current->thread.tm_texasr = 0;
	current->thread.tm_tfiar = 0;
#endif /* CONFIG_PPC_TRANSACTIONAL_MEM */
1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157
}

#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
1158 1159 1160 1161 1162 1163 1164
		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;
		}
1165 1166 1167 1168
#else
		return -EINVAL;
#endif
	}
1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180

	/* 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;
1181 1182 1183 1184 1185 1186 1187 1188 1189
	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
1190 1191 1192 1193
		if (cpu_has_feature(CPU_FTR_SPE))
			val = tsk->thread.fpexc_mode;
		else
			return -EINVAL;
1194 1195 1196 1197 1198 1199 1200 1201
#else
		return -EINVAL;
#endif
	else
		val = __unpack_fe01(tsk->thread.fpexc_mode);
	return put_user(val, (unsigned int __user *) adr);
}

1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245
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);
}

1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256
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);
}

1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280
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;
}

1281
int validate_sp(unsigned long sp, struct task_struct *p,
1282 1283
		       unsigned long nbytes)
{
A
Al Viro 已提交
1284
	unsigned long stack_page = (unsigned long)task_stack_page(p);
1285 1286 1287 1288 1289

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

1290
	return valid_irq_stack(sp, p, nbytes);
1291 1292
}

1293 1294
EXPORT_SYMBOL(validate_sp);

1295 1296 1297 1298 1299 1300 1301 1302 1303
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;
1304
	if (!validate_sp(sp, p, STACK_FRAME_OVERHEAD))
1305 1306 1307 1308
		return 0;

	do {
		sp = *(unsigned long *)sp;
1309
		if (!validate_sp(sp, p, STACK_FRAME_OVERHEAD))
1310 1311
			return 0;
		if (count > 0) {
1312
			ip = ((unsigned long *)sp)[STACK_FRAME_LR_SAVE];
1313 1314 1315 1316 1317 1318
			if (!in_sched_functions(ip))
				return ip;
		}
	} while (count++ < 16);
	return 0;
}
1319

1320
static int kstack_depth_to_print = CONFIG_PRINT_STACK_DEPTH;
1321 1322 1323 1324 1325 1326

void show_stack(struct task_struct *tsk, unsigned long *stack)
{
	unsigned long sp, ip, lr, newsp;
	int count = 0;
	int firstframe = 1;
1327 1328 1329
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
	int curr_frame = current->curr_ret_stack;
	extern void return_to_handler(void);
1330 1331
	unsigned long rth = (unsigned long)return_to_handler;
	unsigned long mrth = -1;
1332
#ifdef CONFIG_PPC64
1333 1334 1335 1336
	extern void mod_return_to_handler(void);
	rth = *(unsigned long *)rth;
	mrth = (unsigned long)mod_return_to_handler;
	mrth = *(unsigned long *)mrth;
1337 1338
#endif
#endif
1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352

	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 {
1353
		if (!validate_sp(sp, tsk, STACK_FRAME_OVERHEAD))
1354 1355 1356 1357
			return;

		stack = (unsigned long *) sp;
		newsp = stack[0];
1358
		ip = stack[STACK_FRAME_LR_SAVE];
1359
		if (!firstframe || ip != lr) {
1360
			printk("["REG"] ["REG"] %pS", sp, ip, (void *)ip);
1361
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
1362
			if ((ip == rth || ip == mrth) && curr_frame >= 0) {
1363 1364 1365 1366 1367
				printk(" (%pS)",
				       (void *)current->ret_stack[curr_frame].ret);
				curr_frame--;
			}
#endif
1368 1369 1370 1371 1372 1373 1374 1375 1376 1377
			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.
		 */
1378 1379
		if (validate_sp(sp, tsk, STACK_INT_FRAME_SIZE)
		    && stack[STACK_FRAME_MARKER] == STACK_FRAME_REGS_MARKER) {
1380 1381 1382
			struct pt_regs *regs = (struct pt_regs *)
				(sp + STACK_FRAME_OVERHEAD);
			lr = regs->link;
1383 1384
			printk("--- Exception: %lx at %pS\n    LR = %pS\n",
			       regs->trap, (void *)regs->nip, (void *)lr);
1385 1386 1387 1388 1389 1390 1391
			firstframe = 1;
		}

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

1392
#ifdef CONFIG_PPC64
1393
/* Called with hard IRQs off */
1394
void notrace __ppc64_runlatch_on(void)
1395
{
1396
	struct thread_info *ti = current_thread_info();
1397 1398
	unsigned long ctrl;

1399 1400 1401
	ctrl = mfspr(SPRN_CTRLF);
	ctrl |= CTRL_RUNLATCH;
	mtspr(SPRN_CTRLT, ctrl);
1402

1403
	ti->local_flags |= _TLF_RUNLATCH;
1404 1405
}

1406
/* Called with hard IRQs off */
1407
void notrace __ppc64_runlatch_off(void)
1408
{
1409
	struct thread_info *ti = current_thread_info();
1410 1411
	unsigned long ctrl;

1412
	ti->local_flags &= ~_TLF_RUNLATCH;
1413

1414 1415 1416
	ctrl = mfspr(SPRN_CTRLF);
	ctrl &= ~CTRL_RUNLATCH;
	mtspr(SPRN_CTRLT, ctrl);
1417
}
1418
#endif /* CONFIG_PPC64 */
1419

1420 1421 1422 1423 1424 1425
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;
}
1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441

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)
{
1442 1443 1444
	unsigned long base = mm->brk;
	unsigned long ret;

1445
#ifdef CONFIG_PPC_STD_MMU_64
1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457
	/*
	 * 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());
1458 1459 1460 1461 1462 1463

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

	return ret;
}
A
Anton Blanchard 已提交
1464 1465 1466 1467 1468 1469 1470 1471 1472 1473

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

	if (ret < base)
		return base;

	return ret;
}