setup_64.c 20.2 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12
/*
 * 
 * Common boot and setup code.
 *
 * Copyright (C) 2001 PPC64 Team, IBM Corp
 *
 *      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.
 */

13
#define DEBUG
14

15
#include <linux/export.h>
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
#include <linux/string.h>
#include <linux/sched.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/reboot.h>
#include <linux/delay.h>
#include <linux/initrd.h>
#include <linux/seq_file.h>
#include <linux/ioport.h>
#include <linux/console.h>
#include <linux/utsname.h>
#include <linux/tty.h>
#include <linux/root_dev.h>
#include <linux/notifier.h>
#include <linux/cpu.h>
#include <linux/unistd.h>
#include <linux/serial.h>
#include <linux/serial_8250.h>
34
#include <linux/bootmem.h>
35
#include <linux/pci.h>
36
#include <linux/lockdep.h>
Y
Yinghai Lu 已提交
37
#include <linux/memblock.h>
38 39
#include <linux/hugetlb.h>

40
#include <asm/io.h>
41
#include <asm/kdump.h>
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61
#include <asm/prom.h>
#include <asm/processor.h>
#include <asm/pgtable.h>
#include <asm/smp.h>
#include <asm/elf.h>
#include <asm/machdep.h>
#include <asm/paca.h>
#include <asm/time.h>
#include <asm/cputable.h>
#include <asm/sections.h>
#include <asm/btext.h>
#include <asm/nvram.h>
#include <asm/setup.h>
#include <asm/rtas.h>
#include <asm/iommu.h>
#include <asm/serial.h>
#include <asm/cache.h>
#include <asm/page.h>
#include <asm/mmu.h>
#include <asm/firmware.h>
P
Paul Mackerras 已提交
62
#include <asm/xmon.h>
D
David Gibson 已提交
63
#include <asm/udbg.h>
64
#include <asm/kexec.h>
65
#include <asm/mmu_context.h>
66
#include <asm/code-patching.h>
67
#include <asm/kvm_ppc.h>
68
#include <asm/hugetlb.h>
69
#include <asm/epapr_hcalls.h>
70 71 72 73 74 75 76

#ifdef DEBUG
#define DBG(fmt...) udbg_printf(fmt)
#else
#define DBG(fmt...)
#endif

77
int spinning_secondaries;
78 79
u64 ppc64_pft_size;

80 81 82 83
/* Pick defaults since we might want to patch instructions
 * before we've read this from the device tree.
 */
struct ppc64_caches ppc64_caches = {
84 85 86 87
	.dline_size = 0x40,
	.log_dline_size = 6,
	.iline_size = 0x40,
	.log_iline_size = 6
88
};
89 90 91 92 93 94 95 96 97 98
EXPORT_SYMBOL_GPL(ppc64_caches);

/*
 * These are used in binfmt_elf.c to put aux entries on the stack
 * for each elf executable being started.
 */
int dcache_bsize;
int icache_bsize;
int ucache_bsize;

99 100 101 102 103
#if defined(CONFIG_PPC_BOOK3E) && defined(CONFIG_SMP)
static void setup_tlb_core_data(void)
{
	int cpu;

104 105
	BUILD_BUG_ON(offsetof(struct tlb_core_data, lock) != 0);

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
	for_each_possible_cpu(cpu) {
		int first = cpu_first_thread_sibling(cpu);

		paca[cpu].tcd_ptr = &paca[first].tcd;

		/*
		 * If we have threads, we need either tlbsrx.
		 * or e6500 tablewalk mode, or else TLB handlers
		 * will be racy and could produce duplicate entries.
		 */
		if (smt_enabled_at_boot >= 2 &&
		    !mmu_has_feature(MMU_FTR_USE_TLBRSRV) &&
		    book3e_htw_mode != PPC_HTW_E6500) {
			/* Should we panic instead? */
			WARN_ONCE("%s: unsupported MMU configuration -- expect problems\n",
				  __func__);
		}
	}
}
#else
static void setup_tlb_core_data(void)
{
}
#endif

131 132
#ifdef CONFIG_SMP

133
static char *smt_enabled_cmdline;
134 135 136 137 138

/* Look for ibm,smt-enabled OF option */
static void check_smt_enabled(void)
{
	struct device_node *dn;
139
	const char *smt_option;
140

141 142
	/* Default to enabling all threads */
	smt_enabled_at_boot = threads_per_core;
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
	/* Allow the command line to overrule the OF option */
	if (smt_enabled_cmdline) {
		if (!strcmp(smt_enabled_cmdline, "on"))
			smt_enabled_at_boot = threads_per_core;
		else if (!strcmp(smt_enabled_cmdline, "off"))
			smt_enabled_at_boot = 0;
		else {
			long smt;
			int rc;

			rc = strict_strtol(smt_enabled_cmdline, 10, &smt);
			if (!rc)
				smt_enabled_at_boot =
					min(threads_per_core, (int)smt);
		}
	} else {
		dn = of_find_node_by_path("/options");
		if (dn) {
			smt_option = of_get_property(dn, "ibm,smt-enabled",
						     NULL);

			if (smt_option) {
				if (!strcmp(smt_option, "on"))
					smt_enabled_at_boot = threads_per_core;
				else if (!strcmp(smt_option, "off"))
					smt_enabled_at_boot = 0;
			}

			of_node_put(dn);
		}
	}
175 176 177 178 179
}

/* Look for smt-enabled= cmdline option */
static int __init early_smt_enabled(char *p)
{
180
	smt_enabled_cmdline = p;
181 182 183 184
	return 0;
}
early_param("smt-enabled", early_smt_enabled);

P
Paul Mackerras 已提交
185 186
#else
#define check_smt_enabled()
187 188
#endif /* CONFIG_SMP */

189 190 191 192 193 194 195 196 197
/** Fix up paca fields required for the boot cpu */
static void fixup_boot_paca(void)
{
	/* The boot cpu is started */
	get_paca()->cpu_start = 1;
	/* Allow percpu accesses to work until we setup percpu data */
	get_paca()->data_offset = 0;
}

198 199 200 201 202 203 204 205 206 207 208 209
static void cpu_ready_for_interrupts(void)
{
	/* Set IR and DR in PACA MSR */
	get_paca()->kernel_msr = MSR_KERNEL;

	/* Enable AIL if supported */
	if (cpu_has_feature(CPU_FTR_ARCH_207S)) {
		unsigned long lpcr = mfspr(SPRN_LPCR);
		mtspr(SPRN_LPCR, lpcr | LPCR_AIL_3);
	}
}

210 211 212 213 214 215
/*
 * Early initialization entry point. This is called by head.S
 * with MMU translation disabled. We rely on the "feature" of
 * the CPU that ignores the top 2 bits of the address in real
 * mode so we can access kernel globals normally provided we
 * only toy with things in the RMO region. From here, we do
Y
Yinghai Lu 已提交
216
 * some early parsing of the device-tree to setup out MEMBLOCK
217 218 219 220 221 222 223 224 225 226 227 228 229 230
 * data structures, and allocate & initialize the hash table
 * and segment tables so we can start running with translation
 * enabled.
 *
 * It is this function which will call the probe() callback of
 * the various platform types and copy the matching one to the
 * global ppc_md structure. Your platform can eventually do
 * some very early initializations from the probe() routine, but
 * this is not recommended, be very careful as, for example, the
 * device-tree is not accessible via normal means at this point.
 */

void __init early_setup(unsigned long dt_ptr)
{
231 232
	static __initdata struct paca_struct boot_paca;

233 234
	/* -------- printk is _NOT_ safe to use here ! ------- */

235
	/* Identify CPU type */
236
	identify_cpu(0, mfspr(SPRN_PVR));
237

238
	/* Assume we're on cpu 0 for now. Don't write to the paca yet! */
239 240
	initialise_paca(&boot_paca, 0);
	setup_paca(&boot_paca);
241
	fixup_boot_paca();
242

243 244 245
	/* Initialize lockdep early or else spinlocks will blow */
	lockdep_init();

246 247
	/* -------- printk is now safe to use ------- */

248 249 250
	/* Enable early debugging if any specified (see udbg.h) */
	udbg_early_init();

251
 	DBG(" -> early_setup(), dt_ptr: 0x%lx\n", dt_ptr);
252 253

	/*
254 255 256
	 * Do early initialization using the flattened device
	 * tree, such as retrieving the physical memory map or
	 * calculating/retrieving the hash table size.
257 258 259
	 */
	early_init_devtree(__va(dt_ptr));

260 261
	epapr_paravirt_early_init();

262
	/* Now we know the logical id of our boot cpu, setup the paca. */
263
	setup_paca(&paca[boot_cpuid]);
264
	fixup_boot_paca();
265

266 267
	/* Probe the machine type */
	probe_machine();
268

269
	setup_kdump_trampoline();
270

271 272
	DBG("Found, Initializing memory management...\n");

273 274
	/* Initialize the hash table or TLB handling */
	early_init_mmu();
275

276 277 278
	/*
	 * At this point, we can let interrupts switch to virtual mode
	 * (the MMU has been setup), so adjust the MSR in the PACA to
279
	 * have IR and DR set and enable AIL if it exists
280
	 */
281
	cpu_ready_for_interrupts();
282 283

	/* Reserve large chunks of memory for use by CMA for KVM */
284 285
	kvm_cma_reserve();

286 287 288 289 290 291 292
	/*
	 * Reserve any gigantic pages requested on the command line.
	 * memblock needs to have been initialized by the time this is
	 * called since this will reserve memory.
	 */
	reserve_hugetlb_gpages();

293
	DBG(" <- early_setup()\n");
294 295 296 297 298 299 300 301 302 303 304 305

#ifdef CONFIG_PPC_EARLY_DEBUG_BOOTX
	/*
	 * This needs to be done *last* (after the above DBG() even)
	 *
	 * Right after we return from this function, we turn on the MMU
	 * which means the real-mode access trick that btext does will
	 * no longer work, it needs to switch to using a real MMU
	 * mapping. This call will ensure that it does
	 */
	btext_map();
#endif /* CONFIG_PPC_EARLY_DEBUG_BOOTX */
306 307
}

308 309 310
#ifdef CONFIG_SMP
void early_setup_secondary(void)
{
311
	/* Mark interrupts enabled in PACA */
312
	get_paca()->soft_enabled = 0;
313

314 315
	/* Initialize the hash table or TLB handling */
	early_init_mmu_secondary();
316 317 318 319 320 321

	/*
	 * At this point, we can let interrupts switch to virtual mode
	 * (the MMU has been setup), so adjust the MSR in the PACA to
	 * have IR and DR set.
	 */
322
	cpu_ready_for_interrupts();
323 324 325
}

#endif /* CONFIG_SMP */
326

327 328 329
#if defined(CONFIG_SMP) || defined(CONFIG_KEXEC)
void smp_release_cpus(void)
{
330
	unsigned long *ptr;
331
	int i;
332 333 334 335 336 337 338

	DBG(" -> smp_release_cpus()\n");

	/* All secondary cpus are spinning on a common spinloop, release them
	 * all now so they can start to spin on their individual paca
	 * spinloops. For non SMP kernels, the secondary cpus never get out
	 * of the common spinloop.
339
	 */
340

341 342
	ptr  = (unsigned long *)((unsigned long)&__secondary_hold_spinloop
			- PHYSICAL_START);
343
	*ptr = __pa(generic_secondary_smp_init);
344 345 346 347 348

	/* And wait a bit for them to catch up */
	for (i = 0; i < 100000; i++) {
		mb();
		HMT_low();
349
		if (spinning_secondaries == 0)
350 351 352
			break;
		udelay(1);
	}
353
	DBG("spinning_secondaries = %d\n", spinning_secondaries);
354 355 356 357 358

	DBG(" <- smp_release_cpus()\n");
}
#endif /* CONFIG_SMP || CONFIG_KEXEC */

359
/*
360 361
 * Initialize some remaining members of the ppc64_caches and systemcfg
 * structures
362 363 364 365 366 367 368 369 370 371 372
 * (at least until we get rid of them completely). This is mostly some
 * cache informations about the CPU that will be used by cache flush
 * routines and/or provided to userland
 */
static void __init initialize_cache_info(void)
{
	struct device_node *np;
	unsigned long num_cpus = 0;

	DBG(" -> initialize_cache_info()\n");

373
	for_each_node_by_type(np, "cpu") {
374 375
		num_cpus += 1;

A
Anton Blanchard 已提交
376 377
		/*
		 * We're assuming *all* of the CPUs have the same
378 379
		 * d-cache and i-cache sizes... -Peter
		 */
A
Anton Blanchard 已提交
380
		if (num_cpus == 1) {
381
			const __be32 *sizep, *lsizep;
382 383 384 385
			u32 size, lsize;

			size = 0;
			lsize = cur_cpu_spec->dcache_bsize;
386
			sizep = of_get_property(np, "d-cache-size", NULL);
387
			if (sizep != NULL)
388
				size = be32_to_cpu(*sizep);
A
Anton Blanchard 已提交
389 390
			lsizep = of_get_property(np, "d-cache-block-size",
						 NULL);
391 392
			/* fallback if block size missing */
			if (lsizep == NULL)
A
Anton Blanchard 已提交
393 394 395
				lsizep = of_get_property(np,
							 "d-cache-line-size",
							 NULL);
396
			if (lsizep != NULL)
397
				lsize = be32_to_cpu(*lsizep);
398
			if (sizep == NULL || lsizep == NULL)
399 400 401
				DBG("Argh, can't find dcache properties ! "
				    "sizep: %p, lsizep: %p\n", sizep, lsizep);

402 403
			ppc64_caches.dsize = size;
			ppc64_caches.dline_size = lsize;
404 405 406 407 408
			ppc64_caches.log_dline_size = __ilog2(lsize);
			ppc64_caches.dlines_per_page = PAGE_SIZE / lsize;

			size = 0;
			lsize = cur_cpu_spec->icache_bsize;
409
			sizep = of_get_property(np, "i-cache-size", NULL);
410
			if (sizep != NULL)
411
				size = be32_to_cpu(*sizep);
A
Anton Blanchard 已提交
412 413
			lsizep = of_get_property(np, "i-cache-block-size",
						 NULL);
414
			if (lsizep == NULL)
A
Anton Blanchard 已提交
415 416 417
				lsizep = of_get_property(np,
							 "i-cache-line-size",
							 NULL);
418
			if (lsizep != NULL)
419
				lsize = be32_to_cpu(*lsizep);
420
			if (sizep == NULL || lsizep == NULL)
421 422 423
				DBG("Argh, can't find icache properties ! "
				    "sizep: %p, lsizep: %p\n", sizep, lsizep);

424 425
			ppc64_caches.isize = size;
			ppc64_caches.iline_size = lsize;
426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
			ppc64_caches.log_iline_size = __ilog2(lsize);
			ppc64_caches.ilines_per_page = PAGE_SIZE / lsize;
		}
	}

	DBG(" <- initialize_cache_info()\n");
}


/*
 * Do some initial setup of the system.  The parameters are those which 
 * were passed in from the bootloader.
 */
void __init setup_system(void)
{
	DBG(" -> setup_system()\n");

443 444
	/* Apply the CPUs-specific and firmware specific fixups to kernel
	 * text (nop out sections not relevant to this CPU or this firmware)
445
	 */
446
	do_feature_fixups(cur_cpu_spec->cpu_features,
447
			  &__start___ftr_fixup, &__stop___ftr_fixup);
448 449
	do_feature_fixups(cur_cpu_spec->mmu_features,
			  &__start___mmu_ftr_fixup, &__stop___mmu_ftr_fixup);
450 451
	do_feature_fixups(powerpc_firmware_features,
			  &__start___fw_ftr_fixup, &__stop___fw_ftr_fixup);
K
Kumar Gala 已提交
452 453
	do_lwsync_fixups(cur_cpu_spec->cpu_features,
			 &__start___lwsync_fixup, &__stop___lwsync_fixup);
454
	do_final_fixups();
455

456 457 458 459 460 461 462
	/*
	 * Unflatten the device-tree passed by prom_init or kexec
	 */
	unflatten_device_tree();

	/*
	 * Fill the ppc64_caches & systemcfg structures with informations
463
 	 * retrieved from the device-tree.
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483
	 */
	initialize_cache_info();

#ifdef CONFIG_PPC_RTAS
	/*
	 * Initialize RTAS if available
	 */
	rtas_initialize();
#endif /* CONFIG_PPC_RTAS */

	/*
	 * Check if we have an initrd provided via the device-tree
	 */
	check_for_initrd();

	/*
	 * Do some platform specific early initializations, that includes
	 * setting up the hash table pointers. It also sets up some interrupt-mapping
	 * related options that will be used by finish_device_tree()
	 */
484 485
	if (ppc_md.init_early)
		ppc_md.init_early();
486

487 488 489 490 491 492 493
 	/*
	 * We can discover serial ports now since the above did setup the
	 * hash table management for us, thus ioremap works. We do that early
	 * so that further code can be debugged
	 */
	find_legacy_serial_ports();

494 495 496 497 498
	/*
	 * Register early console
	 */
	register_early_udbg_console();

499 500 501 502
	/*
	 * Initialize xmon
	 */
	xmon_setup();
503

P
Paul Mackerras 已提交
504
	smp_setup_cpu_maps();
505
	check_smt_enabled();
506
	setup_tlb_core_data();
507

508
#ifdef CONFIG_SMP
509 510 511 512
	/* Release secondary cpus out of their spinloops at 0x60 now that
	 * we can map physical -> logical CPU ids
	 */
	smp_release_cpus();
513
#endif
514

515
	printk("Starting Linux PPC64 %s\n", init_utsname()->version);
516 517

	printk("-----------------------------------------------------\n");
518
	printk("ppc64_pft_size                = 0x%llx\n", ppc64_pft_size);
Y
Yinghai Lu 已提交
519
	printk("physicalMemorySize            = 0x%llx\n", memblock_phys_mem_size());
520 521 522 523 524 525
	if (ppc64_caches.dline_size != 0x80)
		printk("ppc64_caches.dcache_line_size = 0x%x\n",
		       ppc64_caches.dline_size);
	if (ppc64_caches.iline_size != 0x80)
		printk("ppc64_caches.icache_line_size = 0x%x\n",
		       ppc64_caches.iline_size);
526
#ifdef CONFIG_PPC_STD_MMU_64
527 528
	if (htab_address)
		printk("htab_address                  = 0x%p\n", htab_address);
529
	printk("htab_hash_mask                = 0x%lx\n", htab_hash_mask);
530
#endif /* CONFIG_PPC_STD_MMU_64 */
531
	if (PHYSICAL_START > 0)
532 533
		printk("physical_start                = 0x%llx\n",
		       (unsigned long long)PHYSICAL_START);
534 535 536 537 538
	printk("-----------------------------------------------------\n");

	DBG(" <- setup_system()\n");
}

539 540 541 542 543 544
/* This returns the limit below which memory accesses to the linear
 * mapping are guarnateed not to cause a TLB or SLB miss. This is
 * used to allocate interrupt or emergency stacks for which our
 * exception entry path doesn't deal with being interrupted.
 */
static u64 safe_stack_limit(void)
545
{
546 547 548 549 550 551 552 553 554
#ifdef CONFIG_PPC_BOOK3E
	/* Freescale BookE bolts the entire linear mapping */
	if (mmu_has_feature(MMU_FTR_TYPE_FSL_E))
		return linear_map_top;
	/* Other BookE, we assume the first GB is bolted */
	return 1ul << 30;
#else
	/* BookS, the first segment is bolted */
	if (mmu_has_feature(MMU_FTR_1T_SEGMENT))
555 556
		return 1UL << SID_SHIFT_1T;
	return 1UL << SID_SHIFT;
557
#endif
558 559
}

560 561
static void __init irqstack_early_init(void)
{
562
	u64 limit = safe_stack_limit();
563 564 565
	unsigned int i;

	/*
566 567
	 * Interrupt stacks must be in the first segment since we
	 * cannot afford to take SLB misses on them.
568
	 */
569
	for_each_possible_cpu(i) {
570
		softirq_ctx[i] = (struct thread_info *)
Y
Yinghai Lu 已提交
571
			__va(memblock_alloc_base(THREAD_SIZE,
572
					    THREAD_SIZE, limit));
573
		hardirq_ctx[i] = (struct thread_info *)
Y
Yinghai Lu 已提交
574
			__va(memblock_alloc_base(THREAD_SIZE,
575
					    THREAD_SIZE, limit));
576 577 578
	}
}

579 580 581 582
#ifdef CONFIG_PPC_BOOK3E
static void __init exc_lvl_early_init(void)
{
	unsigned int i;
583
	unsigned long sp;
584 585

	for_each_possible_cpu(i) {
586 587 588 589 590 591 592 593 594 595 596
		sp = memblock_alloc(THREAD_SIZE, THREAD_SIZE);
		critirq_ctx[i] = (struct thread_info *)__va(sp);
		paca[i].crit_kstack = __va(sp + THREAD_SIZE);

		sp = memblock_alloc(THREAD_SIZE, THREAD_SIZE);
		dbgirq_ctx[i] = (struct thread_info *)__va(sp);
		paca[i].dbg_kstack = __va(sp + THREAD_SIZE);

		sp = memblock_alloc(THREAD_SIZE, THREAD_SIZE);
		mcheckirq_ctx[i] = (struct thread_info *)__va(sp);
		paca[i].mc_kstack = __va(sp + THREAD_SIZE);
597
	}
598 599

	if (cpu_has_feature(CPU_FTR_DEBUG_LVL_EXC))
600
		patch_exception(0x040, exc_debug_debug_book3e);
601 602 603 604 605
}
#else
#define exc_lvl_early_init()
#endif

606 607
/*
 * Stack space used when we detect a bad kernel stack pointer, and
608 609
 * early in SMP boots before relocation is enabled. Exclusive emergency
 * stack for machine checks.
610 611 612
 */
static void __init emergency_stack_init(void)
{
613
	u64 limit;
614 615 616 617 618 619 620 621 622 623 624
	unsigned int i;

	/*
	 * Emergency stacks must be under 256MB, we cannot afford to take
	 * SLB misses on them. The ABI also requires them to be 128-byte
	 * aligned.
	 *
	 * Since we use these as temporary stacks during secondary CPU
	 * bringup, we need to get at them in real mode. This means they
	 * must also be within the RMO region.
	 */
625
	limit = min(safe_stack_limit(), ppc64_rma_size);
626

627 628
	for_each_possible_cpu(i) {
		unsigned long sp;
Y
Yinghai Lu 已提交
629
		sp  = memblock_alloc_base(THREAD_SIZE, THREAD_SIZE, limit);
630 631
		sp += THREAD_SIZE;
		paca[i].emergency_sp = __va(sp);
632 633 634 635 636 637 638

#ifdef CONFIG_PPC_BOOK3S_64
		/* emergency stack for machine check exception handling. */
		sp  = memblock_alloc_base(THREAD_SIZE, THREAD_SIZE, limit);
		sp += THREAD_SIZE;
		paca[i].mc_emergency_sp = __va(sp);
#endif
639
	}
640 641 642
}

/*
643 644
 * Called into from start_kernel this initializes bootmem, which is used
 * to manage page allocation until mem_init is called.
645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660
 */
void __init setup_arch(char **cmdline_p)
{
	ppc64_boot_msg(0x12, "Setup Arch");

	*cmdline_p = cmd_line;

	/*
	 * Set cache line size based on type of cpu as a default.
	 * Systems with OF can look in the properties on the cpu node(s)
	 * for a possibly more accurate value.
	 */
	dcache_bsize = ppc64_caches.dline_size;
	icache_bsize = ppc64_caches.iline_size;

	if (ppc_md.panic)
661
		setup_panic();
662

663
	init_mm.start_code = (unsigned long)_stext;
664 665 666
	init_mm.end_code = (unsigned long) _etext;
	init_mm.end_data = (unsigned long) _edata;
	init_mm.brk = klimit;
667 668 669
#ifdef CONFIG_PPC_64K_PAGES
	init_mm.context.pte_frag = NULL;
#endif
670
	irqstack_early_init();
671
	exc_lvl_early_init();
672 673
	emergency_stack_init();

674
#ifdef CONFIG_PPC_STD_MMU_64
675
	stabs_alloc();
676
#endif
677 678 679 680
	/* set up the bootmem stuff with available memory */
	do_init_bootmem();
	sparse_init();

681 682 683 684
#ifdef CONFIG_DUMMY_CONSOLE
	conswitchp = &dummy_con;
#endif

685 686
	if (ppc_md.setup_arch)
		ppc_md.setup_arch();
687 688

	paging_init();
689 690 691 692

	/* Initialize the MMU context management stuff */
	mmu_context_init();

693 694 695 696 697
	/* Interrupt code needs to be 64K-aligned */
	if ((unsigned long)_stext & 0xffff)
		panic("Kernelbase not 64K-aligned (0x%lx)!\n",
		      (unsigned long)_stext);

698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725
	ppc64_boot_msg(0x15, "Setup Done");
}


/* ToDo: do something useful if ppc_md is not yet setup. */
#define PPC64_LINUX_FUNCTION 0x0f000000
#define PPC64_IPL_MESSAGE 0xc0000000
#define PPC64_TERM_MESSAGE 0xb0000000

static void ppc64_do_msg(unsigned int src, const char *msg)
{
	if (ppc_md.progress) {
		char buf[128];

		sprintf(buf, "%08X\n", src);
		ppc_md.progress(buf, 0);
		snprintf(buf, 128, "%s", msg);
		ppc_md.progress(buf, 0);
	}
}

/* Print a boot progress message. */
void ppc64_boot_msg(unsigned int src, const char *msg)
{
	ppc64_do_msg(PPC64_LINUX_FUNCTION|PPC64_IPL_MESSAGE|src, msg);
	printk("[boot]%04x %s\n", src, msg);
}

726
#ifdef CONFIG_SMP
727 728 729
#define PCPU_DYN_SIZE		()

static void * __init pcpu_fc_alloc(unsigned int cpu, size_t size, size_t align)
730
{
731 732 733
	return __alloc_bootmem_node(NODE_DATA(cpu_to_node(cpu)), size, align,
				    __pa(MAX_DMA_ADDRESS));
}
734

735 736 737 738
static void __init pcpu_fc_free(void *ptr, size_t size)
{
	free_bootmem(__pa(ptr), size);
}
739

740 741 742 743 744 745 746 747
static int pcpu_cpu_distance(unsigned int from, unsigned int to)
{
	if (cpu_to_node(from) == cpu_to_node(to))
		return LOCAL_DISTANCE;
	else
		return REMOTE_DISTANCE;
}

748 749 750
unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
EXPORT_SYMBOL(__per_cpu_offset);

751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774
void __init setup_per_cpu_areas(void)
{
	const size_t dyn_size = PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE;
	size_t atom_size;
	unsigned long delta;
	unsigned int cpu;
	int rc;

	/*
	 * Linear mapping is one of 4K, 1M and 16M.  For 4K, no need
	 * to group units.  For larger mappings, use 1M atom which
	 * should be large enough to contain a number of units.
	 */
	if (mmu_linear_psize == MMU_PAGE_4K)
		atom_size = PAGE_SIZE;
	else
		atom_size = 1 << 20;

	rc = pcpu_embed_first_chunk(0, dyn_size, atom_size, pcpu_cpu_distance,
				    pcpu_fc_alloc, pcpu_fc_free);
	if (rc < 0)
		panic("cannot initialize percpu area (err=%d)", rc);

	delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start;
775 776 777 778
	for_each_possible_cpu(cpu) {
                __per_cpu_offset[cpu] = delta + pcpu_unit_offsets[cpu];
		paca[cpu].data_offset = __per_cpu_offset[cpu];
	}
779 780
}
#endif
781 782


783
#if defined(CONFIG_PPC_INDIRECT_PIO) || defined(CONFIG_PPC_INDIRECT_MMIO)
784 785
struct ppc_pci_io ppc_pci_io;
EXPORT_SYMBOL(ppc_pci_io);
786
#endif