setup_64.c 15.2 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 31 32 33
/*
 * 
 * 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.
 */

#undef DEBUG

#include <linux/module.h>
#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 <asm/io.h>
37
#include <asm/kdump.h>
38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
#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/system.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/lmb.h>
#include <asm/firmware.h>
P
Paul Mackerras 已提交
60
#include <asm/xmon.h>
D
David Gibson 已提交
61
#include <asm/udbg.h>
62
#include <asm/kexec.h>
63

S
Stephen Rothwell 已提交
64 65
#include "setup.h"

66 67 68 69 70 71 72 73 74 75
#ifdef DEBUG
#define DBG(fmt...) udbg_printf(fmt)
#else
#define DBG(fmt...)
#endif

int have_of = 1;
int boot_cpuid = 0;
u64 ppc64_pft_size;

76 77 78 79
/* Pick defaults since we might want to patch instructions
 * before we've read this from the device tree.
 */
struct ppc64_caches ppc64_caches = {
80 81 82 83
	.dline_size = 0x40,
	.log_dline_size = 6,
	.iline_size = 0x40,
	.log_iline_size = 6
84
};
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102
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;

#ifdef CONFIG_SMP

static int smt_enabled_cmdline;

/* Look for ibm,smt-enabled OF option */
static void check_smt_enabled(void)
{
	struct device_node *dn;
103
	const char *smt_option;
104 105 106 107 108 109 110 111

	/* Allow the command line to overrule the OF option */
	if (smt_enabled_cmdline)
		return;

	dn = of_find_node_by_path("/options");

	if (dn) {
112
		smt_option = of_get_property(dn, "ibm,smt-enabled", NULL);
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139

                if (smt_option) {
			if (!strcmp(smt_option, "on"))
				smt_enabled_at_boot = 1;
			else if (!strcmp(smt_option, "off"))
				smt_enabled_at_boot = 0;
                }
        }
}

/* Look for smt-enabled= cmdline option */
static int __init early_smt_enabled(char *p)
{
	smt_enabled_cmdline = 1;

	if (!p)
		return 0;

	if (!strcmp(p, "on") || !strcmp(p, "1"))
		smt_enabled_at_boot = 1;
	else if (!strcmp(p, "off") || !strcmp(p, "0"))
		smt_enabled_at_boot = 0;

	return 0;
}
early_param("smt-enabled", early_smt_enabled);

P
Paul Mackerras 已提交
140 141
#else
#define check_smt_enabled()
142 143
#endif /* CONFIG_SMP */

144 145 146 147 148 149 150
/* Put the paca pointer into r13 and SPRG3 */
void __init setup_paca(int cpu)
{
	local_paca = &paca[cpu];
	mtspr(SPRN_SPRG3, local_paca);
}

151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
/*
 * 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
 * some early parsing of the device-tree to setup out LMB
 * 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)
{
172
	/* Identify CPU type */
173
	identify_cpu(0, mfspr(SPRN_PVR));
174

175 176 177
	/* Assume we're on cpu 0 for now. Don't write to the paca yet! */
	setup_paca(0);

178 179
	/* Enable early debugging if any specified (see udbg.h) */
	udbg_early_init();
180

181
 	DBG(" -> early_setup(), dt_ptr: 0x%lx\n", dt_ptr);
182 183

	/*
184 185 186
	 * Do early initialization using the flattened device
	 * tree, such as retrieving the physical memory map or
	 * calculating/retrieving the hash table size.
187 188 189
	 */
	early_init_devtree(__va(dt_ptr));

190
	/* Now we know the logical id of our boot cpu, setup the paca. */
191
	setup_paca(boot_cpuid);
192 193 194 195 196 197

	/* Fix up paca fields required for the boot cpu */
	get_paca()->cpu_start = 1;
	get_paca()->stab_real = __pa((u64)&initial_stab);
	get_paca()->stab_addr = (u64)&initial_stab;

198 199
	/* Probe the machine type */
	probe_machine();
200

201
	setup_kdump_trampoline();
202

203 204 205
	DBG("Found, Initializing memory management...\n");

	/*
206 207 208
	 * Initialize the MMU Hash table and create the linear mapping
	 * of memory. Has to be done before stab/slb initialization as
	 * this is currently where the page size encoding is obtained
209
	 */
210
	htab_initialize();
211 212

	/*
213
	 * Initialize stab / SLB management except on iSeries
214
	 */
215 216 217 218
	if (cpu_has_feature(CPU_FTR_SLB))
		slb_initialize();
	else if (!firmware_has_feature(FW_FEATURE_ISERIES))
		stab_initialize(get_paca()->stab_real);
219 220 221 222

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

223 224 225 226 227
#ifdef CONFIG_SMP
void early_setup_secondary(void)
{
	struct paca_struct *lpaca = get_paca();

228 229
	/* Mark interrupts enabled in PACA */
	lpaca->soft_enabled = 0;
230 231 232 233 234 235 236 237 238 239 240 241 242 243 244

	/* Initialize hash table for that CPU */
	htab_initialize_secondary();

	/* Initialize STAB/SLB. We use a virtual address as it works
	 * in real mode on pSeries and we want a virutal address on
	 * iSeries anyway
	 */
	if (cpu_has_feature(CPU_FTR_SLB))
		slb_initialize();
	else
		stab_initialize(lpaca->stab_addr);
}

#endif /* CONFIG_SMP */
245

246 247 248 249
#if defined(CONFIG_SMP) || defined(CONFIG_KEXEC)
void smp_release_cpus(void)
{
	extern unsigned long __secondary_hold_spinloop;
250
	unsigned long *ptr;
251 252 253 254 255 256 257 258 259 260

	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.
	 * This is useless but harmless on iSeries, secondaries are already
	 * waiting on their paca spinloops. */

261 262 263
	ptr  = (unsigned long *)((unsigned long)&__secondary_hold_spinloop
			- PHYSICAL_START);
	*ptr = 1;
264 265 266 267 268 269
	mb();

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

270
/*
271 272
 * Initialize some remaining members of the ppc64_caches and systemcfg
 * structures
273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291
 * (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");

	for (np = NULL; (np = of_find_node_by_type(np, "cpu"));) {
		num_cpus += 1;

		/* We're assuming *all* of the CPUs have the same
		 * d-cache and i-cache sizes... -Peter
		 */

		if ( num_cpus == 1 ) {
292
			const u32 *sizep, *lsizep;
293 294 295 296
			u32 size, lsize;

			size = 0;
			lsize = cur_cpu_spec->dcache_bsize;
297
			sizep = of_get_property(np, "d-cache-size", NULL);
298 299
			if (sizep != NULL)
				size = *sizep;
300 301 302 303
			lsizep = of_get_property(np, "d-cache-block-size", NULL);
			/* fallback if block size missing */
			if (lsizep == NULL)
				lsizep = of_get_property(np, "d-cache-line-size", NULL);
304 305 306 307 308 309
			if (lsizep != NULL)
				lsize = *lsizep;
			if (sizep == 0 || lsizep == 0)
				DBG("Argh, can't find dcache properties ! "
				    "sizep: %p, lsizep: %p\n", sizep, lsizep);

310 311
			ppc64_caches.dsize = size;
			ppc64_caches.dline_size = lsize;
312 313 314 315 316
			ppc64_caches.log_dline_size = __ilog2(lsize);
			ppc64_caches.dlines_per_page = PAGE_SIZE / lsize;

			size = 0;
			lsize = cur_cpu_spec->icache_bsize;
317
			sizep = of_get_property(np, "i-cache-size", NULL);
318 319
			if (sizep != NULL)
				size = *sizep;
320 321 322
			lsizep = of_get_property(np, "i-cache-block-size", NULL);
			if (lsizep == NULL)
				lsizep = of_get_property(np, "i-cache-line-size", NULL);
323 324 325 326 327 328
			if (lsizep != NULL)
				lsize = *lsizep;
			if (sizep == 0 || lsizep == 0)
				DBG("Argh, can't find icache properties ! "
				    "sizep: %p, lsizep: %p\n", sizep, lsizep);

329 330
			ppc64_caches.isize = size;
			ppc64_caches.iline_size = lsize;
331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347
			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");

348 349
	/* Apply the CPUs-specific and firmware specific fixups to kernel
	 * text (nop out sections not relevant to this CPU or this firmware)
350
	 */
351
	do_feature_fixups(cur_cpu_spec->cpu_features,
352
			  &__start___ftr_fixup, &__stop___ftr_fixup);
353 354
	do_feature_fixups(powerpc_firmware_features,
			  &__start___fw_ftr_fixup, &__stop___fw_ftr_fixup);
355

356 357 358 359 360 361 362
	/*
	 * Unflatten the device-tree passed by prom_init or kexec
	 */
	unflatten_device_tree();

	/*
	 * Fill the ppc64_caches & systemcfg structures with informations
363
 	 * retrieved from the device-tree.
364 365 366
	 */
	initialize_cache_info();

367 368 369 370 371
	/*
	 * Initialize irq remapping subsystem
	 */
	irq_early_init();

372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
#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()
	 */
389 390
	if (ppc_md.init_early)
		ppc_md.init_early();
391

392 393 394 395 396 397 398
 	/*
	 * 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();

399 400 401 402 403
	/*
	 * Register early console
	 */
	register_early_udbg_console();

404 405 406 407
	/*
	 * Initialize xmon
	 */
	xmon_setup();
408

P
Paul Mackerras 已提交
409 410
	check_smt_enabled();
	smp_setup_cpu_maps();
411

412
#ifdef CONFIG_SMP
413 414 415 416
	/* Release secondary cpus out of their spinloops at 0x60 now that
	 * we can map physical -> logical CPU ids
	 */
	smp_release_cpus();
417
#endif
418

419
	printk("Starting Linux PPC64 %s\n", init_utsname()->version);
420 421 422

	printk("-----------------------------------------------------\n");
	printk("ppc64_pft_size                = 0x%lx\n", ppc64_pft_size);
423
	printk("physicalMemorySize            = 0x%lx\n", lmb_phys_mem_size());
424 425 426 427 428 429 430 431
	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);
	if (htab_address)
		printk("htab_address                  = 0x%p\n", htab_address);
432
	printk("htab_hash_mask                = 0x%lx\n", htab_hash_mask);
433 434 435
#if PHYSICAL_START > 0
	printk("physical_start                = 0x%x\n", PHYSICAL_START);
#endif
436 437 438 439 440 441 442 443 444 445 446 447 448 449
	printk("-----------------------------------------------------\n");

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

#ifdef CONFIG_IRQSTACKS
static void __init irqstack_early_init(void)
{
	unsigned int i;

	/*
	 * interrupt stacks must be under 256MB, we cannot afford to take
	 * SLB misses on them.
	 */
450
	for_each_possible_cpu(i) {
451 452 453 454 455 456
		softirq_ctx[i] = (struct thread_info *)
			__va(lmb_alloc_base(THREAD_SIZE,
					    THREAD_SIZE, 0x10000000));
		hardirq_ctx[i] = (struct thread_info *)
			__va(lmb_alloc_base(THREAD_SIZE,
					    THREAD_SIZE, 0x10000000));
457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
	}
}
#else
#define irqstack_early_init()
#endif

/*
 * Stack space used when we detect a bad kernel stack pointer, and
 * early in SMP boots before relocation is enabled.
 */
static void __init emergency_stack_init(void)
{
	unsigned long limit;
	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.
	 */
	limit = min(0x10000000UL, lmb.rmo_size);

483
	for_each_possible_cpu(i)
484 485
		paca[i].emergency_sp =
		__va(lmb_alloc_base(HW_PAGE_SIZE, 128, limit)) + HW_PAGE_SIZE;
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
}

/*
 * Called into from start_kernel, after lock_kernel has been called.
 * Initializes bootmem, which is unsed to manage page allocation until
 * mem_init is called.
 */
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;

	/* reboot on panic */
	panic_timeout = 180;

	if (ppc_md.panic)
511
		setup_panic();
512 513 514 515 516 517 518 519 520 521 522 523 524 525 526

	init_mm.start_code = PAGE_OFFSET;
	init_mm.end_code = (unsigned long) _etext;
	init_mm.end_data = (unsigned long) _edata;
	init_mm.brk = klimit;
	
	irqstack_early_init();
	emergency_stack_init();

	stabs_alloc();

	/* set up the bootmem stuff with available memory */
	do_init_bootmem();
	sparse_init();

527 528 529 530
#ifdef CONFIG_DUMMY_CONSOLE
	conswitchp = &dummy_con;
#endif

531 532
	if (ppc_md.setup_arch)
		ppc_md.setup_arch();
533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574

	paging_init();
	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);
}

/* Print a termination message (print only -- does not stop the kernel) */
void ppc64_terminate_msg(unsigned int src, const char *msg)
{
	ppc64_do_msg(PPC64_LINUX_FUNCTION|PPC64_TERM_MESSAGE|src, msg);
	printk("[terminate]%04x %s\n", src, msg);
}

void cpu_die(void)
{
	if (ppc_md.cpu_die)
		ppc_md.cpu_die();
}
575 576 577 578 579 580 581 582 583

#ifdef CONFIG_SMP
void __init setup_per_cpu_areas(void)
{
	int i;
	unsigned long size;
	char *ptr;

	/* Copy section for each CPU (we discard the original) */
584
	size = ALIGN(__per_cpu_end - __per_cpu_start, PAGE_SIZE);
585 586 587 588 589
#ifdef CONFIG_MODULES
	if (size < PERCPU_ENOUGH_ROOM)
		size = PERCPU_ENOUGH_ROOM;
#endif

590
	for_each_possible_cpu(i) {
591
		ptr = alloc_bootmem_pages_node(NODE_DATA(cpu_to_node(i)), size);
592 593 594 595 596 597
		if (!ptr)
			panic("Cannot allocate cpu data for CPU %d\n", i);

		paca[i].data_offset = ptr - __per_cpu_start;
		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
	}
598 599 600

	/* Now that per_cpu is setup, initialize cpu_sibling_map */
	smp_setup_cpu_sibling_map();
601 602
}
#endif
603 604 605 606 607 608 609


#ifdef CONFIG_PPC_INDIRECT_IO
struct ppc_pci_io ppc_pci_io;
EXPORT_SYMBOL(ppc_pci_io);
#endif /* CONFIG_PPC_INDIRECT_IO */