linux/arch/sparc/kernel/leon_smp.c

470 lines
11 KiB
C
Raw Normal View History

/* leon_smp.c: Sparc-Leon SMP support.
*
* based on sun4m_smp.c
* Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu)
* Copyright (C) 2009 Daniel Hellstrom (daniel@gaisler.com) Aeroflex Gaisler AB
* Copyright (C) 2009 Konrad Eisele (konrad@gaisler.com) Aeroflex Gaisler AB
*/
#include <asm/head.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/threads.h>
#include <linux/smp.h>
#include <linux/smp_lock.h>
#include <linux/interrupt.h>
#include <linux/kernel_stat.h>
#include <linux/init.h>
#include <linux/spinlock.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/profile.h>
#include <linux/pm.h>
#include <linux/delay.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/gfp.h>
#include <asm/cacheflush.h>
#include <asm/tlbflush.h>
#include <asm/ptrace.h>
#include <asm/atomic.h>
#include <asm/irq_regs.h>
#include <asm/delay.h>
#include <asm/irq.h>
#include <asm/page.h>
#include <asm/pgalloc.h>
#include <asm/pgtable.h>
#include <asm/oplib.h>
#include <asm/cpudata.h>
#include <asm/asi.h>
#include <asm/leon.h>
#include <asm/leon_amba.h>
#ifdef CONFIG_SPARC_LEON
#include "irq.h"
extern ctxd_t *srmmu_ctx_table_phys;
static int smp_processors_ready;
extern volatile unsigned long cpu_callin_map[NR_CPUS];
extern unsigned char boot_cpu_id;
extern cpumask_t smp_commenced_mask;
void __init leon_configure_cache_smp(void);
static inline unsigned long do_swap(volatile unsigned long *ptr,
unsigned long val)
{
__asm__ __volatile__("swapa [%1] %2, %0\n\t" : "=&r"(val)
: "r"(ptr), "i"(ASI_LEON_DCACHE_MISS)
: "memory");
return val;
}
static void smp_setup_percpu_timer(void);
void __cpuinit leon_callin(void)
{
int cpuid = hard_smpleon_processor_id();
local_flush_cache_all();
local_flush_tlb_all();
leon_configure_cache_smp();
/* Get our local ticker going. */
smp_setup_percpu_timer();
calibrate_delay();
smp_store_cpu_info(cpuid);
local_flush_cache_all();
local_flush_tlb_all();
/*
* Unblock the master CPU _only_ when the scheduler state
* of all secondary CPUs will be up-to-date, so after
* the SMP initialization the master will be just allowed
* to call the scheduler code.
* Allow master to continue.
*/
do_swap(&cpu_callin_map[cpuid], 1);
local_flush_cache_all();
local_flush_tlb_all();
cpu_probe();
/* Fix idle thread fields. */
__asm__ __volatile__("ld [%0], %%g6\n\t" : : "r"(&current_set[cpuid])
: "memory" /* paranoid */);
/* Attach to the address space of init_task. */
atomic_inc(&init_mm.mm_count);
current->active_mm = &init_mm;
while (!cpu_isset(cpuid, smp_commenced_mask))
mb();
local_irq_enable();
cpu_set(cpuid, cpu_online_map);
}
/*
* Cycle through the processors asking the PROM to start each one.
*/
extern struct linux_prom_registers smp_penguin_ctable;
void __init leon_configure_cache_smp(void)
{
unsigned long cfg = sparc_leon3_get_dcachecfg();
int me = smp_processor_id();
if (ASI_LEON3_SYSCTRL_CFG_SSIZE(cfg) > 4) {
printk(KERN_INFO "Note: SMP with snooping only works on 4k cache, found %dk(0x%x) on cpu %d, disabling caches\n",
(unsigned int)ASI_LEON3_SYSCTRL_CFG_SSIZE(cfg),
(unsigned int)cfg, (unsigned int)me);
sparc_leon3_disable_cache();
} else {
if (cfg & ASI_LEON3_SYSCTRL_CFG_SNOOPING) {
sparc_leon3_enable_snooping();
} else {
printk(KERN_INFO "Note: You have to enable snooping in the vhdl model cpu %d, disabling caches\n",
me);
sparc_leon3_disable_cache();
}
}
local_flush_cache_all();
local_flush_tlb_all();
}
void leon_smp_setbroadcast(unsigned int mask)
{
int broadcast =
((LEON3_BYPASS_LOAD_PA(&(leon3_irqctrl_regs->mpstatus)) >>
LEON3_IRQMPSTATUS_BROADCAST) & 1);
if (!broadcast) {
prom_printf("######## !!!! The irqmp-ctrl must have broadcast enabled, smp wont work !!!!! ####### nr cpus: %d\n",
leon_smp_nrcpus());
if (leon_smp_nrcpus() > 1) {
BUG();
} else {
prom_printf("continue anyway\n");
return;
}
}
LEON_BYPASS_STORE_PA(&(leon3_irqctrl_regs->mpbroadcast), mask);
}
unsigned int leon_smp_getbroadcast(void)
{
unsigned int mask;
mask = LEON_BYPASS_LOAD_PA(&(leon3_irqctrl_regs->mpbroadcast));
return mask;
}
int leon_smp_nrcpus(void)
{
int nrcpu =
((LEON3_BYPASS_LOAD_PA(&(leon3_irqctrl_regs->mpstatus)) >>
LEON3_IRQMPSTATUS_CPUNR) & 0xf) + 1;
return nrcpu;
}
void __init leon_boot_cpus(void)
{
int nrcpu = leon_smp_nrcpus();
int me = smp_processor_id();
printk(KERN_INFO "%d:(%d:%d) cpus mpirq at 0x%x\n", (unsigned int)me,
(unsigned int)nrcpu, (unsigned int)NR_CPUS,
(unsigned int)&(leon3_irqctrl_regs->mpstatus));
leon_enable_irq_cpu(LEON3_IRQ_CROSS_CALL, me);
leon_enable_irq_cpu(LEON3_IRQ_TICKER, me);
leon_enable_irq_cpu(LEON3_IRQ_RESCHEDULE, me);
leon_smp_setbroadcast(1 << LEON3_IRQ_TICKER);
leon_configure_cache_smp();
smp_setup_percpu_timer();
local_flush_cache_all();
}
int __cpuinit leon_boot_one_cpu(int i)
{
struct task_struct *p;
int timeout;
/* Cook up an idler for this guy. */
p = fork_idle(i);
current_set[i] = task_thread_info(p);
/* See trampoline.S:leon_smp_cpu_startup for details...
* Initialize the contexts table
* Since the call to prom_startcpu() trashes the structure,
* we need to re-initialize it for each cpu
*/
smp_penguin_ctable.which_io = 0;
smp_penguin_ctable.phys_addr = (unsigned int)srmmu_ctx_table_phys;
smp_penguin_ctable.reg_size = 0;
/* whirrr, whirrr, whirrrrrrrrr... */
printk(KERN_INFO "Starting CPU %d : (irqmp: 0x%x)\n", (unsigned int)i,
(unsigned int)&leon3_irqctrl_regs->mpstatus);
local_flush_cache_all();
LEON_BYPASS_STORE_PA(&(leon3_irqctrl_regs->mpstatus), 1 << i);
/* wheee... it's going... */
for (timeout = 0; timeout < 10000; timeout++) {
if (cpu_callin_map[i])
break;
udelay(200);
}
printk(KERN_INFO "Started CPU %d\n", (unsigned int)i);
if (!(cpu_callin_map[i])) {
printk(KERN_ERR "Processor %d is stuck.\n", i);
return -ENODEV;
} else {
leon_enable_irq_cpu(LEON3_IRQ_CROSS_CALL, i);
leon_enable_irq_cpu(LEON3_IRQ_TICKER, i);
leon_enable_irq_cpu(LEON3_IRQ_RESCHEDULE, i);
}
local_flush_cache_all();
return 0;
}
void __init leon_smp_done(void)
{
int i, first;
int *prev;
/* setup cpu list for irq rotation */
first = 0;
prev = &first;
for (i = 0; i < NR_CPUS; i++) {
if (cpu_online(i)) {
*prev = i;
prev = &cpu_data(i).next;
}
}
*prev = first;
local_flush_cache_all();
/* Free unneeded trap tables */
if (!cpu_isset(1, cpu_present_map)) {
ClearPageReserved(virt_to_page(trapbase_cpu1));
init_page_count(virt_to_page(trapbase_cpu1));
free_page((unsigned long)trapbase_cpu1);
totalram_pages++;
num_physpages++;
}
if (!cpu_isset(2, cpu_present_map)) {
ClearPageReserved(virt_to_page(trapbase_cpu2));
init_page_count(virt_to_page(trapbase_cpu2));
free_page((unsigned long)trapbase_cpu2);
totalram_pages++;
num_physpages++;
}
if (!cpu_isset(3, cpu_present_map)) {
ClearPageReserved(virt_to_page(trapbase_cpu3));
init_page_count(virt_to_page(trapbase_cpu3));
free_page((unsigned long)trapbase_cpu3);
totalram_pages++;
num_physpages++;
}
/* Ok, they are spinning and ready to go. */
smp_processors_ready = 1;
}
void leon_irq_rotate(int cpu)
{
}
static struct smp_funcall {
smpfunc_t func;
unsigned long arg1;
unsigned long arg2;
unsigned long arg3;
unsigned long arg4;
unsigned long arg5;
unsigned long processors_in[NR_CPUS]; /* Set when ipi entered. */
unsigned long processors_out[NR_CPUS]; /* Set when ipi exited. */
} ccall_info;
static DEFINE_SPINLOCK(cross_call_lock);
/* Cross calls must be serialized, at least currently. */
static void leon_cross_call(smpfunc_t func, cpumask_t mask, unsigned long arg1,
unsigned long arg2, unsigned long arg3,
unsigned long arg4)
{
if (smp_processors_ready) {
register int high = NR_CPUS - 1;
unsigned long flags;
spin_lock_irqsave(&cross_call_lock, flags);
{
/* If you make changes here, make sure gcc generates proper code... */
register smpfunc_t f asm("i0") = func;
register unsigned long a1 asm("i1") = arg1;
register unsigned long a2 asm("i2") = arg2;
register unsigned long a3 asm("i3") = arg3;
register unsigned long a4 asm("i4") = arg4;
register unsigned long a5 asm("i5") = 0;
__asm__ __volatile__("std %0, [%6]\n\t"
"std %2, [%6 + 8]\n\t"
"std %4, [%6 + 16]\n\t" : :
"r"(f), "r"(a1), "r"(a2), "r"(a3),
"r"(a4), "r"(a5),
"r"(&ccall_info.func));
}
/* Init receive/complete mapping, plus fire the IPI's off. */
{
register int i;
cpu_clear(smp_processor_id(), mask);
cpus_and(mask, cpu_online_map, mask);
for (i = 0; i <= high; i++) {
if (cpu_isset(i, mask)) {
ccall_info.processors_in[i] = 0;
ccall_info.processors_out[i] = 0;
set_cpu_int(i, LEON3_IRQ_CROSS_CALL);
}
}
}
{
register int i;
i = 0;
do {
if (!cpu_isset(i, mask))
continue;
while (!ccall_info.processors_in[i])
barrier();
} while (++i <= high);
i = 0;
do {
if (!cpu_isset(i, mask))
continue;
while (!ccall_info.processors_out[i])
barrier();
} while (++i <= high);
}
spin_unlock_irqrestore(&cross_call_lock, flags);
}
}
/* Running cross calls. */
void leon_cross_call_irq(void)
{
int i = smp_processor_id();
ccall_info.processors_in[i] = 1;
ccall_info.func(ccall_info.arg1, ccall_info.arg2, ccall_info.arg3,
ccall_info.arg4, ccall_info.arg5);
ccall_info.processors_out[i] = 1;
}
void leon_percpu_timer_interrupt(struct pt_regs *regs)
{
struct pt_regs *old_regs;
int cpu = smp_processor_id();
old_regs = set_irq_regs(regs);
leon_clear_profile_irq(cpu);
profile_tick(CPU_PROFILING);
if (!--prof_counter(cpu)) {
int user = user_mode(regs);
irq_enter();
update_process_times(user);
irq_exit();
prof_counter(cpu) = prof_multiplier(cpu);
}
set_irq_regs(old_regs);
}
static void __init smp_setup_percpu_timer(void)
{
int cpu = smp_processor_id();
prof_counter(cpu) = prof_multiplier(cpu) = 1;
}
void __init leon_blackbox_id(unsigned *addr)
{
int rd = *addr & 0x3e000000;
int rs1 = rd >> 11;
/* patch places where ___b_hard_smp_processor_id appears */
addr[0] = 0x81444000 | rd; /* rd %asr17, reg */
addr[1] = 0x8130201c | rd | rs1; /* srl reg, 0x1c, reg */
addr[2] = 0x01000000; /* nop */
}
void __init leon_blackbox_current(unsigned *addr)
{
int rd = *addr & 0x3e000000;
int rs1 = rd >> 11;
/* patch LOAD_CURRENT macro where ___b_load_current appears */
addr[0] = 0x81444000 | rd; /* rd %asr17, reg */
addr[2] = 0x8130201c | rd | rs1; /* srl reg, 0x1c, reg */
addr[4] = 0x81282002 | rd | rs1; /* sll reg, 0x2, reg */
}
/*
* CPU idle callback function
* See .../arch/sparc/kernel/process.c
*/
void pmc_leon_idle(void)
{
__asm__ volatile ("mov %g0, %asr19");
}
void __init leon_init_smp(void)
{
/* Patch ipi15 trap table */
t_nmi[1] = t_nmi[1] + (linux_trap_ipi15_leon - linux_trap_ipi15_sun4m);
BTFIXUPSET_BLACKBOX(hard_smp_processor_id, leon_blackbox_id);
BTFIXUPSET_BLACKBOX(load_current, leon_blackbox_current);
BTFIXUPSET_CALL(smp_cross_call, leon_cross_call, BTFIXUPCALL_NORM);
BTFIXUPSET_CALL(__hard_smp_processor_id, __leon_processor_id,
BTFIXUPCALL_NORM);
#ifndef PMC_NO_IDLE
/* Assign power management IDLE handler */
pm_idle = pmc_leon_idle;
printk(KERN_INFO "leon: power management initialized\n");
#endif
}
#endif /* CONFIG_SPARC_LEON */