asmlinkage void do_IRQ(int irq, struct pt_regs * regs)
{
struct irqaction *action;
int do_random, cpu;
cpu = smp_processor_id();
irq_enter(cpu);
kstat.irqs[cpu][irq]++;
action = irq_action[irq];
if (action) {
if (!(action->flags & SA_INTERRUPT))
__sti();
action = irq_action[irq];
do_random = 0;
do {
do_random |= action->flags;
action->handler(irq, action->dev_id, regs);
action = action->next;
} while (action);
if (do_random & SA_SAMPLE_RANDOM)
add_interrupt_randomness(irq);
__cli();
}
irq_exit(cpu);
if (softirq_pending(cpu))
do_softirq();
/* unmasking and bottom half handling is done magically for us. */
}
/*
* This should really return information about whether
* we should do bottom half handling etc. Right now we
* end up _always_ checking the bottom half, which is a
* waste of time and is not what some drivers would
* prefer.
*/
int handle_IRQ_event(unsigned int irq, struct pt_regs * regs, struct irqaction * action)
{
int status;
int cpu = smp_processor_id();
irq_enter(cpu, irq);
status = 1; /* Force the "do bottom halves" bit */
if (!(action->flags & SA_INTERRUPT))
__sti();
do {
status |= action->flags;
action->handler(irq, action->dev_id, regs);
action = action->next;
} while (action);
if (status & SA_SAMPLE_RANDOM)
add_interrupt_randomness(irq);
__cli();
irq_exit(cpu, irq);
return status;
}
void __init smp_callin(void)
{
int cpu = current->processor;
smp_store_cpu_info(cpu);
set_dec(tb_ticks_per_jiffy);
cpu_callin_map[cpu] = 1;
smp_ops->setup_cpu(cpu);
init_idle();
/*
* This cpu is now "online". Only set them online
* before they enter the loop below since write access
* to the below variable is _not_ guaranteed to be
* atomic.
* -- Cort <*****@*****.**>
*/
cpu_online_map |= 1UL << smp_processor_id();
while(!smp_commenced)
barrier();
/* see smp_commence for more info */
if (!smp_tb_synchronized && smp_num_cpus == 2) {
smp_software_tb_sync(cpu);
}
__sti();
}
void do_IRQ (int irq /* gets currently not passed: , struct pt_regs * regs*/)
/*
* Note: the code in "interrupt-entry.s" that calls this function
* does leave IRQs disabled until return from this function.
* This function gets entered with the IRQ still disabled. You
* may enable it here.
*/
{
struct irqdesc * desc;
struct irqaction * action;
if (irq >= NR_IRQS) {
// spurious intr
return;
}
desc = irq_desc + irq;
if (desc->mask_ack) {
desc->mask_ack(irq);
action = desc->action;
if (action) {
if (desc->nomask) {
desc->unmask(irq);
}
__sti(); // enable IRQs
do {
action->handler(irq, action->dev_id, 0);
action = action->next;
} while (action);
__cli(); // disable IRQs so that unmask() can be called safely
if (!desc->nomask && desc->enabled) {
desc->unmask(irq);
}
}
}
}
static int inline handle_IRQ_event(unsigned int irq, struct pt_regs * regs, struct irqaction * action)
#endif
{
int status;
int cpu = smp_processor_id();
irq_enter(cpu, irq);
status = 1; /* Force the "do bottom halves" bit */
#if 0
if (!(action->flags & SA_INTERRUPT))
__sti();
#endif
//sc_yang => open shared IRQ
#if 1
//#ifndef JACKSON_NET_WORK
do {
status |= action->flags;
action->handler(irq, action->dev_id, regs);
action = action->next;
} while (action);
if (status & SA_SAMPLE_RANDOM)
add_interrupt_randomness(irq);
#else
action->handler(irq, action->dev_id, regs);
#endif
//__cli();
irq_exit(cpu, irq);
return status;
}
/*
* do_IRQ handles IRQ's that have been installed without the
* SA_INTERRUPT flag: it uses the full signal-handling return
* and runs with other interrupts enabled. All relatively slow
* IRQ's should use this format: notably the keyboard/timer
* routines.
*/
static void do_IRQ(int irq, struct pt_regs * regs)
{
struct irqaction *action;
int do_random, cpu;
cpu = smp_processor_id();
irq_enter(cpu);
kstat.irqs[cpu][irq]++;
mask_irq(irq);
action = *(irq + irq_action);
if (action) {
if (!(action->flags & SA_INTERRUPT))
__sti();
action = *(irq + irq_action);
do_random = 0;
do {
do_random |= action->flags;
action->handler(irq, action->dev_id, regs);
action = action->next;
} while (action);
if (do_random & SA_SAMPLE_RANDOM)
add_interrupt_randomness(irq);
__cli();
} else {
printk("do_IRQ: Unregistered IRQ (0x%X) occured\n", irq);
}
unmask_irq(irq);
irq_exit(cpu);
/* unmasking and bottom half handling is done magically for us. */
}
void __global_sti(void)
{
int cpu = smp_processor_id();
if (!local_irq_count(cpu))
release_irqlock(cpu);
__sti();
}
void smp_send_stop(void)
{
smp_call_function(stop_this_cpu, NULL, 1, 0);
smp_num_cpus = 1;
__cli();
disable_local_APIC();
__sti();
}
static void stop_this_cpu(void *dummy)
{
/*
* Remove this CPU:
*/
clear_bit(smp_processor_id(), &cpu_online_map);
/* May need to service _machine_restart IPI */
__sti();
/* XXXKW wait if available? */
for (;;);
}
void __global_sti(void)
{
int cpu;
preempt_disable();
cpu = smp_processor_id();
if (!local_irq_count(cpu))
release_irqlock(cpu);
preempt_enable();
__sti();
}
/*
* INITIAL C ENTRY POINT.
*/
void _minios_start_kernel(start_info_t *si)
{
bmk_printf_init(minios_putc, NULL);
bmk_core_init(STACK_SIZE_PAGE_ORDER, PAGE_SHIFT);
arch_init(si);
trap_init();
bmk_sched_init();
/* print out some useful information */
minios_printk(" start_info: %p(VA)\n", si);
minios_printk(" nr_pages: 0x%lx\n", si->nr_pages);
minios_printk(" shared_inf: 0x%08lx(MA)\n", si->shared_info);
minios_printk(" pt_base: %p(VA)\n", (void *)si->pt_base);
minios_printk("nr_pt_frames: 0x%lx\n", si->nr_pt_frames);
minios_printk(" mfn_list: %p(VA)\n", (void *)si->mfn_list);
minios_printk(" mod_start: 0x%lx(VA)\n", si->mod_start);
minios_printk(" mod_len: %lu\n", si->mod_len);
minios_printk(" flags: 0x%x\n", (unsigned int)si->flags);
minios_printk(" cmd_line: %s\n",
si->cmd_line ? (const char *)si->cmd_line : "NULL");
/* Set up events. */
init_events();
/* ENABLE EVENT DELIVERY. This is disabled at start of day. */
__sti();
arch_print_info();
setup_xen_features();
/* Init memory management. */
init_mm();
/* Init time and timers. */
init_time();
/* Init the console driver. */
init_console();
/* Init grant tables */
init_gnttab();
/* Init XenBus */
init_xenbus();
/* Init scheduler. */
bmk_sched_startmain(_app_main, &start_info);
bmk_platform_halt("unreachable");
}
void setup_APIC_timer(void * data)
{
unsigned int clocks = (unsigned int) data, slice, t0, t1;
unsigned long flags;
int delta;
__save_flags(flags);
__sti();
/*
* ok, Intel has some smart code in their APIC that knows
* if a CPU was in 'hlt' lowpower mode, and this increases
* its APIC arbitration priority. To avoid the external timer
* IRQ APIC event being in synchron with the APIC clock we
* introduce an interrupt skew to spread out timer events.
*
* The number of slices within a 'big' timeslice is smp_num_cpus+1
*/
slice = clocks / (smp_num_cpus+1);
printk("cpu: %d, clocks: %d, slice: %d\n",
smp_processor_id(), clocks, slice);
/*
* Wait for IRQ0's slice:
*/
wait_8254_wraparound();
__setup_APIC_LVTT(clocks);
t0 = apic_read(APIC_TMICT)*APIC_DIVISOR;
/* Wait till TMCCT gets reloaded from TMICT... */
do {
t1 = apic_read(APIC_TMCCT)*APIC_DIVISOR;
delta = (int)(t0 - t1 - slice*(smp_processor_id()+1));
} while (delta >= 0);
/* Now wait for our slice for real. */
do {
t1 = apic_read(APIC_TMCCT)*APIC_DIVISOR;
delta = (int)(t0 - t1 - slice*(smp_processor_id()+1));
} while (delta < 0);
__setup_APIC_LVTT(clocks);
printk("CPU%d<T0:%d,T1:%d,D:%d,S:%d,C:%d>\n",
smp_processor_id(), t0, t1, delta, slice, clocks);
__restore_flags(flags);
}
void __init setup_APIC_clocks (void)
{
printk("Using local APIC timer interrupts.\n");
using_apic_timer = 1;
__cli();
calibration_result = calibrate_APIC_clock();
/*
* Now set up the timer for real.
*/
setup_APIC_timer((void *)calibration_result);
__sti();
/* and update all other cpus */
smp_call_function(setup_APIC_timer, (void *)calibration_result, 1, 1);
}
void __global_restore_flags(unsigned long flags)
{
switch (flags) {
case 0:
__global_cli();
break;
case 1:
__global_sti();
break;
case 2:
__cli();
break;
case 3:
__sti();
break;
default:
printk("global_restore_flags: %08lx\n", flags);
}
}
/*
* INITIAL C ENTRY POINT.
*/
void start_kernel(start_info_t *si)
{
static char hello[] = "Bootstrapping...\n";
(void)HYPERVISOR_console_io(CONSOLEIO_write, strlen(hello), hello);
arch_init(si);
trap_init();
/* print out some useful information */
printk("Mirage OS!\n");
printk(" start_info: %p(VA)\n", si);
printk(" nr_pages: 0x%lx\n", si->nr_pages);
printk(" shared_inf: 0x%08lx(MA)\n", si->shared_info);
printk(" pt_base: %p(VA)\n", (void *)si->pt_base);
printk("nr_pt_frames: 0x%lx\n", si->nr_pt_frames);
printk(" mfn_list: %p(VA)\n", (void *)si->mfn_list);
printk(" mod_start: 0x%lx(VA)\n", si->mod_start);
printk(" mod_len: %lu\n", si->mod_len);
printk(" flags: 0x%x\n", (unsigned int)si->flags);
printk(" cmd_line: %s\n",
si->cmd_line ? (const char *)si->cmd_line : "NULL");
/* Set up events. */
init_events();
/* ENABLE EVENT DELIVERY. This is disabled at start of day. */
__sti();
arch_print_info();
setup_xen_features();
/* Init memory management. */
init_mm();
/* Init time and timers. */
init_time();
/* Call (possibly overridden) app_main() */
app_main(&start_info);
}
void __init smp_callin(void)
{
int cpu = current->processor;
smp_store_cpu_info(cpu);
set_dec(paca[cpu].default_decr);
cpu_callin_map[cpu] = 1;
ppc_md.smp_setup_cpu(cpu);
init_idle();
set_bit(smp_processor_id(), &cpu_online_map);
while(!smp_commenced) {
barrier();
}
__sti();
}
void __init smp_callin(void)
{
int cpu = current->processor;
smp_store_cpu_info(cpu);
smp_ops->setup_cpu(cpu);
set_dec(tb_ticks_per_jiffy);
cpu_online_map |= 1UL << cpu;
mb();
cpu_callin_map[cpu] = 1;
while(!smp_commenced)
barrier();
/* see smp_commence for more info */
if (!smp_tb_synchronized && smp_num_cpus == 2) {
smp_software_tb_sync(cpu);
}
__sti();
}
void __init setup_APIC_clocks (void)
{
/* Disabled by DMI scan or kernel option? */
if (dont_use_local_apic_timer)
return;
printk("Using local APIC timer interrupts.\n");
using_apic_timer = 1;
__cli();
calibration_result = calibrate_APIC_clock();
/*
* Now set up the timer for real.
*/
setup_APIC_timer((void *)(u64)calibration_result);
__sti();
/* and update all other cpus */
smp_call_function(setup_APIC_timer, (void *)(u64)calibration_result, 1, 1);
}
static inline void wait_on_irq(int cpu)
{
int count = MAXCOUNT;
for (;;) {
/*
* Wait until all interrupts are gone. Wait
* for bottom half handlers unless we're
* already executing in one..
*/
if (!irqs_running())
if (local_bh_count(cpu) || !spin_is_locked(&global_bh_lock))
break;
/* Duh, we have to loop. Release the lock to avoid deadlocks */
spin_unlock(&global_irq_lock);
for (;;) {
if (!--count) {
printk("Count spun out. Huh?\n");
count = ~0;
}
__sti();
SYNC_OTHER_CORES(cpu);
__cli();
if (irqs_running())
continue;
if (spin_is_locked(&global_irq_lock))
continue;
if (!local_bh_count(cpu) && spin_is_locked(&global_bh_lock))
continue;
if (spin_trylock(&global_irq_lock))
break;
}
}
}
asmlinkage void i8259_do_irq(int irq, struct pt_regs *regs)
{
struct irqaction *action;
int do_random, cpu;
cpu = smp_processor_id();
hardirq_enter(cpu);
if (irq >= 16)
goto out;
i8259_mask_and_ack_irq(irq);
kstat.irqs[cpu][irq]++;
action = *(irq + irq_action);
if (!action)
goto out;
if (!(action->flags & SA_INTERRUPT))
__sti();
action = *(irq + irq_action);
do_random = 0;
do {
do_random |= action->flags;
action->handler(irq, action->dev_id, regs);
action = action->next;
} while (action);
if (do_random & SA_SAMPLE_RANDOM)
add_interrupt_randomness(irq);
__cli();
unmask_irq (irq);
out:
hardirq_exit(cpu);
}
/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
* routines.
*
* error_code:
* ****0004 Protection -> Write-Protection (suprression)
* ****0010 Segment translation -> Not present (nullification)
* ****0011 Page translation -> Not present (nullification)
* ****003B Region third exception -> Not present (nullification)
*/
asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code)
{
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct * vma;
unsigned long address;
unsigned long fixup;
int write;
int si_code = SEGV_MAPERR;
int kernel_address = 0;
tsk = current;
mm = tsk->mm;
/*
* Check for low-address protection. This needs to be treated
* as a special case because the translation exception code
* field is not guaranteed to contain valid data in this case.
*/
if ((error_code & 0xff) == 4 && !(S390_lowcore.trans_exc_code & 4)) {
/* Low-address protection hit in kernel mode means
NULL pointer write access in kernel mode. */
if (!(regs->psw.mask & PSW_PROBLEM_STATE)) {
address = 0;
kernel_address = 1;
goto no_context;
}
/* Low-address protection hit in user mode 'cannot happen'. */
die ("Low-address protection", regs, error_code);
do_exit(SIGKILL);
}
/*
* get the failing address
* more specific the segment and page table portion of
* the address
*/
address = S390_lowcore.trans_exc_code&-4096L;
/*
* Check which address space the address belongs to
*/
switch (S390_lowcore.trans_exc_code & 3)
{
case 0: /* Primary Segment Table Descriptor */
kernel_address = 1;
goto no_context;
case 1: /* STD determined via access register */
if (S390_lowcore.exc_access_id == 0)
{
kernel_address = 1;
goto no_context;
}
if (regs && S390_lowcore.exc_access_id < NUM_ACRS)
{
if (regs->acrs[S390_lowcore.exc_access_id] == 0)
{
kernel_address = 1;
goto no_context;
}
if (regs->acrs[S390_lowcore.exc_access_id] == 1)
{
/* user space address */
break;
}
}
die("page fault via unknown access register", regs, error_code);
do_exit(SIGKILL);
break;
case 2: /* Secondary Segment Table Descriptor */
case 3: /* Home Segment Table Descriptor */
/* user space address */
break;
}
/*
* Check whether we have a user MM in the first place.
*/
if (in_interrupt() || !mm || !(regs->psw.mask & _PSW_IO_MASK_BIT))
goto no_context;
/*
* When we get here, the fault happened in the current
* task's user address space, so we can switch on the
* interrupts again and then search the VMAs
*/
__sti();
down_read(&mm->mmap_sem);
vma = find_vma(mm, address);
if (!vma)
goto bad_area;
if (vma->vm_start <= address)
goto good_area;
if (!(vma->vm_flags & VM_GROWSDOWN))
goto bad_area;
if (expand_stack(vma, address))
goto bad_area;
/*
* Ok, we have a good vm_area for this memory access, so
* we can handle it..
*/
good_area:
write = 0;
si_code = SEGV_ACCERR;
switch (error_code & 0xFF) {
case 0x04: /* write, present*/
write = 1;
break;
case 0x10: /* not present*/
case 0x11: /* not present*/
case 0x3B: /* not present*/
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
goto bad_area;
break;
default:
printk("code should be 4, 10 or 11 (%lX) \n",error_code&0xFF);
goto bad_area;
}
survive:
/*
* If for any reason at all we couldn't handle the fault,
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
switch (handle_mm_fault(mm, vma, address, write)) {
case 1:
tsk->min_flt++;
break;
case 2:
tsk->maj_flt++;
break;
case 0:
goto do_sigbus;
default:
goto out_of_memory;
}
up_read(&mm->mmap_sem);
return;
/*
* Something tried to access memory that isn't in our memory map..
* Fix it, but check if it's kernel or user first..
*/
bad_area:
up_read(&mm->mmap_sem);
/* User mode accesses just cause a SIGSEGV */
if (regs->psw.mask & PSW_PROBLEM_STATE) {
tsk->thread.prot_addr = address;
tsk->thread.trap_no = error_code;
#ifndef CONFIG_SYSCTL
#ifdef CONFIG_PROCESS_DEBUG
printk("User process fault: interruption code 0x%lX\n",error_code);
printk("failing address: %lX\n",address);
show_regs(regs);
#endif
#else
if (sysctl_userprocess_debug) {
printk("User process fault: interruption code 0x%lX\n",
error_code);
printk("failing address: %lX\n", address);
show_regs(regs);
}
#endif
force_sigsegv(tsk, si_code, (void *)address);
return;
}
no_context:
/* Are we prepared to handle this kernel fault? */
if ((fixup = search_exception_table(regs->psw.addr)) != 0) {
regs->psw.addr = fixup;
return;
}
/*
* Oops. The kernel tried to access some bad page. We'll have to
* terminate things with extreme prejudice.
*/
if (kernel_address)
printk(KERN_ALERT "Unable to handle kernel pointer dereference"
" at virtual kernel address %016lx\n", address);
else
printk(KERN_ALERT "Unable to handle kernel paging request"
" at virtual user address %016lx\n", address);
die("Oops", regs, error_code);
do_exit(SIGKILL);
/*
* We ran out of memory, or some other thing happened to us that made
* us unable to handle the page fault gracefully.
*/
out_of_memory:
up_read(&mm->mmap_sem);
if (tsk->pid == 1) {
tsk->policy |= SCHED_YIELD;
schedule();
down_read(&mm->mmap_sem);
goto survive;
}
printk("VM: killing process %s\n", tsk->comm);
if (regs->psw.mask & PSW_PROBLEM_STATE)
do_exit(SIGKILL);
goto no_context;
do_sigbus:
up_read(&mm->mmap_sem);
/*
* Send a sigbus, regardless of whether we were in kernel
* or user mode.
*/
tsk->thread.prot_addr = address;
tsk->thread.trap_no = error_code;
force_sig(SIGBUS, tsk);
/* Kernel mode? Handle exceptions or die */
if (!(regs->psw.mask & PSW_PROBLEM_STATE))
goto no_context;
}
/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
* routines.
*
* error_code:
* 04 Protection -> Write-Protection (suprression)
* 10 Segment translation -> Not present (nullification)
* 11 Page translation -> Not present (nullification)
* 3b Region third trans. -> Not present (nullification)
*/
extern inline void do_exception(struct pt_regs *regs, unsigned long error_code)
{
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct * vma;
unsigned long address;
int user_address;
unsigned long fixup;
int si_code = SEGV_MAPERR;
tsk = current;
mm = tsk->mm;
/*
* Check for low-address protection. This needs to be treated
* as a special case because the translation exception code
* field is not guaranteed to contain valid data in this case.
*/
if (error_code == 4 && !(S390_lowcore.trans_exc_code & 4)) {
/* Low-address protection hit in kernel mode means
NULL pointer write access in kernel mode. */
if (!(regs->psw.mask & PSW_PROBLEM_STATE)) {
address = 0;
user_address = 0;
goto no_context;
}
/* Low-address protection hit in user mode 'cannot happen'. */
die ("Low-address protection", regs, error_code);
do_exit(SIGKILL);
}
/*
* get the failing address
* more specific the segment and page table portion of
* the address
*/
address = S390_lowcore.trans_exc_code & -4096L;
user_address = check_user_space(regs, error_code);
/*
* Verify that the fault happened in user space, that
* we are not in an interrupt and that there is a
* user context.
*/
if (user_address == 0 || in_interrupt() || !mm)
goto no_context;
/*
* When we get here, the fault happened in the current
* task's user address space, so we can switch on the
* interrupts again and then search the VMAs
*/
__sti();
down_read(&mm->mmap_sem);
vma = find_vma(mm, address);
if (!vma)
goto bad_area;
if (vma->vm_start <= address)
goto good_area;
if (!(vma->vm_flags & VM_GROWSDOWN))
goto bad_area;
if (expand_stack(vma, address))
goto bad_area;
/*
* Ok, we have a good vm_area for this memory access, so
* we can handle it..
*/
good_area:
si_code = SEGV_ACCERR;
if (error_code != 4) {
/* page not present, check vm flags */
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
goto bad_area;
} else {
if (!(vma->vm_flags & VM_WRITE))
goto bad_area;
}
survive:
/*
* If for any reason at all we couldn't handle the fault,
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
switch (handle_mm_fault(mm, vma, address, error_code == 4)) {
case 1:
tsk->min_flt++;
break;
case 2:
tsk->maj_flt++;
break;
case 0:
goto do_sigbus;
default:
goto out_of_memory;
}
up_read(&mm->mmap_sem);
return;
/*
* Something tried to access memory that isn't in our memory map..
* Fix it, but check if it's kernel or user first..
*/
bad_area:
up_read(&mm->mmap_sem);
/* User mode accesses just cause a SIGSEGV */
if (regs->psw.mask & PSW_PROBLEM_STATE) {
tsk->thread.prot_addr = address;
tsk->thread.trap_no = error_code;
force_sigsegv(regs, error_code, si_code, address);
return;
}
no_context:
/* Are we prepared to handle this kernel fault? */
if ((fixup = search_exception_table(regs->psw.addr)) != 0) {
regs->psw.addr = fixup;
return;
}
/*
* Oops. The kernel tried to access some bad page. We'll have to
* terminate things with extreme prejudice.
*/
if (user_address == 0)
printk(KERN_ALERT "Unable to handle kernel pointer dereference"
" at virtual kernel address %016lx\n", address);
else
printk(KERN_ALERT "Unable to handle kernel paging request"
" at virtual user address %016lx\n", address);
die("Oops", regs, error_code);
do_exit(SIGKILL);
/*
* We ran out of memory, or some other thing happened to us that made
* us unable to handle the page fault gracefully.
*/
out_of_memory:
if (tsk->pid == 1) {
yield();
goto survive;
}
up_read(&mm->mmap_sem);
printk("VM: killing process %s\n", tsk->comm);
if (regs->psw.mask & PSW_PROBLEM_STATE)
do_exit(SIGKILL);
goto no_context;
do_sigbus:
up_read(&mm->mmap_sem);
/*
* Send a sigbus, regardless of whether we were in kernel
* or user mode.
*/
tsk->thread.prot_addr = address;
tsk->thread.trap_no = error_code;
force_sig(SIGBUS, tsk);
/* Kernel mode? Handle exceptions or die */
if (!(regs->psw.mask & PSW_PROBLEM_STATE))
goto no_context;
}
/*
* do_IRQ handles all normal device IRQ's
*/
asmlinkage void do_IRQ(int irq, struct pt_regs * regs)
{
struct irqdesc * desc;
struct irqaction * action;
int cpu;
irq = fixup_irq(irq);
/*
* Some hardware gives randomly wrong interrupts. Rather
* than crashing, do something sensible.
*/
if (irq >= NR_IRQS)
goto bad_irq;
desc = irq_desc + irq;
spin_lock(&irq_controller_lock);
desc->mask_ack(irq);
spin_unlock(&irq_controller_lock);
cpu = smp_processor_id();
irq_enter(cpu, irq);
kstat.irqs[cpu][irq]++;
desc->triggered = 1;
/* Return with this interrupt masked if no action */
action = desc->action;
if (action) {
int status = 0;
if (desc->nomask) {
spin_lock(&irq_controller_lock);
desc->unmask(irq);
spin_unlock(&irq_controller_lock);
}
if (!(action->flags & SA_INTERRUPT))
__sti();
do {
status |= action->flags;
action->handler(irq, action->dev_id, regs);
action = action->next;
} while (action);
if (status & SA_SAMPLE_RANDOM)
add_interrupt_randomness(irq);
__cli();
if (!desc->nomask && desc->enabled) {
spin_lock(&irq_controller_lock);
desc->unmask(irq);
spin_unlock(&irq_controller_lock);
}
}
/*
* Debug measure - hopefully we can continue if an
* IRQ lockup problem occurs...
*/
check_irq_lock(desc, irq, regs);
irq_exit(cpu, irq);
if (softirq_active(cpu) & softirq_mask(cpu))
do_softirq();
return;
bad_irq:
irq_err_count += 1;
printk(KERN_ERR "IRQ: spurious interrupt %d\n", irq);
return;
}
/*
* INITIAL C ENTRY POINT.
*/
void start_kernel(start_info_t *si)
{
static char hello[] = "Bootstrapping...\n";
(void)HYPERVISOR_console_io(CONSOLEIO_write, strlen(hello), hello);
setup_xen_features();
pvh_early_init();
arch_init(si);
trap_init();
/* print out some useful information */
printk("Xen Minimal OS!\n");
printk(" start_info: %p(VA)\n", si);
printk(" nr_pages: 0x%lx\n", si->nr_pages);
printk(" shared_inf: 0x%08lx(MA)\n", si->shared_info);
printk(" pt_base: %p(VA)\n", (void *)si->pt_base);
printk("nr_pt_frames: 0x%lx\n", si->nr_pt_frames);
printk(" mfn_list: %p(VA)\n", (void *)si->mfn_list);
printk(" mod_start: 0x%lx(VA)\n", si->mod_start);
printk(" mod_len: %lu\n", si->mod_len);
printk(" flags: 0x%x\n", (unsigned int)si->flags);
printk(" cmd_line: %s\n",
si->cmd_line ? (const char *)si->cmd_line : "NULL");
/* Set up events. */
init_events();
/* ENABLE EVENT DELIVERY. This is disabled at start of day. */
__sti();
arch_print_info();
/* Init memory management. */
init_mm();
/* Init time and timers. */
init_time();
/* Init the console driver. */
init_console();
/* Init grant tables */
init_gnttab();
/* Init scheduler. */
init_sched();
/* Init XenBus */
init_xenbus();
#ifdef CONFIG_XENBUS
/* Init shutdown thread */
init_shutdown(si);
#endif
/* Call (possibly overridden) app_main() */
app_main(&start_info);
/* Everything initialised, start idle thread */
run_idle_thread();
}
/*
* do_IRQ handles all normal device IRQ's (the special
* SMP cross-CPU interrupts have their own specific
* handlers).
*/
asmlinkage unsigned int __noinstrument do_IRQ(int irq, struct pt_regs *regs)
{
/*
* We ack quickly, we don't want the irq controller
* thinking we're snobs just because some other CPU has
* disabled global interrupts (we have already done the
* INT_ACK cycles, it's too late to try to pretend to the
* controller that we aren't taking the interrupt).
*
* 0 return value means that this irq is already being
* handled by some other CPU. (or is disabled)
*/
int cpu = smp_processor_id();
irq_desc_t *desc = irq_desc + irq;
struct irqaction * action;
unsigned int status;
#ifdef CONFIG_ILATENCY
{
extern void interrupt_overhead_start(void);
interrupt_overhead_start();
}
#endif /* CONFIG_ILATENCY */
preempt_disable();
TRACE_IRQ_ENTRY(irq, !user_mode(regs));
kstat.irqs[cpu][irq]++;
spin_lock(&desc->lock);
desc->handler->ack(irq);
/*
REPLAY is when Linux resends an IRQ that was dropped earlier
WAITING is used by probe to mark irqs that are being tested
*/
status = desc->status & ~(IRQ_REPLAY | IRQ_WAITING);
status |= IRQ_PENDING; /* we _want_ to handle it */
/*
* If the IRQ is disabled for whatever reason, we cannot
* use the action we have.
*/
action = NULL;
if (!(status & (IRQ_DISABLED | IRQ_INPROGRESS))) {
action = desc->action;
status &= ~IRQ_PENDING; /* we commit to handling */
status |= IRQ_INPROGRESS; /* we are handling it */
}
desc->status = status;
/*
* If there is no IRQ handler or it was disabled, exit early.
Since we set PENDING, if another processor is handling
a different instance of this same irq, the other processor
will take care of it.
*/
if (!action)
goto out;
/*
* Edge triggered interrupts need to remember
* pending events.
* This applies to any hw interrupts that allow a second
* instance of the same irq to arrive while we are in do_IRQ
* or in the handler. But the code here only handles the _second_
* instance of the irq, not the third or fourth. So it is mostly
* useful for irq hardware that does not mask cleanly in an
* SMP environment.
*/
#ifdef CONFIG_ILATENCY
{
extern void interrupt_overhead_stop(void);
interrupt_overhead_stop();
}
#endif /* CONFIG_ILATENCY */
for (;;) {
spin_unlock(&desc->lock);
handle_IRQ_event(irq, regs, action);
spin_lock(&desc->lock);
if (!(desc->status & IRQ_PENDING))
break;
desc->status &= ~IRQ_PENDING;
}
desc->status &= ~IRQ_INPROGRESS;
out:
/*
* The ->end() handler has to deal with interrupts which got
* disabled while the handler was running.
*/
desc->handler->end(irq);
spin_unlock(&desc->lock);
TRACE_IRQ_EXIT();
if (softirq_pending(cpu))
do_softirq();
#if defined(CONFIG_PREEMPT)
for(;;) {
preempt_enable_no_resched();
if (preempt_is_disabled() || !need_resched())
break;
db_assert(intr_off());
db_assert(!in_interrupt());
preempt_disable();
__sti();
preempt_schedule();
__cli();
}
#endif
#ifdef CONFIG_ILATENCY
intr_ret_from_exception();
#endif
return 1;
}
/* Stopping processors. */
void smp_stop_cpu_irq(void)
{
__sti();
while(1)
barrier();
}
static int __do_suspend(void *ignore)
{
int i, j, k, fpp, err;
extern unsigned long max_pfn;
extern unsigned long *pfn_to_mfn_frame_list_list;
extern unsigned long *pfn_to_mfn_frame_list[];
extern void time_resume(void);
BUG_ON(smp_processor_id() != 0);
BUG_ON(in_interrupt());
if (xen_feature(XENFEAT_auto_translated_physmap)) {
printk(KERN_WARNING "Cannot suspend in "
"auto_translated_physmap mode.\n");
return -EOPNOTSUPP;
}
err = smp_suspend();
if (err)
return err;
xenbus_suspend();
preempt_disable();
#ifdef __i386__
kmem_cache_shrink(pgd_cache);
#endif
mm_pin_all();
__cli();
preempt_enable();
gnttab_suspend();
HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
clear_fixmap(FIX_SHARED_INFO);
xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn);
xen_start_info->console_mfn = mfn_to_pfn(xen_start_info->console_mfn);
/*
* We'll stop somewhere inside this hypercall. When it returns,
* we'll start resuming after the restore.
*/
HYPERVISOR_suspend(virt_to_mfn(xen_start_info));
shutting_down = SHUTDOWN_INVALID;
set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
memset(empty_zero_page, 0, PAGE_SIZE);
HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
virt_to_mfn(pfn_to_mfn_frame_list_list);
fpp = PAGE_SIZE/sizeof(unsigned long);
for (i = 0, j = 0, k = -1; i < max_pfn; i += fpp, j++) {
if ((j % fpp) == 0) {
k++;
pfn_to_mfn_frame_list_list[k] =
virt_to_mfn(pfn_to_mfn_frame_list[k]);
j = 0;
}
pfn_to_mfn_frame_list[k][j] =
virt_to_mfn(&phys_to_machine_mapping[i]);
}
HYPERVISOR_shared_info->arch.max_pfn = max_pfn;
gnttab_resume();
irq_resume();
time_resume();
switch_idle_mm();
__sti();
xencons_resume();
xenbus_resume();
smp_resume();
return err;
}
void
x86_enable_intr(void)
{
__sti();
}