forked from Minki/linux
c14b84949e
Add a separate IRQ stack. This differs from i386 in having the entire interrupt run on a separate stack rather than starting on the normal kernel stack and switching over once some preparation has been done. The underlying mechanism, is of course, sigaltstack. Another difference is that interrupts that happen in userspace are handled on the normal kernel stack. These cause a wait wakeup instead of a signal delivery so there is no point in trying to switch stacks for these. There's no other stuff on the stack, so there is no extra stack consumption. This quirk makes it possible to have the entire interrupt run on a separate stack - process preemption (and calls to schedule()) happens on a normal kernel stack. If we enable CONFIG_PREEMPT, this will need to be rethought. The IRQ stack for CPU 0 is declared in the same way as the initial kernel stack. IRQ stacks for other CPUs will be allocated dynamically. An extra field was added to the thread_info structure. When the active thread_info is copied to the IRQ stack, the real_thread field points back to the original stack. This makes it easy to tell where to copy the thread_info struct back to when the interrupt is finished. It also serves as a marker of a nested interrupt. It is NULL for the first interrupt on the stack, and non-NULL for any nested interrupts. Care is taken to behave correctly if a second interrupt comes in when the thread_info structure is being set up or taken down. I could just disable interrupts here, but I don't feel like giving up any of the performance gained by not flipping signals on and off. If an interrupt comes in during these critical periods, the handler can't run because it has no idea what shape the stack is in. So, it sets a bit for its signal in a global mask and returns. The outer handler will deal with this signal itself. Atomicity is had with xchg. A nested interrupt that needs to bail out will xchg its signal mask into pending_mask and repeat in case yet another interrupt hit at the same time, until the mask stabilizes. The outermost interrupt will set up the thread_info and xchg a zero into pending_mask when it is done. At this point, nested interrupts will look at ->real_thread and see that no setup needs to be done. They can just continue normally. Similar care needs to be taken when exiting the outer handler. If another interrupt comes in while it is copying the thread_info, it will drop a bit into pending_mask. The outer handler will check this and if it is non-zero, will loop, set up the stack again, and handle the interrupt. Signed-off-by: Jeff Dike <jdike@linux.intel.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
289 lines
6.4 KiB
C
289 lines
6.4 KiB
C
/*
|
|
* Copyright (C) 2004 PathScale, Inc
|
|
* Licensed under the GPL
|
|
*/
|
|
|
|
#include <signal.h>
|
|
#include <stdio.h>
|
|
#include <unistd.h>
|
|
#include <stdlib.h>
|
|
#include <errno.h>
|
|
#include <stdarg.h>
|
|
#include <string.h>
|
|
#include <sys/mman.h>
|
|
#include "user.h"
|
|
#include "signal_kern.h"
|
|
#include "sysdep/sigcontext.h"
|
|
#include "sysdep/barrier.h"
|
|
#include "sigcontext.h"
|
|
#include "mode.h"
|
|
#include "os.h"
|
|
|
|
/* These are the asynchronous signals. SIGVTALRM and SIGARLM are handled
|
|
* together under SIGVTALRM_BIT. SIGPROF is excluded because we want to
|
|
* be able to profile all of UML, not just the non-critical sections. If
|
|
* profiling is not thread-safe, then that is not my problem. We can disable
|
|
* profiling when SMP is enabled in that case.
|
|
*/
|
|
#define SIGIO_BIT 0
|
|
#define SIGIO_MASK (1 << SIGIO_BIT)
|
|
|
|
#define SIGVTALRM_BIT 1
|
|
#define SIGVTALRM_MASK (1 << SIGVTALRM_BIT)
|
|
|
|
#define SIGALRM_BIT 2
|
|
#define SIGALRM_MASK (1 << SIGALRM_BIT)
|
|
|
|
/* These are used by both the signal handlers and
|
|
* block/unblock_signals. I don't want modifications cached in a
|
|
* register - they must go straight to memory.
|
|
*/
|
|
static volatile int signals_enabled = 1;
|
|
static volatile int pending = 0;
|
|
|
|
void sig_handler(int sig, struct sigcontext *sc)
|
|
{
|
|
int enabled;
|
|
|
|
enabled = signals_enabled;
|
|
if(!enabled && (sig == SIGIO)){
|
|
pending |= SIGIO_MASK;
|
|
return;
|
|
}
|
|
|
|
block_signals();
|
|
|
|
CHOOSE_MODE_PROC(sig_handler_common_tt, sig_handler_common_skas,
|
|
sig, sc);
|
|
|
|
set_signals(enabled);
|
|
}
|
|
|
|
static void real_alarm_handler(int sig, struct sigcontext *sc)
|
|
{
|
|
union uml_pt_regs regs;
|
|
|
|
if(sig == SIGALRM)
|
|
switch_timers(0);
|
|
|
|
if(sc != NULL)
|
|
copy_sc(®s, sc);
|
|
regs.skas.is_user = 0;
|
|
unblock_signals();
|
|
timer_handler(sig, ®s);
|
|
|
|
if(sig == SIGALRM)
|
|
switch_timers(1);
|
|
}
|
|
|
|
void alarm_handler(int sig, struct sigcontext *sc)
|
|
{
|
|
int enabled;
|
|
|
|
enabled = signals_enabled;
|
|
if(!signals_enabled){
|
|
if(sig == SIGVTALRM)
|
|
pending |= SIGVTALRM_MASK;
|
|
else pending |= SIGALRM_MASK;
|
|
|
|
return;
|
|
}
|
|
|
|
block_signals();
|
|
|
|
real_alarm_handler(sig, sc);
|
|
set_signals(enabled);
|
|
}
|
|
|
|
void set_sigstack(void *sig_stack, int size)
|
|
{
|
|
stack_t stack = ((stack_t) { .ss_flags = 0,
|
|
.ss_sp = (__ptr_t) sig_stack,
|
|
.ss_size = size - sizeof(void *) });
|
|
|
|
if(sigaltstack(&stack, NULL) != 0)
|
|
panic("enabling signal stack failed, errno = %d\n", errno);
|
|
}
|
|
|
|
void remove_sigstack(void)
|
|
{
|
|
stack_t stack = ((stack_t) { .ss_flags = SS_DISABLE,
|
|
.ss_sp = NULL,
|
|
.ss_size = 0 });
|
|
|
|
if(sigaltstack(&stack, NULL) != 0)
|
|
panic("disabling signal stack failed, errno = %d\n", errno);
|
|
}
|
|
|
|
void (*handlers[_NSIG])(int sig, struct sigcontext *sc);
|
|
|
|
void handle_signal(int sig, struct sigcontext *sc)
|
|
{
|
|
unsigned long pending = 0;
|
|
|
|
do {
|
|
int nested, bail;
|
|
|
|
/*
|
|
* pending comes back with one bit set for each
|
|
* interrupt that arrived while setting up the stack,
|
|
* plus a bit for this interrupt, plus the zero bit is
|
|
* set if this is a nested interrupt.
|
|
* If bail is true, then we interrupted another
|
|
* handler setting up the stack. In this case, we
|
|
* have to return, and the upper handler will deal
|
|
* with this interrupt.
|
|
*/
|
|
bail = to_irq_stack(sig, &pending);
|
|
if(bail)
|
|
return;
|
|
|
|
nested = pending & 1;
|
|
pending &= ~1;
|
|
|
|
while((sig = ffs(pending)) != 0){
|
|
sig--;
|
|
pending &= ~(1 << sig);
|
|
(*handlers[sig])(sig, sc);
|
|
}
|
|
|
|
/* Again, pending comes back with a mask of signals
|
|
* that arrived while tearing down the stack. If this
|
|
* is non-zero, we just go back, set up the stack
|
|
* again, and handle the new interrupts.
|
|
*/
|
|
if(!nested)
|
|
pending = from_irq_stack(nested);
|
|
} while(pending);
|
|
}
|
|
|
|
extern void hard_handler(int sig);
|
|
|
|
void set_handler(int sig, void (*handler)(int), int flags, ...)
|
|
{
|
|
struct sigaction action;
|
|
va_list ap;
|
|
sigset_t sig_mask;
|
|
int mask;
|
|
|
|
handlers[sig] = (void (*)(int, struct sigcontext *)) handler;
|
|
action.sa_handler = hard_handler;
|
|
|
|
sigemptyset(&action.sa_mask);
|
|
|
|
va_start(ap, flags);
|
|
while((mask = va_arg(ap, int)) != -1)
|
|
sigaddset(&action.sa_mask, mask);
|
|
va_end(ap);
|
|
|
|
action.sa_flags = flags;
|
|
action.sa_restorer = NULL;
|
|
if(sigaction(sig, &action, NULL) < 0)
|
|
panic("sigaction failed - errno = %d\n", errno);
|
|
|
|
sigemptyset(&sig_mask);
|
|
sigaddset(&sig_mask, sig);
|
|
if(sigprocmask(SIG_UNBLOCK, &sig_mask, NULL) < 0)
|
|
panic("sigprocmask failed - errno = %d\n", errno);
|
|
}
|
|
|
|
int change_sig(int signal, int on)
|
|
{
|
|
sigset_t sigset, old;
|
|
|
|
sigemptyset(&sigset);
|
|
sigaddset(&sigset, signal);
|
|
sigprocmask(on ? SIG_UNBLOCK : SIG_BLOCK, &sigset, &old);
|
|
return(!sigismember(&old, signal));
|
|
}
|
|
|
|
void block_signals(void)
|
|
{
|
|
signals_enabled = 0;
|
|
/* This must return with signals disabled, so this barrier
|
|
* ensures that writes are flushed out before the return.
|
|
* This might matter if gcc figures out how to inline this and
|
|
* decides to shuffle this code into the caller.
|
|
*/
|
|
mb();
|
|
}
|
|
|
|
void unblock_signals(void)
|
|
{
|
|
int save_pending;
|
|
|
|
if(signals_enabled == 1)
|
|
return;
|
|
|
|
/* We loop because the IRQ handler returns with interrupts off. So,
|
|
* interrupts may have arrived and we need to re-enable them and
|
|
* recheck pending.
|
|
*/
|
|
while(1){
|
|
/* Save and reset save_pending after enabling signals. This
|
|
* way, pending won't be changed while we're reading it.
|
|
*/
|
|
signals_enabled = 1;
|
|
|
|
/* Setting signals_enabled and reading pending must
|
|
* happen in this order.
|
|
*/
|
|
mb();
|
|
|
|
save_pending = pending;
|
|
if(save_pending == 0){
|
|
/* This must return with signals enabled, so
|
|
* this barrier ensures that writes are
|
|
* flushed out before the return. This might
|
|
* matter if gcc figures out how to inline
|
|
* this (unlikely, given its size) and decides
|
|
* to shuffle this code into the caller.
|
|
*/
|
|
mb();
|
|
return;
|
|
}
|
|
|
|
pending = 0;
|
|
|
|
/* We have pending interrupts, so disable signals, as the
|
|
* handlers expect them off when they are called. They will
|
|
* be enabled again above.
|
|
*/
|
|
|
|
signals_enabled = 0;
|
|
|
|
/* Deal with SIGIO first because the alarm handler might
|
|
* schedule, leaving the pending SIGIO stranded until we come
|
|
* back here.
|
|
*/
|
|
if(save_pending & SIGIO_MASK)
|
|
CHOOSE_MODE_PROC(sig_handler_common_tt,
|
|
sig_handler_common_skas, SIGIO, NULL);
|
|
|
|
if(save_pending & SIGALRM_MASK)
|
|
real_alarm_handler(SIGALRM, NULL);
|
|
|
|
if(save_pending & SIGVTALRM_MASK)
|
|
real_alarm_handler(SIGVTALRM, NULL);
|
|
}
|
|
}
|
|
|
|
int get_signals(void)
|
|
{
|
|
return signals_enabled;
|
|
}
|
|
|
|
int set_signals(int enable)
|
|
{
|
|
int ret;
|
|
if(signals_enabled == enable)
|
|
return enable;
|
|
|
|
ret = signals_enabled;
|
|
if(enable)
|
|
unblock_signals();
|
|
else block_signals();
|
|
|
|
return ret;
|
|
}
|