xv6/proc.c
rsc c8919e6537 kernel SMP interruptibility fixes.
Last year, right before I sent xv6 to the printer, I changed the
SETGATE calls so that interrupts would be disabled on entry to
interrupt handlers, and I added the nlock++ / nlock-- in trap()
so that interrupts would stay disabled while the hw handlers
(but not the syscall handler) did their work.  I did this because
the kernel was otherwise causing Bochs to triple-fault in SMP
mode, and time was short.

Robert observed yesterday that something was keeping the SMP
preemption user test from working.  It turned out that when I
simplified the lapic code I swapped the order of two register
writes that I didn't realize were order dependent.  I fixed that
and then since I had everything paged in kept going and tried
to figure out why you can't leave interrupts on during interrupt
handlers.  There are a few issues.

First, there must be some way to keep interrupts from "stacking
up" and overflowing the stack.  Keeping interrupts off the whole
time solves this problem -- even if the clock tick handler runs
long enough that the next clock tick is waiting when it finishes,
keeping interrupts off means that the handler runs all the way
through the "iret" before the next handler begins.  This is not
really a problem unless you are putting too many prints in trap
-- if the OS is doing its job right, the handlers should run
quickly and not stack up.

Second, if xv6 had page faults, then it would be important to
keep interrupts disabled between the start of the interrupt and
the time that cr2 was read, to avoid a scenario like:

   p1 page faults [cr2 set to faulting address]
   p1 starts executing trapasm.S
   clock interrupt, p1 preempted, p2 starts executing
   p2 page faults [cr2 set to another faulting address]
   p2 starts, finishes fault handler
   p1 rescheduled, reads cr2, sees wrong fault address

Alternately p1 could be rescheduled on the other cpu, in which
case it would still see the wrong cr2.  That said, I think cr2
is the only interrupt state that isn't pushed onto the interrupt
stack atomically at fault time, and xv6 doesn't care.  (This isn't
entirely hypothetical -- I debugged this problem on Plan 9.)

Third, and this is the big one, it is not safe to call cpu()
unless interrupts are disabled.  If interrupts are enabled then
there is no guarantee that, between the time cpu() looks up the
cpu id and the time that it the result gets used, the process
has not been rescheduled to the other cpu.  For example, the
very commonly-used expression curproc[cpu()] (aka the macro cp)
can end up referring to the wrong proc: the code stores the
result of cpu() in %eax, gets rescheduled to the other cpu at
just the wrong instant, and then reads curproc[%eax].

We use curproc[cpu()] to get the current process a LOT.  In that
particular case, if we arranged for the current curproc entry
to be addressed by %fs:0 and just use a different %fs on each
CPU, then we could safely get at curproc even with interrupts
disabled, since the read of %fs would be atomic with the read
of %fs:0.  Alternately, we could have a curproc() function that
disables interrupts while computing curproc[cpu()].  I've done
that last one.

Even in the current kernel, with interrupts off on entry to trap,
interrupts are enabled inside release if there are no locks held.
Also, the scheduler's idle loop must be interruptible at times
so that the clock and disk interrupts (which might make processes
runnable) can be handled.

In addition to the rampant use of curproc[cpu()], this little
snippet from acquire is wrong on smp:

  if(cpus[cpu()].nlock == 0)
    cli();
  cpus[cpu()].nlock++;

because if interrupts are off then we might call cpu(), get
rescheduled to a different cpu, look at cpus[oldcpu].nlock, and
wrongly decide not to disable interrupts on the new cpu.  The
fix is to always call cli().  But this is wrong too:

  if(holding(lock))
    panic("acquire");
  cli();
  cpus[cpu()].nlock++;

because holding looks at cpu().  The fix is:

  cli();
  if(holding(lock))
    panic("acquire");
  cpus[cpu()].nlock++;

I've done that, and I changed cpu() to complain the first time
it gets called with interrupts disabled.  (It gets called too
much to complain every time.)

I added new functions splhi and spllo that are like acquire and
release but without the locking:

  void
  splhi(void)
  {
    cli();
    cpus[cpu()].nsplhi++;
  }

  void
  spllo(void)
  {
    if(--cpus[cpu()].nsplhi == 0)
      sti();
  }

and I've used those to protect other sections of code that refer
to cpu() when interrupts would otherwise be disabled (basically
just curproc and setupsegs).  I also use them in acquire/release
and got rid of nlock.

I'm not thrilled with the names, but I think the concept -- a
counted cli/sti -- is sound.  Having them also replaces the
nlock++/nlock-- in trap.c and main.c, which is nice.


Final note: it's still not safe to enable interrupts in
the middle of trap() between lapic_eoi and returning
to user space.  I don't understand why, but we get a
fault on pop %es because 0x10 is a bad segment
descriptor (!) and then the fault faults trying to go into
a new interrupt because 0x8 is a bad segment descriptor too!
Triple fault.  I haven't debugged this yet.
2007-09-27 12:58:42 +00:00

483 lines
10 KiB
C

#include "types.h"
#include "defs.h"
#include "param.h"
#include "mmu.h"
#include "x86.h"
#include "proc.h"
#include "spinlock.h"
struct spinlock proc_table_lock;
struct proc proc[NPROC];
static struct proc *initproc;
int nextpid = 1;
extern void forkret(void);
extern void forkret1(struct trapframe*);
void
pinit(void)
{
initlock(&proc_table_lock, "proc_table");
}
// Look in the process table for an UNUSED proc.
// If found, change state to EMBRYO and return it.
// Otherwise return 0.
static struct proc*
allocproc(void)
{
int i;
struct proc *p;
acquire(&proc_table_lock);
for(i = 0; i < NPROC; i++){
p = &proc[i];
if(p->state == UNUSED){
p->state = EMBRYO;
p->pid = nextpid++;
release(&proc_table_lock);
return p;
}
}
release(&proc_table_lock);
return 0;
}
// Grow current process's memory by n bytes.
// Return old size on success, -1 on failure.
int
growproc(int n)
{
char *newmem, *oldmem;
newmem = kalloc(cp->sz + n);
if(newmem == 0)
return -1;
memmove(newmem, cp->mem, cp->sz);
memset(newmem + cp->sz, 0, n);
oldmem = cp->mem;
cp->mem = newmem;
kfree(oldmem, cp->sz);
cp->sz += n;
setupsegs(cp);
return cp->sz - n;
}
// Set up CPU's segment descriptors and task state for a given process.
// If p==0, set up for "idle" state for when scheduler() is running.
void
setupsegs(struct proc *p)
{
struct cpu *c;
splhi();
c = &cpus[cpu()];
c->ts.ss0 = SEG_KDATA << 3;
if(p)
c->ts.esp0 = (uint)(p->kstack + KSTACKSIZE);
else
c->ts.esp0 = 0xffffffff;
c->gdt[0] = SEG_NULL;
c->gdt[SEG_KCODE] = SEG(STA_X|STA_R, 0, 0x100000 + 64*1024-1, 0);
c->gdt[SEG_KDATA] = SEG(STA_W, 0, 0xffffffff, 0);
c->gdt[SEG_TSS] = SEG16(STS_T32A, (uint)&c->ts, sizeof(c->ts)-1, 0);
c->gdt[SEG_TSS].s = 0;
if(p){
c->gdt[SEG_UCODE] = SEG(STA_X|STA_R, (uint)p->mem, p->sz-1, DPL_USER);
c->gdt[SEG_UDATA] = SEG(STA_W, (uint)p->mem, p->sz-1, DPL_USER);
} else {
c->gdt[SEG_UCODE] = SEG_NULL;
c->gdt[SEG_UDATA] = SEG_NULL;
}
lgdt(c->gdt, sizeof(c->gdt));
ltr(SEG_TSS << 3);
spllo();
}
// Create a new process copying p as the parent.
// Sets up stack to return as if from system call.
// Caller must set state of returned proc to RUNNABLE.
struct proc*
copyproc(struct proc *p)
{
int i;
struct proc *np;
// Allocate process.
if((np = allocproc()) == 0)
return 0;
// Allocate kernel stack.
if((np->kstack = kalloc(KSTACKSIZE)) == 0){
np->state = UNUSED;
return 0;
}
np->tf = (struct trapframe*)(np->kstack + KSTACKSIZE) - 1;
if(p){ // Copy process state from p.
np->parent = p;
memmove(np->tf, p->tf, sizeof(*np->tf));
np->sz = p->sz;
if((np->mem = kalloc(np->sz)) == 0){
kfree(np->kstack, KSTACKSIZE);
np->kstack = 0;
np->state = UNUSED;
return 0;
}
memmove(np->mem, p->mem, np->sz);
for(i = 0; i < NOFILE; i++)
if(p->ofile[i])
np->ofile[i] = filedup(p->ofile[i]);
np->cwd = idup(p->cwd);
}
// Set up new context to start executing at forkret (see below).
memset(&np->context, 0, sizeof(np->context));
np->context.eip = (uint)forkret;
np->context.esp = (uint)np->tf;
// Clear %eax so that fork system call returns 0 in child.
np->tf->eax = 0;
return np;
}
// Set up first user process.
void
userinit(void)
{
struct proc *p;
extern uchar _binary_initcode_start[], _binary_initcode_size[];
p = copyproc(0);
p->sz = PAGE;
p->mem = kalloc(p->sz);
p->cwd = namei("/");
memset(p->tf, 0, sizeof(*p->tf));
p->tf->cs = (SEG_UCODE << 3) | DPL_USER;
p->tf->ds = (SEG_UDATA << 3) | DPL_USER;
p->tf->es = p->tf->ds;
p->tf->ss = p->tf->ds;
p->tf->eflags = FL_IF;
p->tf->esp = p->sz;
// Make return address readable; needed for some gcc.
p->tf->esp -= 4;
*(uint*)(p->mem + p->tf->esp) = 0xefefefef;
// On entry to user space, start executing at beginning of initcode.S.
p->tf->eip = 0;
memmove(p->mem, _binary_initcode_start, (int)_binary_initcode_size);
safestrcpy(p->name, "initcode", sizeof(p->name));
p->state = RUNNABLE;
initproc = p;
}
// Return currently running process.
// XXX comment better
struct proc*
curproc(void)
{
struct proc *p;
splhi();
p = cpus[cpu()].curproc;
spllo();
return p;
}
//PAGEBREAK: 42
// Per-CPU process scheduler.
// Each CPU calls scheduler() after setting itself up.
// Scheduler never returns. It loops, doing:
// - choose a process to run
// - swtch to start running that process
// - eventually that process transfers control
// via swtch back to the scheduler.
void
scheduler(void)
{
struct proc *p;
struct cpu *c;
int i;
for(;;){
// Loop over process table looking for process to run.
acquire(&proc_table_lock);
c = &cpus[cpu()];
for(i = 0; i < NPROC; i++){
p = &proc[i];
if(p->state != RUNNABLE)
continue;
// Switch to chosen process. It is the process's job
// to release proc_table_lock and then reacquire it
// before jumping back to us.
c->curproc = p;
setupsegs(p);
p->state = RUNNING;
swtch(&c->context, &p->context);
// Process is done running for now.
// It should have changed its p->state before coming back.
c->curproc = 0;
setupsegs(0);
}
release(&proc_table_lock);
}
}
// Enter scheduler. Must already hold proc_table_lock
// and have changed curproc[cpu()]->state.
void
sched(void)
{
if(read_eflags()&FL_IF)
panic("sched interruptible");
if(cp->state == RUNNING)
panic("sched running");
if(!holding(&proc_table_lock))
panic("sched proc_table_lock");
if(cpus[cpu()].nsplhi != 1)
panic("sched locks");
swtch(&cp->context, &cpus[cpu()].context);
}
// Give up the CPU for one scheduling round.
void
yield(void)
{
acquire(&proc_table_lock);
cp->state = RUNNABLE;
sched();
release(&proc_table_lock);
}
// A fork child's very first scheduling by scheduler()
// will swtch here. "Return" to user space.
void
forkret(void)
{
// Still holding proc_table_lock from scheduler.
release(&proc_table_lock);
// Jump into assembly, never to return.
forkret1(cp->tf);
}
// Atomically release lock and sleep on chan.
// Reacquires lock when reawakened.
void
sleep(void *chan, struct spinlock *lk)
{
if(cp == 0)
panic("sleep");
if(lk == 0)
panic("sleep without lk");
// Must acquire proc_table_lock in order to
// change p->state and then call sched.
// Once we hold proc_table_lock, we can be
// guaranteed that we won't miss any wakeup
// (wakeup runs with proc_table_lock locked),
// so it's okay to release lk.
if(lk != &proc_table_lock){
acquire(&proc_table_lock);
release(lk);
}
// Go to sleep.
cp->chan = chan;
cp->state = SLEEPING;
sched();
// Tidy up.
cp->chan = 0;
// Reacquire original lock.
if(lk != &proc_table_lock){
release(&proc_table_lock);
acquire(lk);
}
}
//PAGEBREAK!
// Wake up all processes sleeping on chan.
// Proc_table_lock must be held.
static void
wakeup1(void *chan)
{
struct proc *p;
for(p = proc; p < &proc[NPROC]; p++)
if(p->state == SLEEPING && p->chan == chan)
p->state = RUNNABLE;
}
// Wake up all processes sleeping on chan.
// Proc_table_lock is acquired and released.
void
wakeup(void *chan)
{
acquire(&proc_table_lock);
wakeup1(chan);
release(&proc_table_lock);
}
// Kill the process with the given pid.
// Process won't actually exit until it returns
// to user space (see trap in trap.c).
int
kill(int pid)
{
struct proc *p;
acquire(&proc_table_lock);
for(p = proc; p < &proc[NPROC]; p++){
if(p->pid == pid){
p->killed = 1;
// Wake process from sleep if necessary.
if(p->state == SLEEPING)
p->state = RUNNABLE;
release(&proc_table_lock);
return 0;
}
}
release(&proc_table_lock);
return -1;
}
// Exit the current process. Does not return.
// Exited processes remain in the zombie state
// until their parent calls wait() to find out they exited.
void
exit(void)
{
struct proc *p;
int fd;
if(cp == initproc)
panic("init exiting");
// Close all open files.
for(fd = 0; fd < NOFILE; fd++){
if(cp->ofile[fd]){
fileclose(cp->ofile[fd]);
cp->ofile[fd] = 0;
}
}
iput(cp->cwd);
cp->cwd = 0;
acquire(&proc_table_lock);
// Parent might be sleeping in proc_wait.
wakeup1(cp->parent);
// Pass abandoned children to init.
for(p = proc; p < &proc[NPROC]; p++){
if(p->parent == cp){
p->parent = initproc;
if(p->state == ZOMBIE)
wakeup1(initproc);
}
}
// Jump into the scheduler, never to return.
cp->killed = 0;
cp->state = ZOMBIE;
sched();
panic("zombie exit");
}
// Wait for a child process to exit and return its pid.
// Return -1 if this process has no children.
int
wait(void)
{
struct proc *p;
int i, havekids, pid;
acquire(&proc_table_lock);
for(;;){
// Scan through table looking for zombie children.
havekids = 0;
for(i = 0; i < NPROC; i++){
p = &proc[i];
if(p->state == UNUSED)
continue;
if(p->parent == cp){
if(p->state == ZOMBIE){
// Found one.
kfree(p->mem, p->sz);
kfree(p->kstack, KSTACKSIZE);
pid = p->pid;
p->state = UNUSED;
p->pid = 0;
p->parent = 0;
p->name[0] = 0;
release(&proc_table_lock);
return pid;
}
havekids = 1;
}
}
// No point waiting if we don't have any children.
if(!havekids || cp->killed){
release(&proc_table_lock);
return -1;
}
// Wait for children to exit. (See wakeup1 call in proc_exit.)
sleep(cp, &proc_table_lock);
}
}
// Print a process listing to console. For debugging.
// Runs when user types ^P on console.
// No lock to avoid wedging a stuck machine further.
void
procdump(void)
{
static char *states[] = {
[UNUSED] "unused",
[EMBRYO] "embryo",
[SLEEPING] "sleep ",
[RUNNABLE] "runble",
[RUNNING] "run ",
[ZOMBIE] "zombie"
};
int i, j;
struct proc *p;
char *state;
uint pc[10];
for(i = 0; i < NPROC; i++){
p = &proc[i];
if(p->state == UNUSED)
continue;
if(p->state >= 0 && p->state < NELEM(states) && states[p->state])
state = states[p->state];
else
state = "???";
cprintf("%d %s %s", p->pid, state, p->name);
if(p->state == SLEEPING){
getcallerpcs((uint*)p->context.ebp+2, pc);
for(j=0; j<10 && pc[j] != 0; j++)
cprintf(" %p", pc[j]);
}
cprintf("\n");
}
}