i think my cmpxchg use was wrong in acquire

nesting cli/sti: release shouldn't always enable interrupts separate setup of lapic from starting of other cpus, so cpu() works earlier flag to disable locking in console output make locks work even when curproc==0 (still crashes in clock interrupt)

i think my cmpxchg use was wrong in acquire
8148b6ee · rtm · 66432474 · 8148b6ee · 8148b6ee · 8148b6ee
--- a/Notes
+++ b/Notes
@@ -125,6 +125,10 @@ in general, the table locks protect both free-ness and

 why can't i get a lock in console code?
  always triple fault
+  because release turns on interrupts!
+  a bad idea very early in main()
+  but mp_init() calls cprintf
+
 lock code shouldn't call cprintf...
 ide_init doesn't work now?
 and IOAPIC: read from unsupported address

--- a/console.c
+++ b/console.c
@@ -4,6 +4,7 @@
 #include "spinlock.h"

 struct spinlock console_lock;
+int use_printf_lock = 0;

 /*
 * copy console output to parallel port, which you can tell
@@ -29,7 +30,8 @@ cons_putc(int c)
  unsigned short *crt = (unsigned short *) 0xB8000; // base of CGA memory
  int ind;

-  //acquire(&console_lock);
+  if(use_printf_lock)
+    acquire(&console_lock);

  lpt_putc(c);

@@ -62,7 +64,8 @@ cons_putc(int c)
  outb(crtport, 15);
  outb(crtport + 1, ind);

-  //release(&console_lock);
+  if(use_printf_lock)
+    release(&console_lock);
 }

 void
@@ -127,6 +130,8 @@ cprintf(char *fmt, ...)
 void
 panic(char *s)
 {
+  use_printf_lock = 0;
+  cprintf("panic: ");
  cprintf(s, 0);
  cprintf("\n", 0);
  while(1)

--- a/defs.h
+++ b/defs.h
@@ -19,6 +19,8 @@ void wakeup(void *);
 void scheduler(void);
 void proc_exit(void);
 void yield(void);
+void cli(void);
+void sti(void);

 // swtch.S
 struct jmpbuf;
@@ -46,6 +48,7 @@ void pic_init(void);

 // mp.c
 void mp_init(void);
+void mp_startthem(void);
 int cpu(void);
 int mp_isbcpu(void);
 void lapic_init(int);

--- a/kalloc.c
+++ b/kalloc.c
@@ -10,6 +10,9 @@
 #include "param.h"
 #include "types.h"
 #include "defs.h"
+#include "param.h"
+#include "mmu.h"
+#include "proc.h"
 #include "spinlock.h"

 struct spinlock kalloc_lock;

--- a/main.c
+++ b/main.c
@@ -8,6 +8,7 @@
 #include "syscall.h"
 #include "elf.h"
 #include "param.h"
+#include "spinlock.h"

 extern char edata[], end[];
 extern int acpu;
@@ -15,23 +16,33 @@ extern char _binary_user1_start[], _binary_user1_size[];
 extern char _binary_usertests_start[], _binary_usertests_size[];
 extern char _binary_userfs_start[], _binary_userfs_size[];

+extern use_printf_lock;
+
 int
 main()
 {
  struct proc *p;

  if (acpu) {
+    cpus[cpu()].clis = 1;
    cprintf("an application processor\n");
    idtinit(); // CPU's idt
    lapic_init(cpu());
    lapic_timerinit();
    lapic_enableintr();
+    sti();
    scheduler();
  }
  acpu = 1;
+
  // clear BSS
  memset(edata, 0, end - edata);

+  mp_init(); // just set up apic so cpu() works
+  use_printf_lock = 1;
+
+  cpus[cpu()].clis = 1; // cpu starts as if we had called cli()
+
  cprintf("\nxV6\n\n");

  pic_init(); // initialize PIC
@@ -56,7 +67,7 @@ main()
  p->ppid = 0;
  setupsegs(p);

-  mp_init(); // multiprocessor
+  mp_startthem();

  // turn on timer and enable interrupts on the local APIC
  lapic_timerinit();

--- a/mp.c
+++ b/mp.c
@@ -325,8 +325,6 @@ mp_init()
  struct MPCTB *mpctb;
  struct MPPE *proc;
  struct MPBE *bus;
-  int c;
-  extern int main();
  int i;

  ncpu = 0;
@@ -386,13 +384,20 @@ mp_init()
  
  lapic_init(bcpu-cpus);
  cprintf("ncpu: %d boot %d\n", ncpu, bcpu-cpus);
+}

+void
+mp_startthem()
+{
  extern uint8_t _binary_bootother_start[], _binary_bootother_size[];
+  extern int main();
+  int c;
+
  memmove((void *) APBOOTCODE,_binary_bootother_start, 
 	  (uint32_t) _binary_bootother_size);

  for(c = 0; c < ncpu; c++){
-    if (cpus+c == bcpu) continue;
+    if (c == cpu()) continue;
    cprintf ("starting processor %d\n", c);
    *(unsigned *)(APBOOTCODE-4) = (unsigned) (cpus[c].mpstack) + MPSTACK; // tell it what to use for %esp
    *(unsigned *)(APBOOTCODE-8) = (unsigned)&main; // tell it where to jump to

--- a/proc.c
+++ b/proc.c
@@ -148,6 +148,7 @@ scheduler(void)

    if(i < NPROC){
      np->state = RUNNING;
+      release(&proc_table_lock);
      break;
    }
    
@@ -157,8 +158,6 @@ scheduler(void)

  cpus[cpu()].lastproc = np;
  curproc[cpu()] = np;
-  
-  release(&proc_table_lock);

  // h/w sets busy bit in TSS descriptor sometimes, and faults
  // if it's set in LTR. so clear tss descriptor busy bit.
@@ -252,3 +251,25 @@ proc_exit()
  // switch into scheduler
  swtch(ZOMBIE);
 }
+
+// disable interrupts
+void
+cli(void)
+{
+  cpus[cpu()].clis += 1;
+  if(cpus[cpu()].clis == 1)
+    __asm __volatile("cli");
+}
+
+// enable interrupts
+void
+sti(void)
+{
+  if(cpus[cpu()].clis < 1){
+    cprintf("cpu %d clis %d\n", cpu(), cpus[cpu()].clis);
+    panic("sti");
+  }
+  cpus[cpu()].clis -= 1;
+  if(cpus[cpu()].clis < 1)
+    __asm __volatile("sti");
+}
--- a/proc.h
+++ b/proc.h
@@ -69,6 +69,7 @@ struct cpu {
  struct jmpbuf jmpbuf;
  char mpstack[MPSTACK]; // per-cpu start-up stack, only used to get into main()
  struct proc *lastproc;  // last proc scheduled on this cpu (never NULL)
+  int clis; // cli() nesting depth
 };

 extern struct cpu cpus[NCPU];

--- a/spinlock.c
+++ b/spinlock.c
@@ -8,6 +8,8 @@

 #define DEBUG 0

+extern use_printf_lock;
+
 int getcallerpc(void *v) {
  return ((int*)v)[-1];
 }
@@ -15,37 +17,49 @@ int getcallerpc(void *v) {
 void
 acquire(struct spinlock * lock)
 {
-  struct proc * cp = curproc[cpu()];
+  unsigned who;
+
+  if(curproc[cpu()])
+    who = (unsigned) curproc[cpu()];
+  else
+    who = cpu() + 1;

-  // on a real machine there would be a memory barrier here
  if(DEBUG) cprintf("cpu%d: acquiring at %x\n", cpu(), getcallerpc(&lock));
-  if (cp && lock->p == cp && lock->locked){
+
+  if (lock->who == who && lock->locked){
    lock->count += 1;
  } else { 
    cli();
-    while ( cmpxchg(0, 1, &lock->locked) != 1 ) { ; }
+    // if we get the lock, eax will be zero
+    // if we don't get the lock, eax will be one
+    while ( cmpxchg(0, 1, &lock->locked) == 1 ) { ; }
    lock->locker_pc = getcallerpc(&lock);
    lock->count = 1;
-    lock->p = cp;
+    lock->who = who;
  }
+
  if(DEBUG) cprintf("cpu%d: acquired at %x\n", cpu(), getcallerpc(&lock));
 }

 void
 release(struct spinlock * lock)
 {
-  struct proc * cp = curproc[cpu()];
+  unsigned who;
+
+  if(curproc[cpu()])
+    who = (unsigned) curproc[cpu()];
+  else
+    who = cpu() + 1;

  if(DEBUG) cprintf ("cpu%d: releasing at %x\n", cpu(), getcallerpc(&lock));

-  if(lock->p != cp || lock->count < 1 || lock->locked != 1)
+  if(lock->who != who || lock->count < 1 || lock->locked != 1)
    panic("release");

  lock->count -= 1;
  if(lock->count < 1){
-    lock->p = 0;
+    lock->who = 0;
    cmpxchg(1, 0, &lock->locked);
    sti();
-    // on a real machine there would be a memory barrier here
  }
 }
--- a/spinlock.h
+++ b/spinlock.h
 struct spinlock {
  unsigned int locked;
-  struct proc *p;
+  unsigned who;
  int count;
  unsigned locker_pc;
 };
--- a/x86.h
+++ b/x86.h
@@ -304,20 +304,6 @@ read_tsc(void)
        return tsc;
 }

-// disable interrupts
-static __inline void
-cli(void)
-{
-        __asm __volatile("cli");
-}
-
-// enable interrupts
-static __inline void
-sti(void)
-{
-        __asm __volatile("sti");
-}
-
 struct PushRegs {
    /* registers as pushed by pusha */
    uint32_t reg_edi;