Lab5: Copy-on-Write

reference: http://xv6.dgs.zone/labs/answers/lab6.html

fork will copy a page table of parent process for child, but if child don’t use much address of them, it may waste lots of time and memory.
Sharing the common physical address for child is a good idea, when we actually need to write the address, we produce a page fault, then deal with it. e.g. copy at that time, or just kill the process.

1. ref to count how many processes are using a PA

riscv.h
#define PTE_F (1L << 8) // COW bit

kalloc.c
struct ref_stru {
  struct spinlock lock;
  int cnt[PHYSTOP / PGSIZE];  // 引用计数
} ref;

//kalloc.c
void
kinit()
{
  initlock(&kmem.lock, "kmem");
  initlock(&ref.lock, "ref");
  freerange(end, (void*)PHYSTOP);
}

judge whether the PA can be deleted or not. delete when count == 0

void
kfree(void *pa)
{
  struct run *r;

  if(((uint64)pa % PGSIZE) != 0 || (char*)pa < end || (uint64)pa >= PHYSTOP)
    panic("kfree");

  // 只有当引用计数为0了才回收空间
  // 否则只是将引用计数减1
  acquire(&ref.lock);
  if(--ref.cnt[(uint64)pa / PGSIZE] == 0) {
    release(&ref.lock);

    r = (struct run*)pa;

    // Fill with junk to catch dangling refs.
    memset(pa, 1, PGSIZE);

    acquire(&kmem.lock);
    r->next = kmem.freelist;
    kmem.freelist = r;
    release(&kmem.lock);
  } else {
    release(&ref.lock);
  }
}
void *
kalloc(void)
{
  struct run *r;

  acquire(&kmem.lock);
  r = kmem.freelist;
  if(r) {
    kmem.freelist = r->next;
    acquire(&ref.lock);
    ref.cnt[(uint64)r / PGSIZE] = 1;  // 将引用计数初始化为1
    release(&ref.lock);
  }
  release(&kmem.lock);

  if(r)
    memset((char*)r, 5, PGSIZE); // fill with junk
  return (void*)r;
}

2. modify uvmcopy to share PA

//vm.c
int
uvmcopy(pagetable_t old, pagetable_t new, uint64 sz)
{
  pte_t *pte;
  uint64 pa, i;
  uint flags;

  for(i = 0; i < sz; i += PGSIZE){
    if((pte = walk(old, i, 0)) == 0)
      panic("uvmcopy: pte should exist");
    if((*pte & PTE_V) == 0)
      panic("uvmcopy: page not present");
    pa = PTE2PA(*pte);
    flags = PTE_FLAGS(*pte);

    // 仅对可写页面设置COW标记
    if(flags & PTE_W) {
      // 禁用写并设置COW Fork标记
      flags = (flags | PTE_F) & ~PTE_W;
      *pte = PA2PTE(pa) | flags;
    }

    if(mappages(new, i, PGSIZE, pa, flags) != 0) {
      uvmunmap(new, 0, i / PGSIZE, 1);
      return -1;
    }
    // 增加内存的引用计数
    kaddrefcnt((char*)pa);
  }
  return 0;
}

3. modify usertrap to deal with page fault

//trap.c
uint64 cause = r_scause();
if(cause == 8) {
  ...
} else if((which_dev = devintr()) != 0){
  // ok
} else if(cause == 13 || cause == 15) {
  uint64 fault_va = r_stval();  // 获取出错的虚拟地址
  if(fault_va >= p->sz
    || cowpage(p->pagetable, fault_va) != 0
    || cowalloc(p->pagetable, PGROUNDDOWN(fault_va)) == 0)
    p->killed = 1;
} else {
  ...
}

deal with copyout:

uint64 cause = r_scause();
if(cause == 8) {
  ...
} else if((which_dev = devintr()) != 0){
  // ok
} else if(cause == 13 || cause == 15) {
  uint64 fault_va = r_stval();  // 获取出错的虚拟地址
  if(fault_va >= p->sz
    || cowpage(p->pagetable, fault_va) != 0
    || cowalloc(p->pagetable, PGROUNDDOWN(fault_va)) == 0)
    p->killed = 1;
} else {
  ...
}

result:(run out of time, need to improve)

$ usertests
usertests starting
test MAXVAplus: OK
test manywrites: OK
test execout: OK
test copyin: OK
test copyout: OK
test copyinstr1: OK
test copyinstr2: OK
test copyinstr3: OK
test rwsbrk: OK
test truncate1: OK
test truncate2: OK
test truncate3: OK
test reparent2: OK
test pgbug: OK
test sbrkbugs: usertrap(): unexpected scause 0x000000000000000c pid=3269
            sepc=0x00000000000058e0 stval=0x00000000000058e0
usertrap(): unexpected scause 0x000000000000000c pid=3270
            sepc=0x00000000000058e0 stval=0x00000000000058e0
OK
test badarg: OK
test reparent: OK
test twochildren: OK
test forkfork: OK
test forkforkfork: OK
test argptest: OK
test createdelete: OK
test linkunlink: OK
test linktest: OK
test unlinkread: OK
test concreate: OK
test subdir: OK
test fourfiles: OK
test sharedfd: OK
test dirtest: OK
test exectest: OK
test bigargtest: OK
test bigwrite: OK
test bsstest: OK
test sbrkbasic: OK
test sbrkmuch: OK
test kernmem: OK
test sbrkfail: OK
test sbrkarg: OK
test sbrklast: OK
test sbrk8000: OK
test validatetest: OK
test stacktest: OK
test opentest: OK
test writetest: OK
test writebig: OK
test createtest: OK
test openiput: OK
test exitiput: OK
test iput: OK
test mem: OK
test pipe1: OK
test killstatus: OK
test preempt: kill... wait... OK
test exitwait: OK
test rmdot: OK
test fourteen: OK
test bigfile: OK
test dirfile: OK
test iref: OK
test forktest: OK
test bigdir: OK
ALL TESTS PASSED