Scroll to navigation

EPOCH(9) Kernel Developer's Manual EPOCH(9)

NAME

epoch, epoch_context, epoch_alloc, epoch_free, epoch_enter, epoch_exit, epoch_wait, epoch_call, epoch_drain_callbacks, in_epoch, — kernel epoch based reclamation

SYNOPSIS

#include <sys/param.h>
#include <sys/proc.h>
#include <sys/epoch.h>

epoch_t
epoch_alloc(int flags);

void
epoch_enter(epoch_t epoch);

void
epoch_enter_preempt(epoch_t epoch, epoch_tracker_t et);

void
epoch_exit(epoch_t epoch);

void
epoch_exit_preempt(epoch_t epoch, epoch_tracker_t et);

void
epoch_wait(epoch_t epoch);

void
epoch_wait_preempt(epoch_t epoch);

void
epoch_call(epoch_t epoch, epoch_context_t ctx, void (*callback) (epoch_context_t));

void
epoch_drain_callbacks(epoch_t epoch);

int
in_epoch(epoch_t epoch);

DESCRIPTION

Epochs are used to guarantee liveness and immutability of data by deferring reclamation and mutation until a grace period has elapsed. Epochs do not have any lock ordering issues. Entering and leaving an epoch section will never block.

Epochs are allocated with () and freed with (). The flags passed to epoch_alloc determine whether preemption is allowed during a section or not (the default), as specified by EPOCH_PREEMPT. Threads indicate the start of an epoch critical section by calling (). The end of a critical section is indicated by calling (). The _preempt variants can be used around code which requires preemption. A thread can wait until a grace period has elapsed since any threads have entered the epoch by calling epoch_wait() or epoch_wait_preempt(), depending on the epoch_type. The use of a default epoch type allows one to use epoch_wait() which is guaranteed to have much shorter completion times since we know that none of the threads in an epoch section will be preempted before completing its section. If the thread can't sleep or is otherwise in a performance sensitive path it can ensure that a grace period has elapsed by calling epoch_call() with a callback with any work that needs to wait for an epoch to elapse. Only non-sleepable locks can be acquired during a section protected by () and (). INVARIANTS can assert that a thread is in an epoch by using ().

The epoch API currently does not support sleeping in epoch_preempt sections. A caller should never call () in the middle of an epoch section for the same epoch as this will lead to a deadlock.

By default mutexes cannot be held across (). To permit this the epoch must be allocated with EPOCH_LOCKED. When doing this one must be cautious of creating a situation where a deadlock is possible. Note that epochs are not a straight replacement for read locks. Callers must use safe list and tailq traversal routines in an epoch (see ck_queue). When modifying a list referenced from an epoch section safe removal routines must be used and the caller can no longer modify a list entry in place. An item to be modified must be handled with copy on write and frees must be deferred until after a grace period has elapsed.

The () function is used to drain all pending callbacks which have been invoked by prior () function calls on the same epoch. This function is useful when there are shared memory structure(s) referred to by the epoch callback(s) which are not refcounted and are rarely freed. The typical place for calling this function is right before freeing or invalidating the shared resource(s) used by the epoch callback(s). This function can sleep and is not optimized for performance.

RETURN VALUES

in_epoch(curepoch) will return 1 if curthread is in curepoch, 0 otherwise.

CAVEATS

One must be cautious when using epoch_wait_preempt() threads are pinned during epoch sections so if a thread in a section is then preempted by a higher priority compute bound thread on that CPU it can be prevented from leaving the section. Thus the wait time for the waiter is potentially unbounded.

EXAMPLES

Async free example: Thread 1:

int
in_pcbladdr(struct inpcb *inp, struct in_addr *faddr, struct in_laddr *laddr,
    struct ucred *cred)
{
   /* ... */
   epoch_enter(net_epoch);
    CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
        sa = ifa->ifa_addr;
	if (sa->sa_family != AF_INET)
	    continue;
	sin = (struct sockaddr_in *)sa;
	if (prison_check_ip4(cred, &sin->sin_addr) == 0) {
	     ia = (struct in_ifaddr *)ifa;
	     break;
	}
    }
    epoch_exit(net_epoch);
   /* ... */
}
Thread 2:
void
ifa_free(struct ifaddr *ifa)
{

    if (refcount_release(&ifa->ifa_refcnt))
        epoch_call(net_epoch, &ifa->ifa_epoch_ctx, ifa_destroy);
}

void
if_purgeaddrs(struct ifnet *ifp)
{

    /* .... *
    IF_ADDR_WLOCK(ifp);
    CK_STAILQ_REMOVE(&ifp->if_addrhead, ifa, ifaddr, ifa_link);
    IF_ADDR_WUNLOCK(ifp);
    ifa_free(ifa);
}

Thread 1 traverses the ifaddr list in an epoch. Thread 2 unlinks with the corresponding epoch safe macro, marks as logically free, and then defers deletion. More general mutation or a synchronous free would have to follow a call to epoch_wait().

ERRORS

None.

SEE ALSO

locking(9), mtx_pool(9), mutex(9), rwlock(9), sema(9), sleep(9), sx(9), timeout(9)

June 28, 2019 Debian