Invoke address resolution upon subsequent traffic to Failed neighbor

Removes the period of time in which subseqeuent traffic to a Failed neighbor
immediately fails with ErrNoLinkAddress. A Failed neighbor is one in which
address resolution fails; or in other words, the neighbor's IP address cannot
be translated to a MAC address.

This means removing the Failed state for linkAddrCache and allowing transitiong
out of Failed into Incomplete for neighborCache. Previously, both caches would
transition entries to Failed after address resolution fails. In this state, any
subsequent traffic requested within an unreachable time would immediately fail
with ErrNoLinkAddress. This does not follow RFC 4861 section 7.3.3:

  If address resolution fails, the entry SHOULD be deleted, so that subsequent
  traffic to that neighbor invokes the next-hop determination procedure again.
  Invoking next-hop determination at this point ensures that alternate default
  routers are tried.

The API for getting a link address for a given address, whether through the link
address cache or the neighbor table, is updated to optionally take a callback
which will be called when address resolution completes. This allows `Route` to
handle completing link resolution internally, so callers of (*Route).Resolve
(e.g. endpoints) don’t have to keep track of when it completes and update the
Route accordingly.

This change also removes the wakers from LinkAddressCache, NeighborCache, and
Route in favor of the callbacks, and callers that previously used a waker can
now just pass a callback to (*Route).Resolve that will notify the waker on
resolution completion.

Fixes #4796

Startblock:
  has LGTM from sbalana
  and then
  add reviewer ghanan
PiperOrigin-RevId: 348597478

1 files changed, 19 insertions, 13 deletions

diff --git a/pkg/tcpip/stack/registration.go b/pkg/tcpip/stack/registration.go
index b334e27c4..7e83b7fbb 100644
--- a/pkg/tcpip/stack/registration.go
+++ b/pkg/tcpip/stack/registration.go
@@ -17,7 +17,6 @@ package stack
 import (
 	"fmt"
 
-	"gvisor.dev/gvisor/pkg/sleep"
 	"gvisor.dev/gvisor/pkg/tcpip"
 	"gvisor.dev/gvisor/pkg/tcpip/buffer"
 	"gvisor.dev/gvisor/pkg/tcpip/header"
@@ -799,19 +798,26 @@ type LinkAddressCache interface {
 	// AddLinkAddress adds a link address to the cache.
 	AddLinkAddress(nicID tcpip.NICID, addr tcpip.Address, linkAddr tcpip.LinkAddress)
 
-	// GetLinkAddress looks up the cache to translate address to link address (e.g. IP -> MAC).
-	// If the LinkEndpoint requests address resolution and there is a LinkAddressResolver
-	// registered with the network protocol, the cache attempts to resolve the address
-	// and returns ErrWouldBlock. Waker is notified when address resolution is
-	// complete (success or not).
+	// GetLinkAddress finds the link address corresponding to the remote address
+	// (e.g. IP -> MAC).
 	//
-	// If address resolution is required, ErrNoLinkAddress and a notification channel is
-	// returned for the top level caller to block. Channel is closed once address resolution
-	// is complete (success or not).
-	GetLinkAddress(nicID tcpip.NICID, addr, localAddr tcpip.Address, protocol tcpip.NetworkProtocolNumber, w *sleep.Waker) (tcpip.LinkAddress, <-chan struct{}, *tcpip.Error)
-
-	// RemoveWaker removes a waker that has been added in GetLinkAddress().
-	RemoveWaker(nicID tcpip.NICID, addr tcpip.Address, waker *sleep.Waker)
+	// Returns a link address for the remote address, if readily available.
+	//
+	// Returns ErrWouldBlock if the link address is not readily available, along
+	// with a notification channel for the caller to block on. Triggers address
+	// resolution asynchronously.
+	//
+	// If onResolve is provided, it will be called either immediately, if
+	// resolution is not required, or when address resolution is complete, with
+	// the resolved link address and whether resolution succeeded. After any
+	// callbacks have been called, the returned notification channel is closed.
+	//
+	// If specified, the local address must be an address local to the interface
+	// the neighbor cache belongs to. The local address is the source address of
+	// a packet prompting NUD/link address resolution.
+	//
+	// TODO(gvisor.dev/issue/5151): Don't return the link address.
+	GetLinkAddress(nicID tcpip.NICID, addr, localAddr tcpip.Address, protocol tcpip.NetworkProtocolNumber, onResolve func(tcpip.LinkAddress, bool)) (tcpip.LinkAddress, <-chan struct{}, *tcpip.Error)
 }
 
 // RawFactory produces endpoints for writing various types of raw packets.