// Copyright 2020 The gVisor Authors. // // Licensed under the Apache License, Version 2.0 (the "License"); // you may not use this file except in compliance with the License. // You may obtain a copy of the License at // // http://www.apache.org/licenses/LICENSE-2.0 // // Unless required by applicable law or agreed to in writing, software // distributed under the License is distributed on an "AS IS" BASIS, // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. // See the License for the specific language governing permissions and // limitations under the License. package stack import ( "fmt" "gvisor.dev/gvisor/pkg/sync" "gvisor.dev/gvisor/pkg/tcpip" ) var _ GroupAddressableEndpoint = (*AddressableEndpointState)(nil) var _ AddressableEndpoint = (*AddressableEndpointState)(nil) // AddressableEndpointState is an implementation of an AddressableEndpoint. type AddressableEndpointState struct { networkEndpoint NetworkEndpoint // Lock ordering (from outer to inner lock ordering): // // AddressableEndpointState.mu // addressState.mu mu struct { sync.RWMutex endpoints map[tcpip.Address]*addressState primary []*addressState // groups holds the mapping between group addresses and the number of times // they have been joined. groups map[tcpip.Address]uint32 } } // Init initializes the AddressableEndpointState with networkEndpoint. // // Must be called before calling any other function on m. func (a *AddressableEndpointState) Init(networkEndpoint NetworkEndpoint) { a.networkEndpoint = networkEndpoint a.mu.Lock() defer a.mu.Unlock() a.mu.endpoints = make(map[tcpip.Address]*addressState) a.mu.groups = make(map[tcpip.Address]uint32) } // ReadOnlyAddressableEndpointState provides read-only access to an // AddressableEndpointState. type ReadOnlyAddressableEndpointState struct { inner *AddressableEndpointState } // AddrOrMatching returns an endpoint for the passed address that is consisdered // bound to the wrapped AddressableEndpointState. // // If addr is an exact match with an existing address, that address is returned. // Otherwise, f is called with each address and the address that f returns true // for is returned. // // Returns nil of no address matches. func (m ReadOnlyAddressableEndpointState) AddrOrMatching(addr tcpip.Address, spoofingOrPrimiscuous bool, f func(AddressEndpoint) bool) AddressEndpoint { m.inner.mu.RLock() defer m.inner.mu.RUnlock() if ep, ok := m.inner.mu.endpoints[addr]; ok { if ep.IsAssigned(spoofingOrPrimiscuous) && ep.IncRef() { return ep } } for _, ep := range m.inner.mu.endpoints { if ep.IsAssigned(spoofingOrPrimiscuous) && f(ep) && ep.IncRef() { return ep } } return nil } // Lookup returns the AddressEndpoint for the passed address. // // Returns nil if the passed address is not associated with the // AddressableEndpointState. func (m ReadOnlyAddressableEndpointState) Lookup(addr tcpip.Address) AddressEndpoint { m.inner.mu.RLock() defer m.inner.mu.RUnlock() ep, ok := m.inner.mu.endpoints[addr] if !ok { return nil } return ep } // ForEach calls f for each address pair. // // If f returns false, f is no longer be called. func (m ReadOnlyAddressableEndpointState) ForEach(f func(AddressEndpoint) bool) { m.inner.mu.RLock() defer m.inner.mu.RUnlock() for _, ep := range m.inner.mu.endpoints { if !f(ep) { return } } } // ForEachPrimaryEndpoint calls f for each primary address. // // If f returns false, f is no longer be called. func (m ReadOnlyAddressableEndpointState) ForEachPrimaryEndpoint(f func(AddressEndpoint)) { m.inner.mu.RLock() defer m.inner.mu.RUnlock() for _, ep := range m.inner.mu.primary { f(ep) } } // ReadOnly returns a readonly reference to a. func (a *AddressableEndpointState) ReadOnly() ReadOnlyAddressableEndpointState { return ReadOnlyAddressableEndpointState{inner: a} } func (a *AddressableEndpointState) releaseAddressState(addrState *addressState) { a.mu.Lock() defer a.mu.Unlock() a.releaseAddressStateLocked(addrState) } // releaseAddressState removes addrState from s's address state (primary and endpoints list). // // Preconditions: a.mu must be write locked. func (a *AddressableEndpointState) releaseAddressStateLocked(addrState *addressState) { oldPrimary := a.mu.primary for i, s := range a.mu.primary { if s == addrState { a.mu.primary = append(a.mu.primary[:i], a.mu.primary[i+1:]...) oldPrimary[len(oldPrimary)-1] = nil break } } delete(a.mu.endpoints, addrState.addr.Address) } // AddAndAcquirePermanentAddress implements AddressableEndpoint. func (a *AddressableEndpointState) AddAndAcquirePermanentAddress(addr tcpip.AddressWithPrefix, peb PrimaryEndpointBehavior, configType AddressConfigType, deprecated bool) (AddressEndpoint, *tcpip.Error) { a.mu.Lock() defer a.mu.Unlock() ep, err := a.addAndAcquireAddressLocked(addr, peb, configType, deprecated, true /* permanent */) // From https://golang.org/doc/faq#nil_error: // // Under the covers, interfaces are implemented as two elements, a type T and // a value V. // // An interface value is nil only if the V and T are both unset, (T=nil, V is // not set), In particular, a nil interface will always hold a nil type. If we // store a nil pointer of type *int inside an interface value, the inner type // will be *int regardless of the value of the pointer: (T=*int, V=nil). Such // an interface value will therefore be non-nil even when the pointer value V // inside is nil. // // Since addAndAcquireAddressLocked returns a nil value with a non-nil type, // we need to explicitly return nil below if ep is (a typed) nil. if ep == nil { return nil, err } return ep, err } // AddAndAcquireTemporaryAddress adds a temporary address. // // Returns tcpip.ErrDuplicateAddress if the address exists. // // The temporary address's endpoint is acquired and returned. func (a *AddressableEndpointState) AddAndAcquireTemporaryAddress(addr tcpip.AddressWithPrefix, peb PrimaryEndpointBehavior) (AddressEndpoint, *tcpip.Error) { a.mu.Lock() defer a.mu.Unlock() ep, err := a.addAndAcquireAddressLocked(addr, peb, AddressConfigStatic, false /* deprecated */, false /* permanent */) // From https://golang.org/doc/faq#nil_error: // // Under the covers, interfaces are implemented as two elements, a type T and // a value V. // // An interface value is nil only if the V and T are both unset, (T=nil, V is // not set), In particular, a nil interface will always hold a nil type. If we // store a nil pointer of type *int inside an interface value, the inner type // will be *int regardless of the value of the pointer: (T=*int, V=nil). Such // an interface value will therefore be non-nil even when the pointer value V // inside is nil. // // Since addAndAcquireAddressLocked returns a nil value with a non-nil type, // we need to explicitly return nil below if ep is (a typed) nil. if ep == nil { return nil, err } return ep, err } // addAndAcquireAddressLocked adds, acquires and returns a permanent or // temporary address. // // If the addressable endpoint already has the address in a non-permanent state, // and addAndAcquireAddressLocked is adding a permanent address, that address is // promoted in place and its properties set to the properties provided. If the // address already exists in any other state, then tcpip.ErrDuplicateAddress is // returned, regardless the kind of address that is being added. // // Precondition: a.mu must be write locked. func (a *AddressableEndpointState) addAndAcquireAddressLocked(addr tcpip.AddressWithPrefix, peb PrimaryEndpointBehavior, configType AddressConfigType, deprecated, permanent bool) (*addressState, *tcpip.Error) { // attemptAddToPrimary is false when the address is already in the primary // address list. attemptAddToPrimary := true addrState, ok := a.mu.endpoints[addr.Address] if ok { if !permanent { // We are adding a non-permanent address but the address exists. No need // to go any further since we can only promote existing temporary/expired // addresses to permanent. return nil, tcpip.ErrDuplicateAddress } addrState.mu.Lock() if addrState.mu.kind.IsPermanent() { addrState.mu.Unlock() // We are adding a permanent address but a permanent address already // exists. return nil, tcpip.ErrDuplicateAddress } if addrState.mu.refs == 0 { panic(fmt.Sprintf("found an address that should have been released (ref count == 0); address = %s", addrState.addr)) } // We now promote the address. for i, s := range a.mu.primary { if s == addrState { switch peb { case CanBePrimaryEndpoint: // The address is already in the primary address list. attemptAddToPrimary = false case FirstPrimaryEndpoint: if i == 0 { // The address is already first in the primary address list. attemptAddToPrimary = false } else { a.mu.primary = append(a.mu.primary[:i], a.mu.primary[i+1:]...) } case NeverPrimaryEndpoint: a.mu.primary = append(a.mu.primary[:i], a.mu.primary[i+1:]...) default: panic(fmt.Sprintf("unrecognized primary endpoint behaviour = %d", peb)) } break } } } if addrState == nil { addrState = &addressState{ addressableEndpointState: a, addr: addr, } a.mu.endpoints[addr.Address] = addrState addrState.mu.Lock() // We never promote an address to temporary - it can only be added as such. // If we are actaully adding a permanent address, it is promoted below. addrState.mu.kind = Temporary } // At this point we have an address we are either promoting from an expired or // temporary address to permanent, promoting an expired address to temporary, // or we are adding a new temporary or permanent address. // // The address MUST be write locked at this point. defer addrState.mu.Unlock() if permanent { if addrState.mu.kind.IsPermanent() { panic(fmt.Sprintf("only non-permanent addresses should be promoted to permanent; address = %s", addrState.addr)) } // Primary addresses are biased by 1. addrState.mu.refs++ addrState.mu.kind = Permanent } // Acquire the address before returning it. addrState.mu.refs++ addrState.mu.deprecated = deprecated addrState.mu.configType = configType if attemptAddToPrimary { switch peb { case NeverPrimaryEndpoint: case CanBePrimaryEndpoint: a.mu.primary = append(a.mu.primary, addrState) case FirstPrimaryEndpoint: if cap(a.mu.primary) == len(a.mu.primary) { a.mu.primary = append([]*addressState{addrState}, a.mu.primary...) } else { // Shift all the endpoints by 1 to make room for the new address at the // front. We could have just created a new slice but this saves // allocations when the slice has capacity for the new address. primaryCount := len(a.mu.primary) a.mu.primary = append(a.mu.primary, nil) if n := copy(a.mu.primary[1:], a.mu.primary); n != primaryCount { panic(fmt.Sprintf("copied %d elements; expected = %d elements", n, primaryCount)) } a.mu.primary[0] = addrState } default: panic(fmt.Sprintf("unrecognized primary endpoint behaviour = %d", peb)) } } return addrState, nil } // RemovePermanentAddress implements AddressableEndpoint. func (a *AddressableEndpointState) RemovePermanentAddress(addr tcpip.Address) *tcpip.Error { a.mu.Lock() defer a.mu.Unlock() if _, ok := a.mu.groups[addr]; ok { panic(fmt.Sprintf("group address = %s must be removed with LeaveGroup", addr)) } return a.removePermanentAddressLocked(addr) } // removePermanentAddressLocked is like RemovePermanentAddress but with locking // requirements. // // Precondition: a.mu must be write locked. func (a *AddressableEndpointState) removePermanentAddressLocked(addr tcpip.Address) *tcpip.Error { addrState, ok := a.mu.endpoints[addr] if !ok { return tcpip.ErrBadLocalAddress } return a.removePermanentEndpointLocked(addrState) } // RemovePermanentEndpoint removes the passed endpoint if it is associated with // a and permanent. func (a *AddressableEndpointState) RemovePermanentEndpoint(ep AddressEndpoint) *tcpip.Error { addrState, ok := ep.(*addressState) if !ok || addrState.addressableEndpointState != a { return tcpip.ErrInvalidEndpointState } return a.removePermanentEndpointLocked(addrState) } // removePermanentAddressLocked is like RemovePermanentAddress but with locking // requirements. // // Precondition: a.mu must be write locked. func (a *AddressableEndpointState) removePermanentEndpointLocked(addrState *addressState) *tcpip.Error { if !addrState.GetKind().IsPermanent() { return tcpip.ErrBadLocalAddress } addrState.SetKind(PermanentExpired) a.decAddressRefLocked(addrState) return nil } // decAddressRef decrements the address's reference count and releases it once // the reference count hits 0. func (a *AddressableEndpointState) decAddressRef(addrState *addressState) { a.mu.Lock() defer a.mu.Unlock() a.decAddressRefLocked(addrState) } // decAddressRefLocked is like decAddressRef but with locking requirements. // // Precondition: a.mu must be write locked. func (a *AddressableEndpointState) decAddressRefLocked(addrState *addressState) { addrState.mu.Lock() defer addrState.mu.Unlock() if addrState.mu.refs == 0 { panic(fmt.Sprintf("attempted to decrease ref count for AddressEndpoint w/ addr = %s when it is already released", addrState.addr)) } addrState.mu.refs-- if addrState.mu.refs != 0 { return } // A non-expired permanent address must not have its reference count dropped // to 0. if addrState.mu.kind.IsPermanent() { panic(fmt.Sprintf("permanent addresses should be removed through the AddressableEndpoint: addr = %s, kind = %d", addrState.addr, addrState.mu.kind)) } a.releaseAddressStateLocked(addrState) } // MainAddress implements AddressableEndpoint. func (a *AddressableEndpointState) MainAddress() tcpip.AddressWithPrefix { a.mu.RLock() defer a.mu.RUnlock() ep := a.acquirePrimaryAddressRLocked(func(ep *addressState) bool { return ep.GetKind() == Permanent }) if ep == nil { return tcpip.AddressWithPrefix{} } addr := ep.AddressWithPrefix() a.decAddressRefLocked(ep) return addr } // acquirePrimaryAddressRLocked returns an acquired primary address that is // valid according to isValid. // // Precondition: e.mu must be read locked func (a *AddressableEndpointState) acquirePrimaryAddressRLocked(isValid func(*addressState) bool) *addressState { var deprecatedEndpoint *addressState for _, ep := range a.mu.primary { if !isValid(ep) { continue } if !ep.Deprecated() { if ep.IncRef() { // ep is not deprecated, so return it immediately. // // If we kept track of a deprecated endpoint, decrement its reference // count since it was incremented when we decided to keep track of it. if deprecatedEndpoint != nil { a.decAddressRefLocked(deprecatedEndpoint) deprecatedEndpoint = nil } return ep } } else if deprecatedEndpoint == nil && ep.IncRef() { // We prefer an endpoint that is not deprecated, but we keep track of // ep in case a doesn't have any non-deprecated endpoints. // // If we end up finding a more preferred endpoint, ep's reference count // will be decremented. deprecatedEndpoint = ep } } return deprecatedEndpoint } // AcquireAssignedAddress implements AddressableEndpoint. func (a *AddressableEndpointState) AcquireAssignedAddress(localAddr tcpip.Address, allowTemp bool, tempPEB PrimaryEndpointBehavior) AddressEndpoint { a.mu.Lock() defer a.mu.Unlock() if addrState, ok := a.mu.endpoints[localAddr]; ok { if !addrState.IsAssigned(allowTemp) { return nil } if !addrState.IncRef() { panic(fmt.Sprintf("failed to increase the reference count for address = %s", addrState.addr)) } return addrState } if !allowTemp { return nil } addr := localAddr.WithPrefix() ep, err := a.addAndAcquireAddressLocked(addr, tempPEB, AddressConfigStatic, false /* deprecated */, false /* permanent */) if err != nil { // addAndAcquireAddressLocked only returns an error if the address is // already assigned but we just checked above if the address exists so we // expect no error. panic(fmt.Sprintf("a.addAndAcquireAddressLocked(%s, %d, %d, false, false): %s", addr, tempPEB, AddressConfigStatic, err)) } // From https://golang.org/doc/faq#nil_error: // // Under the covers, interfaces are implemented as two elements, a type T and // a value V. // // An interface value is nil only if the V and T are both unset, (T=nil, V is // not set), In particular, a nil interface will always hold a nil type. If we // store a nil pointer of type *int inside an interface value, the inner type // will be *int regardless of the value of the pointer: (T=*int, V=nil). Such // an interface value will therefore be non-nil even when the pointer value V // inside is nil. // // Since addAndAcquireAddressLocked returns a nil value with a non-nil type, // we need to explicitly return nil below if ep is (a typed) nil. if ep == nil { return nil } return ep } // AcquireOutgoingPrimaryAddress implements AddressableEndpoint. func (a *AddressableEndpointState) AcquireOutgoingPrimaryAddress(remoteAddr tcpip.Address, allowExpired bool) AddressEndpoint { a.mu.RLock() defer a.mu.RUnlock() ep := a.acquirePrimaryAddressRLocked(func(ep *addressState) bool { return ep.IsAssigned(allowExpired) }) // From https://golang.org/doc/faq#nil_error: // // Under the covers, interfaces are implemented as two elements, a type T and // a value V. // // An interface value is nil only if the V and T are both unset, (T=nil, V is // not set), In particular, a nil interface will always hold a nil type. If we // store a nil pointer of type *int inside an interface value, the inner type // will be *int regardless of the value of the pointer: (T=*int, V=nil). Such // an interface value will therefore be non-nil even when the pointer value V // inside is nil. // // Since acquirePrimaryAddressRLocked returns a nil value with a non-nil type, // we need to explicitly return nil below if ep is (a typed) nil. if ep == nil { return nil } return ep } // PrimaryAddresses implements AddressableEndpoint. func (a *AddressableEndpointState) PrimaryAddresses() []tcpip.AddressWithPrefix { a.mu.RLock() defer a.mu.RUnlock() var addrs []tcpip.AddressWithPrefix for _, ep := range a.mu.primary { // Don't include tentative, expired or temporary endpoints // to avoid confusion and prevent the caller from using // those. switch ep.GetKind() { case PermanentTentative, PermanentExpired, Temporary: continue } addrs = append(addrs, ep.AddressWithPrefix()) } return addrs } // PermanentAddresses implements AddressableEndpoint. func (a *AddressableEndpointState) PermanentAddresses() []tcpip.AddressWithPrefix { a.mu.RLock() defer a.mu.RUnlock() var addrs []tcpip.AddressWithPrefix for _, ep := range a.mu.endpoints { if !ep.GetKind().IsPermanent() { continue } addrs = append(addrs, ep.AddressWithPrefix()) } return addrs } // JoinGroup implements GroupAddressableEndpoint. func (a *AddressableEndpointState) JoinGroup(group tcpip.Address) (bool, *tcpip.Error) { a.mu.Lock() defer a.mu.Unlock() joins, ok := a.mu.groups[group] if !ok { ep, err := a.addAndAcquireAddressLocked(group.WithPrefix(), NeverPrimaryEndpoint, AddressConfigStatic, false /* deprecated */, true /* permanent */) if err != nil { return false, err } // We have no need for the address endpoint. a.decAddressRefLocked(ep) } a.mu.groups[group] = joins + 1 return !ok, nil } // LeaveGroup implements GroupAddressableEndpoint. func (a *AddressableEndpointState) LeaveGroup(group tcpip.Address) (bool, *tcpip.Error) { a.mu.Lock() defer a.mu.Unlock() joins, ok := a.mu.groups[group] if !ok { return false, tcpip.ErrBadLocalAddress } if joins == 1 { a.removeGroupAddressLocked(group) delete(a.mu.groups, group) return true, nil } a.mu.groups[group] = joins - 1 return false, nil } // IsInGroup implements GroupAddressableEndpoint. func (a *AddressableEndpointState) IsInGroup(group tcpip.Address) bool { a.mu.RLock() defer a.mu.RUnlock() _, ok := a.mu.groups[group] return ok } func (a *AddressableEndpointState) removeGroupAddressLocked(group tcpip.Address) { if err := a.removePermanentAddressLocked(group); err != nil { // removePermanentEndpointLocked would only return an error if group is // not bound to the addressable endpoint, but we know it MUST be assigned // since we have group in our map of groups. panic(fmt.Sprintf("error removing group address = %s: %s", group, err)) } } // Cleanup forcefully leaves all groups and removes all permanent addresses. func (a *AddressableEndpointState) Cleanup() { a.mu.Lock() defer a.mu.Unlock() for group := range a.mu.groups { a.removeGroupAddressLocked(group) } a.mu.groups = make(map[tcpip.Address]uint32) for _, ep := range a.mu.endpoints { // removePermanentEndpointLocked returns tcpip.ErrBadLocalAddress if ep is // not a permanent address. if err := a.removePermanentEndpointLocked(ep); err != nil && err != tcpip.ErrBadLocalAddress { panic(fmt.Sprintf("unexpected error from removePermanentEndpointLocked(%s): %s", ep.addr, err)) } } } var _ AddressEndpoint = (*addressState)(nil) // addressState holds state for an address. type addressState struct { addressableEndpointState *AddressableEndpointState addr tcpip.AddressWithPrefix // Lock ordering (from outer to inner lock ordering): // // AddressableEndpointState.mu // addressState.mu mu struct { sync.RWMutex refs uint32 kind AddressKind configType AddressConfigType deprecated bool } } // NetworkEndpoint implements AddressEndpoint. func (a *addressState) NetworkEndpoint() NetworkEndpoint { return a.addressableEndpointState.networkEndpoint } // AddressWithPrefix implements AddressEndpoint. func (a *addressState) AddressWithPrefix() tcpip.AddressWithPrefix { return a.addr } // GetKind implements AddressEndpoint. func (a *addressState) GetKind() AddressKind { a.mu.RLock() defer a.mu.RUnlock() return a.mu.kind } // SetKind implements AddressEndpoint. func (a *addressState) SetKind(kind AddressKind) { a.mu.Lock() defer a.mu.Unlock() a.mu.kind = kind } // IsAssigned implements AddressEndpoint. func (a *addressState) IsAssigned(allowExpired bool) bool { if !a.addressableEndpointState.networkEndpoint.Enabled() { return false } switch a.GetKind() { case PermanentTentative: return false case PermanentExpired: return allowExpired default: return true } } // IncRef implements AddressEndpoint. func (a *addressState) IncRef() bool { a.mu.Lock() defer a.mu.Unlock() if a.mu.refs == 0 { return false } a.mu.refs++ return true } // DecRef implements AddressEndpoint. func (a *addressState) DecRef() { a.addressableEndpointState.decAddressRef(a) } // ConfigType implements AddressEndpoint. func (a *addressState) ConfigType() AddressConfigType { a.mu.RLock() defer a.mu.RUnlock() return a.mu.configType } // SetDeprecated implements AddressEndpoint. func (a *addressState) SetDeprecated(d bool) { a.mu.Lock() defer a.mu.Unlock() a.mu.deprecated = d } // Deprecated implements AddressEndpoint. func (a *addressState) Deprecated() bool { a.mu.RLock() defer a.mu.RUnlock() return a.mu.deprecated }