ssh: ensure that handshakeTransport goroutines have finished before Close returns
This fixes a data race in the tests for x/crypto/ssh, which expects to be able to examine a transport's read and write counters without locking after closing it. (Given the number of goroutines, channels, and mutexes used in this package, I wouldn't be surprised if other concurrency bugs remain. I would suggest simplifying the concurrency in this package, but I don't intend to follow up on that myself at the moment.) Fixes golang/go#56957. Change-Id: Ib1f1390b66707c66a3608e48f3f52483cff3c1f5 Reviewed-on: https://go-review.googlesource.com/c/crypto/+/456758 Reviewed-by: Roland Shoemaker <roland@golang.org> TryBot-Result: Gopher Robot <gobot@golang.org> Auto-Submit: Bryan Mills <bcmills@google.com> Run-TryBot: Bryan Mills <bcmills@google.com>
This commit is contained in:
committed by
Gopher Robot
parent
f495dc37d5
commit
23edec0b38
+31
-19
@@ -58,11 +58,13 @@ type handshakeTransport struct {
|
|||||||
incoming chan []byte
|
incoming chan []byte
|
||||||
readError error
|
readError error
|
||||||
|
|
||||||
mu sync.Mutex
|
mu sync.Mutex
|
||||||
writeError error
|
writeError error
|
||||||
sentInitPacket []byte
|
sentInitPacket []byte
|
||||||
sentInitMsg *kexInitMsg
|
sentInitMsg *kexInitMsg
|
||||||
pendingPackets [][]byte // Used when a key exchange is in progress.
|
pendingPackets [][]byte // Used when a key exchange is in progress.
|
||||||
|
writePacketsLeft uint32
|
||||||
|
writeBytesLeft int64
|
||||||
|
|
||||||
// If the read loop wants to schedule a kex, it pings this
|
// If the read loop wants to schedule a kex, it pings this
|
||||||
// channel, and the write loop will send out a kex
|
// channel, and the write loop will send out a kex
|
||||||
@@ -71,7 +73,8 @@ type handshakeTransport struct {
|
|||||||
|
|
||||||
// If the other side requests or confirms a kex, its kexInit
|
// If the other side requests or confirms a kex, its kexInit
|
||||||
// packet is sent here for the write loop to find it.
|
// packet is sent here for the write loop to find it.
|
||||||
startKex chan *pendingKex
|
startKex chan *pendingKex
|
||||||
|
kexLoopDone chan struct{} // closed (with writeError non-nil) when kexLoop exits
|
||||||
|
|
||||||
// data for host key checking
|
// data for host key checking
|
||||||
hostKeyCallback HostKeyCallback
|
hostKeyCallback HostKeyCallback
|
||||||
@@ -86,12 +89,10 @@ type handshakeTransport struct {
|
|||||||
// Algorithms agreed in the last key exchange.
|
// Algorithms agreed in the last key exchange.
|
||||||
algorithms *algorithms
|
algorithms *algorithms
|
||||||
|
|
||||||
|
// Counters exclusively owned by readLoop.
|
||||||
readPacketsLeft uint32
|
readPacketsLeft uint32
|
||||||
readBytesLeft int64
|
readBytesLeft int64
|
||||||
|
|
||||||
writePacketsLeft uint32
|
|
||||||
writeBytesLeft int64
|
|
||||||
|
|
||||||
// The session ID or nil if first kex did not complete yet.
|
// The session ID or nil if first kex did not complete yet.
|
||||||
sessionID []byte
|
sessionID []byte
|
||||||
}
|
}
|
||||||
@@ -108,7 +109,8 @@ func newHandshakeTransport(conn keyingTransport, config *Config, clientVersion,
|
|||||||
clientVersion: clientVersion,
|
clientVersion: clientVersion,
|
||||||
incoming: make(chan []byte, chanSize),
|
incoming: make(chan []byte, chanSize),
|
||||||
requestKex: make(chan struct{}, 1),
|
requestKex: make(chan struct{}, 1),
|
||||||
startKex: make(chan *pendingKex, 1),
|
startKex: make(chan *pendingKex),
|
||||||
|
kexLoopDone: make(chan struct{}),
|
||||||
|
|
||||||
config: config,
|
config: config,
|
||||||
}
|
}
|
||||||
@@ -340,16 +342,17 @@ write:
|
|||||||
t.mu.Unlock()
|
t.mu.Unlock()
|
||||||
}
|
}
|
||||||
|
|
||||||
// drain startKex channel. We don't service t.requestKex
|
|
||||||
// because nobody does blocking sends there.
|
|
||||||
go func() {
|
|
||||||
for init := range t.startKex {
|
|
||||||
init.done <- t.writeError
|
|
||||||
}
|
|
||||||
}()
|
|
||||||
|
|
||||||
// Unblock reader.
|
// Unblock reader.
|
||||||
t.conn.Close()
|
t.conn.Close()
|
||||||
|
|
||||||
|
// drain startKex channel. We don't service t.requestKex
|
||||||
|
// because nobody does blocking sends there.
|
||||||
|
for request := range t.startKex {
|
||||||
|
request.done <- t.getWriteError()
|
||||||
|
}
|
||||||
|
|
||||||
|
// Mark that the loop is done so that Close can return.
|
||||||
|
close(t.kexLoopDone)
|
||||||
}
|
}
|
||||||
|
|
||||||
// The protocol uses uint32 for packet counters, so we can't let them
|
// The protocol uses uint32 for packet counters, so we can't let them
|
||||||
@@ -545,7 +548,16 @@ func (t *handshakeTransport) writePacket(p []byte) error {
|
|||||||
}
|
}
|
||||||
|
|
||||||
func (t *handshakeTransport) Close() error {
|
func (t *handshakeTransport) Close() error {
|
||||||
return t.conn.Close()
|
// Close the connection. This should cause the readLoop goroutine to wake up
|
||||||
|
// and close t.startKex, which will shut down kexLoop if running.
|
||||||
|
err := t.conn.Close()
|
||||||
|
|
||||||
|
// Wait for the kexLoop goroutine to complete.
|
||||||
|
// At that point we know that the readLoop goroutine is complete too,
|
||||||
|
// because kexLoop itself waits for readLoop to close the startKex channel.
|
||||||
|
<-t.kexLoopDone
|
||||||
|
|
||||||
|
return err
|
||||||
}
|
}
|
||||||
|
|
||||||
func (t *handshakeTransport) enterKeyExchange(otherInitPacket []byte) error {
|
func (t *handshakeTransport) enterKeyExchange(otherInitPacket []byte) error {
|
||||||
|
|||||||
Reference in New Issue
Block a user