mirror of
https://github.com/slackhq/nebula.git
synced 2026-11-15 19:36:52 +01:00
Caleb JasikandClaude Opus 5.5
61947ff1c2
Passthrough.Flush hands its whole lane to a new tio.BatchWriter in one WriteBatch call when the queue implements it, and NewSingleQueue exposes WriteBatch when its source does. On darwin, tun.WriteBatch writes up to 64 packets per sendmsg_x on the utun control socket, which is a connected datagram socket, so xnu takes the batched sendit_x path rather than one sendit per message. utun registers no send_list handler, so the kernel still passes each packet to utun on its own: this saves syscalls and fd locking, not per-packet kernel work. Error handling follows what sendmsg_x reports: - A short count with no error (a packet past the send buffer mid-call, or a later send failing with ENOBUFS): the remainder goes out in the next sendmsg_x. - EMSGSIZE: only returned when the first packet is too large, before the kernel takes any, so that packet alone goes through Write, which reports it, and the rest are batched again. - ENOSYS, EPERM, EOPNOTSUPP: sendmsg_x is private API, so WriteBatch logs once and falls back to writev per packet for good. - Anything else: sendmsg_x returns the error without the count it took, so the call is reported rather than retried, which could deliver packets twice. One loss mode has no userspace fix. sendit_x counts each packet as sent once it is copied into an mbuf; if a later allocation fails, it frees the copied prefix unsent, returns its length, and sendmsg_x drops the ENOBUFS because the count is nonzero. That looks exactly like a real short count, so under mbuf exhaustion WriteBatch silently drops packets that writev would have delivered. The iOS tun path (tun_ios.go) is unchanged. Inside a macOS Network Extension, download goes from 640 to 832 Mbps against the recvmsg_x change alone in three paired rounds, at 15% less CPU per Mbps; the likely cause is that its utun charges more per write than a raw one. Raw dnclient already runs download near line rate, and there the CPU difference is within noise. TestTunWriteBatch* round-trip through a connected AF_UNIX datagram socketpair, which takes sendmsg_x the same way, over both the batched and the fallback path: order, AF prefixes, skipped invalid packets, and an oversize packet mid-call and first-in-call. Dropping Namelen from msghdrX fails them. Mutation testing with gremlins leaves only equivalent survivors in WriteBatch, plus which error is reported first when several fail. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
50 lines
1.0 KiB
Go
50 lines
1.0 KiB
Go
package batch
|
|
|
|
import (
|
|
"io"
|
|
|
|
"github.com/slackhq/nebula/overlay/tio"
|
|
)
|
|
|
|
// Passthrough is MultiCoalescer's verbatim lane: no coalescing, packets are written at Flush in the
|
|
// order enqueued, in one WriteBatch when out implements tio.BatchWriter.
|
|
type Passthrough struct {
|
|
out io.Writer
|
|
bw tio.BatchWriter // nil: Flush writes one packet per Write
|
|
slots [][]byte
|
|
}
|
|
|
|
func NewPassthrough(w io.Writer) *Passthrough {
|
|
bw, _ := w.(tio.BatchWriter)
|
|
return &Passthrough{
|
|
out: w,
|
|
bw: bw,
|
|
slots: make([][]byte, 0, 128),
|
|
}
|
|
}
|
|
|
|
// enqueue accepts one packet, already sorted into transmission order by dispatch.
|
|
func (p *Passthrough) enqueue(pkt []byte) error {
|
|
p.slots = append(p.slots, pkt)
|
|
return nil
|
|
}
|
|
|
|
func (p *Passthrough) Flush() error {
|
|
var firstErr error
|
|
if p.bw != nil {
|
|
if len(p.slots) > 0 {
|
|
firstErr = p.bw.WriteBatch(p.slots)
|
|
}
|
|
} else {
|
|
for _, s := range p.slots {
|
|
_, err := p.out.Write(s)
|
|
if err != nil && firstErr == nil {
|
|
firstErr = err
|
|
}
|
|
}
|
|
}
|
|
clear(p.slots)
|
|
p.slots = p.slots[:0]
|
|
return firstErr
|
|
}
|