| Message ID | 20190528183425.31691-1-sergej.benilov@googlemail.com |
|---|---|
| State | Rejected |
| Delegated to: | David Miller |
| Headers | show |
| Series | v3.19.8: tcp: re-enable high throughput for low pacing rate | expand |
On 5/28/19 11:34 AM, Sergej Benilov wrote: > Since commit 605ad7f184b60cfaacbc038aa6c55ee68dee3c89 "tcp: refine TSO autosizing", > the TSQ limit is computed as the smaller of > sysctl_tcp_limit_output_bytes and max(2 * skb->truesize, sk->sk_pacing_rate >> 10). > For low pacing rates, this approach sets a low limit, reducing throughput dramatically. ... > > Signed-off-by: Sergej Benilov <sergej.benilov@googlemail.com> > --- > net/ipv4/tcp_output.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c > index e625be56..71efca72 100644 > --- a/net/ipv4/tcp_output.c > +++ b/net/ipv4/tcp_output.c > @@ -2054,7 +2054,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, > * One example is wifi aggregation (802.11 AMPDU) > */ > limit = max(2 * skb->truesize, sk->sk_pacing_rate >> 10); > - limit = min_t(u32, limit, sysctl_tcp_limit_output_bytes); > + limit = max_t(u32, limit, sysctl_tcp_limit_output_bytes); > > if (atomic_read(&sk->sk_wmem_alloc) > limit) { > set_bit(TSQ_THROTTLED, &tp->tsq_flags); > NACK again, for the same reasons.
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index e625be56..71efca72 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2054,7 +2054,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, * One example is wifi aggregation (802.11 AMPDU) */ limit = max(2 * skb->truesize, sk->sk_pacing_rate >> 10); - limit = min_t(u32, limit, sysctl_tcp_limit_output_bytes); + limit = max_t(u32, limit, sysctl_tcp_limit_output_bytes); if (atomic_read(&sk->sk_wmem_alloc) > limit) { set_bit(TSQ_THROTTLED, &tp->tsq_flags);
Since commit 605ad7f184b60cfaacbc038aa6c55ee68dee3c89 "tcp: refine TSO autosizing", the TSQ limit is computed as the smaller of sysctl_tcp_limit_output_bytes and max(2 * skb->truesize, sk->sk_pacing_rate >> 10). For low pacing rates, this approach sets a low limit, reducing throughput dramatically. Compute the limit as the greater of sysctl_tcp_limit_output_bytes and max(2 * skb->truesize, sk->sk_pacing_rate >> 10). Test: netperf -H remote -l -2000000 -- -s 1000000 before patch: MIGRATED TCP STREAM TEST from 0.0.0.0 () port 0 AF_INET to remote () port 0 AF_INET : demo Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 327680 327680 250.17 0.06 after patch: MIGRATED TCP STREAM TEST from 0.0.0.0 () port 0 AF_INET to remote () port 0 AF_INET : demo Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 327680 327680 1.29 12.54 Signed-off-by: Sergej Benilov <sergej.benilov@googlemail.com> --- net/ipv4/tcp_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)