Re: [PATCH] sis900: re-enable high throughput

2019-06-09 Thread David Miller
From: Eric Dumazet Date: Fri, 7 Jun 2019 12:18:21 -0700 > Note that this suggests the driver is not performing TX completion > fast enough. > > Looking at the driver, I do not see anything requesting interrupt > mitigation, so this might also be caused by a race in the driver > (some skbs being

Re: [PATCH] sis900: re-enable high throughput

2019-06-07 Thread Eric Dumazet
On 6/7/19 10:31 AM, Sergej Benilov wrote: > On Fri, 7 Jun 2019 at 19:26, 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->truesiz

Re: [PATCH] sis900: re-enable high throughput

2019-06-07 Thread Sergej Benilov
On Fri, 7 Jun 2019 at 19:26, 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 some connection

[PATCH] sis900: re-enable high throughput

2019-06-07 Thread Sergej Benilov
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 some connections this approach sets a low limit, reducing throughput dramaticall