Re: regression with napi/softirq ?

2019-07-19 Thread Sudip Mukherjee
On Thu, Jul 18, 2019 at 4:08 PM Eric Dumazet wrote: > > > > On 7/18/19 2:55 PM, Sudip Mukherjee wrote: > > > Thanks Eric. But there is no improvement in delay between > > softirq_raise and softirq_entry with this change. > > But moving to a later kernel (linus master branch? ) like Thomas has > >

Re: regression with napi/softirq ?

2019-07-18 Thread Eric Dumazet
On 7/18/19 2:55 PM, Sudip Mukherjee wrote: > Thanks Eric. But there is no improvement in delay between > softirq_raise and softirq_entry with this change. > But moving to a later kernel (linus master branch? ) like Thomas has > said in the other mail might be difficult atm. I can definitely >

Re: regression with napi/softirq ?

2019-07-18 Thread Sudip Mukherjee
On Thu, Jul 18, 2019 at 12:42 PM Eric Dumazet wrote: > > > > On 7/18/19 1:18 PM, Sudip Mukherjee wrote: > > Hi Eric, > > > > On Thu, Jul 18, 2019 at 7:58 AM Eric Dumazet wrote: > >> > >> > >> > >> On 7/17/19 11:52 PM, Thomas Gleixner wrote: > >>> Sudip, > >>> > >>> On Wed, 17 Jul 2019, Sudip

Re: regression with napi/softirq ?

2019-07-18 Thread Thomas Gleixner
On Thu, 18 Jul 2019, Sudip Mukherjee wrote: > On Thu, Jul 18, 2019 at 7:58 AM Eric Dumazet wrote: > > ksoftirqd might be spuriously scheduled from tx path, when > > __qdisc_run() also reacts to need_resched(). > > > > By raising NET_TX while we are processing NET_RX (say we send a TCP ACK > >

Re: regression with napi/softirq ?

2019-07-18 Thread Eric Dumazet
On 7/18/19 1:18 PM, Sudip Mukherjee wrote: > Hi Eric, > > On Thu, Jul 18, 2019 at 7:58 AM Eric Dumazet wrote: >> >> >> >> On 7/17/19 11:52 PM, Thomas Gleixner wrote: >>> Sudip, >>> >>> On Wed, 17 Jul 2019, Sudip Mukherjee wrote: On Wed, Jul 17, 2019 at 9:53 PM Thomas Gleixner wrote:

Re: regression with napi/softirq ?

2019-07-18 Thread Sudip Mukherjee
Hi Eric, On Thu, Jul 18, 2019 at 7:58 AM Eric Dumazet wrote: > > > > On 7/17/19 11:52 PM, Thomas Gleixner wrote: > > Sudip, > > > > On Wed, 17 Jul 2019, Sudip Mukherjee wrote: > >> On Wed, Jul 17, 2019 at 9:53 PM Thomas Gleixner wrote: > >>> You can hack ksoftirq_running() to return always

Re: regression with napi/softirq ?

2019-07-18 Thread Eric Dumazet
On 7/17/19 11:52 PM, Thomas Gleixner wrote: > Sudip, > > On Wed, 17 Jul 2019, Sudip Mukherjee wrote: >> On Wed, Jul 17, 2019 at 9:53 PM Thomas Gleixner wrote: >>> You can hack ksoftirq_running() to return always false to avoid this, but >>> that might cause application starvation and a huge

Re: regression with napi/softirq ?

2019-07-17 Thread Thomas Gleixner
Sudip, On Wed, 17 Jul 2019, Sudip Mukherjee wrote: > On Wed, Jul 17, 2019 at 9:53 PM Thomas Gleixner wrote: > > You can hack ksoftirq_running() to return always false to avoid this, but > > that might cause application starvation and a huge packet buffer backlog > > when the amount of incoming

Re: regression with napi/softirq ?

2019-07-17 Thread Sudip Mukherjee
Hi, On Wed, Jul 17, 2019 at 9:53 PM Thomas Gleixner wrote: > > On Wed, 17 Jul 2019, Sudip Mukherjee wrote: > > I am using v4.14.55 on an Intel Atom based board and I am seeing network > > packet drops frequently on wireshark logs. After lots of debugging it > > seems that when this happens

Re: regression with napi/softirq ?

2019-07-17 Thread Thomas Gleixner
On Wed, 17 Jul 2019, Sudip Mukherjee wrote: > I am using v4.14.55 on an Intel Atom based board and I am seeing network > packet drops frequently on wireshark logs. After lots of debugging it > seems that when this happens softirq is taking huge time to start after > it has been raised. This is a

regression with napi/softirq ?

2019-07-17 Thread Sudip Mukherjee
Hi All, I am using v4.14.55 on an Intel Atom based board and I am seeing network packet drops frequently on wireshark logs. After lots of debugging it seems that when this happens softirq is taking huge time to start after it has been raised. This is a small snippet from ftrace: