Re: Moving away from softnet interrupts

2016-05-02 Thread Alexander Bluhm
On Mon, May 02, 2016 at 11:27:36AM +0200, Martin Pieuchot wrote: > On 18/04/16(Mon) 10:50, Martin Pieuchot wrote: > > The current goal of the Network SMP effort is to have a single CPU > > process the IP forwarding path in a process context without holding > > the KERNEL_LOCK(). To achieve this

Re: Moving away from softnet interrupts

2016-05-02 Thread Mark Kettenis
> Date: Mon, 2 May 2016 11:27:36 +0200 > From: Martin Pieuchot > > On 18/04/16(Mon) 10:50, Martin Pieuchot wrote: > > The current goal of the Network SMP effort is to have a single CPU > > process the IP forwarding path in a process context without holding > > the

Re: Moving away from softnet interrupts

2016-05-02 Thread Martin Pieuchot
On 18/04/16(Mon) 10:50, Martin Pieuchot wrote: > The current goal of the Network SMP effort is to have a single CPU > process the IP forwarding path in a process context without holding > the KERNEL_LOCK(). To achieve this goal we're progressively moving > code from the softnet interrupt context

Re: Moving away from softnet interrupts

2016-04-26 Thread Janne Johansson
2016-04-25 9:59 GMT+02:00 Martin Pieuchot : > > > The current goal of the Network SMP effort is to have a single CPU > > > process the IP forwarding path in a process context without holding > > > the KERNEL_LOCK(). To achieve this goal we're progressively moving > > > code

Re: Moving away from softnet interrupts

2016-04-25 Thread Martin Pieuchot
On 20/04/16(Wed) 09:33, Dimitris Papastamos wrote: > On Mon, Apr 18, 2016 at 10:50:46AM +0200, Martin Pieuchot wrote: > > The current goal of the Network SMP effort is to have a single CPU > > process the IP forwarding path in a process context without holding > > the KERNEL_LOCK(). To achieve

Re: Moving away from softnet interrupts

2016-04-20 Thread Dimitris Papastamos
On Mon, Apr 18, 2016 at 10:50:46AM +0200, Martin Pieuchot wrote: > The current goal of the Network SMP effort is to have a single CPU > process the IP forwarding path in a process context without holding > the KERNEL_LOCK(). To achieve this goal we're progressively moving > code from the softnet

Re: Moving away from softnet interrupts

2016-04-18 Thread Hrvoje Popovski
On 18.4.2016. 15:31, Hrvoje Popovski wrote: > On 18.4.2016. 10:50, Martin Pieuchot wrote: >> The current goal of the Network SMP effort is to have a single CPU >> process the IP forwarding path in a process context without holding >> the KERNEL_LOCK(). To achieve this goal we're progressively

Re: Moving away from softnet interrupts

2016-04-18 Thread Hrvoje Popovski
On 18.4.2016. 10:50, Martin Pieuchot wrote: > The current goal of the Network SMP effort is to have a single CPU > process the IP forwarding path in a process context without holding > the KERNEL_LOCK(). To achieve this goal we're progressively moving > code from the softnet interrupt context to

Re: Moving away from softnet interrupts

2016-04-18 Thread Martin Pieuchot
On 18/04/16(Mon) 11:33, Mark Kettenis wrote: > [...] > However, this diff uses the same taskq for both the "unlocked" and > "locked" parts. My experience with crypto code was that frequently > grabbing and releasing the kernel lock did affect latency in a bad > way. Some time ago dlg@ had a

Re: Moving away from softnet interrupts

2016-04-18 Thread Mark Kettenis
> Date: Mon, 18 Apr 2016 10:50:46 +0200 > From: Martin Pieuchot > > The current goal of the Network SMP effort is to have a single CPU > process the IP forwarding path in a process context without holding > the KERNEL_LOCK(). To achieve this goal we're progressively moving >