Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-04-16 Thread Sebastian Andrzej Siewior
On 2018-03-08 10:57:39 [+0100], To Mauro Carvalho Chehab wrote: > On 2018-02-27 14:39:34 [-0300], Mauro Carvalho Chehab wrote: > > Hi Sebastian, > Hi Mauro, > > > Sorry for taking some time to test it, has been busy those days... > :) > > > Anyway, I tested it today. Didn't work. It keep losing

Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-03-08 Thread Sebastian Andrzej Siewior
On 2018-02-27 14:39:34 [-0300], Mauro Carvalho Chehab wrote: > Hi Sebastian, Hi Mauro, > Sorry for taking some time to test it, has been busy those days... :) > Anyway, I tested it today. Didn't work. It keep losing data. Okay, this was unexpected. What I learned from the thread is that you use

Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-02-27 Thread Mauro Carvalho Chehab
Hi Sebastian, Em Fri, 16 Feb 2018 18:04:50 +0100 Sebastian Andrzej Siewior escreveu: > I've been going over Frederic's softirq patches and it seems that there > were two problems. One was network related, the other was Mauro's USB > dvb-[stc] device which was not able to

Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-02-22 Thread Alan Stern
On Thu, 22 Feb 2018, Sebastian Andrzej Siewior wrote: > On 2018-02-16 16:46:41 [-0500], Alan Stern wrote: > > > The theaded interrupt runs SCHED_FIFO priority 50 by default. The only > > > thing that can interrupt it are interrupts, a softirq (not ksoftirqd) > > > and other tasks with a higher

Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-02-22 Thread Sebastian Andrzej Siewior
On 2018-02-16 16:46:41 [-0500], Alan Stern wrote: > > The theaded interrupt runs SCHED_FIFO priority 50 by default. The only > > thing that can interrupt it are interrupts, a softirq (not ksoftirqd) > > and other tasks with a higher priority than 50. > > There should be no downside performance

Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-02-16 Thread Alan Stern
On Fri, 16 Feb 2018, Sebastian Andrzej Siewior wrote: > On 2018-02-16 13:29:01 [-0500], Alan Stern wrote: > > We originally used tasklets because we didn't want to incur the delays > > associated with running in a process context. It seems odd to be > > reversing that decision now. > > The

Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-02-16 Thread Sebastian Andrzej Siewior
On 2018-02-16 13:29:01 [-0500], Alan Stern wrote: > We originally used tasklets because we didn't want to incur the delays > associated with running in a process context. It seems odd to be > reversing that decision now. The theaded interrupt runs SCHED_FIFO priority 50 by default. The only

Re: [RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-02-16 Thread Alan Stern
On Fri, 16 Feb 2018, Sebastian Andrzej Siewior wrote: > I've been going over Frederic's softirq patches and it seems that there > were two problems. One was network related, the other was Mauro's USB > dvb-[stc] device which was not able to stream properly over time. > > Here is an attempt to

[RFC PATCH] usb: hcd: complete URBs in threaded-IRQ context instead of tasklet

2018-02-16 Thread Sebastian Andrzej Siewior
I've been going over Frederic's softirq patches and it seems that there were two problems. One was network related, the other was Mauro's USB dvb-[stc] device which was not able to stream properly over time. Here is an attempt to let the URB complete in the threaded handler instead of going to