Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-04-24 Thread Miroslav Lichvar
On Wed, Apr 23, 2014 at 09:22:45PM -0700, John Stultz wrote: > On 02/12/2014 07:42 AM, Miroslav Lichvar wrote: > > You can see in this test it takes about 2500 updates to correct the > > initial ntp error and settle down. That's with 1GHz clocksource. In > > some tests I did with smaller clock

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-04-24 Thread Miroslav Lichvar
On Wed, Apr 23, 2014 at 09:22:45PM -0700, John Stultz wrote: On 02/12/2014 07:42 AM, Miroslav Lichvar wrote: You can see in this test it takes about 2500 updates to correct the initial ntp error and settle down. That's with 1GHz clocksource. In some tests I did with smaller clock

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-04-23 Thread John Stultz
Hey Miroslav! Once again, a few months pass and I finally get some more time to look at this. :( Sorry for how slow this has been going. On 02/12/2014 07:42 AM, Miroslav Lichvar wrote: > On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: >> Got a few cycles to take another look at

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-04-23 Thread John Stultz
Hey Miroslav! Once again, a few months pass and I finally get some more time to look at this. :( Sorry for how slow this has been going. On 02/12/2014 07:42 AM, Miroslav Lichvar wrote: On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: Got a few cycles to take another look at

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-02-12 Thread Miroslav Lichvar
On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: > Got a few cycles to take another look at this, and tried to address > Miroslav's latest comments. Please let me know if you have further > thoughts! In the simulations this version of the patch does indeed work better than the

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-02-12 Thread Miroslav Lichvar
On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: Got a few cycles to take another look at this, and tried to address Miroslav's latest comments. Please let me know if you have further thoughts! In the simulations this version of the patch does indeed work better than the previous

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-02-07 Thread John Stultz
On 02/07/2014 03:45 AM, Miroslav Lichvar wrote: > On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: >> Got a few cycles to take another look at this, and tried to address >> Miroslav's latest comments. Please let me know if you have further >> thoughts! > I've had finally some time to

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-02-07 Thread Miroslav Lichvar
On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: > Got a few cycles to take another look at this, and tried to address > Miroslav's latest comments. Please let me know if you have further > thoughts! I've had finally some time to look at this, sorry for the delay. > I also dropped

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-02-07 Thread Miroslav Lichvar
On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: Got a few cycles to take another look at this, and tried to address Miroslav's latest comments. Please let me know if you have further thoughts! I've had finally some time to look at this, sorry for the delay. I also dropped the

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-02-07 Thread John Stultz
On 02/07/2014 03:45 AM, Miroslav Lichvar wrote: On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: Got a few cycles to take another look at this, and tried to address Miroslav's latest comments. Please let me know if you have further thoughts! I've had finally some time to look at

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-29 Thread John Stultz
On Tue, Jan 28, 2014 at 9:58 AM, Richard Cochran wrote: > I tested for a regression using the patched kernel with the nohz=off > command line option... > > On Mon, Jan 13, 2014 at 10:15:18AM -0800, John Stultz wrote: >> On 01/13/2014 09:51 AM, Richard Cochran wrote: >> > >> > - Linux

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-29 Thread John Stultz
On Tue, Jan 28, 2014 at 9:58 AM, Richard Cochran richardcoch...@gmail.com wrote: I tested for a regression using the patched kernel with the nohz=off command line option... On Mon, Jan 13, 2014 at 10:15:18AM -0800, John Stultz wrote: On 01/13/2014 09:51 AM, Richard Cochran wrote: - Linux

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-28 Thread Richard Cochran
I tested for a regression using the patched kernel with the nohz=off command line option... On Mon, Jan 13, 2014 at 10:15:18AM -0800, John Stultz wrote: > On 01/13/2014 09:51 AM, Richard Cochran wrote: > > > > - Linux 3.12.7-nohz-plain-20140106nohz-plain.log > > - Linux

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-28 Thread Richard Cochran
I tested for a regression using the patched kernel with the nohz=off command line option... On Mon, Jan 13, 2014 at 10:15:18AM -0800, John Stultz wrote: On 01/13/2014 09:51 AM, Richard Cochran wrote: - Linux 3.12.7-nohz-plain-20140106nohz-plain.log - Linux

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-13 Thread Richard Cochran
On Mon, Jan 13, 2014 at 10:15:18AM -0800, John Stultz wrote: > That's great to hear! Thanks so much, I really appreciate the testing! > And this is with HZ=? HZ=1000 > If you do get a chance to look again, I'd also be interested if running > with nohz=off w/ the fix doesn't show any regression

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-13 Thread John Stultz
On 01/13/2014 09:51 AM, Richard Cochran wrote: > On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: >> I still think this is probably 3.15 or later material, but I'd be >> very interested in feedback, thoughts, and testing. > Over the weekend I did a short test of this new approach. I

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-13 Thread Richard Cochran
On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: > > I still think this is probably 3.15 or later material, but I'd be > very interested in feedback, thoughts, and testing. Over the weekend I did a short test of this new approach. I compiled two kernels, one plain v3.12.7 and one

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-13 Thread Richard Cochran
On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: I still think this is probably 3.15 or later material, but I'd be very interested in feedback, thoughts, and testing. Over the weekend I did a short test of this new approach. I compiled two kernels, one plain v3.12.7 and one with

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-13 Thread John Stultz
On 01/13/2014 09:51 AM, Richard Cochran wrote: On Mon, Jan 06, 2014 at 07:57:03PM -0800, John Stultz wrote: I still think this is probably 3.15 or later material, but I'd be very interested in feedback, thoughts, and testing. Over the weekend I did a short test of this new approach. I compiled

Re: [PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-13 Thread Richard Cochran
On Mon, Jan 13, 2014 at 10:15:18AM -0800, John Stultz wrote: That's great to hear! Thanks so much, I really appreciate the testing! And this is with HZ=? HZ=1000 If you do get a chance to look again, I'd also be interested if running with nohz=off w/ the fix doesn't show any regression

[PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-06 Thread John Stultz
Got a few cycles to take another look at this, and tried to address Miroslav's latest comments. Please let me know if you have further thoughts! thanks -john The existing timekeeping_adjust logic has always been complicated to understand. Further, since it was developed prior to NOHZ becoming

[PATCH] [RFC] timekeeping: Rework frequency adjustments to work better w/ nohz

2014-01-06 Thread John Stultz
Got a few cycles to take another look at this, and tried to address Miroslav's latest comments. Please let me know if you have further thoughts! thanks -john The existing timekeeping_adjust logic has always been complicated to understand. Further, since it was developed prior to NOHZ becoming