On Tue, Nov 3, 2015 at 11:18 AM, Stanton, Kevin B
wrote:
> On Wed, 21 Oct 2015, Thomas Gleixner wrote:
>> On Tue, 20 Oct 2015, John Stultz wrote:
>>> Being able to have various hardware sharing a time base is quite
>>> useful, and methods for correlating timestamps
On Wed, 21 Oct 2015, Thomas Gleixner wrote:
> On Tue, 20 Oct 2015, John Stultz wrote:
>> Being able to have various hardware sharing a time base is quite
>> useful, and methods for correlating timestamps together are useful.
>> But I don't yet really understand why its important that we can
>>
On Tue, 20 Oct 2015, John Stultz wrote:
> On Tue, Oct 20, 2015 at 12:11 PM, Thomas Gleixner wrote:
> > On Tue, 20 Oct 2015, Richard Cochran wrote:
> >> On Tue, Oct 20, 2015 at 01:51:13PM +0200, Richard Cochran wrote:
> >> > You can, in fact, achieve "proper" correlation by
On Mon, Oct 19, 2015 at 05:36:56PM -0700, John Stultz wrote:
> If we're only tracking 4ms of history, how does this solution
> measurably improve the error over using the timestamps to generate
> MONOTONIC_RAW clock deltas (which doesn't require keeping any history)
> and using
On Tue, 20 Oct 2015, Richard Cochran wrote:
> On Mon, Oct 19, 2015 at 05:36:56PM -0700, John Stultz wrote:
> > If we're only tracking 4ms of history, how does this solution
> > measurably improve the error over using the timestamps to generate
> > MONOTONIC_RAW clock deltas (which doesn't require
On Tue, Oct 20, 2015 at 12:48:03PM +0200, Thomas Gleixner wrote:
> On Tue, 20 Oct 2015, Richard Cochran wrote:
>
> > On Mon, Oct 19, 2015 at 05:36:56PM -0700, John Stultz wrote:
> > > If we're only tracking 4ms of history, how does this solution
> > > measurably improve the error over using the
On Tue, 20 Oct 2015, Richard Cochran wrote:
> On Tue, Oct 20, 2015 at 01:51:13PM +0200, Richard Cochran wrote:
> > You can, in fact, achieve "proper" correlation by sampling. As John
> > said, the question is whether the method in the patch set "measurably
> > improves the error" over using
On Tue, Oct 20, 2015 at 09:11:21PM +0200, Thomas Gleixner wrote:
> Darn, we don't want to have that kind of sampling in every driver
> which has this kind of problem even if it looks like the simpler
> choice for this particular use case. This is going to be something
> which next generation chips
On Tue, Oct 20, 2015 at 12:11 PM, Thomas Gleixner wrote:
> On Tue, 20 Oct 2015, Richard Cochran wrote:
>> On Tue, Oct 20, 2015 at 01:51:13PM +0200, Richard Cochran wrote:
>> > You can, in fact, achieve "proper" correlation by sampling. As John
>> > said, the question is
On Tue, Oct 20, 2015 at 01:51:13PM +0200, Richard Cochran wrote:
> You can, in fact, achieve "proper" correlation by sampling. As John
> said, the question is whether the method in the patch set "measurably
> improves the error" over using another, simpler method.
Here is a short example to put
Thomas,
On Thu, 15 Oct 2015 01:15:57 -0700, Thomas Gleixner
wrote:
>
> > +#define SHADOW_HISTORY_DEPTH 7
>
> And that number is 7 because?
Due to power of 2 it will be 8 instead. As above the useful history is
8-2*1
ms (1 ms is the minimum jiffy length). Array size 4
On Mon, Oct 19, 2015 at 5:18 PM, Christopher Hall
wrote:
> On Thu, 15 Oct 2015 01:15:57 -0700, Thomas Gleixner
> wrote:
>>>
>>> >
>>> > > +#define SHADOW_HISTORY_DEPTH 7
>>> >
>>> > And that number is 7 because?
>>>
>>> Due to power of 2 it will
On Wed, 14 Oct 2015, Christopher Hall wrote:
> On Tue, 13 Oct 2015 12:42:52 -0700, Thomas Gleixner
> wrote:
> > On Mon, 12 Oct 2015, Christopher S. Hall wrote:
> > > audio.
> >
> > This wants to be a seperate patch, really.
>
> OK. This makes sense, I'll do this the next
On Thu, 15 Oct 2015, Richard Cochran wrote:
> Thomas seems to say that there are *other* applications that will want
> to transform device time into system time, but why does your audio
> application use the system time, when the audio-to-ptp time is
> directly available, without any man in the
On Wed, Oct 14, 2015 at 04:22:03PM +0200, Thomas Gleixner wrote:
> It's not only the DSP. There is a bunch of usre space software
> involved as well. Care to think about the full picture and not just
> about some randomly chosen single problem out of the full problem
> space.
I would surely like
On Wed, 14 Oct 2015, Richard Cochran wrote:
> On Wed, Oct 14, 2015 at 09:21:42AM +0200, Thomas Gleixner wrote:
> > The firmware gives you an ART/audio timestamp pair. The firmware does
> > neither know about system time nor about PTP time.
> >
> > So we have to do correlation in software.
>
>
On Tue, 13 Oct 2015, Richard Cochran wrote:
> On Tue, Oct 13, 2015 at 09:15:51PM +0200, Thomas Gleixner wrote:
> > That's not working. The firmware is not going to change, no matter
> > what.
>
> Can we at least have a explanation of how the firmware operates? How
> are (ART,sys) pairs are
On Wed, Oct 14, 2015 at 09:21:42AM +0200, Thomas Gleixner wrote:
> The firmware gives you an ART/audio timestamp pair. The firmware does
> neither know about system time nor about PTP time.
>
> So we have to do correlation in software.
Ok, so give me the ART/audio and two recent ART/ptp times*
Thomas,
On Tue, 13 Oct 2015 12:42:52 -0700, Thomas Gleixner
wrote:
On Mon, 12 Oct 2015, Christopher S. Hall wrote:
audio.
This wants to be a seperate patch, really.
OK. This makes sense, I'll do this the next time.
+/* This needs to be 3 or greater for backtracking
Richard,
On Tue, 13 Oct 2015 14:12:24 -0700, Richard Cochran
wrote:
On Tue, Oct 13, 2015 at 09:15:51PM +0200, Thomas Gleixner wrote:
Can we at least have a explanation of how the firmware operates? How
are (ART,sys) pairs are generated, and how they are supposed
On Wed, Oct 14, 2015 at 07:34:03PM -0700, Christopher Hall wrote:
> I hope this is helpful. Thanks.
So the DSP does not produce or consume system time stamps. Fine.
Still I fail to understand why you need the system time.
Thomas seems to say that there are *other* applications that will want
to
On Wed, Oct 14, 2015 at 06:57:33PM -0700, Christopher Hall wrote:
> >>+#define SHADOW_HISTORY_DEPTH 7
> >
> >And that number is 7 because?
>
> Due to power of 2 it will be 8 instead. As above the useful history is 8-2*1
> ms (1 ms is the minimum jiffy length). Array size 4 would not be enough
>
On Tue, Oct 13, 2015 at 09:51:02AM +0200, Thomas Gleixner wrote:
>
> You are restricting the problem space to this particular use
> case. There are other use cases where PTP is not available or not the
> relevant reference, but you still want to correlate time domains to
> ART.
They may well be
On Tue, 13 Oct 2015, Richard Cochran wrote:
> On Mon, Oct 12, 2015 at 11:45:19AM -0700, Christopher S. Hall wrote:
> > The transforms such an application would
> > perform are:
> >
> > System Clock <-> Audio clock
> > System Clock <-> Network Device Clock [<-> PTP Master Clock]
>
> This is extra
Now that I am starting to understand what this code is trying to
achieve...
On Mon, Oct 12, 2015 at 11:45:19AM -0700, Christopher S. Hall wrote:
> The modification to the original patch accomodates these
> slow devices by adding the option of providing an ART value outside
> of the retry loop and
On Tue, 13 Oct 2015, Richard Cochran wrote:
> On Tue, Oct 13, 2015 at 09:51:02AM +0200, Thomas Gleixner wrote:
> >
> > You are restricting the problem space to this particular use
> > case. There are other use cases where PTP is not available or not the
> > relevant reference, but you still want
On Mon, 12 Oct 2015, Christopher S. Hall wrote:
> Another representative use case of time sync and the correlated
> clocksource (in addition to PTP noted above) is PTP synchronized
> audio.
This wants to be a seperate patch, really.
> +/* This needs to be 3 or greater for backtracking to be
On Tue, Oct 13, 2015 at 09:15:51PM +0200, Thomas Gleixner wrote:
> That's not working. The firmware is not going to change, no matter
> what.
Can we at least have a explanation of how the firmware operates? How
are (ART,sys) pairs are generated, and how they are supposed to get
into the DSP?
From: Thomas Gleixner
Modern Intel hardware provides the so called Always Running Timer
(ART). The TSC which is usually used for timekeeping is derived from
ART and runs with a fixed frequency ratio to it. ART is routed to
devices and allows to take atomic timestamp samples
On Mon, Oct 12, 2015 at 11:45:19AM -0700, Christopher S. Hall wrote:
> Another representative use case of time sync and the correlated
> clocksource (in addition to PTP noted above) is PTP synchronized
> audio.
The added explanations of the audio use case do help. However, you
did not address my
On Mon, Oct 12, 2015 at 11:45:19AM -0700, Christopher S. Hall wrote:
> +int get_correlated_timestamp(struct correlated_ts *crt,
> + struct correlated_cs *crs)
> +{
> + struct timekeeper *tk = _core.timekeeper;
> + unsigned long seq;
> + cycles_t cycles,
31 matches
Mail list logo