Thread (28 messages) flat view 28 messages, 6 authors, 2015-09-05

RE: [PATCH v3 1/4] Add correlated clocksource deriving system time from an auxiliary clocksource

From: Hall, Christopher S <hidden>
Date: 2015-09-04 21:12:50
Also in: intel-wired-lan, lkml

-----Original Message-----
From: Peter Zijlstra [mailto:peterz@infradead.org]
Sent: Friday, September 04, 2015 7:28 AM
To: Richard Cochran
Cc: Hall, Christopher S; Thomas Gleixner; Kirsher, Jeffrey T;
hpa@zytor.com; mingo@redhat.com; john.stultz@linaro.org; x86@kernel.org;
linux-kernel@vger.kernel.org; netdev@vger.kernel.org; intel-wired-
lan@lists.osuosl.org
Subject: Re: [PATCH v3 1/4] Add correlated clocksource deriving system time
from an auxiliary clocksource
quoted
quoted
(several milliseconds) and the result will be out of date by some
fraction of that
quoted
quoted
amount.
Why does it take milliseconds to read one audio time stamp?
So what I suspect, but please correct me if I'm wrong Chris, is that a
DSP will buffer and process audio signals, and only later wake up the
main CPU.

So by the time the CPU is made aware of the data, it's 'old'.
That's about right.  The DSP runs on a 1 ms cadence.  Any access to registers controlled by the DSP will take 1-2 DSP ticks to access.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help