Kernel PPS processing

[Gary E. Miller]

Yo Achim!

On Tue, 05 Jul 2016 07:35:40 +0200
Achim Gratz <Stromeko at nexgo.de> wrote:

> Gary E. Miller writes:
> > Nothing at all about GPIO or timestamping.
> >
> > Did I miss something?  
> 
> Maybe the fact that all the GPIO registers and the timestamp counter
> are on the VC4 side of the SoC.  The Linux kernel just reads those
> when it gets interrupted by the VC4, the ARM subsystem doesn't have
> any connections of its own to the outside.

I looked for any GPIO timestamp counter in those docs.  I could not find
it.  Got a page number?  Or maybe see where the gpio driver reads such a
thing?

> On another tangent back to NTP, I'm wondering if it wouldn't make
> sense to offload the timestamp filtering at least to the VC4.  Most
> NTP boxes would run headless anyway, so there'd be 16 processors
> sitting idle for that sort of thing.

Gack.  It would be non-portable and very host specific.  Once the
timestamp is put on the PPS pulse by the kernel driver there
is no time critical task left to do, and the PPS thread is pretty 
small.

Maybe if once gcc gets a good generic GPU offloading mode, but I suspect
the context switching would swamp any benefits.

> >> There seems to be nothing up yet for the BCM2837 (Pi3), but it's
> >> basically just replacing the A7 cluster in the BCM2836 with an A8
> >> cluster.  
> >
> > Yeah, I'm wondering why the dealy in Linux kernel for 64 bit A8?  
> 
> It wouldn't buy anyone anything of immediate use except having a
> complete additional distro to build

I use Gentoo.  Gentoo is a source distro, so I'm rebuilding it on every
host anyway.

> and maintain and more memory
> pressure to deal with.

Yes, memory usage is worse on a 64 bit processor.

On the flip side, it makes it a lot easier for the kernel to deal
with RAM over 2 GB, so maybe Pi's would get more RAM.

The big thing for NTP and gpsd would be the 64 bit math.  Both do a lot
of 64 bit math.

Going to native A53 mode from A7 would also buy us better floating
point, better SIMD, more GP registers, crypto extensions and security
extensions.  And maybe even hope for big.LITTLE mode, which would likely
not help the time keeping application

>  I suspect that eventually a 64bit port will be
> added anyway to tick a checkbox somewhere.

Linux is trying to kill off x86 (32bit) mode.  So I expect 32 bit will
get less love in the future.


RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
	gem at rellim.com  Tel:+1 541 382 8588
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <http://lists.ntpsec.org/pipermail/devel/attachments/20160705/64cf64f3/attachment.bin>