Re: [RFC PATCH 0/2]: hw timestamp support

["Mcmillan, Scott A" <scott.a.mcmillan () intel com>]

Guy,

There isn't a way currently to ask a Linux driver what timestamp options it supports.  That doesn't mean something 
couldn't be done, and I think this would be a very nice feature.  Rather than add this capability on a per driver 
basis, I think changes should be pushed to ethtool to do something like this.  That, however, isn't an immediate type 
of fix.

What timestamp information would you like to get from the driver and in what format?  The hw timestamp capabilities may 
depend on the protocol, RX/TX, and so on, so I think at a minimum it would need to be a bit field.  I can imagine many 
variations of hw timestamp capabilities, so extensibility is probably important.  

Would you like to know the current driver timestamp configuration as well as the hw timestamp capabilities of the 
driver?

Is there value in knowing the hw clock resolution?

Thanks,
Scott

-----Original Message-----
From: tcpdump-workers-owner () lists tcpdump org [mailto:tcpdump-workers-owner () lists tcpdump org] On Behalf Of Guy 
Harris
Sent: Friday, May 28, 2010 10:28 PM
To: tcpdump-workers () lists tcpdump org
Subject: Re: [tcpdump-workers] [RFC PATCH 0/2]: hw timestamp support


On May 24, 2010, at 7:26 AM, Mcmillan, Scott A wrote:

> This patch adds the capability to select the packet timestamp source.  It also adds support for the PACKET_TIMESTAMP 
> Linux kernel setting to specify the source of packet timestamps.  The corresponding Linux kernel patch is being 
> submitted concurrently.

Is there a way to ask the driver (e.g., an ioctl) which, if any, time stamp options it supports?

That would be useful in an pcap API to get that, so that a GUI-based packet analyzer would know whether to offer a 
choice of packet timestamp types and, if so, which types to offer.

-
This is the tcpdump-workers list.
Visit https://cod.sandelman.ca/ to unsubscribe.
-
This is the tcpdump-workers list.
Visit https://cod.sandelman.ca/ to unsubscribe.