From: "Christopher Hall" <christopher.s.hall@intel.com>
To: "Richard Cochran" <richardcochran@gmail.com>,
"Thomas Gleixner" <tglx@linutronix.de>
Cc: "Kirsher, Jeffrey T" <jeffrey.t.kirsher@intel.com>,
"hpa@zytor.com" <hpa@zytor.com>,
"mingo@redhat.com" <mingo@redhat.com>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"john.stultz@linaro.org" <john.stultz@linaro.org>,
"peterz@infradead.org" <peterz@infradead.org>,
"x86@kernel.org" <x86@kernel.org>,
"intel-wired-lan@lists.osuosl.org"
<intel-wired-lan@lists.osuosl.org>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"Stanton, Kevin B" <kevin.b.stanton@intel.com>
Subject: Re: [PATCH v4 3/4] Add PTP_SYS_OFFSET_PRECISE for driver crosstimestamping
Date: Fri, 06 Nov 2015 18:15:21 -0800 [thread overview]
Message-ID: <op.x7pkjvdk1774gr@chall-mobl2.amr.corp.intel.com> (raw)
In-Reply-To: <20151013135925.GB5422@localhost.localdomain>
Richard/Thomas:
On Tue, 13 Oct 2015 06:59:26 -0700, Richard Cochran
<richardcochran@gmail.com> wrote:
> On Mon, Oct 12, 2015 at 11:45:21AM -0700, Christopher S. Hall wrote:
>>
>> +struct ptp_sys_offset_precise {
>> + unsigned int rsv[4]; /* Reserved for future use. */
>> + struct ptp_clock_time dev;
>> + struct ptp_clock_time sys;
>> +};
>> +
>
> Please put the reserved field at the bottom. Also, since we reading
> the raw monotonic time under the hood, we might as well return it in
> this struct too. It costs us almost nothing, and having that value
> can be useful for characterizing the system oscillator.
Below is my proposal to implement your suggested changes adding monotonic
raw to the PTP_SYS_OFFSET_PRECISE ioctl:
Move the reserved field to the end and add monotonic raw to
ptp_sys_offset_precise:
struct ptp_sys_offset_precise {
struct ptp_clock_time dev;
struct ptp_clock_time sys_real;
struct ptp_clock_time sys_raw;
unsigned int rsv[4]; /* Reserved for future use. */
};
In my previous patch, the getsynctime() prototype is:
int (*getsynctime64)(struct ptp_clock_info *ptp, struct timespec64 *dev,
struct timespec64 *sys);
To simplify the PTP callback arguments (not add an additional raw
argument), it seems better to replace all of the arguments with a single
argument:
int (*getsynctime)(struct ptp_clock_info *ptp, struct correlated_ts *cts);
This requires adding a device time field to the correlated timestamp
struct:
struct correlated_ts {
int (*get_ts)(struct correlated_ts *ts);
u64 system_ts;
u64 device_ts;
ktime_t system_real;
ktime_t system_raw;
+ ktime_t device;
void *private;
};
The device time field is filled out by the driver "on the way back"
(following completion of get_correlated_timestamp()) to the PTP driver.
The call flow:
PTP driver-> get_synctime() (Ethernet driver)-> get_correlated_timestamp()
(Timekeeping)->get_ts (Ethernet driver)
The timestamp argument is the same (type struct correlated_ts *) for all
of these calls.
Thanks,
Chris
next prev parent reply other threads:[~2015-11-07 2:15 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-12 18:45 [PATCH v4 0/4] Patchset enabling hardware based cross-timestamps for next gen Intel platforms Christopher S. Hall
2015-10-12 18:45 ` [PATCH v4 1/4] Produce system time from correlated clocksource Christopher S. Hall
2015-10-13 4:58 ` Richard Cochran
2015-10-13 7:51 ` Thomas Gleixner
2015-10-13 8:31 ` Richard Cochran
2015-10-13 19:15 ` Thomas Gleixner
2015-10-13 21:12 ` Richard Cochran
2015-10-14 7:21 ` Thomas Gleixner
2015-10-14 9:29 ` Richard Cochran
2015-10-14 14:22 ` Thomas Gleixner
2015-10-14 16:18 ` Richard Cochran
2015-10-15 2:34 ` Christopher Hall
2015-10-15 5:41 ` Richard Cochran
2015-10-15 8:13 ` Thomas Gleixner
2015-10-13 5:26 ` Richard Cochran
2015-10-13 13:50 ` Richard Cochran
2015-10-13 19:42 ` Thomas Gleixner
2015-10-15 1:57 ` Christopher Hall
2015-10-15 5:57 ` Richard Cochran
2015-10-15 8:15 ` Thomas Gleixner
2015-10-20 0:18 ` Christopher Hall
2015-10-20 0:36 ` John Stultz
2015-10-20 8:54 ` Richard Cochran
2015-10-20 10:48 ` Thomas Gleixner
2015-10-20 11:51 ` Richard Cochran
2015-10-20 14:55 ` Richard Cochran
2015-10-20 19:11 ` Thomas Gleixner
2015-10-20 19:36 ` Richard Cochran
2015-10-20 20:16 ` John Stultz
2015-10-21 7:44 ` Thomas Gleixner
2015-11-03 19:18 ` Stanton, Kevin B
2015-11-09 21:17 ` John Stultz
2015-10-12 18:45 ` [PATCH v4 2/4] Always running timer " Christopher S. Hall
2015-10-13 2:03 ` kbuild test robot
2015-11-18 23:53 ` Jacob Pan
2015-10-12 18:45 ` [PATCH v4 3/4] Add PTP_SYS_OFFSET_PRECISE for driver crosstimestamping Christopher S. Hall
2015-10-13 13:59 ` Richard Cochran
2015-10-15 2:47 ` Christopher Hall
2015-11-07 2:15 ` Christopher Hall [this message]
2015-10-12 18:45 ` [PATCH v4 4/4] Adds hardware supported cross timestamp Christopher S. Hall
2015-10-13 2:10 ` kbuild test robot
2015-10-13 2:11 ` kbuild test robot
2015-10-13 2:31 ` David Miller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=op.x7pkjvdk1774gr@chall-mobl2.amr.corp.intel.com \
--to=christopher.s.hall@intel.com \
--cc=hpa@zytor.com \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=jeffrey.t.kirsher@intel.com \
--cc=john.stultz@linaro.org \
--cc=kevin.b.stanton@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=richardcochran@gmail.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®