[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <op.ycxuonx01774gr@chall-mobl2.amr.corp.intel.com>
Date: Tue, 16 Feb 2016 10:23:01 -0800
From: "Christopher Hall" <christopher.s.hall@...el.com>
To: "Richard Cochran" <richardcochran@...il.com>
Cc: tglx@...utronix.de, mingo@...hat.com, john.stultz@...aro.org,
hpa@...or.com, jeffrey.t.kirsher@...el.com, x86@...nel.org,
linux-kernel@...r.kernel.org, intel-wired-lan@...ts.osuosl.org,
netdev@...r.kernel.org, kevin.b.stanton@...el.com,
kevin.j.clarke@...el.com
Subject: Re: [PATCH v7 3/8] time: Remove duplicated code in
ktime_get_raw_and_real()
On Mon, 15 Feb 2016 23:52:10 -0800, Richard Cochran
<richardcochran@...il.com> wrote:
>> @@ -888,6 +888,8 @@ void ktime_get_snapshot(struct system_time_snapshot
>> *systime_snapshot)
>> s64 nsec_real;
>> cycle_t now;
>>
>> + WARN_ON(timekeeping_suspended);
> ...
>> - WARN_ON_ONCE(timekeeping_suspended);
>
> Is this change intentional?
Yes it is. The code I replaced in pps_kernel.h (pps_get_ts()) has two call
paths depending on whether CONFIG_NTP_PPS is set. On one call path
(realtime only, config unset) WARN_ON(), from getnstimeofday64, is used,
on the other WARN_ON_ONCE(), from ktime_get_raw_and_real_ts64(), is used.
I opted for the greater verbosity of the two when I combined the code.
Is WARN_ON_ONCE() better here?
Thanks,
Chris
Powered by blists - more mailing lists