Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-10 Thread Joao Martins
On 06/10/2016 10:29 AM, Jan Beulich wrote: On 10.06.16 at 08:59, wrote: >> Actually, since local time gets seeded from platform time in >> init_percpu_time(), I don't think we can do away with >> maintaining platform time. > Yeah, I agree. But the case of my previous

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-10 Thread Jan Beulich
>>> On 10.06.16 at 08:59, wrote: > Actually, since local time gets seeded from platform time in > init_percpu_time(), I don't think we can do away with > maintaining platform time. And it looks like this seeding is where much of the remaining backwards deltas are coming from:

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-10 Thread Jan Beulich
>>> On 09.06.16 at 20:19, wrote: > [changing Dario address to citrix.com as it was bouncing for me ] > > On 06/09/2016 04:52 PM, Jan Beulich wrote: > On 09.06.16 at 17:00, wrote: >>> On 06/09/2016 01:57 PM, Jan Beulich wrote: >>> On

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-09 Thread Joao Martins
[changing Dario address to citrix.com as it was bouncing for me ] On 06/09/2016 04:52 PM, Jan Beulich wrote: On 09.06.16 at 17:00, wrote: >> On 06/09/2016 01:57 PM, Jan Beulich wrote: >> On 09.06.16 at 14:11, wrote: >>> So in effect

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-09 Thread Jan Beulich
>>> On 09.06.16 at 17:00, wrote: > On 06/09/2016 01:57 PM, Jan Beulich wrote: > On 09.06.16 at 14:11, wrote: >> So in effect for the fast path the patch >> changes the situation from c->stime_local_stamp being effectively >> unused to

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-09 Thread Jan Beulich
>>> On 09.06.16 at 14:11, wrote: > On 06/09/2016 01:01 PM, Jan Beulich wrote: >> This looks like a copy and paste mistake in commit 1b6a99892d ("x86: >> Simpler time handling when TSC is constant across all power saving >> states"), responsible for occasional

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-09 Thread Wei Liu
On Thu, Jun 09, 2016 at 06:01:18AM -0600, Jan Beulich wrote: > This looks like a copy and paste mistake in commit 1b6a99892d ("x86: > Simpler time handling when TSC is constant across all power saving > states"), responsible for occasional many-microsecond cross-CPU skew of > what NOW() returns. >

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-09 Thread Joao Martins
On 06/09/2016 01:01 PM, Jan Beulich wrote: > This looks like a copy and paste mistake in commit 1b6a99892d ("x86: > Simpler time handling when TSC is constant across all power saving > states"), responsible for occasional many-microsecond cross-CPU skew of > what NOW() returns. > > Also improve

Re: [Xen-devel] [PATCH] x86/time: use correct (local) time stamp in constant-TSC calibration fast path

2016-06-09 Thread Andrew Cooper
On 09/06/16 13:01, Jan Beulich wrote: > This looks like a copy and paste mistake in commit 1b6a99892d ("x86: > Simpler time handling when TSC is constant across all power saving > states"), responsible for occasional many-microsecond cross-CPU skew of > what NOW() returns. > > Also improve the