On 06/28/2018 06:43 AM, Thomas Gleixner wrote:
>
> For enhanced fun the early calibration stuff was moved from right after
> init_hypervisor_platform() to the place where it is now in commit
> ccb64941f375a6 ("x86/timers: Move simple_udelay_calibration() past
> kvmclock_init()") to speed up KVM
On 06/28/2018 06:43 AM, Thomas Gleixner wrote:
>
> For enhanced fun the early calibration stuff was moved from right after
> init_hypervisor_platform() to the place where it is now in commit
> ccb64941f375a6 ("x86/timers: Move simple_udelay_calibration() past
> kvmclock_init()") to speed up KVM
On Fri, 29 Jun 2018, Pavel Tatashin wrote:
> On 18-06-29 09:30:10, Thomas Gleixner wrote:
> For now, in my series, I would like check in tsc_early_init() if we are
> running xen, and set tsc_khz to 0 if so. Later in tsc_init() we will get
> a proper calibration. The question is how to make this
On Fri, 29 Jun 2018, Pavel Tatashin wrote:
> On 18-06-29 09:30:10, Thomas Gleixner wrote:
> For now, in my series, I would like check in tsc_early_init() if we are
> running xen, and set tsc_khz to 0 if so. Later in tsc_init() we will get
> a proper calibration. The question is how to make this
On Fri, Jun 29, 2018 at 09:30:10AM +0200, Thomas Gleixner wrote:
> On Thu, 28 Jun 2018, Pavel Tatashin wrote:
> > On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
> > Hi Thomas,
> >
> > In addition to above, we have xen hvm:
> >
> > setup_arch()
> > ...
> >
On Fri, Jun 29, 2018 at 09:30:10AM +0200, Thomas Gleixner wrote:
> On Thu, 28 Jun 2018, Pavel Tatashin wrote:
> > On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
> > Hi Thomas,
> >
> > In addition to above, we have xen hvm:
> >
> > setup_arch()
> > ...
> >
On Fri, 2018-06-29 at 17:30 +0300, Andy Shevchenko wrote:
> On Thu, 2018-06-28 at 12:43 +0200, Thomas Gleixner wrote:
> > On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> > >
> Taking above into consideration, I think we may just remove the legacy
> code from mfld.c and mrfld.c and see what happen.
On Fri, 2018-06-29 at 17:30 +0300, Andy Shevchenko wrote:
> On Thu, 2018-06-28 at 12:43 +0200, Thomas Gleixner wrote:
> > On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> > >
> Taking above into consideration, I think we may just remove the legacy
> code from mfld.c and mrfld.c and see what happen.
On Thu, 2018-06-28 at 12:43 +0200, Thomas Gleixner wrote:
> On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> > I still want to document the unholy mess of what is initialized and
> > available when. We have 5 hypervisors and 3 different points in
> > early boot
> > where the calibrate_* callbacks are
On Thu, 2018-06-28 at 12:43 +0200, Thomas Gleixner wrote:
> On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> > I still want to document the unholy mess of what is initialized and
> > available when. We have 5 hypervisors and 3 different points in
> > early boot
> > where the calibrate_* callbacks are
On 18-06-29 09:30:10, Thomas Gleixner wrote:
> On Thu, 28 Jun 2018, Pavel Tatashin wrote:
> > On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
> > Hi Thomas,
> >
> > In addition to above, we have xen hvm:
> >
> > setup_arch()
> > ...
> > init_hypervisor_platform();
> >
On 18-06-29 09:30:10, Thomas Gleixner wrote:
> On Thu, 28 Jun 2018, Pavel Tatashin wrote:
> > On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
> > Hi Thomas,
> >
> > In addition to above, we have xen hvm:
> >
> > setup_arch()
> > ...
> > init_hypervisor_platform();
> >
On Thu, 28 Jun 2018, Pavel Tatashin wrote:
> On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
> Hi Thomas,
>
> In addition to above, we have xen hvm:
>
> setup_arch()
> ...
> init_hypervisor_platform();
> x86_init.hyper.init_platform();
> xen_hvm_guest_init()
On Thu, 28 Jun 2018, Pavel Tatashin wrote:
> On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
> Hi Thomas,
>
> In addition to above, we have xen hvm:
>
> setup_arch()
> ...
> init_hypervisor_platform();
> x86_init.hyper.init_platform();
> xen_hvm_guest_init()
On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
>
> On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> > I still want to document the unholy mess of what is initialized and
> > available when. We have 5 hypervisors and 3 different points in early boot
> > where the calibrate_* callbacks are
On Thu, Jun 28, 2018 at 11:23 AM Thomas Gleixner wrote:
>
> On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> > I still want to document the unholy mess of what is initialized and
> > available when. We have 5 hypervisors and 3 different points in early boot
> > where the calibrate_* callbacks are
On Thu, 28 Jun 2018, Peter Zijlstra wrote:
> On Thu, Jun 28, 2018 at 12:43:59PM +0200, Thomas Gleixner wrote:
> > init_hypervisor_platform()
> >vmware:
> >Retrieves frequency and store it for the
> >calibration function
> >
> >khz =
On Thu, 28 Jun 2018, Peter Zijlstra wrote:
> On Thu, Jun 28, 2018 at 12:43:59PM +0200, Thomas Gleixner wrote:
> > init_hypervisor_platform()
> >vmware:
> >Retrieves frequency and store it for the
> >calibration function
> >
> >khz =
On Thu, Jun 28, 2018 at 12:43:59PM +0200, Thomas Gleixner wrote:
> init_hypervisor_platform()
> vmware:
> Retrieves frequency and store it for the
> calibration function
>
> khz = vmware_get_khz_magic()
>
On Thu, Jun 28, 2018 at 12:43:59PM +0200, Thomas Gleixner wrote:
> init_hypervisor_platform()
> vmware:
> Retrieves frequency and store it for the
> calibration function
>
> khz = vmware_get_khz_magic()
>
On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> I still want to document the unholy mess of what is initialized and
> available when. We have 5 hypervisors and 3 different points in early boot
> where the calibrate_* callbacks are overwritten. The XEN PV one is actually
> post tsc_init_early() for
On Thu, 28 Jun 2018, Thomas Gleixner wrote:
> I still want to document the unholy mess of what is initialized and
> available when. We have 5 hypervisors and 3 different points in early boot
> where the calibrate_* callbacks are overwritten. The XEN PV one is actually
> post tsc_init_early() for
On Tue, 26 Jun 2018, Pavel Tatashin wrote:
> On Tue, Jun 26, 2018 at 11:44 AM Thomas Gleixner wrote:
> > I have an idea how to distangle it and we'll end up in a staged approach,
> > which looks like this:
> >
> > 1) Earliest one (not sure how early yet)
> >
> >Attempt to use
On Tue, 26 Jun 2018, Pavel Tatashin wrote:
> On Tue, Jun 26, 2018 at 11:44 AM Thomas Gleixner wrote:
> > I have an idea how to distangle it and we'll end up in a staged approach,
> > which looks like this:
> >
> > 1) Earliest one (not sure how early yet)
> >
> >Attempt to use
On Tue, Jun 26, 2018 at 2:42 PM Pavel Tatashin
wrote:
>
> Hi Thomas,
>
> On Tue, Jun 26, 2018 at 11:44 AM Thomas Gleixner wrote:
> >
> > Pavel,
> >
> > first of all, sorry for my last outburst. I just was in a lousy mood after
> > staring into too much half baken stuff and failed to make myself
On Tue, Jun 26, 2018 at 2:42 PM Pavel Tatashin
wrote:
>
> Hi Thomas,
>
> On Tue, Jun 26, 2018 at 11:44 AM Thomas Gleixner wrote:
> >
> > Pavel,
> >
> > first of all, sorry for my last outburst. I just was in a lousy mood after
> > staring into too much half baken stuff and failed to make myself
Hi Thomas,
On Tue, Jun 26, 2018 at 11:44 AM Thomas Gleixner wrote:
>
> Pavel,
>
> first of all, sorry for my last outburst. I just was in a lousy mood after
> staring into too much half baken stuff and failed to make myself stay away
> from the computer.
Thank you.
>
> On Sun, 24 Jun 2018,
Hi Thomas,
On Tue, Jun 26, 2018 at 11:44 AM Thomas Gleixner wrote:
>
> Pavel,
>
> first of all, sorry for my last outburst. I just was in a lousy mood after
> staring into too much half baken stuff and failed to make myself stay away
> from the computer.
Thank you.
>
> On Sun, 24 Jun 2018,
Pavel,
first of all, sorry for my last outburst. I just was in a lousy mood after
staring into too much half baken stuff and failed to make myself stay away
from the computer.
On Sun, 24 Jun 2018, Thomas Gleixner wrote:
> On Sat, 23 Jun 2018, Pavel Tatashin wrote:
> And this early init sequence
Pavel,
first of all, sorry for my last outburst. I just was in a lousy mood after
staring into too much half baken stuff and failed to make myself stay away
from the computer.
On Sun, 24 Jun 2018, Thomas Gleixner wrote:
> On Sat, 23 Jun 2018, Pavel Tatashin wrote:
> And this early init sequence
On Sat, 23 Jun 2018, Pavel Tatashin wrote:
> > > As soon as sched_clock() starts output non-zero values, we start
> > > output time without correcting the output as it is done in
> > > sched_clock_local() where unstable TSC and backward motion are
> > > detected. But, since early in boot
On Sat, 23 Jun 2018, Pavel Tatashin wrote:
> > > As soon as sched_clock() starts output non-zero values, we start
> > > output time without correcting the output as it is done in
> > > sched_clock_local() where unstable TSC and backward motion are
> > > detected. But, since early in boot
Hi Thomas,
Thank you for your feedback. My comments below:
> > As soon as sched_clock() starts output non-zero values, we start
> > output time without correcting the output as it is done in
> > sched_clock_local() where unstable TSC and backward motion are
> > detected. But, since early in boot
Hi Thomas,
Thank you for your feedback. My comments below:
> > As soon as sched_clock() starts output non-zero values, we start
> > output time without correcting the output as it is done in
> > sched_clock_local() where unstable TSC and backward motion are
> > detected. But, since early in boot
On Sat, 23 Jun 2018, Pavel Tatashin wrote:
> > So you forgot to answer this question. I did not find a system yet, which
> > actually exposes this behaviour on mainline.
> >
> > Is this an artifact of your early sched clock thing?
> >
>
> Yes, it is. Let me explain how it happens.
>
> So, the
On Sat, 23 Jun 2018, Pavel Tatashin wrote:
> > So you forgot to answer this question. I did not find a system yet, which
> > actually exposes this behaviour on mainline.
> >
> > Is this an artifact of your early sched clock thing?
> >
>
> Yes, it is. Let me explain how it happens.
>
> So, the
> So you forgot to answer this question. I did not find a system yet, which
> actually exposes this behaviour on mainline.
>
> Is this an artifact of your early sched clock thing?
>
Yes, it is. Let me explain how it happens.
So, the problem is introduced in patch "sched: early boot clock" by
> So you forgot to answer this question. I did not find a system yet, which
> actually exposes this behaviour on mainline.
>
> Is this an artifact of your early sched clock thing?
>
Yes, it is. Let me explain how it happens.
So, the problem is introduced in patch "sched: early boot clock" by
On Sat, 23 Jun 2018, Thomas Gleixner wrote:
> On Thu, 21 Jun 2018, Pavel Tatashin wrote:
> > We will change sched_clock() to be called early.
>
> Why is this relevant? Does the issue only appear with that change?
So you forgot to answer this question. I did not find a system yet, which
actually
On Sat, 23 Jun 2018, Thomas Gleixner wrote:
> On Thu, 21 Jun 2018, Pavel Tatashin wrote:
> > We will change sched_clock() to be called early.
>
> Why is this relevant? Does the issue only appear with that change?
So you forgot to answer this question. I did not find a system yet, which
actually
> Let me give you an example:
>
> When tsc_init() enables the usage of TSC for sched_clock() the
> initialization of the tsc sched clock conversion data starts from zero
> and not from the current jiffies based sched_clock() value. This makes
> the timestamps jump backwards:
>
> [
> Let me give you an example:
>
> When tsc_init() enables the usage of TSC for sched_clock() the
> initialization of the tsc sched clock conversion data starts from zero
> and not from the current jiffies based sched_clock() value. This makes
> the timestamps jump backwards:
>
> [
On Thu, 21 Jun 2018, Pavel Tatashin wrote:
> We will change sched_clock() to be called early.
Why is this relevant? Does the issue only appear with that change?
> But, during boot sched_clock() changes its output without notifying us
> about the change of clock source.
Why would the sched clock
On Thu, 21 Jun 2018, Pavel Tatashin wrote:
> We will change sched_clock() to be called early.
Why is this relevant? Does the issue only appear with that change?
> But, during boot sched_clock() changes its output without notifying us
> about the change of clock source.
Why would the sched clock
We will change sched_clock() to be called early. But, during boot
sched_clock() changes its output without notifying us about the change of
clock source.
This happens in tsc_init(), when static_branch_enable(&__use_tsc) is
called.
native_sched_clock() changes from outputing jiffies to reading
We will change sched_clock() to be called early. But, during boot
sched_clock() changes its output without notifying us about the change of
clock source.
This happens in tsc_init(), when static_branch_enable(&__use_tsc) is
called.
native_sched_clock() changes from outputing jiffies to reading
46 matches
Mail list logo