_exit as
r12 contains the trap number
Signed-off-by: Gautham R. Shenoy
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 4d70df2..f0d7c54 100644
--- a/arch/powerp
igned-off-by: Gautham R. Shenoy
---
arch/powerpc/include/asm/kvm_ppc.h | 11 +++
arch/powerpc/platforms/powernv/subcore.c | 4 ++--
2 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h
b/arch/powerpc/include/asm/kvm_ppc.h
index c6
Hi Michael,
On Tue, Aug 04, 2015 at 08:08:58PM +1000, Michael Ellerman wrote:
> On Tue, 2015-04-08 at 08:30:58 UTC, "Gautham R. Shenoy" wrote:
> > Section 3.7 of Version 1.2 of the Power8 Processor User's Manual
> > prescribes that updates to HID0 be prec
path.
Signed-off-by: Gautham R. Shenoy
---
[v1--> v2: Moved defn of update_hid0 to reg.h from kvm_ppc.h]
arch/powerpc/include/asm/reg.h | 13 +
arch/powerpc/platforms/powernv/subcore.c | 4 ++--
2 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc
Hi Segher,
Thanks for the suggestions. I will rename the function to
update_power8_hid0() and use asm volatile.
On Tue, Aug 04, 2015 at 09:30:57PM -0500, Segher Boessenkool wrote:
> On Tue, Aug 04, 2015 at 08:08:58PM +1000, Michael Ellerman wrote:
> > > +static inline void update_hid0(unsigned l
core path.
Signed-off-by: Gautham R. Shenoy
---
[v1 --> v2: Moved defn of update_hid0 to reg.h from kvm_ppc.h]
[v2 --> v3: Renamed to update_power8_hid0 and used asm volatile]
arch/powerpc/include/asm/reg.h | 9 +
arch/powerpc/platforms/powernv/subcore.c | 4 ++--
2 fi
Hi Sam,
On Fri, Aug 14, 2015 at 03:07:28PM +1000, Sam Bobroff wrote:
> On Thu, May 21, 2015 at 01:57:04PM +0530, Gautham R. Shenoy wrote:
> > In guest_exit_cont we call kvmhv_commence_exit which expects the trap
> > number as the argument. However r3 doesn't contain the tra
he pstate_id to go out of bounds ?
Otherwise it looks ok.
Acked-by: Gautham R. Shenoy
> + }
>
> return powernv_freqs[i].frequency;
> }
--
Thanks and Regards
gautham.
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
From: "Gautham R. Shenoy"
Move the code that computes the cpumask corresponding to the
thread-siblings of a cpu to a new method named
powernv_cpu_to_core_mask() so that it can be used by other places in
the code.
Signed-off-by: Gautham R. Shenoy
---
drivers/cpufreq/powernv-cpuf
From: "Gautham R. Shenoy"
Create a helper routine that can return the cpu-frequency for the
corresponding pstate_id.
Also, cache the values of the pstate_max, pstate_min and
pstate_nominal and nr_pstates in a static structure so that they can
be reused in the future to perform any v
From: "Gautham R. Shenoy"
Create a driver attribute named cpuinfo_nominal_frequency which
creates a sysfs read-only file named cpuinfo_nominal_frequency. Export
the frequency corresponding to the nominal_pstate through this
interface.
Signed-off-by: Gautham R. Shenoy
---
drive
From: "Gautham R. Shenoy"
Hi,
On IBM POWERNV platforms[1], presently we do not report the current
operating frequency of the processor through the sysfs interface
"cpuinfo_cur_freq" since the cpu frequency driver[1] has not
implemented the ->get(unsigned int c
From: "Gautham R. Shenoy"
The current frequency of a cpu is reported through the sysfs file
cpuinfo_cur_freq. This requires the driver to implement a
"->get(unsigned int cpu)" method which will return the current
operating frequency.
Implement a function named powe
Hello Ben,
On Fri, Mar 07, 2014 at 01:57:31PM +1100, Benjamin Herrenschmidt wrote:
> On Fri, 2014-02-14 at 13:22 +0530, Srivatsa S. Bhat wrote:
> > Subsystems that want to register CPU hotplug callbacks, as well as perform
> > initialization for the CPUs that are already online, often do it as sho
From: "Gautham R. Shenoy"
Hi,
This is the v2 of the consolidated patchset consisting
patches for enabling cpufreq on IBM POWERNV platforms
along with some enhancements.
The v1 of these patches have been previously
submitted on linuxppc-dev [1][2].
- This patchset contains co
From: "Gautham R. Shenoy"
Create a driver attribute named cpuinfo_nominal_freq which
creates a sysfs read-only file named cpuinfo_nominal_freq. Export
the frequency corresponding to the nominal_pstate through this
interface.
Nominal frequency is the highest non-turbo frequency for th
idyanathan Srinivasan
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Anton Blanchard
Signed-off-by: Gautham R. Shenoy
---
arch/powerpc/include/asm/reg.h | 4 +
arch/powerpc/platforms/powernv/Kconfig | 1 +
drivers/cpufreq/Kconfig| 1 +
drivers/cpufreq/Kconfig.powerpc
er-CPU lock and use the ones that belongs to the first
thread sibling of the core.
cpu_first_thread_sibling() macro is used to find the *common* lock for
all thread siblings belonging to a core.
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Vaidyanathan Srinivasan
Signed-off-by: Ga
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Gautham R. Shenoy
---
drivers/cpufreq/powernv-cpufreq.c | 24 ++--
1 file changed, 18 insertions(+), 6 deletions(-)
diff --git a/drivers/cpufreq/powernv-cpufreq.c
b/drivers/cpufreq/powernv-cpufreq.c
index ab1551f..4cad727 100644
--- a/dr
From: "Gautham R. Shenoy"
The current frequency of a cpu is reported through the sysfs file
cpuinfo_cur_freq. This requires the driver to implement a
"->get(unsigned int cpu)" method which will return the current
operating frequency.
Implement a function named powernv_cp
From: "Gautham R. Shenoy"
Create a helper routine that can return the cpu-frequency for the
corresponding pstate_id.
Also, cache the values of the pstate_max, pstate_min and
pstate_nominal and nr_pstates in a static structure so that they can
be reused in the future to perform any v
From: "Gautham R. Shenoy"
Currently, the code in setup-common.c for powerpc assumes that all
clock rates are same in a smp system. This value is cached in the
variable named ppc_proc_freq and is the value that is reported in
/proc/cpuinfo.
However on the PowerNV platform, the clock ra
From: "Gautham R. Shenoy"
Implement a method named pnv_get_proc_freq(unsigned int cpu) which
returns the current clock rate on the 'cpu' in Hz to be reported in
/proc/cpuinfo. This method uses the value reported by cpufreq when
such a value is sane. Otherwise it fall
From: "Gautham R. Shenoy"
Hi,
Currently, the code in setup-common.c for powerpc assumes that all
clock rates are same in a smp system. This value is cached in the
variable named ppc_proc_freq and is the value that is reported in
/proc/cpuinfo.
However on the PowerNV platform, the clo
On Wed, Mar 19, 2014 at 12:05:20PM +0530, Srivatsa S. Bhat wrote:
> On 03/19/2014 05:07 AM, Benjamin Herrenschmidt wrote:
> > On Mon, 2014-03-10 at 16:40 +0530, Gautham R. Shenoy wrote:
> >> From: "Srivatsa S. Bhat"
> >>
> >> Create a helper method
From: "Gautham R. Shenoy"
Hi,
This is the v3 of the consolidated patchset consisting
patches for enabling cpufreq on IBM POWERNV platforms
along with some enhancements. I have incorporated the review
for v2 (which can be found at [1]).
- This patchset contains code for the platform
]
Signed-off-by: Vaidyanathan Srinivasan
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Anton Blanchard
Signed-off-by: Gautham R. Shenoy
---
arch/powerpc/include/asm/reg.h | 4 +
arch/powerpc/platforms/powernv/Kconfig | 1 +
drivers/cpufreq/Kconfig| 1 +
drivers/cpufreq
er-CPU lock and use the ones that belongs to the first
thread sibling of the core.
cpu_first_thread_sibling() macro is used to find the *common* lock for
all thread siblings belonging to a core.
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Vaidyanathan Srinivasan
Signed-off-by: Ga
From: "Gautham R. Shenoy"
Create a helper routine that can return the cpu-frequency for the
corresponding pstate_id.
Also, cache the values of the pstate_max, pstate_min and
pstate_nominal and nr_pstates in a static structure so that they can
be reused in the future to perform any v
From: "Gautham R. Shenoy"
Create a driver attribute named cpuinfo_nominal_freq which
creates a sysfs read-only file named cpuinfo_nominal_freq. Export
the frequency corresponding to the nominal_pstate through this
interface.
Nominal frequency is the highest non-turbo frequency for th
From: "Gautham R. Shenoy"
The current frequency of a cpu is reported through the sysfs file
cpuinfo_cur_freq. This requires the driver to implement a
"->get(unsigned int cpu)" method which will return the current
operating frequency.
Implement a function named powernv_cp
On Thu, Mar 20, 2014 at 05:40:57PM +0530, Gautham R. Shenoy wrote:
> From: "Srivatsa S. Bhat"
>
> On POWER systems, the CPU frequency is controlled at a core-level and
> hence we need to serialize so that only one of the threads in the core
> switches the core's fre
On Thu, Mar 20, 2014 at 05:40:58PM +0530, Gautham R. Shenoy wrote:
> From: "Gautham R. Shenoy"
>
> Create a helper routine that can return the cpu-frequency for the
> corresponding pstate_id.
>
> Also, cache the values of the pstate_max, pstate_min and
> pst
On Thu, Mar 20, 2014 at 05:41:00PM +0530, Gautham R. Shenoy wrote:
> From: "Gautham R. Shenoy"
>
> The current frequency of a cpu is reported through the sysfs file
> cpuinfo_cur_freq. This requires the driver to implement a
> "->get(unsigned int cpu)"
Hi Viresh,
On Fri, Mar 21, 2014 at 01:16:03PM +0530, Viresh Kumar wrote:
> On Thu, Mar 20, 2014 at 5:40 PM, Gautham R. Shenoy
> wrote:
> > This is the v3 of the consolidated patchset consisting
> > patches for enabling cpufreq on IBM POWERNV platforms
> > along with some
On Fri, Mar 21, 2014 at 02:17:28PM +0530, Viresh Kumar wrote:
> > +static ssize_t show_cpuinfo_nominal_freq(struct cpufreq_policy *policy,
> > + char *buf)
> > +{
> > + int nominal_freq;
> > + nominal_freq =
> > pstate_id_to_freq(powernv_pstate_inf
Hi Viresh,
On Fri, Mar 21, 2014 at 02:11:32PM +0530, Viresh Kumar wrote:
> On Thu, Mar 20, 2014 at 5:40 PM, Gautham R. Shenoy
> wrote:
> > From: Vaidyanathan Srinivasan
>
> Hi Vaidy,
>
> > diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
> &
On Fri, Mar 21, 2014 at 03:01:29PM +0530, Viresh Kumar wrote:
> On Thu, Mar 20, 2014 at 5:41 PM, Gautham R. Shenoy
> wrote:
> > From: "Gautham R. Shenoy"
> >
> > The current frequency of a cpu is reported through the sysfs file
> > cpuinfo_cur_freq.
On Fri, Mar 21, 2014 at 12:01:07PM +, David Laight wrote:
> From: Viresh Kumar
>
> > On 21 March 2014 16:34, Gautham R Shenoy wrote:
> > > Heh! Well, that wasn't the reason why this was sent out as a separate
> > > patch, but never mind. Though
On Fri, Mar 21, 2014 at 05:15:34PM +0530, Viresh Kumar wrote:
> On 21 March 2014 16:34, Gautham R Shenoy wrote:
> > Heh! Well, that wasn't the reason why this was sent out as a separate
> > patch, but never mind. Though I don't understand why it would be
> > diffi
Hi Viresh,
On Fri, Mar 21, 2014 at 04:24:27PM +0530, Viresh Kumar wrote:
> On 21 March 2014 16:13, Gautham R Shenoy wrote:
> > On Fri, Mar 21, 2014 at 02:11:32PM +0530, Viresh Kumar wrote:
>
> >> > + pr_debug("PState i
On Fri, Mar 21, 2014 at 06:42:50PM +0530, Viresh Kumar wrote:
> On 21 March 2014 18:34, Gautham R Shenoy wrote:
> > Consider the case when pmspr = 0x00fe;
> >
> > We are interested in extracting the value 'fe'. And ensure that when
> > we store th
On Fri, Mar 21, 2014 at 04:24:27PM +0530, Viresh Kumar wrote:
> >> > +static int powernv_cpufreq_cpu_init(struct cpufreq_policy *policy)
> >> > +{
> >> > + int base, i;
> >> > +
> >> > + base = cpu_first_thread_sibling(policy->cpu);
> >> > +
> >> > + for (i = 0; i < threads_per_co
Hi Ben,
On Sat, Mar 22, 2014 at 09:56:30AM +1100, Benjamin Herrenschmidt wrote:
> On Fri, 2014-03-21 at 16:34 +0530, Gautham R Shenoy wrote:
>
> > > >
> > > > +/*
> > > > + * Computes the current frequency on this cpu
> > > > + * and stores
From: "Gautham R. Shenoy"
Hi,
This is the v4 of the patchset to enable Dynamic Frequency Scaling
on IBM PowerNV Platforms. I have incorporated the review comments
from the previous version (can be found at [1]).
In this version,
* Multiple patches from the previous version have
Murthy
Signed-off-by: Vaidyanathan Srinivasan
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Anton Blanchard
Signed-off-by: Gautham R. Shenoy
---
arch/powerpc/configs/pseries_defconfig| 1 +
arch/powerpc/configs/pseries_le_defconfig | 1 +
arch/powerpc/include/asm/reg.h| 4 +
a
On Thu, Mar 27, 2014 at 12:09:53PM +0530, Viresh Kumar wrote:
> Cc'ing Rafael.
>
Thanks. It was a lapse on my part by not Cc'ing Rafael in the first
place.
> On 26 March 2014 22:25, Gautham R. Shenoy wrote:
> > From: Vaidyanathan Srinivasan
> >
> > Back
On Thu, Mar 27, 2014 at 03:29:36PM +0530, Viresh Kumar wrote:
> On 27 March 2014 15:00, Gautham R Shenoy wrote:
> > As of now, I prefer this patch be based on code that is in the -next
> > tree. I'll get rid of the per-core locking once the serialization
> > pat
From: "Gautham R. Shenoy"
Hi,
This is v5 of the patchset to enable dynamic frequency scaling on IBM
PowerNV platforms. This patchset does address all the review comments
obtained for v4 (which can be found at [1]).
Changes from v4:
* Created a separate patch to select the CPUFr
_freq by setting powernv_cpufreq_driver.get to
powernv_cpufreq_get(). [Authored by e...@linux.vnet.ibm.com]
[Change log updated by e...@linux.vnet.ibm.com]
Reviewed-by: Preeti U Murthy
Signed-off-by: Vaidyanathan Srinivasan
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Anton Blanchard
Signed-off-by: G
From: "Gautham R. Shenoy"
Enable CPUFreq for PowerNV. Select "performance", "powersave",
"userspace" and "ondemand" governors. Choose "ondemand" to be the
default governor.
Signed-off-by: Gautham R. Shenoy
Signed-off-by: Srivatsa
From: "Gautham R. Shenoy"
The .driver_data field in the cpufreq_frequency_table was supposed to
be private to the drivers. However at some later point, it was being
used to indicate if the particular frequency in the table is the
BOOST_FREQUENCY. After patches [1] and [2], the .driv
_freq by setting powernv_cpufreq_driver.get to
powernv_cpufreq_get(). [Authored by e...@linux.vnet.ibm.com]
[Change log updated by e...@linux.vnet.ibm.com]
Reviewed-by: Preeti U Murthy
Signed-off-by: Vaidyanathan Srinivasan
Signed-off-by: Srivatsa S. Bhat
Signed-off-by: Anton Blanchard
Signed-off-by: G
From: "Gautham R. Shenoy"
The .driver_data field in the cpufreq_frequency_table was supposed to
be private to the drivers. However at some later point, it was being
used to indicate if the particular frequency in the table is the
BOOST_FREQUENCY. After patches [1] and [2], the .driv
From: "Gautham R. Shenoy"
Enable CPUFreq for PowerNV. Select "performance", "powersave",
"userspace" and "ondemand" governors. Choose "ondemand" to be the
default governor.
Signed-off-by: Gautham R. Shenoy
Signed-off-by: Srivatsa
), we
can unconditionally clear the doorbell message in
kvmppc_check_wake_reason.
Signed-off-by: Gautham R. Shenoy
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 17 ++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
b/arch/powerp
Hi Shilpa,
Just saw this resend!
On Tue, Jan 12, 2016 at 04:24:26AM -0600, Shilpasri G Bhat wrote:
> Record the throttle event with a trace print replacing the printk,
> except for events like throttling below nominal and occ reset
> event which print a warning message.
>
> Signed-off-by: Shilpa
Hi Shilpa,
On Tue, Jan 12, 2016 at 04:24:27AM -0600, Shilpasri G Bhat wrote:
> +static inline int get_chip_index(struct kobject *kobj)
Probably have "get_chip_index(int id)". See the reason below.
> +{
> + int i, id;
> +
> + i = kstrtoint(kobj->name + 4, 0, &id);
> + if (i)
> +
On Thu, Jan 21, 2016 at 03:08:59PM +0530, Shilpasri G Bhat wrote:
> Signed-off-by: Shilpasri G Bhat
Reviewed-by: Gautham R. Shenoy
--
Thanks and Regards
gautham.
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.
ary overhead
> > in a hot path. So instead of calling cpu_to_chip_id() everytime cache
> > the chip ids for all cores in the array 'core_to_chip_map' and use it
> > in the hotpath.
> >
> > Reported-by: Anton Blanchard
> > Signed-off-by: Shilpasri
On Thu, Jan 28, 2016 at 12:55:36PM +0530, Shilpasri G Bhat wrote:
> This will free the dynamically allocated memory of'chips' on
> module exit.
>
> Signed-off-by: Shilpasri G Bhat
Reviewed-by: Gautham R. Shenoy
--
Than
chip ids for all cores in the array 'core_to_chip_map' and use it
> in the hotpath.
>
> Reported-by: Anton Blanchard
> Signed-off-by: Shilpasri G Bhat
Reviewed-by: Gautham R. Shenoy
--
Thanks and Regards
gautham.
___
Linu
ts
> like throttle below nominal frequency and OCC_RESET are reduced to
> pr_warn/pr_warn_once as pointed by MFG to not mark them as critical
> messages. This patch adds 'throttle_reason' to struct chip to store the
> throttle reason.
>
> Signed-off-by: Shilpasri G
Hi Shilpa,
A minor nit.
On Thu, Jan 28, 2016 at 12:55:41PM +0530, Shilpasri G Bhat wrote:
[..snip..]
> +
> +What:
> /sys/devices/system/cpu/cpufreq/chip*/throttle_reasons/
> +Date:Jan 2016
> +Contact: Linux kernel mailing list
> + Linux for PowerPC
Hi Viresh,
>
> What I can suggest is:
> - Move this directory inside cpuX/cpufreq/ directory, in a similar way
> as to how we create 'stats' directory today.
> - You can then get policy->cpu, to get chip->id out of it.
> - The only disadvantage here is that the same chip directory will be
> r
Hello Thiago,
On Fri, Feb 22, 2019 at 07:57:52PM -0300, Thiago Jung Bauermann wrote:
> When testing DLPAR CPU add/remove on a system under stress,
> pseries_cpu_die() doesn't wait long enough for a CPU to die:
>
> [ 446.983944] cpu 148 (hwid 148) Ready to die...
> [ 446.984062] cpu 149 (hwid 14
Hi Shilpa,
On Thu, Feb 28, 2019 at 11:25:25AM +0530, Shilpasri G Bhat wrote:
> Hi,
>
> On 02/28/2019 10:14 AM, Daniel Axtens wrote:
> > Shilpasri G Bhat writes:
> >
> >> In POWER9, OCC(On-Chip-Controller) provides for hard and soft system
> >> powercapping range. The hard powercap range is guar
longer use paca to save IAMR, instead use _DAR (thanks mpe)
Looks good to me. Once we move to Nick Piggin's C-based save/restore
code, we will be saving all these SPR values on the stack anyway.
Reviewed-by: Gautham R. Shenoy
--
Thanks and Regards
gautham.
From: "Gautham R. Shenoy"
In cpu_to_drc_index() in the case when FW_FEATURE_DRC_INFO is absent,
we currently use of_read_property() to obtain the pointer to the array
corresponding to the property "ibm,drc-indexes". The elements of this
array are of type __be32, but are
ffirmative, the maximum time spent
> inside the loop was was 10 ms. This patch loops for 20 ms just be sure.
>
> Signed-off-by: Thiago Jung Bauermann
Thanks for this version. I have tested the patch and we no longer see
the "Querying DEAD? cpu X (Y) shows 2" message.
Tested-and-Reviewed-by: Gautham R. Shenoy
--
Thanks and Regards
gautham.
Hello Daniel,
On Thu, Apr 4, 2019 at 3:52 PM Daniel Lezcano wrote:
>
>
> Hi Abhishek,
>
> thanks for taking the time to test the different scenario and give us
> the numbers.
>
> On 01/04/2019 07:11, Abhishek wrote:
> >
[.. snip..]
>
> In case of POWER, this is problematic, when the predic
Hello Nicholas,
On Tue, Apr 2, 2019 at 4:57 PM Nicholas Piggin wrote:
>
> Using a jiffies timer creates a dependency on the tick_do_timer_cpu
> incrementing jiffies. If that CPU has locked up and jiffies is not
> incrementing, the watchdog heartbeat timer for all CPUs stops and
> creates false po
stops and
> creates false positives and confusing warnings on local CPUs, and
> also causes the SMP detector to stop, so the root cause is never
> detected.
>
> Fix this by using hrtimer based timers for the watchdog heartbeat,
> like the generic kernel hardlockup detector.
>
> Cc:
From: "Gautham R. Shenoy"
On Pseries LPARs, to calculate utilization, we need to know the
[S]PURR ticks when the CPUs were busy or idle.
The total PURR and SPURR ticks are already exposed via the per-cpu
sysfs files "purr" and "spurr". This patch adds support
From: "Gautham R. Shenoy"
Currently when CPU goes idle, we take a snapshot of PURR via
pseries_idle_prolog() which is used at the CPU idle exit to compute
the idle PURR cycles via the function pseries_idle_epilog(). Thus,
the value of idle PURR cycle thus read before pseries_idle_pr
From: "Gautham R. Shenoy"
On Pseries LPARs, to calculate utilization, we need to know the
[S]PURR ticks when the CPUs were busy or idle.
Via pseries_idle_prolog(), pseries_idle_epilog(), we track the idle
PURR ticks in the VPA variable "wait_state_cycles". This patch ex
From: "Gautham R. Shenoy"
Add documentation for the following sysfs interfaces:
/sys/devices/system/cpu/cpuX/purr
/sys/devices/system/cpu/cpuX/spurr
/sys/devices/system/cpu/cpuX/idle_purr
/sys/devices/system/cpu/cpuX/idle_spurr
Signed-off-by: Gautham R. Shenoy
---
Documentation/A
From: "Gautham R. Shenoy"
Currently prior to entering an idle state on a Linux Guest, the
pseries cpuidle driver implement an idle_loop_prolog() and
idle_loop_epilog() functions which ensure that idle_purr is correctly
computed, and the hypervisor is informed that the CPU cycles
From: "Gautham R. Shenoy"
Hi,
This is the third version of the patches to track and expose idle PURR
and SPURR ticks. These patches are required by tools such as lparstat
to compute system utilization for capacity planning purposes.
The previous versions can be found here:
v2: https:
From: "Gautham R. Shenoy"
Currently purr, spurr, idle_purr, idle_spurr are exposed for every CPU
via the sysfs interface
/sys/devices/system/cpu/cpuX/[idle_][s]purr. Each sysfs read currently
generates an IPI to obtain the desired value from the target CPU X.
Since these aforementi
pu frequency to policy->cur
> on unthrottling")
> Signed-off-by: Pratik Rajesh Sampat
Reviewed-by: Gautham R. Shenoy
> ---
> drivers/cpufreq/powernv-cpufreq.c | 13 -
> 1 file changed, 8 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/cpufreq/
From: "Gautham R. Shenoy"
Hi,
This is the fourth version of the patches to track and expose idle PURR
and SPURR ticks. These patches are required by tools such as lparstat
to compute system utilization for capacity planning purposes.
The previous versions can be found here:
From: "Gautham R. Shenoy"
Currently when CPU goes idle, we take a snapshot of PURR via
pseries_idle_prolog() which is used at the CPU idle exit to compute
the idle PURR cycles via the function pseries_idle_epilog(). Thus,
the value of idle PURR cycle thus read before pseries_idle_pr
From: "Gautham R. Shenoy"
On Pseries LPARs, to calculate utilization, we need to know the
[S]PURR ticks when the CPUs were busy or idle.
Via pseries_idle_prolog(), pseries_idle_epilog(), we track the idle
PURR ticks in the VPA variable "wait_state_cycles". This patch ex
From: "Gautham R. Shenoy"
Currently prior to entering an idle state on a Linux Guest, the
pseries cpuidle driver implement an idle_loop_prolog() and
idle_loop_epilog() functions which ensure that idle_purr is correctly
computed, and the hypervisor is informed that the CPU cycles
From: "Gautham R. Shenoy"
Add documentation for the following sysfs interfaces:
/sys/devices/system/cpu/cpuX/purr
/sys/devices/system/cpu/cpuX/spurr
/sys/devices/system/cpu/cpuX/idle_purr
/sys/devices/system/cpu/cpuX/idle_spurr
Signed-off-by: Gautham R. Shenoy
---
Documentation/A
From: "Gautham R. Shenoy"
Currently purr, spurr, idle_purr, idle_spurr are exposed for every CPU
via the sysfs interface
/sys/devices/system/cpu/cpuX/[idle_][s]purr. Each sysfs read currently
generates an IPI to obtain the desired value from the target CPU X.
Since these aforementioned
From: "Gautham R. Shenoy"
On Pseries LPARs, to calculate utilization, we need to know the
[S]PURR ticks when the CPUs were busy or idle.
The total PURR and SPURR ticks are already exposed via the per-cpu
sysfs files "purr" and "spurr". This patch adds support
From: "Gautham R. Shenoy"
If a guest executes a stop instruction when the hypervisor has set the
PSSCR[ESL|EC] bits, the processor will throw an Hypervisor Facility
Unavailable exception. Currently when we receive this exception, we
only check if the exeception is generated due to
From: "Gautham R. Shenoy"
ISA v3.0 allows the guest to execute a stop instruction. For this, the
PSSCR[ESL|EC] bits need to be cleared by the hypervisor before
scheduling in the guest vCPU.
Currently we always schedule in a vCPU with PSSCR[ESL|EC] bits
set. This patch changes the be
From: "Gautham R. Shenoy"
*** RFC Only. Not intended for inclusion
Motivation
~~~
The POWER ISA v3.0 allows stop instruction to be executed from a Guest
Kernel (HV=0,PR=0) context. If the hypervisor has cleared
PSSCR[ESL|EC] bits, then the stop instru
From: "Gautham R. Shenoy"
The POWER ISA v3.0 allows stop instruction to be executed from a
HV=0,PR=0 context. If the PSSCR[ESL|EC] bits are cleared, then the
stop instruction thus executed will cause the thread to pause, thereby
donating its cycles to the other threads in the core
On Tue, Mar 31, 2020 at 05:40:55PM +0530, Gautham R. Shenoy wrote:
> From: "Gautham R. Shenoy"
>
>
> *** RFC Only. Not intended for inclusion
>
> Motivation
> ~~~
>
> The POWER ISA v3.0 allows stop instruction to be executed from
Hello Naveen,
On Wed, Apr 01, 2020 at 03:28:48PM +0530, Naveen N. Rao wrote:
> Gautham R. Shenoy wrote:
> >From: "Gautham R. Shenoy"
> >
[..snip..]
> >-static DEVICE_ATTR(spurr, 0400, show_spurr, NULL);
> >-static DEVICE_ATTR(purr, 0400, show_purr, st
On Wed, Apr 01, 2020 at 03:12:53PM +0530, Naveen N. Rao wrote:
> Hi Gautham,
>
> Gautham R. Shenoy wrote:
> >From: "Gautham R. Shenoy"
> >
> >Currently when CPU goes idle, we take a snapshot of PURR via
> >pseries_idle_prolog() which is used at th
Hi Naveen,
On Thu, Apr 02, 2020 at 01:04:34PM +0530, Naveen N. Rao wrote:
[..snip..]
> >
> >It does reduce it to 10ms window. I am not sure if anyone samples PURR
> >etc faster than that rate.
> >
> >I measured how much time it takes to read the purr, spurr, idle_purr,
> >idle_spurr files back-to
On Fri, Apr 03, 2020 at 04:04:56PM +0530, Naveen N. Rao wrote:
> Gautham R Shenoy wrote:
> >On Wed, Apr 01, 2020 at 03:12:53PM +0530, Naveen N. Rao wrote:
> >>Hi Gautham,
> >>
> >>Gautham R. Shenoy wrote:
> >>>From: "Gautham R. Shenoy"
>
On Fri, Apr 03, 2020 at 12:20:26PM +1000, Nicholas Piggin wrote:
> Gautham R. Shenoy's on March 31, 2020 10:10 pm:
> > From: "Gautham R. Shenoy"
> >
> > ISA v3.0 allows the guest to execute a stop instruction. For this, the
> > PSSCR[ESL|EC] bits n
From: "Gautham R. Shenoy"
This is the v2 of the fix to change the default behaviour of cede_offline.
The previous version can be found here: https://lkml.org/lkml/2019/9/12/222
The main change from v1 is that the patch2 to create a sysfs file to
report and control th
From: "Gautham R. Shenoy"
Currently on PSeries Linux guests, the offlined CPU can be put to one
of the following two states:
- Long term processor cede (also called extended cede)
- Returned to the hypervisor via RTAS "stop-self" call.
This is controlled by the k
1 - 100 of 626 matches
Mail list logo
