of cpuidle
states in the driver by the user. Will add the capability of this
driver to be used as a module in subsequent patches.
Original idea and discussion for this patch can be found at:
https://lkml.org/lkml/2019/12/17/655
Signed-off-by: Abhishek Goel
---
v1->v2 : Changes adapted f
set of cpuidle
states in the driver. Will add the capability of this driver to be used
as a module in subsequent patches.
Original idea and discussion for this patch can be found at:
https://lkml.org/lkml/2019/12/17/655
Signed-off-by: Abhishek Goel
---
v1-> v2 : Includes minor changes adap
set of cpuidle
states in the driver. Will add the capability of this driver to be used
as a module in subsequent patches.
Original idea and discussion for this patch can be found at:
https://lkml.org/lkml/2019/12/17/655
Signed-off-by: Abhishek Goel
---
drivers/cpuidle/Kconfig| 9
Commit 1961acad2f88559c2cdd2ef67c58c3627f1f6e54 removes usage of
function "validate_dt_prop_sizes". This patch removes this unused
function.
Signed-off-by: Abhishek Goel
---
drivers/cpuidle/cpuidle-powernv.c | 14 --
1 file changed, 14 deletions(-)
diff --git a/drive
if firmware-stop-supported property is present.
Earlier part of this patch was posted in this series :
https://lkml.org/lkml/2020/3/4/589
Signed-off-by: Abhishek Goel
Signed-off-by: Pratik Rajesh Sampat
---
v1->v2 : Combined patch 2 and 3 from previous iteration and rebased it.
arch/powe
, are then handled in opal,
from where we return successfully back to kernel.
Signed-off-by: Abhishek Goel
---
v1->v2 : Rebased the patch on Nick's Opal V4 OS patchset
arch/powerpc/include/asm/opal-api.h| 4 +++-
arch/powerpc/include/asm/opal.h| 1 +
arch/powerpc/p
info
Subcommand not supported on POWER
Same result for set subcommand.
This patch does not affect results on a intel box.
Signed-off-by: Abhishek Goel
Acked-by: Thomas Renninger
---
v1 -> v2 : Instead of bailing out early in set and info commands, in V2,
we are cutting out support f
tches that was found out in recent experiments.
No change in patch 2 and 3.
Abhishek Goel (3):
cpuidle-powernv : forced wakeup for stop states
cpuidle : Add callback whenever a state usage is enabled/disabled
cpuidle-powernv : Recompute the idle-state timeouts when state usage
: Abhishek Goel
---
drivers/cpuidle/sysfs.c | 15 ++-
include/linux/cpuidle.h | 3 +++
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c
index 2bb2683b4..6c9bf2f7b 100644
--- a/drivers/cpuidle/sysfs.c
+++ b/drivers/cpuidle
idle path.
We also use the computed timeout to get timeout for snooze, thus getting
rid of get_snooze_timeout for snooze loop.
Signed-off-by: Abhishek Goel
---
drivers/cpuidle/cpuidle-powernv.c | 35 +++
include/linux/cpuidle.h | 1 +
2 files changed, 13
essentially train the
governor to select a deeper state for that cpu, as the timer here
corresponds to the next available cpuidle state residency. Thus, cpu will
eventually end up in the deepest possible state.
Signed-off-by: Abhishek Goel
---
Auto-promotion
v1 : started as auto promotion logic for
info
Supported commands are:
frequency-info
frequency-set
idle-info
idle-set
monitor
help
Same result for set subcommand.
This patch does not affect results on a intel box.
Signed-off-by: Abhishek Goel
Acked-by: Thomas Renninger
---
changes fr
Cpupower tool has set and info options which are not being used by
POWER machines. For powerpc, we will return directly for these two
subcommands. This removes the ambiguous error message while using set
option in case of power systems.
Signed-off-by: Abhishek Goel
---
tools/power/cpupower
info.
Signed-off-by: Abhishek Goel
---
arch/powerpc/include/asm/processor.h | 5 +-
arch/powerpc/platforms/powernv/idle.c | 50 ---
drivers/cpuidle/cpuidle-powernv.c | 69 +--
3 files changed, 55 insertions(+), 69 deletions(-)
diff --git a/arch/powerpc
Removed threshold latency which was being used to decide if a state
is cpuidle type or not. This decision can be taken using flags, as this
information has been encapsulated in the state->flags and being read
from idle device-tree.
Signed-off-by: Abhishek Goel
---
arch/powerpc/include/asm/o
residency-ns = <0x989680>;
psscr = <0x0 0x300374>;
...
};
...
stop11 {
...
compatible = "stop11,v1",
...
};
bility of the code.
Stop handle corresponding to each state can be called directly since
state pointer is being passed now.
Signed-off-by: Abhishek Goel
---
arch/powerpc/include/asm/cpuidle.h| 8 +-
arch/powerpc/platforms/powernv/idle.c | 331 +++---
2 files changed,
idle path.
We also use the computed timeout to get timeout for snooze, thus getting
rid of get_snooze_timeout for snooze loop.
Signed-off-by: Abhishek Goel
---
drivers/cpuidle/cpuidle-powernv.c | 35 +++
include/linux/cpuidle.h | 1 +
2 files changed, 13
ame of set/reset decrementer function.
Handled irq work pending in try_set_dec_before_idle.
No change in patch 2 and 3.
Abhishek Goel (3):
cpuidle-powernv : forced wakeup for stop states
cpuidle : Add callback whenever a state usage is enabled/disabled
cpuidle-powernv : Recomput
-by: Abhishek Goel
---
drivers/cpuidle/sysfs.c | 15 ++-
include/linux/cpuidle.h | 4
2 files changed, 18 insertions(+), 1 deletion(-)
diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c
index eb20adb5de23..141671a53967 100644
--- a/drivers/cpuidle/sysfs.c
+++ b/drivers
essentially train the
governor to select a deeper state for that cpu, as the timer here
corresponds to the next available cpuidle state residency. Thus, cpu will
eventually end up in the deepest possible state.
Signed-off-by: Abhishek Goel
---
Auto-promotion
v1 : started as auto promotion logic for
-by: Abhishek Goel
---
drivers/cpuidle/sysfs.c | 15 ++-
include/linux/cpuidle.h | 4
2 files changed, 18 insertions(+), 1 deletion(-)
diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c
index eb20adb5d..141671a53 100644
--- a/drivers/cpuidle/sysfs.c
+++ b/drivers
whenever a
state is enbaled or disabled instead of computing everytime in fast
idle path.
3) Use disable callback to recompute timeout whenever state usage
is changed for a state. Also, cleaned up the get_snooze_timeout
function.
Abhishek Goel (3):
cpuid
essentially train the
governor to select a deeper state for that cpu, as the timer here
corresponds to the next available cpuidle state residency. Thus, cpu will
eventually end up in the deepest possible state.
Signed-off-by: Abhishek Goel
---
Auto-promotion
v1 : started as auto promotion logic for
idle path.
We also use the computed timeout to get timeout for snooze, thus getting
rid of get_snooze_timeout for snooze loop.
Signed-off-by: Abhishek Goel
---
drivers/cpuidle/cpuidle-powernv.c | 35 +++
include/linux/cpuidle.h | 1 +
2 files changed, 13
essentially train the
governor to select a deeper state for that cpu, as the timer here
corresponds to the next available cpuidle state residency. Thus, cpu will
eventually end up in the deepest possible state.
Signed-off-by: Abhishek Goel
---
Auto-promotion
v1 : started as auto promotion logic for
ced wakeup instead of auto-promotion
v2 of forced-wakeup : Extended the forced wakeup logic for all states.
Setting the decrementer instead of queuing up a hrtimer to implement the
logic.
Abhishek Goel (1):
cpuidle-powernv : forced wakeup for stop states
drivers/cpuidle/cpuidle-powe
rved to be missing the last cpu in related cpu
list. This patch fixes the problem.
Signed-off-by: Abhishek Goel
---
tools/power/cpupower/utils/cpufreq-set.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tools/power/cpupower/utils/cpufreq-set.c
b/tools/power/cpupower/utils/cpufreq-set.c
index
. Few such iterations will essentially train the governor to
select a deeper state for that cpu, as the timer here corresponds to the
next available cpuidle state residency. Cpu will be kicked out of the lite
state and end up in a non-lite state.
Signed-off-by: Abhishek Goel
---
arch/powerpc/include
down power consumption. We may extend this
model for other states in future.
Abhishek Goel (1):
cpuidle-powernv : forced wakeup for stop lite states
arch/powerpc/include/asm/opal-api.h | 1 +
drivers/cpuidle/cpuidle-powernv.c | 71 -
2 files changed, 71 insertions(+),
forms
will rely on "scaling_boost_frequency" file to display boost frequency.
Signed-off-by: Abhishek Goel
---
tools/power/cpupower/lib/cpufreq.c| 19 +-
tools/power/cpupower/lib/cpufreq.h| 16 -
tools/power/cpupower/utils/cpufreq-info.c | 42
:~# cpupower frequency-set -f 206
root@ubuntu:~# cpupower frequency-set -f 2061000
Signed-off-by: Abhishek Goel
Signed-off-by: Bharath Thodla
Tested-by: Thomas Renninger
---
tools/power/cpupower/Makefile | 6 +-
tools/power/cpupower/cpupower-completion.sh | 128
l" or "cpupower idle-info".
Signed-off-by: Abhishek Goel
---
The skiboot patch which adds description for idle states in device tree
can be found here: https://patchwork.ozlabs.org/patch/921637/
drivers/cpuidle/cpuidle-powernv.c | 17 +
include/linux/cpuidle.h
0.00| 1.58| 0.58| 0.42| 8.55| 0.09| 21.11| 0.99| 63.32
1| 0.00| 1.26| 0.88| 0.43| 9.00| 0.02| 7.78| 4.65| 71.91
3| 0.00| 0.30| 0.42| 0.06| 13.62| 0.21| 30.29| 0.00| 52.45
Signed-off-by: Abhishek Goel
---
Previous attempt was made by Seeteena Thoufeek. This is the pa
y.
Signed-off-by: Abhishek Goel
---
Skiboot patch required for the corresponding device-tree changes have been
posted here : http://patchwork.ozlabs.org/patch/862256/
drivers/cpufreq/powernv-cpufreq.c | 104 ++
1 file changed, 95 insertions(+), 9 deletion
y.
Signed-off-by: Abhishek Goel
---
drivers/cpufreq/powernv-cpufreq.c | 110 ++
1 file changed, 100 insertions(+), 10 deletions(-)
diff --git a/drivers/cpufreq/powernv-cpufreq.c
b/drivers/cpufreq/powernv-cpufreq.c
index b6d7c4c..fd642bc 100644
--- a/drive
y.
Signed-off-by: Abhishek Goel
---
drivers/cpufreq/powernv-cpufreq.c | 95 +++
1 file changed, 87 insertions(+), 8 deletions(-)
diff --git a/drivers/cpufreq/powernv-cpufreq.c
b/drivers/cpufreq/powernv-cpufreq.c
index b6d7c4c..9384110 100644
--- a/drive
cpuidle_monitor used to assume that cpu0 is always online which is not
a valid assumption on POWER machines. This patch fixes this by getting
the cpu on which the current thread is running, instead of always using
cpu0 for monitoring which may not be online.
Signed-off-by: Abhishek Goel
cpupower_is_cpu_online was incorrectly checking for 0. This patch fixes
this by checking for 1 when the cpu is online.
Signed-off-by: Abhishek Goel
---
v2: Commit message rephrased.
---
tools/power/cpupower/bench/system.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools
cpuidle_monitor used to assume that cpu0 is always online which is not
a valid assumption on POWER machines. This patch fixes this by searching
for the first online cpu and uses it, instead of always using cpu0 for
monitoring which may not be online.
Signed-off-by: Abhishek Goel
---
v2: Commit
cpuidle_monitor used to assume that cpu0 is always online. Now the
cpuidle_monitor function searches for the first online cpu and use
it, instead of always using cpu0 which may not be online.
Signed-off-by: Abhishek Goel
---
tools/power/cpupower/utils/idle_monitor/cpuidle_sysfs.c | 14
cpuidle_monitor used to assume that cpu0 is always online. Now the
cpuidle_monitor function searches for the first online cpu and use
it, instead of always using cpu0 which maynot be online.
Signed-off-by: Abhishek Goel
---
tools/power/cpupower/utils/idle_monitor/cpuidle_sysfs.c | 6 +++---
1
cpupower_is_cpu_online returns 1 when cpu is online. But in function
set_cpufreq_governor, it was incorrectly checking for 0 in case cpu is
online.
Signed-off-by: Abhishek Goel
---
tools/power/cpupower/bench/system.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools
43 matches
Mail list logo
