LinuxLists.cc - [PATCH v4 0/3] cpufreq/amd-pstate: Fix null pointer dereference when frequency invariance is disabled

2023-05-17 16:37:45

Subject: [PATCH v4 0/3] cpufreq/amd-pstate: Fix null pointer dereference when frequency invariance is disabled

Currently, amd_pstate only uses `adjust_perf` and `target` callback
functions to get the frequency/effective utilization data from scaling
governors. But the governor can switch back to fast_switch (as
fast_switch_possible is set) when frequency invariance is disabled. In
that case, if the driver didn't set fast_switch function pointer, null
pointer dereference can happen and the kernel will crash.

Currently, frequency invariance is turned off when any anomaly is
detected with aperf/mperf readings. Which triggers this problem.

Fix this by adding `fast_switch` function in the amd_pstate driver.

The same problem could arise with any scaling driver which doesn't set
fast_switch but sets fast_switch_possible. Return failure while loading
such drivers.

Changelog:
v3 -> v4:
- Add extra patch to remove fast_switch_possible flag from amd_pstate
active mode driver.
- Fail to load such drivers which sets fast_switch_possible flag but not
fast_switch.

v2 -> v3
- Add fast_switch callback in amd_pstate and warn if any driver
sets fast_swich_possible flag but not fast_switch function pointer.

v1 -> v2
- Remove fast_switch_enabled flag dependency for adjust_perf callback

v1: https://lore.kernel.org/linux-pm/[email protected]/
v2: https://lore.kernel.org/linux-pm/[email protected]/
v3: https://lore.kernel.org/lkml/[email protected]/

Gautham R. Shenoy (1):
amd_pstate: Add ->fast_switch() callback

Wyes Karny (2):
cpufreq/amd-pstate: Remove fast_switch_possible flag from active
driver
cpufreq: Return failure if fast_switch is not set and
fast_switch_possible is set

drivers/cpufreq/amd-pstate.c | 38 ++++++++++++++++++++++++++++--------
drivers/cpufreq/cpufreq.c | 5 +++++
include/linux/cpufreq.h | 4 +++-
3 files changed, 38 insertions(+), 9 deletions(-)

--
2.34.1

2023-05-17 16:38:52

by Wyes Karny

[permalink] [raw]

Subject: [PATCH v4 2/3] cpufreq/amd-pstate: Remove fast_switch_possible flag from active driver

amd_pstate active mode driver is only compatible with static governors.
Therefore it doesn't need fast_switch functionality. Remove
fast_switch_possible flag from amd_pstate active mode driver.

Fixes: ffa5096a7c33 ("cpufreq: amd-pstate: implement Pstate EPP support for the AMD processors")
Signed-off-by: Wyes Karny <[email protected]>
---
drivers/cpufreq/amd-pstate.c | 2 --
1 file changed, 2 deletions(-)

diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
index 45711fc0a856..ac54779f5a49 100644
--- a/drivers/cpufreq/amd-pstate.c
+++ b/drivers/cpufreq/amd-pstate.c
@@ -1102,7 +1102,6 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy)
policy->policy = CPUFREQ_POLICY_POWERSAVE;

if (boot_cpu_has(X86_FEATURE_CPPC)) {
- policy->fast_switch_possible = true;
ret = rdmsrl_on_cpu(cpudata->cpu, MSR_AMD_CPPC_REQ, &value);
if (ret)
return ret;
@@ -1125,7 +1124,6 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy)
static int amd_pstate_epp_cpu_exit(struct cpufreq_policy *policy)
{
pr_debug("CPU %d exiting\n", policy->cpu);
- policy->fast_switch_possible = false;
return 0;
}

--
2.34.1

2023-05-17 16:41:27

by Wyes Karny

[permalink] [raw]

Subject: [PATCH v4 3/3] cpufreq: Return failure if fast_switch is not set and fast_switch_possible is set

If fast_switch_possible flag is set by the scaling driver, the governor
is free to select fast_switch function even if adjust_perf is set. Some
scaling drivers which use adjust_perf don't set fast_switch thinking
that the governor would never fall back to fast_switch. But the governor
can fall back to fast_switch even in runtime if frequency invariance is
disabled due to some reason. This could crash the kernel if the driver
didn't set the fast_switch function pointer.

Therefore, return failure in cpufreq_online function if fast_switch is
not set and fast_switch_possible is set.

Signed-off-by: Wyes Karny <[email protected]>
---
drivers/cpufreq/cpufreq.c | 5 +++++
include/linux/cpufreq.h | 4 +++-
2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 6b52ebe5a890..7835ba4fa34c 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1376,6 +1376,11 @@ static int cpufreq_online(unsigned int cpu)
goto out_free_policy;
}

+ if (policy->fast_switch_possible && !cpufreq_driver->fast_switch) {
+ pr_err("fast_switch_possible is enabled but fast_switch callback is not set\n");
+ ret = -EINVAL;
+ goto out_destroy_policy;
+ }
/*
* The initialization has succeeded and the policy is online.
* If there is a problem with its frequency table, take it
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 26e2eb399484..8cdf77bb3bc1 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -340,7 +340,9 @@ struct cpufreq_driver {
/*
* ->fast_switch() replacement for drivers that use an internal
* representation of performance levels and can pass hints other than
- * the target performance level to the hardware.
+ * the target performance level to the hardware. If driver is setting this,
+ * then it needs to set fast_switch also. Because in certain scenario scale
+ * invariance could be disabled and governor can switch back to fast_switch.
*/
void (*adjust_perf)(unsigned int cpu,
unsigned long min_perf,
--
2.34.1

2023-05-17 16:49:31

by Wyes Karny

[permalink] [raw]

Subject: [PATCH v4 1/3] amd_pstate: Add ->fast_switch() callback

From: "Gautham R. Shenoy" <[email protected]>

Schedutil normally calls the adjust_perf callback for drivers with
adjust_perf callback available and fast_switch_possible flag set.
However, when frequency invariance is disabled and schedutil tries to
invoke fast_switch. So, there is a chance of kernel crash if this
function pointer is not set. To protect against this scenario add
fast_switch callback to amd_pstate driver.

Fixes: 1d215f0319c2 ("cpufreq: amd-pstate: Add fast switch function for AMD P-State")

Signed-off-by: Wyes Karny <[email protected]>
Signed-off-by: Gautham R. Shenoy <[email protected]>
---
drivers/cpufreq/amd-pstate.c | 36 ++++++++++++++++++++++++++++++------
1 file changed, 30 insertions(+), 6 deletions(-)

diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
index 5a3d4aa0f45a..45711fc0a856 100644
--- a/drivers/cpufreq/amd-pstate.c
+++ b/drivers/cpufreq/amd-pstate.c
@@ -444,9 +444,8 @@ static int amd_pstate_verify(struct cpufreq_policy_data *policy)
return 0;
}

-static int amd_pstate_target(struct cpufreq_policy *policy,
- unsigned int target_freq,
- unsigned int relation)
+static int amd_pstate_update_freq(struct cpufreq_policy *policy,
+ unsigned int target_freq, bool fast_switch)
{
struct cpufreq_freqs freqs;
struct amd_cpudata *cpudata = policy->driver_data;
@@ -465,14 +464,37 @@ static int amd_pstate_target(struct cpufreq_policy *policy,
des_perf = DIV_ROUND_CLOSEST(target_freq * cap_perf,
cpudata->max_freq);

- cpufreq_freq_transition_begin(policy, &freqs);
+ WARN_ON(fast_switch && !policy->fast_switch_enabled);
+ /*
+ * If fast_switch is desired, then there aren't any registered
+ * transition notifiers. See comment for
+ * cpufreq_enable_fast_switch().
+ */
+ if (!fast_switch)
+ cpufreq_freq_transition_begin(policy, &freqs);
+
amd_pstate_update(cpudata, min_perf, des_perf,
- max_perf, false, policy->governor->flags);
- cpufreq_freq_transition_end(policy, &freqs, false);
+ max_perf, fast_switch, policy->governor->flags);
+
+ if (!fast_switch)
+ cpufreq_freq_transition_end(policy, &freqs, false);

return 0;
}

+static int amd_pstate_target(struct cpufreq_policy *policy,
+ unsigned int target_freq,
+ unsigned int relation)
+{
+ return amd_pstate_update_freq(policy, target_freq, false);
+}
+
+static unsigned int amd_pstate_fast_switch(struct cpufreq_policy *policy,
+ unsigned int target_freq)
+{
+ return amd_pstate_update_freq(policy, target_freq, true);
+}
+
static void amd_pstate_adjust_perf(unsigned int cpu,
unsigned long _min_perf,
unsigned long target_perf,
@@ -715,6 +737,7 @@ static int amd_pstate_cpu_exit(struct cpufreq_policy *policy)

freq_qos_remove_request(&cpudata->req[1]);
freq_qos_remove_request(&cpudata->req[0]);
+ policy->fast_switch_possible = false;
kfree(cpudata);

return 0;
@@ -1309,6 +1332,7 @@ static struct cpufreq_driver amd_pstate_driver = {
.flags = CPUFREQ_CONST_LOOPS | CPUFREQ_NEED_UPDATE_LIMITS,
.verify = amd_pstate_verify,
.target = amd_pstate_target,
+ .fast_switch = amd_pstate_fast_switch,
.init = amd_pstate_cpu_init,
.exit = amd_pstate_cpu_exit,
.suspend = amd_pstate_cpu_suspend,
--
2.34.1

2023-05-22 08:58:21

by kernel test robot

[permalink] [raw]

Subject: Re: [PATCH v4 3/3] cpufreq: Return failure if fast_switch is not set and fast_switch_possible is set

Hello,

kernel test robot

commit: 278dfbd17c5d076783 url: In addition to that,
+------------------+------ | testcase: change | test machine | test parameters | +------------------+------ | testcase: change | test machine | test parameters | | disk=1HDD | | fs=xfs | +------------------+------ | testcase: change | test machine | test parameters | | option_a=ffs | +------------------+------ | testcase: change | test machine | test parameters | | +------------------+------ | testcase: change | test machine | test parameters | | option_a=3 | +------------------+------ | testcase: change | test machine | test parameters | | option_a=Medium | +------------------+------ | testcase: change | test machine | test parameters | | +------------------+------ | testcase: change | test machine | test parameters | | need_x=true | | +------------------+------

If you fix the issue, | Reported-by: kernel | Closes:
noticed a 302.3% regression of phoronix-test-suite.glibc-bench.sin.nanoseconds on:
4471d7fea71d3127c3ecfb ("[PATCH v4 3/3] cpufreq: Return failure if fast_switch is not set and fast_switch_possible is set")
hub.com/intel-lab-lkp/linux/commits/Wyes-Karny/amd_pstate-Add-fast_switch-callback/20230518-010448">https://github.com/intel-lab-lkp/linux/commits/Wyes-Karny/amd_pstate-Add-fast_switch-callback/20230518-010448
t.kernel.org/cgit/linux/kernel/git/rafael/linux-pm.git">https://git.kernel.org/cgit/linux/kernel/git/rafael/linux-pm.git linux-next
s://lore.kernel.org/all/20230517162817.8538-4-wyes.karny@amd.com/">https://lore.kernel.org/all/[email protected]/
v4 3/3] cpufreq: Return failure if fast_switch is not set and fast_switch_possible is set
threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
The Phoronix Test Suite is the most comprehensive testing and benchmarking platform available that provides an extensible framework for which new tests can be easily added.
/www.phoronix-test-suite.com/">http://www.phoronix-test-suite.com/
the commit also has significant impact on the following tests:
-----------------------------------------------------------------------------------------------------------------------------------+
| phoronix-test-suite: phoronix-test-suite.n-queens.0.seconds 313.6% regression |
| 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory |
| cpufreq_governor=performance |
| test=n-queens-1.2.1 |
-----------------------------------------------------------------------------------------------------------------------------------+
| filebench: filebench.sum_operations/s -72.1% regression |
| 96 threads 2 sockets (Ice Lake) with 256G memory |
| cpufreq_governor=performance |
|
|
| test=cvar_example.f |
-----------------------------------------------------------------------------------------------------------------------------------+
| phoronix-test-suite: phoronix-test-suite.glibc-bench.ffs.nanoseconds 301.2% regression |
| 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory |
| cpufreq_governor=performance |
|
| test=glibc-bench-1.6.1 |
-----------------------------------------------------------------------------------------------------------------------------------+
| phoronix-test-suite: phoronix-test-suite.glibc-bench.pthread_once.nanoseconds 300.8% regression |
| 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory |
| cpufreq_governor=performance |
| option_a=pthread_once |
| test=glibc-bench-1.6.1 |
-----------------------------------------------------------------------------------------------------------------------------------+
| phoronix-test-suite: phoronix-test-suite.cyclictest.clock_nanosleepTIME_ABSTIME,Interval500ms,100000Loops.ms_average 5550.0% regression |
| 96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory |
| cpufreq_governor=performance |
|
| test=cyclictest-1.0.0 |
-----------------------------------------------------------------------------------------------------------------------------------+
| phoronix-test-suite: phoronix-test-suite.astcenc.Medium.seconds 106.6% regression |
| 16 threads 1 sockets Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz (Coffee Lake) with 32G memory |
| cpufreq_governor=performance |
|
| test=astcenc-1.1.0 |
-----------------------------------------------------------------------------------------------------------------------------------+
| phoronix-test-suite: phoronix-test-suite.pyperformance.pickle_pure_python.milliseconds -99.4% regression |
| 16 threads 1 sockets Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz (Coffee Lake) with 32G memory |
| cpufreq_governor=performance |
| option_a=pickle_pure_python |
| test=pyperformance-1.0.2 |
-----------------------------------------------------------------------------------------------------------------------------------+
| phoronix-test-suite: phoronix-test-suite.x11perf.500pxCopyFromWindowToWindow.operations___second -7.3% regression |
| 12 threads 1 sockets Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz (Coffee Lake) with 32G memory |
| cpufreq_governor=performance |
|
| option_a=500px Copy From Window To Window |
| test=x11perf-1.1.1 |
-----------------------------------------------------------------------------------------------------------------------------------+
kindly add following tag
test robot <[email protected]>
//lore.kernel.org/oe-lkp/202305221612.200d8fb1-oliver.sang@intel.com">https://lore.kernel.org/oe-lkp/[email protected]
------------------------------------------------------------------------>
href="https://github.com/intel/lkp-tests.git">https://github.com/intel/lkp-tests.git
# job file is attached in this email
--compatible job.yaml # generate the yaml file for lkp run
run generated-yaml-file
across any failure that blocks the test,
remove ~/.lkp and /lkp dir to run from a clean state.
===============================================================
kconfig/option_a/rootfs/tbox_group/test/testcase:
4-rhel-8.3/sin/debian-x86_64-phoronix/lkp-csl-2sp7/glibc-bench-1.6.1/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
289.69 phoronix-test-suite.glibc-bench.sin.nanoseconds
98.84 phoronix-test-suite.time.elapsed_time
98.84 phoronix-test-suite.time.elapsed_time.max
1525 ? 3% phoronix-test-suite.time.involuntary_context_switches
+5.6% 78.50 phoronix-test-suite.time.percent_of_cpu_this_job_got
69.52 phoronix-test-suite.time.user_time
10593 meminfo.Active(anon)
+9.8% 9.003e+09 cpuidle..time
+12.1% 9564836 cpuidle..usage
48852 ? 28% numa-meminfo.node0.AnonHugePages
65375 ? 24% numa-meminfo.node1.AnonHugePages
+13.8% 164.46 uptime.boot
14439 uptime.idle
2.00 vmstat.procs.r
4888 vmstat.system.cs
-2.3% 93730 vmstat.system.in
0.03 ? 4% mpstat.cpu.all.iowait%
3.59 mpstat.cpu.all.irq%
0.21 ? 2% mpstat.cpu.all.soft%
0.28 mpstat.cpu.all.sys%
1.68 mpstat.cpu.all.usr%
2648 proc-vmstat.nr_active_anon
-4.0% 105659 proc-vmstat.nr_anon_pages
-1.6% 281968 proc-vmstat.nr_inactive_anon
-1.1% 52049 proc-vmstat.nr_slab_unreclaimable
2648 proc-vmstat.nr_zone_active_anon
-1.6% 281968 proc-vmstat.nr_zone_inactive_anon
+8.7% 557738 proc-vmstat.numa_hit
+9.0% 538756 proc-vmstat.numa_local
+9.6% 649301 proc-vmstat.pgalloc_normal
+10.6% 754452 proc-vmstat.pgfault
+10.6% 544510 proc-vmstat.pgfree
+16.7% 104272 proc-vmstat.pgreuse
+13.6% 948992 proc-vmstat.unevictable_pgs_scanned
8.90 turbostat.Busy%
-56.4% 700.00 turbostat.Bzy_MHz
217899 ? 7% turbostat.C1E
0.30 ? 2% turbostat.C1E%
+10.9% 9254503 turbostat.C6
56.44 turbostat.CPU%c1
34.65 turbostat.CPU%c6
38.17 turbostat.CoreTmp
+11.7% 9542613 turbostat.IRQ
1375 ? 38% turbostat.POLL
8.06 ? 3% turbostat.Pkg%pc2
0.66 ? 4% turbostat.Pkg%pc6
38.50 turbostat.PkgTmp
70.00 turbostat.PkgWatt
37.52 ? 22% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
37.52 ? 22% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.42 ?155% perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.__mmput.exit_mm.do_exit
2.42 ?155% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.exit_mmap.__mmput.exit_mm
1.26 ?145% perf-profile.calltrace.cycles-pp.asm_exc_page_fault.fault_in_readable.fault_in_iov_iter_readable.generic_perform_write.__generic_file_write_iter
4.09 ?178% perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap.__mmput
4.09 ?178% perf-profile.calltrace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap
0.00 perf-profile.calltrace.cycles-pp.page_remove_rmap.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas
0.50 ?223% perf-profile.calltrace.cycles-pp.__do_fault.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault
2.17 ?169% perf-profile.calltrace.cycles-pp.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault
2.17 ?169% perf-profile.calltrace.cycles-pp.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault
37.52 ? 22% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
37.52 ? 22% perf-profile.children.cycles-pp.do_syscall_64
4.09 ?178% perf-profile.children.cycles-pp.unmap_vmas
4.09 ?178% perf-profile.children.cycles-pp.unmap_page_range
4.09 ?178% perf-profile.children.cycles-pp.zap_pmd_range
4.09 ?178% perf-profile.children.cycles-pp.zap_pte_range
0.00 perf-profile.children.cycles-pp.page_remove_rmap
0.50 ?223% perf-profile.children.cycles-pp.__do_fault
2.17 ?169% perf-profile.children.cycles-pp.do_fault
2.17 ?169% perf-profile.children.cycles-pp.do_read_fault
164728 ?213% sched_debug.cfs_rq:/.load.avg
15373401 ?219% sched_debug.cfs_rq:/.load.max
1566546 ?218% sched_debug.cfs_rq:/.load.stddev
24.09 ? 46% sched_debug.cfs_rq:/.load_avg.avg
95.47 ? 50% sched_debug.cfs_rq:/.load_avg.stddev
56368 ? 5% sched_debug.cfs_rq:/.min_vruntime.max
8.76 ?126% sched_debug.cfs_rq:/.removed.load_avg.avg
63.64 ?106% sched_debug.cfs_rq:/.removed.load_avg.stddev
4.16 ?123% sched_debug.cfs_rq:/.removed.runnable_avg.avg
4.16 ?123% sched_debug.cfs_rq:/.removed.util_avg.avg
-37.7% 133.71 ? 6% sched_debug.cfs_rq:/.runnable_avg.avg
-17.4% 205.07 ? 4% sched_debug.cfs_rq:/.runnable_avg.stddev
-37.7% 133.59 ? 6% sched_debug.cfs_rq:/.util_avg.avg
-17.4% 205.03 ? 4% sched_debug.cfs_rq:/.util_avg.stddev
-41.8% 416.50 ? 25% sched_debug.cfs_rq:/.util_est_enqueued.max
67.78 ? 24% sched_debug.cfs_rq:/.util_est_enqueued.stddev
-49.6% 1201480 ? 12% sched_debug.cpu.avg_idle.max
43615 ? 2% sched_debug.cpu.clock_task.min
14136 ? 82% sched_debug.cpu.max_idle_balance_cost.stddev
0.00 ? 63% sched_debug.cpu.next_balance.stddev
60475 sched_debug.sched_clk
38.19 perf-stat.i.MPKI
-45.2% 4.33e+08 perf-stat.i.branch-instructions
5.96 perf-stat.i.branch-miss-rate%
+64.8% 25060061 perf-stat.i.branch-misses
+68.1% 3438514 perf-stat.i.cache-misses
+69.3% 83372455 perf-stat.i.cache-references
4601 perf-stat.i.context-switches
2.32 perf-stat.i.cpi
-6.3% 5.218e+09 perf-stat.i.cpu-cycles
1571 perf-stat.i.cycles-between-cache-misses
0.52 perf-stat.i.dTLB-load-miss-rate%
+70.0% 2497301 perf-stat.i.dTLB-load-misses
-52.2% 5.343e+08 perf-stat.i.dTLB-loads
0.11 perf-stat.i.dTLB-store-miss-rate%
+63.8% 230910 perf-stat.i.dTLB-store-misses
-46.6% 2.354e+08 perf-stat.i.dTLB-stores
73.84 perf-stat.i.iTLB-load-miss-rate%
-3.9% 768539 perf-stat.i.iTLB-load-misses
-45.3% 273344 ? 3% perf-stat.i.iTLB-loads
-55.2% 2.504e+09 perf-stat.i.instructions
3281 perf-stat.i.instructions-per-iTLB-miss
0.47 perf-stat.i.ipc
0.86 ? 6% perf-stat.i.major-faults
0.05 perf-stat.i.metric.GHz
+40.6% 869.20 perf-stat.i.metric.K/sec
12.53 perf-stat.i.metric.M/sec
79.68 perf-stat.i.node-load-miss-rate%
+36.8% 126293 ? 3% perf-stat.i.node-load-misses
44601 ? 3% perf-stat.i.node-loads
60176 ? 9% perf-stat.i.node-store-misses
33.29 perf-stat.overall.MPKI
5.79 perf-stat.overall.branch-miss-rate%
2.08 perf-stat.overall.cpi
1517 perf-stat.overall.cycles-between-cache-misses
0.47 perf-stat.overall.dTLB-load-miss-rate%
0.10 perf-stat.overall.dTLB-store-miss-rate%
73.77 perf-stat.overall.iTLB-load-miss-rate%
3258 perf-stat.overall.instructions-per-iTLB-miss
0.48 perf-stat.overall.ipc
73.87 perf-stat.overall.node-load-miss-rate%
51.80 ? 8% perf-stat.overall.node-store-miss-rate%
-45.1% 4.286e+08 perf-stat.ps.branch-instructions
+65.0% 24799436 perf-stat.ps.branch-misses
+68.3% 3402668 perf-stat.ps.cache-misses
+69.4% 82502207 perf-stat.ps.cache-references
4554 perf-stat.ps.context-switches
-6.2% 5.164e+09 perf-stat.ps.cpu-cycles
+70.1% 2471118 perf-stat.ps.dTLB-load-misses
-52.1% 5.288e+08 perf-stat.ps.dTLB-loads
+64.0% 228525 perf-stat.ps.dTLB-store-misses
-46.5% 2.33e+08 perf-stat.ps.dTLB-stores
-3.7% 760575 perf-stat.ps.iTLB-load-misses
-45.2% 270481 ? 3% perf-stat.ps.iTLB-loads
-55.1% 2.479e+09 perf-stat.ps.instructions
0.85 ? 6% perf-stat.ps.major-faults
+37.0% 125023 ? 3% perf-stat.ps.node-load-misses
44171 ? 3% perf-stat.ps.node-loads
59560 ? 9% perf-stat.ps.node-store-misses
-49.8% 2.401e+11 perf-stat.total.instructions
*************************************************************************
threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
===============================================================
kconfig/rootfs/tbox_group/test/testcase:
4-rhel-8.3/debian-x86_64-phoronix/lkp-csl-2sp7/n-queens-1.2.1/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
17.10 phoronix-test-suite.n-queens.0.seconds
80.28 phoronix-test-suite.time.elapsed_time
80.28 phoronix-test-suite.time.elapsed_time.max
-3.0% 27994 ? 2% phoronix-test-suite.time.maximum_resident_set_size
+22.0% 2.769e+09 cpuidle..time
+16.5% 2841419 cpuidle..usage
10166 meminfo.Active(anon)
+13.6% 2416662 meminfo.Committed_AS
3240 ? 9% numa-meminfo.node0.PageTables
10489 ? 2% numa-meminfo.node1.KernelStack
2885 ? 9% numa-meminfo.node1.PageTables
-16.6% 810.00 ? 9% numa-vmstat.node0.nr_page_table_pages
10494 ? 3% numa-vmstat.node1.nr_kernel_stack
+49.8% 721.83 ? 9% numa-vmstat.node1.nr_page_table_pages
35.05 mpstat.cpu.all.idle%
0.04 mpstat.cpu.all.iowait%
2.84 mpstat.cpu.all.irq%
0.11 mpstat.cpu.all.soft%
0.29 mpstat.cpu.all.sys%
61.67 mpstat.cpu.all.usr%
36.00 vmstat.cpu.id
59.17 vmstat.cpu.us
1069 ? 3% vmstat.io.bi
60.17 vmstat.procs.r
4927 vmstat.system.cs
-2.6% 94264 vmstat.system.in
-53.7% 454.00 turbostat.Avg_MHz
65.74 turbostat.Busy%
-73.8% 700.00 turbostat.Bzy_MHz
19387 ? 7% turbostat.C1
0.01 turbostat.C1%
108951 ? 11% turbostat.C1E
0.17 ? 9% turbostat.C1E%
+16.6% 2687774 turbostat.C6
34.39 turbostat.C6%
20.16 turbostat.CPU%c1
14.11 turbostat.CPU%c6
38.67 turbostat.CoreTmp
+113.2% 7921708 turbostat.IRQ
-96.2% 509.50 ? 24% turbostat.POLL
3.30 ? 11% turbostat.Pkg%pc2
0.16 ? 5% turbostat.Pkg%pc6
39.17 turbostat.PkgTmp
83.38 turbostat.PkgWatt
+3.3% 143.74 turbostat.RAMWatt
2539 proc-vmstat.nr_active_anon
+3.9% 27867 proc-vmstat.nr_active_file
-4.1% 103038 proc-vmstat.nr_anon_pages
-1.6% 279280 proc-vmstat.nr_inactive_anon
+3.8% 21344 proc-vmstat.nr_kernel_stack
1526 proc-vmstat.nr_page_table_pages
+1.0% 25919 proc-vmstat.nr_slab_reclaimable
2539 proc-vmstat.nr_zone_active_anon
+3.9% 27867 proc-vmstat.nr_zone_active_file
-1.6% 279280 proc-vmstat.nr_zone_inactive_anon
+25.2% 489291 proc-vmstat.numa_hit
+26.5% 470277 proc-vmstat.numa_local
-90.3% 628.33 ? 59% proc-vmstat.numa_pages_migrated
5294 ? 67% proc-vmstat.numa_pte_updates
+24.3% 573395 proc-vmstat.pgalloc_normal
+33.7% 647486 proc-vmstat.pgfault
+29.2% 467634 proc-vmstat.pgfree
-90.3% 628.33 ? 59% proc-vmstat.pgmigrate_success
88315 ? 2% proc-vmstat.pgreuse
678912 proc-vmstat.unevictable_pgs_scanned
6184 ? 32% sched_debug.cfs_rq:/.load.avg
22488 ? 53% sched_debug.cfs_rq:/.load.stddev
24.17 ? 32% sched_debug.cfs_rq:/.load_avg.avg
-19.2% 824.58 ? 25% sched_debug.cfs_rq:/.load_avg.max
-49.2% 102.31 ? 30% sched_debug.cfs_rq:/.load_avg.stddev
59568 ?128% sched_debug.cfs_rq:/.min_vruntime.stddev
7.68 ? 95% sched_debug.cfs_rq:/.removed.load_avg.avg
64.59 ? 83% sched_debug.cfs_rq:/.removed.load_avg.stddev
3.82 ? 96% sched_debug.cfs_rq:/.removed.runnable_avg.avg
32.11 ? 84% sched_debug.cfs_rq:/.removed.runnable_avg.stddev
3.82 ? 96% sched_debug.cfs_rq:/.removed.util_avg.avg
32.11 ? 84% sched_debug.cfs_rq:/.removed.util_avg.stddev
-44.8% 124.22 ? 24% sched_debug.cfs_rq:/.runnable_avg.avg
-21.5% 198.91 ? 13% sched_debug.cfs_rq:/.runnable_avg.stddev
59563 ?128% sched_debug.cfs_rq:/.spread0.stddev
-44.8% 124.11 ? 24% sched_debug.cfs_rq:/.util_avg.avg
-21.5% 198.83 ? 13% sched_debug.cfs_rq:/.util_avg.stddev
-42.9% 447.67 ? 49% sched_debug.cfs_rq:/.util_est_enqueued.max
65.67 ? 33% sched_debug.cfs_rq:/.util_est_enqueued.stddev
8.67 ? 5% sched_debug.cpu.clock.stddev
53120 ? 26% sched_debug.cpu.clock_task.min
+57.5% 220.89 ? 8% sched_debug.cpu.curr->pid.avg
+27.1% 667.27 ? 3% sched_debug.cpu.curr->pid.stddev
0.00 ? 17% sched_debug.cpu.next_balance.stddev
0.10 ? 13% sched_debug.cpu.nr_running.avg
69957 ? 20% sched_debug.sched_clk
0.00 perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.search_binary_handler.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve
0.00 perf-profile.calltrace.cycles-pp.load_elf_binary.search_binary_handler.exec_binprm.bprm_execve.do_execveat_common
0.00 perf-profile.calltrace.cycles-pp.__mmput.exec_mmap.begin_new_exec.load_elf_binary.search_binary_handler
0.00 perf-profile.calltrace.cycles-pp.exit_mmap.__mmput.exec_mmap.begin_new_exec.load_elf_binary
0.00 perf-profile.calltrace.cycles-pp.begin_new_exec.load_elf_binary.search_binary_handler.exec_binprm.bprm_execve
0.00 perf-profile.calltrace.cycles-pp.exec_mmap.begin_new_exec.load_elf_binary.search_binary_handler.exec_binprm
0.00 perf-profile.calltrace.cycles-pp.free_pgtables.exit_mmap.__mmput.exec_mmap.begin_new_exec
6.21 ? 52% perf-profile.calltrace.cycles-pp.perf_release.__fput.task_work_run.do_exit.do_group_exit
6.21 ? 52% perf-profile.calltrace.cycles-pp.perf_event_release_kernel.perf_release.__fput.task_work_run.do_exit
6.90 ? 46% perf-profile.calltrace.cycles-pp.task_work_run.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart
6.90 ? 46% perf-profile.calltrace.cycles-pp.__fput.task_work_run.do_exit.do_group_exit.get_signal
0.00 perf-profile.children.cycles-pp.__x64_sys_execve
0.00 perf-profile.children.cycles-pp.do_execveat_common
0.00 perf-profile.children.cycles-pp.bprm_execve
0.00 perf-profile.children.cycles-pp.exec_binprm
0.00 perf-profile.children.cycles-pp.search_binary_handler
0.00 perf-profile.children.cycles-pp.load_elf_binary
0.00 perf-profile.children.cycles-pp.begin_new_exec
0.00 perf-profile.children.cycles-pp.exec_mmap
1.11 ?223% perf-profile.children.cycles-pp.free_pgtables
6.21 ? 52% perf-profile.children.cycles-pp.perf_release
6.21 ? 52% perf-profile.children.cycles-pp.perf_event_release_kernel
7.74 ? 41% perf-profile.children.cycles-pp.task_work_run
7.74 ? 41% perf-profile.children.cycles-pp.__fput
0.70 ?223% perf-profile.self.cycles-pp.zap_pte_range
12.62 ? 5% perf-stat.i.MPKI
-54.1% 3.721e+09 perf-stat.i.branch-instructions
16.33 perf-stat.i.branch-miss-rate%
-55.4% 7.401e+08 perf-stat.i.branch-misses
9.79 ? 4% perf-stat.i.cache-miss-rate%
-18.4% 1936172 ? 2% perf-stat.i.cache-misses
-25.1% 33764433 perf-stat.i.cache-references
4657 perf-stat.i.context-switches
1.42 ? 2% perf-stat.i.cpi
-54.8% 4.488e+10 perf-stat.i.cpu-cycles
+12.2% 141.01 perf-stat.i.cpu-migrations
68500 ? 5% perf-stat.i.cycles-between-cache-misses
0.17 ? 4% perf-stat.i.dTLB-load-miss-rate%
-25.3% 972044 perf-stat.i.dTLB-load-misses
-54.6% 4.867e+09 perf-stat.i.dTLB-loads
0.04 ? 2% perf-stat.i.dTLB-store-miss-rate%
+9.5% 219785 perf-stat.i.dTLB-store-misses
-55.3% 4.684e+09 perf-stat.i.dTLB-stores
70.51 perf-stat.i.iTLB-load-miss-rate%
-34.9% 544154 ? 3% perf-stat.i.iTLB-load-misses
-22.1% 219530 perf-stat.i.iTLB-loads
-55.3% 4.576e+10 perf-stat.i.instructions
0.85 perf-stat.i.ipc
1.10 ? 5% perf-stat.i.major-faults
0.47 perf-stat.i.metric.GHz
-46.4% 357.69 perf-stat.i.metric.K/sec
-54.7% 138.14 perf-stat.i.metric.M/sec
4935 perf-stat.i.minor-faults
-11.2% 127226 ? 3% perf-stat.i.node-load-misses
46661 ? 4% perf-stat.i.node-loads
67.39 perf-stat.i.node-store-miss-rate%
42369 ? 3% perf-stat.i.node-stores
4936 perf-stat.i.page-faults
0.71 perf-stat.overall.MPKI
19.92 perf-stat.overall.branch-miss-rate%
5.76 ? 2% perf-stat.overall.cache-miss-rate%
0.98 perf-stat.overall.cpi
24012 ? 3% perf-stat.overall.cycles-between-cache-misses
0.02 perf-stat.overall.dTLB-load-miss-rate%
0.00 perf-stat.overall.dTLB-store-miss-rate%
71.17 perf-stat.overall.iTLB-load-miss-rate%
86229 ? 3% perf-stat.overall.instructions-per-iTLB-miss
1.02 perf-stat.overall.ipc
73.36 perf-stat.overall.node-load-miss-rate%
62.86 perf-stat.overall.node-store-miss-rate%
-52.8% 3.722e+09 perf-stat.ps.branch-instructions
-54.0% 7.415e+08 perf-stat.ps.branch-misses
-18.8% 1872089 ? 2% perf-stat.ps.cache-misses
-25.8% 32479193 perf-stat.ps.cache-references
4600 perf-stat.ps.context-switches
+1.6% 94785 perf-stat.ps.cpu-clock
-53.4% 4.491e+10 perf-stat.ps.cpu-cycles
+14.4% 139.72 perf-stat.ps.cpu-migrations
-26.1% 934112 perf-stat.ps.dTLB-load-misses
-53.2% 4.871e+09 perf-stat.ps.dTLB-loads
+11.1% 216552 perf-stat.ps.dTLB-store-misses
-53.9% 4.692e+09 perf-stat.ps.dTLB-stores
-34.4% 532304 ? 3% perf-stat.ps.iTLB-load-misses
-21.3% 215398 perf-stat.ps.iTLB-loads
-53.9% 4.583e+10 perf-stat.ps.instructions
1.05 ? 4% perf-stat.ps.major-faults
4827 perf-stat.ps.minor-faults
-10.6% 124398 ? 3% perf-stat.ps.node-load-misses
45167 ? 4% perf-stat.ps.node-loads
41093 ? 3% perf-stat.ps.node-stores
4828 perf-stat.ps.page-faults
+1.6% 94785 perf-stat.ps.task-clock
+1.6% 3.593e+12 perf-stat.total.instructions
*************************************************************************
threads 2 sockets (Ice Lake) with 256G memory
===============================================================
disk/fs/kconfig/rootfs/tbox_group/test/testcase:
xfs/x86_64-rhel-8.3/debian-11.1-x86_64-20220510.cgz/lkp-icl-2sp1/cvar_example.f/filebench
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
-72.1% 788.78 filebench.sum_bytes_mb/s
-72.1% 24234692 filebench.sum_operations
-72.1% 403880 filebench.sum_operations/s
-72.1% 134626 filebench.sum_reads/s
0.00 filebench.sum_time_ms/op
6919 ? 13% filebench.time.involuntary_context_switches
+2.1% 14680 filebench.time.minor_page_faults
-1.1% 92.00 filebench.time.percent_of_cpu_this_job_got
158.88 ? 40% boot-time.boot
14065 ? 39% boot-time.idle
-1.7% 95.12 iostat.cpu.idle
4.17 iostat.cpu.system
95941 ? 8% numa-meminfo.node1.AnonHugePages
+62.8% 320714 ? 20% numa-meminfo.node1.Inactive
+61.8% 228.60 ? 28% uptime.boot
20437 ? 27% uptime.idle
5562 vmstat.system.cs
-1.8% 92687 vmstat.system.in
3193 meminfo.Active
3185 meminfo.Active(anon)
+45.0% 108373 ? 4% meminfo.AnonHugePages
+28.7% 10738346 ? 19% meminfo.DirectMap2M
3.27 mpstat.cpu.all.irq%
0.17 ? 6% mpstat.cpu.all.soft%
0.76 mpstat.cpu.all.sys%
0.70 mpstat.cpu.all.usr%
5.39 ? 26% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
5.39 ? 26% perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
5.39 ? 26% perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt
5.39 ? 26% perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt
6.24 ? 46% perf-profile.calltrace.cycles-pp.show_interrupts.seq_read_iter.proc_reg_read_iter.vfs_read.ksys_read
7.16 ? 60% perf-profile.calltrace.cycles-pp.proc_reg_read_iter.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
7.16 ? 60% perf-profile.calltrace.cycles-pp.seq_read_iter.proc_reg_read_iter.vfs_read.ksys_read.do_syscall_64
5.39 ? 26% perf-profile.children.cycles-pp.__hrtimer_run_queues
5.39 ? 26% perf-profile.children.cycles-pp.tick_sched_timer
5.39 ? 26% perf-profile.children.cycles-pp.tick_sched_handle
5.39 ? 26% perf-profile.children.cycles-pp.update_process_times
5.83 ? 54% perf-profile.children.cycles-pp.show_interrupts
7.16 ? 60% perf-profile.children.cycles-pp.proc_reg_read_iter
+88.9% 796.33 proc-vmstat.nr_active_anon
+2.3% 18864 proc-vmstat.nr_kernel_stack
+3.4% 12017 proc-vmstat.nr_shmem
-2.0% 54811 proc-vmstat.nr_slab_unreclaimable
+88.9% 796.33 proc-vmstat.nr_zone_active_anon
2710 ? 9% proc-vmstat.numa_hint_faults_local
-54.4% 509544 ? 2% proc-vmstat.numa_hit
-59.0% 422532 ? 2% proc-vmstat.numa_local
-66.1% 623575 ? 3% proc-vmstat.pgalloc_normal
+4.3% 321665 proc-vmstat.pgfault
-74.1% 427598 ? 5% proc-vmstat.pgfree
14531 proc-vmstat.pgreuse
+11.8% 570880 proc-vmstat.unevictable_pgs_scanned
62.33 turbostat.Avg_MHz
6.29 turbostat.Busy%
1000 turbostat.Bzy_MHz
80757 ? 2% turbostat.C1
66640 ? 2% turbostat.C1E
5.30 ? 5% turbostat.CPU%c1
45.33 ? 5% turbostat.CoreTmp
-96.8% 545.33 ? 30% turbostat.POLL
30.63 ? 3% turbostat.Pkg%pc2
1.17 ? 2% turbostat.Pkg%pc6
46.83 ? 4% turbostat.PkgTmp
-41.0% 123.86 turbostat.PkgWatt
-10.2% 145.96 turbostat.RAMWatt
28.98 ? 21% sched_debug.cfs_rq:/.load_avg.avg
-55.7% 120.65 ? 19% sched_debug.cfs_rq:/.load_avg.stddev
12715 ? 19% sched_debug.cfs_rq:/.min_vruntime.avg
9.91 ? 54% sched_debug.cfs_rq:/.removed.load_avg.avg
87.63 ? 46% sched_debug.cfs_rq:/.removed.load_avg.stddev
3.99 ? 73% sched_debug.cfs_rq:/.removed.runnable_avg.avg
-41.3% 310.17 ? 56% sched_debug.cfs_rq:/.removed.runnable_avg.max
34.42 ? 61% sched_debug.cfs_rq:/.removed.runnable_avg.stddev
3.99 ? 73% sched_debug.cfs_rq:/.removed.util_avg.avg
-41.3% 310.17 ? 56% sched_debug.cfs_rq:/.removed.util_avg.max
34.42 ? 61% sched_debug.cfs_rq:/.removed.util_avg.stddev
-29.3% 220.52 ? 4% sched_debug.cfs_rq:/.runnable_avg.avg
1024 sched_debug.cfs_rq:/.runnable_avg.max
-26.1% 249.36 ? 3% sched_debug.cfs_rq:/.runnable_avg.stddev
-29.2% 220.40 ? 4% sched_debug.cfs_rq:/.util_avg.avg
1024 sched_debug.cfs_rq:/.util_avg.max
-25.9% 249.28 ? 3% sched_debug.cfs_rq:/.util_avg.stddev
+57.5% 791.33 sched_debug.cfs_rq:/.util_est_enqueued.max
6880 ? 6% sched_debug.cpu.avg_idle.min
165880 ? 38% sched_debug.cpu.clock.avg
165890 ? 38% sched_debug.cpu.clock.max
165870 ? 38% sched_debug.cpu.clock.min
5.68 ? 5% sched_debug.cpu.clock.stddev
162504 ? 38% sched_debug.cpu.clock_task.avg
165076 ? 38% sched_debug.cpu.clock_task.max
74192 ? 10% sched_debug.cpu.clock_task.min
165870 ? 38% sched_debug.cpu_clk
164658 ? 39% sched_debug.ktime
91223 ? 9% sched_debug.sched_clk
31.43 perf-stat.i.MPKI
-60.9% 6.35e+08 perf-stat.i.branch-instructions
2.47 ? 2% perf-stat.i.branch-miss-rate%
0.68 ? 5% perf-stat.i.cache-miss-rate%
+11.1% 624178 ? 4% perf-stat.i.cache-misses
-5.8% 89223187 perf-stat.i.cache-references
5298 perf-stat.i.context-switches
1.57 perf-stat.i.cpi
-33.0% 4.536e+09 perf-stat.i.cpu-cycles
-2.3% 102.97 perf-stat.i.cpu-migrations
9673 ? 6% perf-stat.i.cycles-between-cache-misses
0.35 perf-stat.i.dTLB-load-miss-rate%
+3.2% 2556155 perf-stat.i.dTLB-load-misses
-60.6% 7.691e+08 perf-stat.i.dTLB-loads
0.06 ? 5% perf-stat.i.dTLB-store-miss-rate%
-63.3% 4.259e+08 perf-stat.i.dTLB-stores
-60.4% 3.08e+09 perf-stat.i.instructions
0.66 perf-stat.i.ipc
0.01 ?141% perf-stat.i.major-faults
0.05 perf-stat.i.metric.GHz
19.08 perf-stat.i.metric.M/sec
+31.6% 141162 ? 5% perf-stat.i.node-load-misses
28.98 perf-stat.overall.MPKI
2.47 ? 2% perf-stat.overall.branch-miss-rate%
0.70 ? 4% perf-stat.overall.cache-miss-rate%
1.47 perf-stat.overall.cpi
7285 ? 4% perf-stat.overall.cycles-between-cache-misses
0.33 ? 2% perf-stat.overall.dTLB-load-miss-rate%
0.06 ? 5% perf-stat.overall.dTLB-store-miss-rate%
0.68 perf-stat.overall.ipc
-60.9% 6.242e+08 perf-stat.ps.branch-instructions
+11.2% 613416 ? 4% perf-stat.ps.cache-misses
-5.9% 87736015 perf-stat.ps.cache-references
5212 perf-stat.ps.context-switches
-33.0% 4.46e+09 perf-stat.ps.cpu-cycles
-2.4% 101.20 perf-stat.ps.cpu-migrations
+3.2% 2513516 perf-stat.ps.dTLB-load-misses
-60.7% 7.56e+08 perf-stat.ps.dTLB-loads
-63.3% 4.188e+08 perf-stat.ps.dTLB-stores
-60.4% 3.028e+09 perf-stat.ps.instructions
0.01 ?141% perf-stat.ps.major-faults
+31.5% 138766 ? 5% perf-stat.ps.node-load-misses
-61.5% 1.85e+11 perf-stat.total.instructions
*************************************************************************
threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
===============================================================
kconfig/option_a/rootfs/tbox_group/test/testcase:
4-rhel-8.3/ffs/debian-x86_64-phoronix/lkp-csl-2sp7/glibc-bench-1.6.1/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
10.31 phoronix-test-suite.glibc-bench.ffs.nanoseconds
68.06 phoronix-test-suite.time.elapsed_time
68.06 phoronix-test-suite.time.elapsed_time.max
1496 ? 2% phoronix-test-suite.time.involuntary_context_switches
69.83 phoronix-test-suite.time.percent_of_cpu_this_job_got
11.36 phoronix-test-suite.time.system_time
36.42 phoronix-test-suite.time.user_time
+17.9% 6.228e+09 cpuidle..time
+18.5% 6564240 cpuidle..usage
9970 meminfo.Active(anon)
-16.3% 439748 ? 11% meminfo.DirectMap4k
+19.2% 136.55 ? 7% uptime.boot
11822 ? 6% uptime.idle
1177 ? 6% vmstat.io.bi
2.00 vmstat.procs.r
4983 vmstat.system.cs
-3.3% 92477 vmstat.system.in
0.05 ? 18% mpstat.cpu.all.iowait%
3.56 mpstat.cpu.all.irq%
0.21 mpstat.cpu.all.soft%
0.40 mpstat.cpu.all.sys%
1.87 mpstat.cpu.all.usr%
-28.9% 110.84 ? 34% sched_debug.cfs_rq:/.load_avg.stddev
24289 ? 9% sched_debug.cfs_rq:/.min_vruntime.avg
57491 ? 4% sched_debug.cfs_rq:/.min_vruntime.max
-18.3% 199.54 ? 8% sched_debug.cfs_rq:/.runnable_avg.stddev
34998 ? 7% sched_debug.cfs_rq:/.spread0.max
-18.3% 199.53 ? 8% sched_debug.cfs_rq:/.util_avg.stddev
8.82 ? 4% sched_debug.cpu.clock.stddev
42534 sched_debug.cpu.clock_task.min
+74.1% 259.73 ? 31% sched_debug.cpu.curr->pid.avg
+29.1% 684.71 ? 13% sched_debug.cpu.curr->pid.stddev
59506 sched_debug.sched_clk
2492 proc-vmstat.nr_active_anon
-6.0% 102231 proc-vmstat.nr_anon_pages
-2.4% 278339 proc-vmstat.nr_inactive_anon
-1.6% 51613 proc-vmstat.nr_slab_unreclaimable
2492 proc-vmstat.nr_zone_active_anon
-2.4% 278339 proc-vmstat.nr_zone_inactive_anon
+9.3% 487223 proc-vmstat.numa_hit
+9.7% 468279 proc-vmstat.numa_local
+9.1% 560135 proc-vmstat.pgalloc_normal
+12.7% 642729 proc-vmstat.pgfault
+9.4% 458954 proc-vmstat.pgfree
83936 proc-vmstat.pgreuse
+21.5% 722944 proc-vmstat.unevictable_pgs_scanned
9.07 turbostat.Busy%
-55.4% 700.00 turbostat.Bzy_MHz
47362 ? 2% turbostat.C1
0.02 turbostat.C1%
153009 ? 9% turbostat.C1E
+18.0% 6343412 turbostat.C6
56.35 turbostat.CPU%c1
34.58 turbostat.CPU%c6
37.50 turbostat.CoreTmp
+18.2% 6565797 turbostat.IRQ
1008 ? 14% turbostat.POLL
5.38 ? 4% turbostat.Pkg%pc2
0.19 ? 3% turbostat.Pkg%pc6
38.00 turbostat.PkgTmp
69.85 turbostat.PkgWatt
+1.4% 143.42 turbostat.RAMWatt
7.12 ? 56% perf-profile.calltrace.cycles-pp.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
7.12 ? 56% perf-profile.calltrace.cycles-pp.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64
7.12 ? 56% perf-profile.calltrace.cycles-pp.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write
7.91 ? 60% perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
7.91 ? 60% perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.79 ?223% perf-profile.calltrace.cycles-pp.do_filp_open.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.79 ?223% perf-profile.calltrace.cycles-pp.path_openat.do_filp_open.do_sys_openat2.__x64_sys_openat.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.39 ?143% perf-profile.calltrace.cycles-pp.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.39 ?143% perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.75 ?159% perf-profile.calltrace.cycles-pp.shmem_write_begin.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write
1.96 ?223% perf-profile.calltrace.cycles-pp.shmem_alloc_and_acct_folio.shmem_get_folio_gfp.shmem_write_begin.generic_perform_write.__generic_file_write_iter
2.75 ?159% perf-profile.calltrace.cycles-pp.shmem_get_folio_gfp.shmem_write_begin.generic_perform_write.__generic_file_write_iter.generic_file_write_iter
7.12 ? 56% perf-profile.children.cycles-pp.generic_file_write_iter
7.12 ? 56% perf-profile.children.cycles-pp.__generic_file_write_iter
7.12 ? 56% perf-profile.children.cycles-pp.generic_perform_write
7.91 ? 60% perf-profile.children.cycles-pp.ksys_write
7.91 ? 60% perf-profile.children.cycles-pp.vfs_write
0.79 ?223% perf-profile.children.cycles-pp.do_filp_open
0.79 ?223% perf-profile.children.cycles-pp.path_openat
0.00 perf-profile.children.cycles-pp.do_vmi_align_munmap
0.00 perf-profile.children.cycles-pp.__do_sys_brk
1.39 ?143% perf-profile.children.cycles-pp.__x64_sys_openat
1.39 ?143% perf-profile.children.cycles-pp.do_sys_openat2
0.79 ?223% perf-profile.children.cycles-pp.__alloc_pages
2.75 ?159% perf-profile.children.cycles-pp.shmem_write_begin
1.96 ?223% perf-profile.children.cycles-pp.shmem_alloc_and_acct_folio
2.75 ?159% perf-profile.children.cycles-pp.shmem_get_folio_gfp
0.79 ?223% perf-profile.self.cycles-pp.fault_in_readable
0.60 ?223% perf-profile.self.cycles-pp.zap_pte_range
+6.2% 37.47 perf-stat.i.MPKI
-49.5% 5.566e+08 perf-stat.i.branch-instructions
5.80 perf-stat.i.branch-miss-rate%
+51.2% 30154000 perf-stat.i.branch-misses
4.25 perf-stat.i.cache-miss-rate%
+57.9% 3651013 perf-stat.i.cache-misses
+63.0% 85336312 perf-stat.i.cache-references
4703 perf-stat.i.context-switches
2.31 perf-stat.i.cpi
-5.3% 5.353e+09 perf-stat.i.cpu-cycles
1536 perf-stat.i.cycles-between-cache-misses
0.48 perf-stat.i.dTLB-load-miss-rate%
+63.8% 2527374 perf-stat.i.dTLB-load-misses
-44.3% 5.669e+08 perf-stat.i.dTLB-loads
0.09 perf-stat.i.dTLB-store-miss-rate%
+54.3% 234914 perf-stat.i.dTLB-store-misses
-51.9% 2.885e+08 perf-stat.i.dTLB-stores
76.32 perf-stat.i.iTLB-load-miss-rate%
-32.8% 901087 perf-stat.i.iTLB-load-misses
-42.8% 278743 ? 2% perf-stat.i.iTLB-loads
-39.9% 2.455e+09 perf-stat.i.instructions
2730 perf-stat.i.instructions-per-iTLB-miss
0.45 perf-stat.i.ipc
1.22 perf-stat.i.major-faults
0.06 perf-stat.i.metric.GHz
+27.4% 879.69 perf-stat.i.metric.K/sec
14.72 perf-stat.i.metric.M/sec
5749 perf-stat.i.minor-faults
76.09 perf-stat.i.node-load-miss-rate%
+25.1% 145514 ? 4% perf-stat.i.node-load-misses
5751 perf-stat.i.page-faults
34.76 perf-stat.overall.MPKI
5.42 perf-stat.overall.branch-miss-rate%
4.28 perf-stat.overall.cache-miss-rate%
2.18 perf-stat.overall.cpi
1466 perf-stat.overall.cycles-between-cache-misses
0.44 perf-stat.overall.dTLB-load-miss-rate%
0.08 perf-stat.overall.dTLB-store-miss-rate%
76.38 perf-stat.overall.iTLB-load-miss-rate%
2725 perf-stat.overall.instructions-per-iTLB-miss
0.46 perf-stat.overall.ipc
-49.4% 5.48e+08 perf-stat.ps.branch-instructions
+51.8% 29688966 perf-stat.ps.branch-misses
+58.4% 3594809 perf-stat.ps.cache-misses
+63.4% 84023854 perf-stat.ps.cache-references
4630 perf-stat.ps.context-switches
-5.0% 5.271e+09 perf-stat.ps.cpu-cycles
+64.2% 2488466 perf-stat.ps.dTLB-load-misses
-44.1% 5.582e+08 perf-stat.ps.dTLB-loads
+54.7% 231314 perf-stat.ps.dTLB-store-misses
-51.7% 2.841e+08 perf-stat.ps.dTLB-stores
-32.6% 887291 perf-stat.ps.iTLB-load-misses
-42.6% 274431 ? 2% perf-stat.ps.iTLB-loads
-39.7% 2.418e+09 perf-stat.ps.instructions
1.20 perf-stat.ps.major-faults
5661 perf-stat.ps.minor-faults
+25.6% 143299 ? 4% perf-stat.ps.node-load-misses
5662 perf-stat.ps.page-faults
-28.0% 1.596e+11 perf-stat.total.instructions
*************************************************************************
threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
===============================================================
kconfig/option_a/rootfs/tbox_group/test/testcase:
4-rhel-8.3/pthread_once/debian-x86_64-phoronix/lkp-csl-2sp7/glibc-bench-1.6.1/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
10.40 phoronix-test-suite.glibc-bench.pthread_once.nanoseconds
68.01 phoronix-test-suite.time.elapsed_time
68.01 phoronix-test-suite.time.elapsed_time.max
1480 ? 2% phoronix-test-suite.time.involuntary_context_switches
69.67 phoronix-test-suite.time.percent_of_cpu_this_job_got
11.24 phoronix-test-suite.time.system_time
36.46 phoronix-test-suite.time.user_time
9980 meminfo.Active(anon)
9356 ? 44% numa-vmstat.node1.nr_slab_reclaimable
+16.4% 6.223e+09 cpuidle..time
+17.4% 6571074 cpuidle..usage
37426 ? 44% numa-meminfo.node1.KReclaimable
37426 ? 44% numa-meminfo.node1.SReclaimable
+91.5% 239.55 ? 50% uptime.boot
20560 ? 49% uptime.idle
167.84 ? 72% boot-time.boot
142.14 ? 85% boot-time.dhcp
14100 ? 72% boot-time.idle
10.52 ? 98% boot-time.smp_boot
1202 ? 5% vmstat.io.bi
2.00 vmstat.procs.r
5008 vmstat.system.cs
-3.3% 92603 vmstat.system.in
0.04 ? 3% mpstat.cpu.all.iowait%
3.55 mpstat.cpu.all.irq%
0.21 mpstat.cpu.all.soft%
0.40 mpstat.cpu.all.sys%
1.89 mpstat.cpu.all.usr%
2495 proc-vmstat.nr_active_anon
-5.6% 102534 proc-vmstat.nr_anon_pages
-2.2% 278915 proc-vmstat.nr_inactive_anon
-1.5% 51635 proc-vmstat.nr_slab_unreclaimable
2495 proc-vmstat.nr_zone_active_anon
-2.2% 278915 proc-vmstat.nr_zone_inactive_anon
+6.8% 490091 proc-vmstat.numa_hit
+7.3% 472049 proc-vmstat.numa_local
+6.8% 565786 proc-vmstat.pgalloc_normal
+9.0% 645487 proc-vmstat.pgfault
+7.1% 462298 proc-vmstat.pgfree
84416 proc-vmstat.pgreuse
+18.5% 722688 proc-vmstat.unevictable_pgs_scanned
9.10 turbostat.Busy%
-55.6% 700.00 turbostat.Bzy_MHz
47384 ? 2% turbostat.C1
162332 ? 3% turbostat.C1E
0.32 turbostat.C1E%
+16.4% 6329781 turbostat.C6
56.16 turbostat.CPU%c1
34.74 turbostat.CPU%c6
37.83 turbostat.CoreTmp
+17.2% 6561866 turbostat.IRQ
-96.9% 792.83 ? 7% turbostat.POLL
6.19 ? 27% turbostat.Pkg%pc2
0.20 ? 2% turbostat.Pkg%pc6
38.50 turbostat.PkgTmp
69.92 turbostat.PkgWatt
+1.4% 143.45 turbostat.RAMWatt
0.13 ? 7% sched_debug.cfs_rq:/.h_nr_running.avg
0.34 ? 3% sched_debug.cfs_rq:/.h_nr_running.stddev
24543 ? 12% sched_debug.cfs_rq:/.min_vruntime.avg
0.13 ? 10% sched_debug.cfs_rq:/.nr_running.avg
0.33 ? 4% sched_debug.cfs_rq:/.nr_running.stddev
6.29 ?160% sched_debug.cfs_rq:/.removed.load_avg.avg
42.53 ?145% sched_debug.cfs_rq:/.removed.load_avg.stddev
2.62 ?190% sched_debug.cfs_rq:/.removed.runnable_avg.avg
-78.2% 112.33 ?157% sched_debug.cfs_rq:/.removed.runnable_avg.max
16.12 ?173% sched_debug.cfs_rq:/.removed.runnable_avg.stddev
2.62 ?190% sched_debug.cfs_rq:/.removed.util_avg.avg
-78.2% 112.17 ?157% sched_debug.cfs_rq:/.removed.util_avg.max
16.11 ?173% sched_debug.cfs_rq:/.removed.util_avg.stddev
-32.6% 138.25 ? 7% sched_debug.cfs_rq:/.runnable_avg.avg
-15.7% 211.41 ? 5% sched_debug.cfs_rq:/.runnable_avg.stddev
2634 ?154% sched_debug.cfs_rq:/.spread0.avg
-32.7% 138.10 ? 7% sched_debug.cfs_rq:/.util_avg.avg
-15.7% 211.37 ? 5% sched_debug.cfs_rq:/.util_avg.stddev
-38.4% 468.33 ? 11% sched_debug.cfs_rq:/.util_est_enqueued.max
+17.3% 1055048 ? 12% sched_debug.cpu.avg_idle.avg
170398 ? 71% sched_debug.cpu.clock.avg
170415 ? 71% sched_debug.cpu.clock.max
170385 ? 71% sched_debug.cpu.clock.min
8.43 ? 9% sched_debug.cpu.clock.stddev
164149 ? 70% sched_debug.cpu.clock_task.avg
168044 ? 71% sched_debug.cpu.clock_task.max
42655 ? 2% sched_debug.cpu.clock_task.min
12494 ? 95% sched_debug.cpu.clock_task.stddev
278.63 ? 10% sched_debug.cpu.curr->pid.avg
+42.3% 715.86 ? 4% sched_debug.cpu.curr->pid.stddev
0.00 ? 29% sched_debug.cpu.next_balance.stddev
0.13 ? 11% sched_debug.cpu.nr_running.avg
0.34 ? 4% sched_debug.cpu.nr_running.stddev
170386 ? 71% sched_debug.cpu_clk
169670 ? 71% sched_debug.ktime
60008 sched_debug.sched_clk
-52.8% 5.957e+08 perf-stat.i.branch-instructions
5.57 perf-stat.i.branch-miss-rate%
+52.5% 30265786 perf-stat.i.branch-misses
4.27 perf-stat.i.cache-miss-rate%
+59.1% 3633877 perf-stat.i.cache-misses
+62.4% 84585057 perf-stat.i.cache-references
4705 perf-stat.i.context-switches
2.25 perf-stat.i.cpi
-4.1% 5.374e+09 perf-stat.i.cpu-cycles
-2.3% 99.54 perf-stat.i.cpu-migrations
1545 perf-stat.i.cycles-between-cache-misses
0.43 perf-stat.i.dTLB-load-miss-rate%
+65.2% 2549066 ? 2% perf-stat.i.dTLB-load-misses
-55.7% 6.799e+08 perf-stat.i.dTLB-loads
0.10 ? 3% perf-stat.i.dTLB-store-miss-rate%
+59.7% 243190 ? 3% perf-stat.i.dTLB-store-misses
-39.1% 2.517e+08 perf-stat.i.dTLB-stores
76.38 perf-stat.i.iTLB-load-miss-rate%
-32.7% 884328 perf-stat.i.iTLB-load-misses
-44.6% 273276 ? 2% perf-stat.i.iTLB-loads
-42.3% 2.542e+09 perf-stat.i.instructions
2873 perf-stat.i.instructions-per-iTLB-miss
0.47 perf-stat.i.ipc
1.22 perf-stat.i.major-faults
0.06 perf-stat.i.metric.GHz
+26.2% 880.12 perf-stat.i.metric.K/sec
15.91 perf-stat.i.metric.M/sec
5771 perf-stat.i.minor-faults
75.78 perf-stat.i.node-load-miss-rate%
+24.2% 145903 ? 2% perf-stat.i.node-load-misses
59761 ? 6% perf-stat.i.node-loads
5772 perf-stat.i.page-faults
33.29 perf-stat.overall.MPKI
5.08 perf-stat.overall.branch-miss-rate%
2.11 perf-stat.overall.cpi
1479 perf-stat.overall.cycles-between-cache-misses
0.37 perf-stat.overall.dTLB-load-miss-rate%
0.10 ? 3% perf-stat.overall.dTLB-store-miss-rate%
76.40 perf-stat.overall.iTLB-load-miss-rate%
2874 perf-stat.overall.instructions-per-iTLB-miss
0.47 perf-stat.overall.ipc
-52.7% 5.865e+08 perf-stat.ps.branch-instructions
+52.9% 29789495 perf-stat.ps.branch-misses
+59.5% 3577319 perf-stat.ps.cache-misses
+62.8% 83278509 perf-stat.ps.cache-references
4632 perf-stat.ps.context-switches
-3.9% 5.291e+09 perf-stat.ps.cpu-cycles
-2.1% 98.01 perf-stat.ps.cpu-migrations
+65.6% 2509672 ? 2% perf-stat.ps.dTLB-load-misses
-55.6% 6.693e+08 perf-stat.ps.dTLB-loads
+60.1% 239445 ? 3% perf-stat.ps.dTLB-store-misses
-39.0% 2.478e+08 perf-stat.ps.dTLB-stores
-32.5% 870773 perf-stat.ps.iTLB-load-misses
-44.4% 269019 ? 2% perf-stat.ps.iTLB-loads
-42.1% 2.502e+09 perf-stat.ps.instructions
1.20 perf-stat.ps.major-faults
5681 perf-stat.ps.minor-faults
+24.5% 143643 ? 2% perf-stat.ps.node-load-misses
58833 ? 6% perf-stat.ps.node-loads
5683 perf-stat.ps.page-faults
-32.0% 1.646e+11 perf-stat.total.instructions
3.08 ? 72% perf-profile.calltrace.cycles-pp.exit_mmap.__mmput.exit_mm.do_exit.do_group_exit
2.16 ?100% perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.__mmput.exit_mm.do_exit
2.16 ?100% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.exit_mmap.__mmput.exit_mm
1.59 ?144% perf-profile.calltrace.cycles-pp.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
0.70 ?223% perf-profile.calltrace.cycles-pp.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.70 ?223% perf-profile.calltrace.cycles-pp.do_vmi_align_munmap.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.70 ?223% perf-profile.calltrace.cycles-pp.unmap_region.do_vmi_align_munmap.__do_sys_brk.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.67 ?223% perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.67 ?223% perf-profile.calltrace.cycles-pp.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.67 ?223% perf-profile.calltrace.cycles-pp.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.fault_in_readable
1.59 ?144% perf-profile.calltrace.cycles-pp.handle_mm_fault.do_user_addr_fault.exc_page_fault.asm_exc_page_fault
0.00 perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault.exc_page_fault
0.00 perf-profile.calltrace.cycles-pp.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault.do_user_addr_fault
0.00 perf-profile.calltrace.cycles-pp.__do_fault.do_read_fault.do_fault.__handle_mm_fault.handle_mm_fault
0.79 ?223% perf-profile.calltrace.cycles-pp.exc_page_fault.asm_exc_page_fault.fault_in_readable.fault_in_iov_iter_readable.generic_perform_write
0.79 ?223% perf-profile.calltrace.cycles-pp.do_user_addr_fault.exc_page_fault.asm_exc_page_fault.fault_in_readable.fault_in_iov_iter_readable
0.00 perf-profile.calltrace.cycles-pp.exit_mm.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.__mmput.exit_mm.do_exit.do_group_exit.__x64_sys_exit_group
0.00 perf-profile.calltrace.cycles-pp.page_remove_rmap.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas
0.67 ?223% perf-profile.calltrace.cycles-pp.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64
0.67 ?223% perf-profile.calltrace.cycles-pp.search_binary_handler.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve
1.46 ?142% perf-profile.calltrace.cycles-pp.rep_movs_alternative.copyin.copy_page_from_iter_atomic.generic_perform_write.__generic_file_write_iter
1.46 ?142% perf-profile.calltrace.cycles-pp.copyin.copy_page_from_iter_atomic.generic_perform_write.__generic_file_write_iter.generic_file_write_iter
8.42 ? 42% perf-profile.calltrace.cycles-pp.proc_reg_read_iter.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
8.42 ? 42% perf-profile.calltrace.cycles-pp.seq_read_iter.proc_reg_read_iter.vfs_read.ksys_read.do_syscall_64
10.30 ? 57% perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
10.30 ? 57% perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.16 ?100% perf-profile.children.cycles-pp.unmap_vmas
3.08 ? 72% perf-profile.children.cycles-pp.__mmput
3.08 ? 72% perf-profile.children.cycles-pp.exit_mmap
2.16 ?100% perf-profile.children.cycles-pp.unmap_page_range
3.08 ? 72% perf-profile.children.cycles-pp.exit_mm
1.59 ?144% perf-profile.children.cycles-pp.handle_mm_fault
3.31 ? 72% perf-profile.children.cycles-pp.exc_page_fault
3.31 ? 72% perf-profile.children.cycles-pp.do_user_addr_fault
1.59 ?144% perf-profile.children.cycles-pp.__handle_mm_fault
0.70 ?223% perf-profile.children.cycles-pp.__do_sys_brk
0.70 ?223% perf-profile.children.cycles-pp.do_vmi_align_munmap
0.70 ?223% perf-profile.children.cycles-pp.unmap_region
0.67 ?223% perf-profile.children.cycles-pp.__x64_sys_execve
0.67 ?223% perf-profile.children.cycles-pp.do_execveat_common
0.67 ?223% perf-profile.children.cycles-pp.bprm_execve
0.00 perf-profile.children.cycles-pp.__x64_sys_exit_group
0.00 perf-profile.children.cycles-pp.do_fault
0.00 perf-profile.children.cycles-pp.do_read_fault
0.00 perf-profile.children.cycles-pp.__do_fault
0.00 perf-profile.children.cycles-pp.page_remove_rmap
2.18 ?149% perf-profile.children.cycles-pp.__alloc_pages
0.67 ?223% perf-profile.children.cycles-pp.exec_binprm
0.67 ?223% perf-profile.children.cycles-pp.search_binary_handler
1.46 ?142% perf-profile.children.cycles-pp.rep_movs_alternative
1.46 ?142% perf-profile.children.cycles-pp.copyin
8.42 ? 42% perf-profile.children.cycles-pp.proc_reg_read_iter
8.42 ? 42% perf-profile.children.cycles-pp.seq_read_iter
10.30 ? 57% perf-profile.children.cycles-pp.ksys_read
10.30 ? 57% perf-profile.children.cycles-pp.vfs_read
0.70 ?223% perf-profile.self.cycles-pp.zap_pte_range
1.46 ?142% perf-profile.self.cycles-pp.rep_movs_alternative
*************************************************************************
threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz (Cascade Lake) with 512G memory
===============================================================
kconfig/option_a/rootfs/tbox_group/test/testcase:
4-rhel-8.3/3/debian-x86_64-phoronix/lkp-csl-2sp7/cyclictest-1.0.0/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
56.50 ? 90% phoronix-test-suite.cyclictest.clock_nanosleepTIME_ABSTIME,Interval500ms,100000Loops.ms_average
75.83 phoronix-test-suite.time.elapsed_time
75.83 phoronix-test-suite.time.elapsed_time.max
1500 ? 3% phoronix-test-suite.time.involuntary_context_switches
18.50 ? 2% phoronix-test-suite.time.percent_of_cpu_this_job_got
-28.7% 1.588e+08 cpuidle..usage
10129 meminfo.Active(anon)
1069 ? 3% vmstat.io.bi
1.33 ? 35% vmstat.procs.r
7576 vmstat.system.cs
-3.0% 94332 vmstat.system.in
0.04 ? 5% mpstat.cpu.all.iowait%
2.47 mpstat.cpu.all.irq%
0.16 ? 2% mpstat.cpu.all.soft%
0.34 ? 2% mpstat.cpu.all.sys%
1.36 mpstat.cpu.all.usr%
2531 proc-vmstat.nr_active_anon
-4.6% 103460 proc-vmstat.nr_anon_pages
-1.8% 279599 proc-vmstat.nr_inactive_anon
-1.2% 51832 proc-vmstat.nr_slab_unreclaimable
2531 proc-vmstat.nr_zone_active_anon
-1.8% 279599 proc-vmstat.nr_zone_inactive_anon
+6.5% 505695 proc-vmstat.numa_hit
+6.5% 486800 proc-vmstat.numa_local
+4.8% 578226 proc-vmstat.pgalloc_normal
+9.0% 671086 proc-vmstat.pgfault
+4.3% 476256 proc-vmstat.pgfree
88541 ? 2% proc-vmstat.pgreuse
+12.5% 750208 proc-vmstat.unevictable_pgs_scanned
-77.0% 467.83 turbostat.Avg_MHz
-74.8% 700.00 turbostat.Bzy_MHz
25182 ? 8% turbostat.C1
0.01 turbostat.C1%
91233 ? 7% turbostat.C1E
+31.5% 2364347 turbostat.C6
32.32 turbostat.C6%
13.30 turbostat.CPU%c6
39.83 turbostat.CoreTmp
-29.3% 1.563e+08 turbostat.POLL
60.97 turbostat.POLL%
2.12 ? 7% turbostat.Pkg%pc2
0.18 ? 6% turbostat.Pkg%pc6
40.17 turbostat.PkgTmp
81.80 turbostat.PkgWatt
5020 ? 11% sched_debug.cfs_rq:/.load.avg
14243 ? 5% sched_debug.cfs_rq:/.load.stddev
25.56 ? 30% sched_debug.cfs_rq:/.load_avg.avg
-50.5% 104.97 ? 34% sched_debug.cfs_rq:/.load_avg.stddev
58197 ? 2% sched_debug.cfs_rq:/.min_vruntime.max
5231 ? 3% sched_debug.cfs_rq:/.min_vruntime.stddev
10.16 ? 58% sched_debug.cfs_rq:/.removed.load_avg.avg
89.17 ? 48% sched_debug.cfs_rq:/.removed.load_avg.stddev
5.03 ? 58% sched_debug.cfs_rq:/.removed.runnable_avg.avg
44.09 ? 48% sched_debug.cfs_rq:/.removed.runnable_avg.stddev
5.02 ? 58% sched_debug.cfs_rq:/.removed.util_avg.avg
44.07 ? 48% sched_debug.cfs_rq:/.removed.util_avg.stddev
-32.3% 135.14 ? 10% sched_debug.cfs_rq:/.runnable_avg.avg
-16.2% 206.81 ? 7% sched_debug.cfs_rq:/.runnable_avg.stddev
5239 ? 3% sched_debug.cfs_rq:/.spread0.stddev
-32.4% 135.03 ? 10% sched_debug.cfs_rq:/.util_avg.avg
-16.2% 206.77 ? 7% sched_debug.cfs_rq:/.util_avg.stddev
-45.7% 441.00 ? 17% sched_debug.cfs_rq:/.util_est_enqueued.max
69.51 ? 16% sched_debug.cfs_rq:/.util_est_enqueued.stddev
11264 ? 12% sched_debug.cpu.avg_idle.min
8.96 ? 17% sched_debug.cpu.clock.stddev
42576 ? 2% sched_debug.cpu.clock_task.min
60013 sched_debug.sched_clk
0.00 perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.search_binary_handler.exec_binprm.bprm_execve.do_execveat_common.__x64_sys_execve
0.00 perf-profile.calltrace.cycles-pp.load_elf_binary.search_binary_handler.exec_binprm.bprm_execve.do_execveat_common
1.55 ?141% perf-profile.calltrace.cycles-pp.page_remove_rmap.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas
5.63 ? 28% perf-profile.calltrace.cycles-pp.__fput.task_work_run.do_exit.do_group_exit.get_signal
5.63 ? 28% perf-profile.calltrace.cycles-pp.task_work_run.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart
5.63 ? 28% perf-profile.calltrace.cycles-pp.perf_release.__fput.task_work_run.do_exit.do_group_exit
5.63 ? 28% perf-profile.calltrace.cycles-pp.perf_event_release_kernel.perf_release.__fput.task_work_run.do_exit
10.50 ? 15% perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
10.50 ? 15% perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
10.50 ? 15% perf-profile.calltrace.cycles-pp.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64
10.50 ? 15% perf-profile.calltrace.cycles-pp.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode
10.50 ? 15% perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare
10.50 ? 15% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop
13.60 ? 27% perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
44.92 ? 8% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
44.92 ? 8% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
45.84 ? 5% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
47.65 ? 3% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
47.65 ? 3% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify
47.65 ? 3% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
47.65 ? 3% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
0.00 perf-profile.children.cycles-pp.inode_permission
0.00 perf-profile.children.cycles-pp.__x64_sys_execve
0.00 perf-profile.children.cycles-pp.do_execveat_common
0.00 perf-profile.children.cycles-pp.bprm_execve
0.00 perf-profile.children.cycles-pp.exec_binprm
0.00 perf-profile.children.cycles-pp.search_binary_handler
0.00 perf-profile.children.cycles-pp.override_creds
0.00 perf-profile.children.cycles-pp.load_elf_binary
1.55 ?141% perf-profile.children.cycles-pp.page_remove_rmap
5.63 ? 28% perf-profile.children.cycles-pp.perf_release
5.63 ? 28% perf-profile.children.cycles-pp.perf_event_release_kernel
10.50 ? 15% perf-profile.children.cycles-pp.arch_do_signal_or_restart
10.50 ? 15% perf-profile.children.cycles-pp.get_signal
11.42 ? 24% perf-profile.children.cycles-pp.exit_to_user_mode_prepare
11.42 ? 24% perf-profile.children.cycles-pp.exit_to_user_mode_loop
13.60 ? 27% perf-profile.children.cycles-pp.syscall_exit_to_user_mode
44.92 ? 8% perf-profile.children.cycles-pp.cpuidle_enter
44.92 ? 8% perf-profile.children.cycles-pp.cpuidle_enter_state
45.84 ? 5% perf-profile.children.cycles-pp.cpuidle_idle_call
47.65 ? 3% perf-profile.children.cycles-pp.secondary_startup_64_no_verify
47.65 ? 3% perf-profile.children.cycles-pp.cpu_startup_entry
47.65 ? 3% perf-profile.children.cycles-pp.do_idle
47.65 ? 3% perf-profile.children.cycles-pp.start_secondary
0.76 ?223% perf-profile.self.cycles-pp.fault_in_readable
0.00 perf-profile.self.cycles-pp.override_creds
1.52 ?223% perf-profile.self.cycles-pp.show_interrupts
-71.5% 2.61e+09 perf-stat.i.branch-instructions
2.48 perf-stat.i.branch-miss-rate%
-28.1% 26816122 perf-stat.i.branch-misses
8.20 ? 3% perf-stat.i.cache-miss-rate%
+56.3% 1826809 perf-stat.i.cache-misses
+74.1% 31696003 perf-stat.i.cache-references
7477 perf-stat.i.context-switches
4.37 perf-stat.i.cpi
-76.6% 4.725e+10 perf-stat.i.cpu-cycles
89226 ? 3% perf-stat.i.cycles-between-cache-misses
+85.2% 857410 ? 2% perf-stat.i.dTLB-load-misses
-64.5% 1.966e+09 perf-stat.i.dTLB-loads
92464 ? 3% perf-stat.i.dTLB-store-misses
-29.3% 5.971e+08 perf-stat.i.dTLB-stores
82.29 perf-stat.i.iTLB-load-miss-rate%
+37.7% 208592 ? 2% perf-stat.i.iTLB-loads
-68.2% 9.452e+09 perf-stat.i.instructions
9882 ? 2% perf-stat.i.instructions-per-iTLB-miss
0.28 perf-stat.i.ipc
1.20 ? 4% perf-stat.i.major-faults
0.49 perf-stat.i.metric.GHz
+37.5% 334.64 perf-stat.i.metric.K/sec
53.88 perf-stat.i.metric.M/sec
80.11 perf-stat.i.node-load-miss-rate%
+32.3% 140953 ? 7% perf-stat.i.node-load-misses
51045 ? 2% perf-stat.i.node-loads
69365 ? 8% perf-stat.i.node-store-misses
3.37 ? 2% perf-stat.overall.MPKI
1.03 perf-stat.overall.branch-miss-rate%
5.76 ? 2% perf-stat.overall.cache-miss-rate%
5.00 perf-stat.overall.cpi
25771 perf-stat.overall.cycles-between-cache-misses
0.04 perf-stat.overall.dTLB-load-miss-rate%
0.02 ? 2% perf-stat.overall.dTLB-store-miss-rate%
82.02 perf-stat.overall.iTLB-load-miss-rate%
9918 perf-stat.overall.instructions-per-iTLB-miss
0.20 perf-stat.overall.ipc
62.77 ? 4% perf-stat.overall.node-store-miss-rate%
-71.5% 2.571e+09 perf-stat.ps.branch-instructions
-27.9% 26457021 perf-stat.ps.branch-misses
+56.6% 1806281 perf-stat.ps.cache-misses
+74.3% 31357377 perf-stat.ps.cache-references
7371 perf-stat.ps.context-switches
-76.6% 4.653e+10 perf-stat.ps.cpu-cycles
+85.4% 848579 ? 2% perf-stat.ps.dTLB-load-misses
-64.5% 1.937e+09 perf-stat.ps.dTLB-loads
91486 ? 3% perf-stat.ps.dTLB-store-misses
-29.2% 5.884e+08 perf-stat.ps.dTLB-stores
+37.7% 205884 ? 2% perf-stat.ps.iTLB-loads
-68.2% 9.31e+09 perf-stat.ps.instructions
1.19 ? 4% perf-stat.ps.major-faults
+32.5% 139219 ? 7% perf-stat.ps.node-load-misses
50497 ? 2% perf-stat.ps.node-loads
68472 ? 8% perf-stat.ps.node-store-misses
-65.3% 6.875e+11 perf-stat.total.instructions
*************************************************************************
1 sockets Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz (Coffee Lake) with 32G memory
===============================================================
kconfig/option_a/rootfs/tbox_group/test/testcase:
4-rhel-8.3/Medium/debian-x86_64-phoronix/lkp-cfl-e1/astcenc-1.1.0/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
13.45 phoronix-test-suite.astcenc.Medium.seconds
132.82 phoronix-test-suite.time.elapsed_time
132.82 phoronix-test-suite.time.elapsed_time.max
19360 ? 6% phoronix-test-suite.time.involuntary_context_switches
+56.9% 371.33 phoronix-test-suite.time.percent_of_cpu_this_job_got
489.06 phoronix-test-suite.time.user_time
+161.8% 1.497e+09 cpuidle..time
+172.7% 1661621 cpuidle..usage
7902 meminfo.Active(anon)
+73.9% 344256 ? 2% meminfo.AnonHugePages
+19.9% 562889 meminfo.AnonPages
70.05 mpstat.cpu.all.idle%
0.14 ? 4% mpstat.cpu.all.iowait%
2.94 mpstat.cpu.all.irq%
0.16 ? 3% mpstat.cpu.all.soft%
0.70 mpstat.cpu.all.sys%
26.00 mpstat.cpu.all.usr%
70.00 vmstat.cpu.id
25.00 vmstat.cpu.us
-70.2% 769.17 ? 2% vmstat.io.bi
4.00 vmstat.procs.r
1984 vmstat.system.cs
+4.2% 16461 vmstat.system.in
1976 proc-vmstat.nr_active_anon
+6.8% 30122 proc-vmstat.nr_active_file
+19.9% 140853 proc-vmstat.nr_anon_pages
+74.2% 167.83 ? 2% proc-vmstat.nr_anon_transparent_hugepages
+8.1% 320871 proc-vmstat.nr_inactive_anon
6764 proc-vmstat.nr_kernel_stack
+1.2% 18332 proc-vmstat.nr_mapped
1460 proc-vmstat.nr_page_table_pages
-1.9% 15472 proc-vmstat.nr_slab_unreclaimable
1976 proc-vmstat.nr_zone_active_anon
+6.8% 30122 proc-vmstat.nr_zone_active_file
+8.1% 320871 proc-vmstat.nr_zone_inactive_anon
+40.6% 408922 proc-vmstat.numa_hit
+40.7% 409089 proc-vmstat.numa_local
+4.7% 19409 proc-vmstat.pgactivate
+17.9% 976790 proc-vmstat.pgalloc_normal
+56.3% 528417 proc-vmstat.pgfault
+19.0% 917408 proc-vmstat.pgfree
65408 proc-vmstat.pgreuse
+150.6% 1167488 proc-vmstat.unevictable_pgs_scanned
94.59 turbostat.Any%C0
-63.6% 247.50 turbostat.Avg_MHz
30.92 turbostat.Busy%
-80.7% 800.00 turbostat.Bzy_MHz
35467 ? 4% turbostat.C1
0.02 ? 20% turbostat.C1%
0.04 ? 23% turbostat.C10%
31828 ? 2% turbostat.C1E
0.13 ? 3% turbostat.C1E%
0.13 ? 2% turbostat.C3%
743779 turbostat.C6
32.04 turbostat.C6%
835400 turbostat.C8
12.89 turbostat.CPU%c1
0.18 ? 4% turbostat.CPU%c3
16.88 turbostat.CPU%c7
2.69 turbostat.CorWatt
30.33 turbostat.CoreTmp
+220.7% 2236189 turbostat.IRQ
1320 ? 14% turbostat.POLL
0.54 ? 13% turbostat.Pkg%pc2
4.51 turbostat.Pkg%pc3
31.17 turbostat.PkgTmp
3.17 turbostat.PkgWatt
+92.8% 320.39 turbostat.Totl%C0
69438 ? 21% sched_debug.cfs_rq:/.load.avg
367447 ? 47% sched_debug.cfs_rq:/.load.max
117549 ? 33% sched_debug.cfs_rq:/.load.stddev
98.31 ? 17% sched_debug.cfs_rq:/.load_avg.avg
-52.4% 575.11 ? 40% sched_debug.cfs_rq:/.load_avg.max
4.25 ? 67% sched_debug.cfs_rq:/.load_avg.min
-48.4% 165.60 ? 34% sched_debug.cfs_rq:/.load_avg.stddev
161504 ? 38% sched_debug.cfs_rq:/.min_vruntime.avg
196783 ? 35% sched_debug.cfs_rq:/.min_vruntime.max
145426 ? 40% sched_debug.cfs_rq:/.min_vruntime.min
13357 ? 29% sched_debug.cfs_rq:/.min_vruntime.stddev
0.42 ? 8% sched_debug.cfs_rq:/.nr_running.stddev
-53.0% 345.89 ? 11% sched_debug.cfs_rq:/.runnable_avg.avg
60.78 ? 31% sched_debug.cfs_rq:/.runnable_avg.min
43070 ? 53% sched_debug.cfs_rq:/.spread0.max
13358 ? 29% sched_debug.cfs_rq:/.spread0.stddev
-52.8% 333.26 ? 11% sched_debug.cfs_rq:/.util_avg.avg
51.31 ? 29% sched_debug.cfs_rq:/.util_est_enqueued.avg
-57.0% 340.19 ? 44% sched_debug.cfs_rq:/.util_est_enqueued.max
96.73 ? 35% sched_debug.cfs_rq:/.util_est_enqueued.stddev
+47.6% 813196 ? 9% sched_debug.cpu.avg_idle.avg
83557 ? 17% sched_debug.cpu.clock_task.min
+27.4% 617.87 ? 9% sched_debug.cpu.curr->pid.avg
2311 ? 12% sched_debug.cpu.curr->pid.max
+54.7% 814.24 ? 12% sched_debug.cpu.curr->pid.stddev
10611 ? 17% sched_debug.cpu.nr_switches.avg
20797 ? 13% sched_debug.cpu.nr_switches.max
3959 ? 27% sched_debug.cpu.nr_switches.min
4792 ? 22% sched_debug.cpu.nr_switches.stddev
94104 ? 16% sched_debug.sched_clk
38.49 perf-stat.i.MPKI
-64.6% 5.725e+08 perf-stat.i.branch-instructions
4.59 perf-stat.i.branch-miss-rate%
-61.2% 25072928 perf-stat.i.branch-misses
6.47 perf-stat.i.cache-miss-rate%
-46.7% 3654271 perf-stat.i.cache-misses
-63.0% 1.691e+08 perf-stat.i.cache-references
1899 perf-stat.i.context-switches
0.96 perf-stat.i.cpi
-65.4% 3.82e+09 perf-stat.i.cpu-cycles
32.81 perf-stat.i.cpu-migrations
1675 perf-stat.i.cycles-between-cache-misses
0.15 ? 8% perf-stat.i.dTLB-load-miss-rate%
-37.4% 503607 ? 8% perf-stat.i.dTLB-load-misses
-67.1% 1.236e+09 perf-stat.i.dTLB-loads
0.05 ? 13% perf-stat.i.dTLB-store-miss-rate%
55137 ? 9% perf-stat.i.dTLB-store-misses
-65.9% 3.495e+08 perf-stat.i.dTLB-stores
69.74 perf-stat.i.iTLB-load-miss-rate%
-15.4% 161312 perf-stat.i.iTLB-load-misses
-66.3% 4.476e+09 perf-stat.i.instructions
35109 ? 5% perf-stat.i.instructions-per-iTLB-miss
1.19 perf-stat.i.ipc
1.27 perf-stat.i.major-faults
0.00 perf-stat.i.metric.G/sec
0.24 perf-stat.i.metric.GHz
39.59 ? 2% perf-stat.i.metric.K/sec
-48.8% 145.40 perf-stat.i.metric.M/sec
3513 perf-stat.i.minor-faults
0.46 ? 33% perf-stat.i.node-load-misses
97050 perf-stat.i.node-loads
0.39 ? 38% perf-stat.i.node-store-misses
-64.6% 409216 ? 3% perf-stat.i.node-stores
3514 perf-stat.i.page-faults
+9.6% 37.74 perf-stat.overall.MPKI
4.37 perf-stat.overall.branch-miss-rate%
2.17 perf-stat.overall.cache-miss-rate%
0.85 perf-stat.overall.cpi
1043 perf-stat.overall.cycles-between-cache-misses
0.04 ? 9% perf-stat.overall.dTLB-load-miss-rate%
0.02 ? 9% perf-stat.overall.dTLB-store-miss-rate%
69.67 perf-stat.overall.iTLB-load-miss-rate%
27693 perf-stat.overall.instructions-per-iTLB-miss
1.17 perf-stat.overall.ipc
-64.1% 5.665e+08 perf-stat.ps.branch-instructions
-60.7% 24776524 perf-stat.ps.branch-misses
-45.9% 3620492 perf-stat.ps.cache-misses
-62.6% 1.671e+08 perf-stat.ps.cache-references
1881 perf-stat.ps.context-switches
+1.7% 15880 perf-stat.ps.cpu-clock
-64.9% 3.778e+09 perf-stat.ps.cpu-cycles
32.52 perf-stat.ps.cpu-migrations
-36.4% 499506 ? 8% perf-stat.ps.dTLB-load-misses
-66.6% 1.223e+09 perf-stat.ps.dTLB-loads
54670 ? 9% perf-stat.ps.dTLB-store-misses
-65.4% 3.463e+08 perf-stat.ps.dTLB-stores
-14.1% 159974 perf-stat.ps.iTLB-load-misses
-65.9% 4.429e+09 perf-stat.ps.instructions
1.25 perf-stat.ps.major-faults
3478 perf-stat.ps.minor-faults
0.45 ? 33% perf-stat.ps.node-load-misses
96125 perf-stat.ps.node-loads
0.39 ? 38% perf-stat.ps.node-store-misses
-64.0% 405385 ? 4% perf-stat.ps.node-stores
3479 perf-stat.ps.page-faults
+1.7% 15880 perf-stat.ps.task-clock
+8.0% 5.918e+11 perf-stat.total.instructions
6.60 ? 24% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.70 ? 25% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
0.23 ?141% perf-profile.calltrace.cycles-pp.exit_mmap.__mmput.exit_mm.do_exit.do_group_exit
0.45 ? 72% perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.24 ?142% perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.24 ?142% perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.24 ?142% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap.__mmput
0.00 perf-profile.calltrace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap
0.00 perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare
0.00 perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop
0.00 perf-profile.calltrace.cycles-pp.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode
0.00 perf-profile.calltrace.cycles-pp.exit_mm.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart
0.00 perf-profile.calltrace.cycles-pp.__mmput.exit_mm.do_exit.do_group_exit.get_signal
0.00 perf-profile.calltrace.cycles-pp.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write
0.00 perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.__mmput.exit_mm.do_exit
0.00 perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.exit_mmap.__mmput.exit_mm
0.23 ?141% perf-profile.calltrace.cycles-pp.__mmput.exit_mm.do_exit.do_group_exit.__x64_sys_exit_group
0.23 ?141% perf-profile.calltrace.cycles-pp.exit_mm.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.page_remove_rmap.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas
0.00 perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.__mmput.exec_mmap.begin_new_exec
0.20 ?142% perf-profile.calltrace.cycles-pp.__mmput.exec_mmap.begin_new_exec.load_elf_binary.search_binary_handler
0.20 ?142% perf-profile.calltrace.cycles-pp.exit_mmap.__mmput.exec_mmap.begin_new_exec.load_elf_binary
0.83 ? 34% perf-profile.calltrace.cycles-pp.lapic_next_deadline.clockevents_program_event.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
0.97 ? 30% perf-profile.calltrace.cycles-pp.tick_nohz_get_sleep_length.menu_select.cpuidle_idle_call.do_idle.cpu_startup_entry
0.99 ? 20% perf-profile.calltrace.cycles-pp.update_rq_clock_task.scheduler_tick.update_process_times.tick_sched_handle.tick_sched_timer
1.10 ? 26% perf-profile.calltrace.cycles-pp.account_user_time.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
1.17 ? 34% perf-profile.calltrace.cycles-pp.update_curr.task_tick_fair.scheduler_tick.update_process_times.tick_sched_handle
1.26 ? 35% perf-profile.calltrace.cycles-pp.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
1.28 ? 36% perf-profile.calltrace.cycles-pp.clockevents_program_event.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
1.34 ? 35% perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
1.41 ? 32% perf-profile.calltrace.cycles-pp.perf_rotate_context.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt
1.45 ? 38% perf-profile.calltrace.cycles-pp.sync_regs.asm_sysvec_apic_timer_interrupt
1.52 ? 48% perf-profile.calltrace.cycles-pp.drm_fb_helper_damage_work.process_one_work.worker_thread.kthread.ret_from_fork
1.52 ? 48% perf-profile.calltrace.cycles-pp.drm_fbdev_generic_helper_fb_dirty.drm_fb_helper_damage_work.process_one_work.worker_thread.kthread
1.60 ? 43% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork
1.63 ? 25% perf-profile.calltrace.cycles-pp.update_load_avg.task_tick_fair.scheduler_tick.update_process_times.tick_sched_handle
1.65 ? 41% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork
1.86 ? 26% perf-profile.calltrace.cycles-pp.asm_sysvec_irq_work
1.86 ? 26% perf-profile.calltrace.cycles-pp.sysvec_irq_work.asm_sysvec_irq_work
1.86 ? 26% perf-profile.calltrace.cycles-pp.__sysvec_irq_work.sysvec_irq_work.asm_sysvec_irq_work
1.86 ? 26% perf-profile.calltrace.cycles-pp.irq_work_run.__sysvec_irq_work.sysvec_irq_work.asm_sysvec_irq_work
2.21 ? 16% perf-profile.calltrace.cycles-pp.asm_sysvec_irq_work.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_irq_work
2.21 ? 16% perf-profile.calltrace.cycles-pp.sysvec_irq_work.asm_sysvec_irq_work.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode
2.21 ? 16% perf-profile.calltrace.cycles-pp.__sysvec_irq_work.sysvec_irq_work.asm_sysvec_irq_work.exit_to_user_mode_loop.exit_to_user_mode_prepare
2.21 ? 16% perf-profile.calltrace.cycles-pp.irq_work_run.__sysvec_irq_work.sysvec_irq_work.asm_sysvec_irq_work.exit_to_user_mode_loop
2.21 ? 16% perf-profile.calltrace.cycles-pp.irqentry_exit_to_user_mode.asm_sysvec_irq_work
2.21 ? 16% perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_irq_work
2.21 ? 16% perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.irqentry_exit_to_user_mode.asm_sysvec_irq_work
2.60 ? 17% perf-profile.calltrace.cycles-pp.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
2.96 ? 19% perf-profile.calltrace.cycles-pp.io_serial_in.wait_for_lsr.serial8250_console_write.console_flush_all.console_unlock
3.62 ? 19% perf-profile.calltrace.cycles-pp.wait_for_lsr.serial8250_console_write.console_flush_all.console_unlock.vprintk_emit
4.19 ? 20% perf-profile.calltrace.cycles-pp.serial8250_console_write.console_flush_all.console_unlock.vprintk_emit._printk
4.76 ? 20% perf-profile.calltrace.cycles-pp.irq_work_run_list.irq_work_run.__sysvec_irq_work.sysvec_irq_work.asm_sysvec_irq_work
4.76 ? 20% perf-profile.calltrace.cycles-pp.irq_work_single.irq_work_run_list.irq_work_run.__sysvec_irq_work.sysvec_irq_work
4.76 ? 20% perf-profile.calltrace.cycles-pp._printk.irq_work_single.irq_work_run_list.irq_work_run.__sysvec_irq_work
4.76 ? 20% perf-profile.calltrace.cycles-pp.vprintk_emit._printk.irq_work_single.irq_work_run_list.irq_work_run
4.76 ? 20% perf-profile.calltrace.cycles-pp.console_unlock.vprintk_emit._printk.irq_work_single.irq_work_run_list
4.76 ? 20% perf-profile.calltrace.cycles-pp.console_flush_all.console_unlock.vprintk_emit._printk.irq_work_single
6.43 ? 9% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
6.70 ? 7% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
9.56 ? 11% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
15.57 ? 23% perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
10.91 ? 13% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
21.96 ? 21% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
20.04 ? 25% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
20.64 ? 25% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
41.42 ? 23% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
22.76 ? 26% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
26.14 ? 27% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt
48.20 ? 17% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
50.92 ? 17% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
52.83 ? 18% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
51.54 ? 16% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
51.55 ? 16% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
51.55 ? 16% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify
57.38 ? 16% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
6.60 ? 24% perf-profile.children.cycles-pp.do_syscall_64
6.70 ? 25% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
0.50 ? 33% perf-profile.children.cycles-pp.do_group_exit
0.50 ? 33% perf-profile.children.cycles-pp.do_exit
0.50 ? 49% perf-profile.children.cycles-pp.syscall_exit_to_user_mode
0.00 perf-profile.children.cycles-pp.arch_do_signal_or_restart
0.00 perf-profile.children.cycles-pp.get_signal
0.08 ? 11% perf-profile.children.cycles-pp.perf_event_mmap
0.10 ? 38% perf-profile.children.cycles-pp.do_anonymous_page
0.10 ? 41% perf-profile.children.cycles-pp.run_timer_softirq
0.12 ? 61% perf-profile.children.cycles-pp.__run_timers
0.12 ? 39% perf-profile.children.cycles-pp.mas_alloc_nodes
0.13 ? 46% perf-profile.children.cycles-pp.rcu_gp_fqs_loop
0.13 ? 69% perf-profile.children.cycles-pp.flush_smp_call_function_queue
0.14 ? 28% perf-profile.children.cycles-pp.irq_work_needs_cpu
0.14 ? 49% perf-profile.children.cycles-pp.schedule_timeout
0.14 ? 46% perf-profile.children.cycles-pp.perf_adjust_freq_unthr_context
0.14 ? 68% perf-profile.children.cycles-pp.select_task_rq
0.14 ? 61% perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore
0.15 ? 39% perf-profile.children.cycles-pp.profile_tick
0.16 ? 94% perf-profile.children.cycles-pp.enqueue_entity
0.16 ? 15% perf-profile.children.cycles-pp.io_serial_out
0.16 ? 45% perf-profile.children.cycles-pp.rcu_gp_kthread
0.16 ? 73% perf-profile.children.cycles-pp.tick_nohz_irq_exit
0.16 ? 73% perf-profile.children.cycles-pp.do_vmi_munmap
0.16 ? 73% perf-profile.children.cycles-pp.do_vmi_align_munmap
0.16 ? 43% perf-profile.children.cycles-pp.futex_wait
0.17 ? 50% perf-profile.children.cycles-pp.hrtimer_forward
0.17 ? 52% perf-profile.children.cycles-pp.drm_gem_vunmap_unlocked
0.17 ? 52% perf-profile.children.cycles-pp.drm_gem_vunmap
0.18 ? 38% perf-profile.children.cycles-pp.hrtimer_wakeup
0.18 ? 50% perf-profile.children.cycles-pp.__x64_sys_futex
0.18 ? 50% perf-profile.children.cycles-pp.do_futex
0.20 ? 47% perf-profile.children.cycles-pp.kmem_cache_alloc
0.20 ? 18% perf-profile.children.cycles-pp.cgroup_rstat_updated
0.21 ? 53% perf-profile.children.cycles-pp.drm_client_buffer_vmap
0.21 ? 41% perf-profile.children.cycles-pp.number
0.22 ? 49% perf-profile.children.cycles-pp.hrtimer_next_event_without
0.22 ? 41% perf-profile.children.cycles-pp.tick_check_oneshot_broadcast_this_cpu
0.22 ? 68% perf-profile.children.cycles-pp.error_entry
0.22 ? 50% perf-profile.children.cycles-pp.autoremove_wake_function
0.23 ? 36% perf-profile.children.cycles-pp.__irqentry_text_end
0.24 ? 41% perf-profile.children.cycles-pp.native_apic_mem_write
0.24 ? 47% perf-profile.children.cycles-pp.nohz_balancer_kick
0.25 ? 40% perf-profile.children.cycles-pp.irqentry_enter
0.25 ? 46% perf-profile.children.cycles-pp.__wake_up_common
0.26 ? 49% perf-profile.children.cycles-pp.hrtimer_active
0.26 ? 28% perf-profile.children.cycles-pp.cpuidle_governor_latency_req
0.26 ? 38% perf-profile.children.cycles-pp.task_tick_mm_cid
0.27 ? 37% perf-profile.children.cycles-pp.acct_account_cputime
0.27 ? 64% perf-profile.children.cycles-pp.enqueue_task_fair
0.28 ? 48% perf-profile.children.cycles-pp.rb_insert_color
0.28 ? 55% perf-profile.children.cycles-pp.activate_task
0.28 ? 39% perf-profile.children.cycles-pp.__wake_up_common_lock
0.28 ? 64% perf-profile.children.cycles-pp.trigger_load_balance
0.28 ? 51% perf-profile.children.cycles-pp.timekeeping_advance
0.28 ? 51% perf-profile.children.cycles-pp.update_wall_time
0.29 ? 53% perf-profile.children.cycles-pp.irqtime_account_process_tick
0.29 ? 97% perf-profile.children.cycles-pp.drm_atomic_helper_prepare_planes
0.29 ? 97% perf-profile.children.cycles-pp.drm_gem_fb_vmap
0.30 ? 52% perf-profile.children.cycles-pp.__do_huge_pmd_anonymous_page
0.30 ? 52% perf-profile.children.cycles-pp.clear_huge_page
0.30 ? 48% perf-profile.children.cycles-pp.account_process_tick
0.30 ? 50% perf-profile.children.cycles-pp.update_rq_clock
0.31 ? 29% perf-profile.children.cycles-pp.__calc_delta
0.31 ? 49% perf-profile.children.cycles-pp.ttwu_do_activate
0.32 ? 66% perf-profile.children.cycles-pp.ct_idle_exit
0.33 ? 62% perf-profile.children.cycles-pp.get_next_timer_interrupt
0.33 ? 72% perf-profile.children.cycles-pp.sprintf
0.34 ? 50% perf-profile.children.cycles-pp.clear_page_erms
0.37 ? 28% perf-profile.children.cycles-pp.pipe_write
0.37 ? 34% perf-profile.children.cycles-pp.irq_work_tick
0.37 ? 52% perf-profile.children.cycles-pp.check_cpu_stall
0.38 ? 50% perf-profile.children.cycles-pp.rb_erase
0.38 ? 62% perf-profile.children.cycles-pp.show_time_in_state
0.38 ? 62% perf-profile.children.cycles-pp.show
0.38 ? 23% perf-profile.children.cycles-pp.__cgroup_account_cputime_field
0.39 ? 60% perf-profile.children.cycles-pp.vsnprintf
0.40 ? 42% perf-profile.children.cycles-pp._raw_spin_lock_irq
0.41 ? 59% perf-profile.children.cycles-pp.perf_event_task_tick
0.41 ? 42% perf-profile.children.cycles-pp.hrtimer_update_next_event
0.42 ? 27% perf-profile.children.cycles-pp.wait_for_xmitr
0.42 ? 62% perf-profile.children.cycles-pp.sysfs_kf_seq_show
0.42 ? 75% perf-profile.children.cycles-pp.drm_gem_get_pages
0.42 ? 75% perf-profile.children.cycles-pp.shmem_read_mapping_page_gfp
0.43 ? 25% perf-profile.children.cycles-pp.dup_mmap
0.45 ? 23% perf-profile.children.cycles-pp.dup_mm
0.46 ? 40% perf-profile.children.cycles-pp.rb_next
0.47 ? 27% perf-profile.children.cycles-pp.try_to_wake_up
0.47 ? 57% perf-profile.children.cycles-pp.traverse
0.48 ? 11% perf-profile.children.cycles-pp.__hrtimer_next_event_base
0.49 ? 63% perf-profile.children.cycles-pp.run_posix_cpu_timers
0.49 ? 80% perf-profile.children.cycles-pp.delay_tsc
0.50 ? 62% perf-profile.children.cycles-pp.drm_gem_vmap_unlocked
0.50 ? 62% perf-profile.children.cycles-pp.drm_gem_vmap
0.52 ? 31% perf-profile.children.cycles-pp.lf
0.52 ? 31% perf-profile.children.cycles-pp.con_scroll
0.52 ? 31% perf-profile.children.cycles-pp.fbcon_scroll
0.52 ? 31% perf-profile.children.cycles-pp.fbcon_redraw
0.52 ? 40% perf-profile.children.cycles-pp.fbcon_putcs
0.54 ? 50% perf-profile.children.cycles-pp.rcu_core
0.57 ? 30% perf-profile.children.cycles-pp.vt_console_print
0.58 ? 38% perf-profile.children.cycles-pp.__update_load_avg_se
0.60 ? 45% perf-profile.children.cycles-pp.update_cfs_group
0.62 ? 28% perf-profile.children.cycles-pp.__update_load_avg_cfs_rq
0.63 ? 31% perf-profile.children.cycles-pp.copy_process
0.64 ? 29% perf-profile.children.cycles-pp.__do_sys_clone
0.64 ? 29% perf-profile.children.cycles-pp.kernel_clone
0.66 ? 47% perf-profile.children.cycles-pp.tick_sched_do_timer
0.67 ? 47% perf-profile.children.cycles-pp.seq_read_iter
0.73 ? 32% perf-profile.children.cycles-pp.tick_nohz_next_event
0.73 ? 37% perf-profile.children.cycles-pp.sched_clock_cpu
0.75 ? 32% perf-profile.children.cycles-pp.native_sched_clock
0.76 ? 67% perf-profile.children.cycles-pp.drm_atomic_helper_commit_tail_rpm
0.76 ? 67% perf-profile.children.cycles-pp.drm_atomic_helper_commit_planes
0.76 ? 67% perf-profile.children.cycles-pp.ast_primary_plane_helper_atomic_update
0.76 ? 67% perf-profile.children.cycles-pp.drm_fb_memcpy
0.76 ? 67% perf-profile.children.cycles-pp.memcpy_toio
0.77 ? 27% perf-profile.children.cycles-pp.rebalance_domains
0.82 ? 36% perf-profile.children.cycles-pp._raw_spin_lock
0.85 ? 61% perf-profile.children.cycles-pp.commit_tail
0.88 ? 23% perf-profile.children.cycles-pp.timerqueue_add
0.88 ? 53% perf-profile.children.cycles-pp.rcu_pending
0.94 ? 39% perf-profile.children.cycles-pp.tick_irq_enter
0.96 ? 35% perf-profile.children.cycles-pp.ktime_get_update_offsets_now
0.96 ? 22% perf-profile.children.cycles-pp.read_tsc
0.98 ? 30% perf-profile.children.cycles-pp.irqtime_account_irq
1.00 ? 36% perf-profile.children.cycles-pp.timerqueue_del
1.00 ? 27% perf-profile.children.cycles-pp.enqueue_hrtimer
1.02 ? 15% perf-profile.children.cycles-pp.update_irq_load_avg
1.03 ? 31% perf-profile.children.cycles-pp.tick_nohz_get_sleep_length
1.06 ? 29% perf-profile.children.cycles-pp.arch_scale_freq_tick
1.10 ? 26% perf-profile.children.cycles-pp.account_user_time
1.14 ? 64% perf-profile.children.cycles-pp.drm_atomic_helper_commit
1.15 ? 62% perf-profile.children.cycles-pp.drm_atomic_helper_dirtyfb
1.16 ? 28% perf-profile.children.cycles-pp.irq_enter_rcu
1.16 ? 23% perf-profile.children.cycles-pp.lapic_next_deadline
1.33 ? 17% perf-profile.children.cycles-pp.native_irq_return_iret
1.36 ? 45% perf-profile.children.cycles-pp.rcu_sched_clock_irq
1.52 ? 48% perf-profile.children.cycles-pp.drm_fb_helper_damage_work
1.52 ? 48% perf-profile.children.cycles-pp.drm_fbdev_generic_helper_fb_dirty
1.57 ? 22% perf-profile.children.cycles-pp.perf_rotate_context
1.58 ? 25% perf-profile.children.cycles-pp.clockevents_program_event
1.60 ? 43% perf-profile.children.cycles-pp.process_one_work
1.65 ? 41% perf-profile.children.cycles-pp.worker_thread
1.66 ? 39% perf-profile.children.cycles-pp.sync_regs
1.88 ? 23% perf-profile.children.cycles-pp.update_load_avg
2.74 ? 20% perf-profile.children.cycles-pp.perf_mux_hrtimer_handler
3.56 ? 16% perf-profile.children.cycles-pp.io_serial_in
3.90 ? 32% perf-profile.children.cycles-pp.irqentry_exit_to_user_mode
4.04 ? 20% perf-profile.children.cycles-pp.wait_for_lsr
4.19 ? 20% perf-profile.children.cycles-pp.serial8250_console_write
4.76 ? 20% perf-profile.children.cycles-pp.asm_sysvec_irq_work
4.76 ? 20% perf-profile.children.cycles-pp.sysvec_irq_work
4.76 ? 20% perf-profile.children.cycles-pp.__sysvec_irq_work
4.76 ? 20% perf-profile.children.cycles-pp.irq_work_run
4.76 ? 20% perf-profile.children.cycles-pp.irq_work_single
4.76 ? 20% perf-profile.children.cycles-pp._printk
4.76 ? 20% perf-profile.children.cycles-pp.vprintk_emit
4.78 ? 20% perf-profile.children.cycles-pp.console_unlock
4.78 ? 20% perf-profile.children.cycles-pp.console_flush_all
4.89 ? 20% perf-profile.children.cycles-pp.irq_work_run_list
15.80 ? 23% perf-profile.children.cycles-pp.tick_sched_timer
22.32 ? 21% perf-profile.children.cycles-pp.__hrtimer_run_queues
26.72 ? 20% perf-profile.children.cycles-pp.hrtimer_interrupt
41.41 ? 23% perf-profile.children.cycles-pp.intel_idle
27.60 ? 19% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
32.68 ? 17% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
36.88 ? 18% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
51.55 ? 16% perf-profile.children.cycles-pp.start_secondary
53.68 ? 18% perf-profile.children.cycles-pp.cpuidle_enter_state
53.82 ? 18% perf-profile.children.cycles-pp.cpuidle_enter
56.72 ? 17% perf-profile.children.cycles-pp.cpuidle_idle_call
57.37 ? 16% perf-profile.children.cycles-pp.do_idle
57.38 ? 16% perf-profile.children.cycles-pp.secondary_startup_64_no_verify
57.38 ? 16% perf-profile.children.cycles-pp.cpu_startup_entry
0.02 ?223% perf-profile.self.cycles-pp.page_remove_rmap
0.11 ? 69% perf-profile.self.cycles-pp.timekeeping_advance
0.11 ? 40% perf-profile.self.cycles-pp.irq_work_needs_cpu
0.13 ? 40% perf-profile.self.cycles-pp.irq_work_run_list
0.14 ? 68% perf-profile.self.cycles-pp._raw_spin_unlock_irqrestore
0.14 ? 46% perf-profile.self.cycles-pp.perf_adjust_freq_unthr_context
0.14 ? 20% perf-profile.self.cycles-pp.tick_irq_enter
0.14 ? 71% perf-profile.self.cycles-pp.nohz_balancer_kick
0.15 ? 56% perf-profile.self.cycles-pp.number
0.15 ? 39% perf-profile.self.cycles-pp.profile_tick
0.16 ? 15% perf-profile.self.cycles-pp.io_serial_out
0.17 ? 50% perf-profile.self.cycles-pp.hrtimer_forward
0.20 ? 51% perf-profile.self.cycles-pp.tick_nohz_next_event
0.20 ? 18% perf-profile.self.cycles-pp.cgroup_rstat_updated
0.21 ? 38% perf-profile.self.cycles-pp.timerqueue_del
0.22 ? 41% perf-profile.self.cycles-pp.tick_check_oneshot_broadcast_this_cpu
0.22 ? 68% perf-profile.self.cycles-pp.error_entry
0.23 ? 45% perf-profile.self.cycles-pp.reweight_entity
0.23 ? 36% perf-profile.self.cycles-pp.__irqentry_text_end
0.24 ? 41% perf-profile.self.cycles-pp.native_apic_mem_write
0.25 ? 47% perf-profile.self.cycles-pp.hrtimer_active
0.25 ? 62% perf-profile.self.cycles-pp.rb_insert_color
0.25 ? 38% perf-profile.self.cycles-pp.task_tick_mm_cid
0.27 ? 37% perf-profile.self.cycles-pp.acct_account_cputime
0.27 ? 69% perf-profile.self.cycles-pp.trigger_load_balance
0.28 ? 28% perf-profile.self.cycles-pp.perf_mux_hrtimer_handler
0.29 ? 53% perf-profile.self.cycles-pp.irqtime_account_process_tick
0.30 ? 48% perf-profile.self.cycles-pp.account_process_tick
0.31 ? 29% perf-profile.self.cycles-pp.__calc_delta
0.31 ? 37% perf-profile.self.cycles-pp.asm_sysvec_apic_timer_interrupt
0.32 ? 62% perf-profile.self.cycles-pp.cpuidle_idle_call
0.32 ? 39% perf-profile.self.cycles-pp.irq_work_tick
0.34 ? 50% perf-profile.self.cycles-pp.clear_page_erms
0.36 ? 52% perf-profile.self.cycles-pp._raw_spin_lock_irq
0.37 ? 52% perf-profile.self.cycles-pp.check_cpu_stall
0.38 ? 50% perf-profile.self.cycles-pp.rb_erase
0.39 ? 54% perf-profile.self.cycles-pp.rcu_sched_clock_irq
0.41 ? 59% perf-profile.self.cycles-pp.rcu_pending
0.42 ? 13% perf-profile.self.cycles-pp.__hrtimer_next_event_base
0.44 ? 70% perf-profile.self.cycles-pp.task_tick_fair
0.44 ? 33% perf-profile.self.cycles-pp.memcpy_toio
0.46 ? 40% perf-profile.self.cycles-pp.rb_next
0.46 ? 26% perf-profile.self.cycles-pp.irqtime_account_irq
0.49 ? 63% perf-profile.self.cycles-pp.run_posix_cpu_timers
0.49 ? 80% perf-profile.self.cycles-pp.delay_tsc
0.52 ? 56% perf-profile.self.cycles-pp.update_process_times
0.55 ? 41% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.57 ? 40% perf-profile.self.cycles-pp.__update_load_avg_se
0.60 ? 24% perf-profile.self.cycles-pp.__update_load_avg_cfs_rq
0.60 ? 45% perf-profile.self.cycles-pp.update_cfs_group
0.63 ? 26% perf-profile.self.cycles-pp.timerqueue_add
0.63 ? 26% perf-profile.self.cycles-pp.account_user_time
0.65 ? 21% perf-profile.self.cycles-pp.__sysvec_apic_timer_interrupt
0.67 ? 17% perf-profile.self.cycles-pp.perf_rotate_context
0.73 ? 33% perf-profile.self.cycles-pp.ktime_get_update_offsets_now
0.73 ? 30% perf-profile.self.cycles-pp.native_sched_clock
0.75 ? 35% perf-profile.self.cycles-pp.update_load_avg
0.81 ? 35% perf-profile.self.cycles-pp._raw_spin_lock
0.84 ? 40% perf-profile.self.cycles-pp.scheduler_tick
0.89 ? 28% perf-profile.self.cycles-pp.hrtimer_interrupt
0.94 ? 25% perf-profile.self.cycles-pp.read_tsc
1.02 ? 15% perf-profile.self.cycles-pp.update_irq_load_avg
1.06 ? 29% perf-profile.self.cycles-pp.arch_scale_freq_tick
1.16 ? 23% perf-profile.self.cycles-pp.lapic_next_deadline
1.25 ? 25% perf-profile.self.cycles-pp.__hrtimer_run_queues
1.27 ? 23% perf-profile.self.cycles-pp.cpuidle_enter_state
1.33 ? 17% perf-profile.self.cycles-pp.native_irq_return_iret
1.66 ? 39% perf-profile.self.cycles-pp.sync_regs
3.56 ? 16% perf-profile.self.cycles-pp.io_serial_in
41.41 ? 23% perf-profile.self.cycles-pp.intel_idle
*************************************************************************
1 sockets Intel(R) Xeon(R) E-2278G CPU @ 3.40GHz (Coffee Lake) with 32G memory
===============================================================
kconfig/option_a/rootfs/tbox_group/test/testcase:
4-rhel-8.3/pickle_pure_python/debian-x86_64-phoronix/lkp-cfl-e1/pyperformance-1.0.2/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
2.22 phoronix-test-suite.pyperformance.pickle_pure_python.milliseconds
+70.0% 156.34 phoronix-test-suite.time.elapsed_time
+70.0% 156.34 phoronix-test-suite.time.elapsed_time.max
84.00 phoronix-test-suite.time.percent_of_cpu_this_job_got
+78.7% 122.45 phoronix-test-suite.time.user_time
+9.1% 48.74 boot-time.boot
+6.0% 612.59 boot-time.idle
+57.1% 2.211e+09 cpuidle..time
+64.8% 2460130 cpuidle..usage
8177 meminfo.Active(anon)
+14.4% 138735 ? 2% meminfo.AnonHugePages
+49.3% 208.46 uptime.boot
2859 uptime.idle
-5.7% 87.67 vmstat.cpu.id
-40.6% 660.50 ? 2% vmstat.io.bi
2026 vmstat.system.cs
0.17 ? 2% mpstat.cpu.all.iowait%
3.06 mpstat.cpu.all.irq%
0.20 mpstat.cpu.all.soft%
0.83 mpstat.cpu.all.sys%
7.59 mpstat.cpu.all.usr%
2044 proc-vmstat.nr_active_anon
-1.7% 90534 proc-vmstat.nr_anon_pages
6789 proc-vmstat.nr_kernel_stack
-2.6% 15569 proc-vmstat.nr_slab_unreclaimable
2044 proc-vmstat.nr_zone_active_anon
+15.3% 796970 proc-vmstat.numa_hit
+15.3% 796942 proc-vmstat.numa_local
+4.3% 17256 proc-vmstat.pgactivate
+17.2% 876972 proc-vmstat.pgalloc_normal
+14.9% 1279681 proc-vmstat.pgfault
+18.2% 823764 proc-vmstat.pgfree
+8.8% 378312 proc-vmstat.pgreuse
+57.1% 1366784 proc-vmstat.unevictable_pgs_scanned
95.53 turbostat.Any%C0
-62.2% 105.50 turbostat.Avg_MHz
13.20 turbostat.Busy%
-82.5% 800.00 turbostat.Bzy_MHz
50860 turbostat.C1
0.03 turbostat.C1%
48630 turbostat.C1E
22838 ? 3% turbostat.C3
0.15 turbostat.C3%
+124.0% 1118904 turbostat.C6
40.94 turbostat.C6%
+30.0% 1218418 turbostat.C8
46.02 turbostat.C8%
0.20 ? 2% turbostat.CPU%c3
49.69 turbostat.CPU%c6
21.11 turbostat.CPU%c7
1.95 turbostat.CorWatt
30.00 turbostat.CoreTmp
+68.8% 2564565 turbostat.IRQ
1721 ? 24% turbostat.POLL
0.44 ? 4% turbostat.Pkg%pc2
3.76 ? 2% turbostat.Pkg%pc3
30.67 turbostat.PkgTmp
2.43 turbostat.PkgWatt
+99.1% 193.14 turbostat.Totl%C0
0.42 ? 8% sched_debug.cfs_rq:/.h_nr_running.stddev
24290 ? 8% sched_debug.cfs_rq:/.min_vruntime.avg
57957 ? 15% sched_debug.cfs_rq:/.min_vruntime.max
10806 ? 6% sched_debug.cfs_rq:/.min_vruntime.min
14213 ? 19% sched_debug.cfs_rq:/.min_vruntime.stddev
-47.1% 352.68 ? 21% sched_debug.cfs_rq:/.runnable_avg.avg
50.08 ? 41% sched_debug.cfs_rq:/.runnable_avg.min
31063 ? 41% sched_debug.cfs_rq:/.spread0.max
-16116 sched_debug.cfs_rq:/.spread0.min
14220 ? 19% sched_debug.cfs_rq:/.spread0.stddev
-46.9% 338.25 ? 21% sched_debug.cfs_rq:/.util_avg.avg
44.39 ? 50% sched_debug.cfs_rq:/.util_avg.min
-73.8% 191.67 ? 23% sched_debug.cfs_rq:/.util_est_enqueued.max
62.29 ? 21% sched_debug.cfs_rq:/.util_est_enqueued.stddev
+89.7% 502726 ? 12% sched_debug.cpu.avg_idle.min
91142 ? 16% sched_debug.cpu.clock.avg
91144 ? 16% sched_debug.cpu.clock.max
91140 ? 16% sched_debug.cpu.clock.min
88500 ? 15% sched_debug.cpu.clock_task.avg
89762 ? 15% sched_debug.cpu.clock_task.max
79135 ? 17% sched_debug.cpu.clock_task.min
2290 ? 12% sched_debug.cpu.curr->pid.max
+53.7% 791.50 ? 13% sched_debug.cpu.curr->pid.stddev
10204 ? 16% sched_debug.cpu.nr_switches.avg
24267 ? 14% sched_debug.cpu.nr_switches.max
3713 ? 17% sched_debug.cpu.nr_switches.min
5425 ? 15% sched_debug.cpu.nr_switches.stddev
91140 ? 16% sched_debug.cpu_clk
90265 ? 16% sched_debug.ktime
89499 ? 15% sched_debug.sched_clk
38.20 perf-stat.i.MPKI
-76.9% 4.446e+08 perf-stat.i.branch-instructions
2.78 perf-stat.i.branch-miss-rate%
-33.9% 11845681 perf-stat.i.branch-misses
5.04 perf-stat.i.cache-miss-rate%
-6.9% 3053819 perf-stat.i.cache-misses
-41.7% 60884324 perf-stat.i.cache-references
1974 perf-stat.i.context-switches
0.84 perf-stat.i.cpi
-65.5% 1.515e+09 perf-stat.i.cpu-cycles
-5.4% 23.57 ? 2% perf-stat.i.cpu-migrations
-66.8% 640.60 perf-stat.i.cycles-between-cache-misses
0.13 ? 2% perf-stat.i.dTLB-load-miss-rate%
-11.4% 503424 perf-stat.i.dTLB-load-misses
-79.0% 5.784e+08 perf-stat.i.dTLB-loads
0.03 perf-stat.i.dTLB-store-miss-rate%
64183 perf-stat.i.dTLB-store-misses
-79.2% 2.639e+08 perf-stat.i.dTLB-stores
46.14 perf-stat.i.iTLB-load-miss-rate%
-52.0% 263472 perf-stat.i.iTLB-load-misses
-62.6% 336446 perf-stat.i.iTLB-loads
-76.8% 2.07e+09 perf-stat.i.instructions
7907 perf-stat.i.instructions-per-iTLB-miss
1.35 perf-stat.i.ipc
1.26 ? 2% perf-stat.i.major-faults
0.09 perf-stat.i.metric.GHz
47.79 perf-stat.i.metric.K/sec
84.21 perf-stat.i.metric.M/sec
7775 perf-stat.i.minor-faults
4.73 ? 4% perf-stat.i.node-load-misses
-34.2% 116568 perf-stat.i.node-loads
5.13 ? 12% perf-stat.i.node-store-misses
-32.5% 246258 perf-stat.i.node-stores
7776 perf-stat.i.page-faults
29.48 perf-stat.overall.MPKI
2.66 perf-stat.overall.branch-miss-rate%
5.01 perf-stat.overall.cache-miss-rate%
0.73 perf-stat.overall.cpi
-63.0% 496.35 perf-stat.overall.cycles-between-cache-misses
0.09 perf-stat.overall.dTLB-load-miss-rate%
0.02 perf-stat.overall.dTLB-store-miss-rate%
43.87 perf-stat.overall.iTLB-load-miss-rate%
7839 perf-stat.overall.instructions-per-iTLB-miss
1.37 perf-stat.overall.ipc
-76.9% 4.412e+08 perf-stat.ps.branch-instructions
-33.7% 11756414 perf-stat.ps.branch-misses
-6.5% 3031172 perf-stat.ps.cache-misses
-41.4% 60556271 perf-stat.ps.cache-references
1959 perf-stat.ps.context-switches
-65.3% 1.504e+09 perf-stat.ps.cpu-cycles
-4.8% 23.46 ? 2% perf-stat.ps.cpu-migrations
-11.0% 500237 perf-stat.ps.dTLB-load-misses
-78.9% 5.74e+08 perf-stat.ps.dTLB-loads
63822 perf-stat.ps.dTLB-store-misses
-79.1% 2.62e+08 perf-stat.ps.dTLB-stores
-51.7% 262035 perf-stat.ps.iTLB-load-misses
-62.4% 335305 perf-stat.ps.iTLB-loads
-76.7% 2.054e+09 perf-stat.ps.instructions
1.25 ? 2% perf-stat.ps.major-faults
7730 perf-stat.ps.minor-faults
4.70 ? 5% perf-stat.ps.node-load-misses
-33.9% 115833 perf-stat.ps.node-loads
5.11 ? 13% perf-stat.ps.node-store-misses
-31.9% 245454 perf-stat.ps.node-stores
7731 perf-stat.ps.page-faults
-60.7% 3.233e+11 perf-stat.total.instructions
4.20 ? 10% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
4.22 ? 10% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exit_mmap.__mmput.exit_mm.do_exit.do_group_exit
0.00 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.56 ? 8% perf-profile.calltrace.cycles-pp.asm_exc_page_fault
0.00 perf-profile.calltrace.cycles-pp.zap_pte_range.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap
0.00 perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare
0.00 perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop
0.00 perf-profile.calltrace.cycles-pp.zap_pmd_range.unmap_page_range.unmap_vmas.exit_mmap.__mmput
0.47 ? 71% perf-profile.calltrace.cycles-pp.bprm_execve.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exit_mm.do_exit.do_group_exit.get_signal.arch_do_signal_or_restart
0.00 perf-profile.calltrace.cycles-pp.__mmput.exit_mm.do_exit.do_group_exit.get_signal
0.73 ? 18% perf-profile.calltrace.cycles-pp.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.73 ? 18% perf-profile.calltrace.cycles-pp.do_execveat_common.__x64_sys_execve.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.get_signal.arch_do_signal_or_restart.exit_to_user_mode_loop.exit_to_user_mode_prepare.syscall_exit_to_user_mode
0.00 perf-profile.calltrace.cycles-pp.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.00 perf-profile.calltrace.cycles-pp.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write.do_syscall_64
0.00 perf-profile.calltrace.cycles-pp.generic_perform_write.__generic_file_write_iter.generic_file_write_iter.vfs_write.ksys_write
0.00 perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.__mmput.exit_mm.do_exit
0.00 perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.exit_mmap.__mmput.exit_mm
0.70 ? 6% perf-profile.calltrace.cycles-pp.timerqueue_del.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
0.71 ? 20% perf-profile.calltrace.cycles-pp.rcu_sched_clock_irq.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
0.71 ? 19% perf-profile.calltrace.cycles-pp.enqueue_hrtimer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
0.77 ? 16% perf-profile.calltrace.cycles-pp.load_balance.rebalance_domains.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt
0.79 ? 24% perf-profile.calltrace.cycles-pp.tick_nohz_next_event.tick_nohz_get_sleep_length.menu_select.cpuidle_idle_call.do_idle
0.80 ? 16% perf-profile.calltrace.cycles-pp.update_blocked_averages.run_rebalance_domains.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt
0.80 ? 23% perf-profile.calltrace.cycles-pp.perf_rotate_context.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt
0.85 ? 16% perf-profile.calltrace.cycles-pp.run_rebalance_domains.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
0.97 ? 16% perf-profile.calltrace.cycles-pp.rebalance_domains.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
1.20 ? 25% perf-profile.calltrace.cycles-pp.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
1.41 ? 36% perf-profile.calltrace.cycles-pp.serial8250_console_write.console_flush_all.console_unlock.vprintk_emit._printk
1.42 ? 26% perf-profile.calltrace.cycles-pp.tick_nohz_get_sleep_length.menu_select.cpuidle_idle_call.do_idle.cpu_startup_entry
1.46 ? 20% perf-profile.calltrace.cycles-pp.tick_irq_enter.irq_enter_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
1.50 ? 18% perf-profile.calltrace.cycles-pp.irq_enter_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
1.71 ? 23% perf-profile.calltrace.cycles-pp.vprintk_emit._printk.irq_work_single.irq_work_run_list.irq_work_run
1.71 ? 23% perf-profile.calltrace.cycles-pp.console_unlock.vprintk_emit._printk.irq_work_single.irq_work_run_list
1.71 ? 23% perf-profile.calltrace.cycles-pp.console_flush_all.console_unlock.vprintk_emit._printk.irq_work_single
1.71 ? 23% perf-profile.calltrace.cycles-pp.irq_work_run_list.irq_work_run.__sysvec_irq_work.sysvec_irq_work.asm_sysvec_irq_work
1.71 ? 23% perf-profile.calltrace.cycles-pp.irq_work_single.irq_work_run_list.irq_work_run.__sysvec_irq_work.sysvec_irq_work
1.71 ? 23% perf-profile.calltrace.cycles-pp._printk.irq_work_single.irq_work_run_list.irq_work_run.__sysvec_irq_work
2.32 ? 10% perf-profile.calltrace.cycles-pp.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
2.37 ? 19% perf-profile.calltrace.cycles-pp.scheduler_tick.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
2.85 ? 10% perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
3.52 ? 20% perf-profile.calltrace.cycles-pp.menu_select.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
4.27 ? 9% perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt
4.76 ? 8% perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt
5.42 ? 6% perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
5.62 ? 82% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.rest_init
5.83 ? 79% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.rest_init.arch_call_rest_init
5.90 ? 79% perf-profile.calltrace.cycles-pp.cpu_startup_entry.rest_init.arch_call_rest_init.start_kernel.x86_64_start_reservations
5.90 ? 79% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.rest_init.arch_call_rest_init.start_kernel
5.90 ? 79% perf-profile.calltrace.cycles-pp.x86_64_start_kernel.secondary_startup_64_no_verify
5.90 ? 79% perf-profile.calltrace.cycles-pp.x86_64_start_reservations.x86_64_start_kernel.secondary_startup_64_no_verify
5.90 ? 79% perf-profile.calltrace.cycles-pp.start_kernel.x86_64_start_reservations.x86_64_start_kernel.secondary_startup_64_no_verify
5.90 ? 79% perf-profile.calltrace.cycles-pp.arch_call_rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel.secondary_startup_64_no_verify
5.90 ? 79% perf-profile.calltrace.cycles-pp.rest_init.arch_call_rest_init.start_kernel.x86_64_start_reservations.x86_64_start_kernel
9.21 ? 2% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
11.19 ? 3% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
11.74 ? 3% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
18.08 perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
20.55 ? 3% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
61.39 ? 2% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
79.24 ? 5% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
83.24 perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
83.74 ? 5% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
84.78 ? 4% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
84.82 ? 4% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
84.82 ? 4% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify
90.72 perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
4.21 ? 10% perf-profile.children.cycles-pp.do_syscall_64
4.22 ? 10% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
0.28 ? 20% perf-profile.children.cycles-pp.do_group_exit
0.28 ? 20% perf-profile.children.cycles-pp.do_exit
0.51 ? 10% perf-profile.children.cycles-pp.__mmput
0.51 ? 10% perf-profile.children.cycles-pp.exit_mmap
0.22 ? 30% perf-profile.children.cycles-pp.exit_mm
0.39 ? 30% perf-profile.children.cycles-pp.syscall_exit_to_user_mode
1.78 ? 12% perf-profile.children.cycles-pp.asm_exc_page_fault
0.00 perf-profile.children.cycles-pp.get_signal
0.01 ?223% perf-profile.children.cycles-pp.arch_do_signal_or_restart
0.62 ? 21% perf-profile.children.cycles-pp.bprm_execve
0.73 ? 18% perf-profile.children.cycles-pp.__x64_sys_execve
0.73 ? 18% perf-profile.children.cycles-pp.do_execveat_common
0.00 perf-profile.children.cycles-pp.generic_perform_write
0.01 ?223% perf-profile.children.cycles-pp.__generic_file_write_iter
0.01 ?223% perf-profile.children.cycles-pp.generic_file_write_iter
0.09 ? 24% perf-profile.children.cycles-pp.mas_store_prealloc
0.10 ? 38% perf-profile.children.cycles-pp.hrtimer_get_next_event
0.10 ? 19% perf-profile.children.cycles-pp.sprintf
0.11 ? 23% perf-profile.children.cycles-pp.show
0.11 ? 23% perf-profile.children.cycles-pp.show_time_in_state
0.12 ? 24% perf-profile.children.cycles-pp.traverse
0.12 ? 28% perf-profile.children.cycles-pp.hrtimer_wakeup
0.13 ? 17% perf-profile.children.cycles-pp._raw_spin_unlock_irqrestore
0.13 ? 19% perf-profile.children.cycles-pp.irq_work_needs_cpu
0.13 ? 28% perf-profile.children.cycles-pp.calc_global_load_tick
0.14 ? 36% perf-profile.children.cycles-pp.pm_qos_read_value
0.14 ? 48% perf-profile.children.cycles-pp.cpuidle_not_available
0.14 ? 33% perf-profile.children.cycles-pp.schedule_idle
0.14 ? 34% perf-profile.children.cycles-pp.sysfs_kf_seq_show
0.14 ? 45% perf-profile.children.cycles-pp.rcu_nocb_flush_deferred_wakeup
0.15 ? 45% perf-profile.children.cycles-pp.trigger_load_balance
0.16 ? 43% perf-profile.children.cycles-pp.enqueue_entity
0.16 ? 38% perf-profile.children.cycles-pp.call_cpuidle
0.17 ? 53% perf-profile.children.cycles-pp.ct_irq_enter
0.17 ? 39% perf-profile.children.cycles-pp.x86_pmu_disable
0.18 ? 48% perf-profile.children.cycles-pp._raw_spin_lock_irq
0.18 ? 43% perf-profile.children.cycles-pp.update_load_avg
0.19 ? 33% perf-profile.children.cycles-pp.tick_nohz_tick_stopped
0.19 ? 22% perf-profile.children.cycles-pp.__update_load_avg_cfs_rq
0.20 ? 47% perf-profile.children.cycles-pp.task_tick_mm_cid
0.20 ? 39% perf-profile.children.cycles-pp.enqueue_task_fair
0.21 ? 36% perf-profile.children.cycles-pp.activate_task
0.22 ? 54% perf-profile.children.cycles-pp.update_wall_time
0.22 ? 54% perf-profile.children.cycles-pp.timekeeping_advance
0.22 ? 27% perf-profile.children.cycles-pp.rb_insert_color
0.22 ? 30% perf-profile.children.cycles-pp.ct_kernel_exit
0.22 ? 27% perf-profile.children.cycles-pp.ct_nmi_enter
0.23 ? 24% perf-profile.children.cycles-pp.rb_next
0.23 ? 27% perf-profile.children.cycles-pp.ttwu_do_activate
0.24 ? 76% perf-profile.children.cycles-pp.copy_page_to_iter
0.24 ? 76% perf-profile.children.cycles-pp._copy_to_iter
0.24 ? 76% perf-profile.children.cycles-pp.copyout
0.25 ? 32% perf-profile.children.cycles-pp.tick_nohz_irq_exit
0.25 ? 32% perf-profile.children.cycles-pp.rcu_core
0.26 ? 30% perf-profile.children.cycles-pp.irq_work_tick
0.27 ? 29% perf-profile.children.cycles-pp.try_to_wake_up
0.27 ? 47% perf-profile.children.cycles-pp.get_cpu_device
0.28 ? 27% perf-profile.children.cycles-pp.check_cpu_stall
0.30 ? 50% perf-profile.children.cycles-pp.irqtime_account_process_tick
0.31 ? 73% perf-profile.children.cycles-pp.perf_event_task_tick
0.31 ? 27% perf-profile.children.cycles-pp.__intel_pmu_enable_all
0.31 ? 42% perf-profile.children.cycles-pp.tsc_verify_tsc_adjust
0.32 ? 33% perf-profile.children.cycles-pp.ktime_get_update_offsets_now
0.32 ? 38% perf-profile.children.cycles-pp.arch_cpu_idle_enter
0.34 ? 36% perf-profile.children.cycles-pp.tick_sched_do_timer
0.35 ? 9% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.36 ? 42% perf-profile.children.cycles-pp.ct_kernel_enter
0.37 ? 43% perf-profile.children.cycles-pp.native_apic_mem_write
0.39 ? 23% perf-profile.children.cycles-pp.get_next_timer_interrupt
0.39 ? 17% perf-profile.children.cycles-pp.nr_iowait_cpu
0.40 ? 23% perf-profile.children.cycles-pp.rb_erase
0.40 ? 31% perf-profile.children.cycles-pp.local_clock
0.41 ? 30% perf-profile.children.cycles-pp.tick_check_oneshot_broadcast_this_cpu
0.50 ? 14% perf-profile.children.cycles-pp.update_sd_lb_stats
0.46 ? 21% perf-profile.children.cycles-pp.run_posix_cpu_timers
0.55 ? 15% perf-profile.children.cycles-pp.find_busiest_group
0.47 ? 20% perf-profile.children.cycles-pp.sched_clock_cpu
0.48 ? 26% perf-profile.children.cycles-pp.__update_blocked_fair
0.49 ? 36% perf-profile.children.cycles-pp.hrtimer_update_next_event
0.50 ? 25% perf-profile.children.cycles-pp.irqentry_enter
0.51 ? 22% perf-profile.children.cycles-pp.lapic_next_deadline
0.52 ? 33% perf-profile.children.cycles-pp.hrtimer_next_event_without
0.52 ? 29% perf-profile.children.cycles-pp.update_irq_load_avg
0.54 ? 26% perf-profile.children.cycles-pp.cpuidle_governor_latency_req
0.55 ? 11% perf-profile.children.cycles-pp.arch_scale_freq_tick
0.56 ? 29% perf-profile.children.cycles-pp.ct_idle_exit
0.57 ? 18% perf-profile.children.cycles-pp.tick_nohz_stop_idle
0.57 ? 22% perf-profile.children.cycles-pp.read_tsc
0.58 ? 28% perf-profile.children.cycles-pp.rcu_pending
0.59 ? 20% perf-profile.children.cycles-pp.native_sched_clock
0.61 ? 6% perf-profile.children.cycles-pp.__hrtimer_next_event_base
0.68 ? 25% perf-profile.children.cycles-pp.timerqueue_add
0.74 ? 5% perf-profile.children.cycles-pp.timerqueue_del
0.75 ? 13% perf-profile.children.cycles-pp.irqtime_account_irq
0.76 ? 25% perf-profile.children.cycles-pp.update_rq_clock_task
0.84 ? 15% perf-profile.children.cycles-pp.load_balance
0.76 ? 25% perf-profile.children.cycles-pp.clockevents_program_event
0.78 ? 17% perf-profile.children.cycles-pp.enqueue_hrtimer
0.80 ? 22% perf-profile.children.cycles-pp.rcu_sched_clock_irq
0.89 ? 17% perf-profile.children.cycles-pp.update_blocked_averages
0.89 ? 18% perf-profile.children.cycles-pp.perf_rotate_context
0.92 ? 24% perf-profile.children.cycles-pp.tick_nohz_next_event
0.95 ? 19% perf-profile.children.cycles-pp.run_rebalance_domains
1.04 ? 14% perf-profile.children.cycles-pp.rebalance_domains
1.05 ? 17% perf-profile.children.cycles-pp.ktime_get
1.34 ? 18% perf-profile.children.cycles-pp.perf_mux_hrtimer_handler
1.55 ? 22% perf-profile.children.cycles-pp.tick_nohz_get_sleep_length
1.60 ? 18% perf-profile.children.cycles-pp.tick_irq_enter
1.64 ? 16% perf-profile.children.cycles-pp.irq_enter_rcu
2.54 ? 10% perf-profile.children.cycles-pp.__do_softirq
2.46 ? 7% perf-profile.children.cycles-pp.asm_sysvec_irq_work
2.46 ? 7% perf-profile.children.cycles-pp.sysvec_irq_work
2.46 ? 7% perf-profile.children.cycles-pp.__sysvec_irq_work
2.46 ? 7% perf-profile.children.cycles-pp.irq_work_run
2.46 ? 7% perf-profile.children.cycles-pp.irq_work_single
2.46 ? 7% perf-profile.children.cycles-pp._printk
2.52 ? 6% perf-profile.children.cycles-pp.irq_work_run_list
2.76 ? 20% perf-profile.children.cycles-pp.scheduler_tick
3.13 ? 11% perf-profile.children.cycles-pp.__irq_exit_rcu
3.76 ? 20% perf-profile.children.cycles-pp.menu_select
4.92 ? 8% perf-profile.children.cycles-pp.update_process_times
5.41 ? 7% perf-profile.children.cycles-pp.tick_sched_handle
5.90 ? 79% perf-profile.children.cycles-pp.x86_64_start_kernel
5.90 ? 79% perf-profile.children.cycles-pp.x86_64_start_reservations
5.90 ? 79% perf-profile.children.cycles-pp.start_kernel
5.90 ? 79% perf-profile.children.cycles-pp.arch_call_rest_init
5.90 ? 79% perf-profile.children.cycles-pp.rest_init
6.13 ? 4% perf-profile.children.cycles-pp.tick_sched_timer
9.85 ? 3% perf-profile.children.cycles-pp.__hrtimer_run_queues
11.92 ? 3% perf-profile.children.cycles-pp.hrtimer_interrupt
12.48 ? 2% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
18.80 perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
20.24 ? 2% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
61.40 ? 2% perf-profile.children.cycles-pp.intel_idle
84.64 perf-profile.children.cycles-pp.cpuidle_enter_state
84.82 ? 4% perf-profile.children.cycles-pp.start_secondary
84.86 perf-profile.children.cycles-pp.cpuidle_enter
89.64 perf-profile.children.cycles-pp.cpuidle_idle_call
90.72 perf-profile.children.cycles-pp.secondary_startup_64_no_verify
90.72 perf-profile.children.cycles-pp.cpu_startup_entry
90.72 perf-profile.children.cycles-pp.do_idle
0.13 ? 36% perf-profile.self.cycles-pp.pm_qos_read_value
0.14 ? 48% perf-profile.self.cycles-pp.cpuidle_not_available
0.14 ? 37% perf-profile.self.cycles-pp.trigger_load_balance
0.14 ? 33% perf-profile.self.cycles-pp.cpuidle_governor_latency_req
0.14 ? 45% perf-profile.self.cycles-pp.rcu_nocb_flush_deferred_wakeup
0.15 ? 27% perf-profile.self.cycles-pp.perf_mux_hrtimer_handler
0.15 ? 42% perf-profile.self.cycles-pp.call_cpuidle
0.15 ? 32% perf-profile.self.cycles-pp.timerqueue_del
0.16 ? 33% perf-profile.self.cycles-pp.irqentry_enter
0.16 ? 40% perf-profile.self.cycles-pp.x86_pmu_disable
0.18 ? 48% perf-profile.self.cycles-pp.task_tick_mm_cid
0.18 ? 48% perf-profile.self.cycles-pp._raw_spin_lock_irq
0.19 ? 22% perf-profile.self.cycles-pp.__update_load_avg_cfs_rq
0.20 ? 50% perf-profile.self.cycles-pp.rcu_sched_clock_irq
0.20 ? 52% perf-profile.self.cycles-pp.tsc_verify_tsc_adjust
0.20 ? 42% perf-profile.self.cycles-pp.do_idle
0.20 ? 46% perf-profile.self.cycles-pp.tick_nohz_stop_idle
0.21 ? 47% perf-profile.self.cycles-pp.ct_kernel_enter
0.21 ? 32% perf-profile.self.cycles-pp.ct_nmi_enter
0.22 ? 27% perf-profile.self.cycles-pp.rb_insert_color
0.22 ? 20% perf-profile.self.cycles-pp.cpuidle_enter
0.22 ? 34% perf-profile.self.cycles-pp.tick_irq_enter
0.22 ? 29% perf-profile.self.cycles-pp.rb_next
0.22 ? 37% perf-profile.self.cycles-pp.load_balance
0.23 ? 25% perf-profile.self.cycles-pp.irq_work_tick
0.24 ? 30% perf-profile.self.cycles-pp.__irq_exit_rcu
0.24 ? 9% perf-profile.self.cycles-pp.rebalance_domains
0.24 ? 25% perf-profile.self.cycles-pp.update_rq_clock_task
0.24 ? 28% perf-profile.self.cycles-pp.get_next_timer_interrupt
0.26 ? 23% perf-profile.self.cycles-pp.hrtimer_interrupt
0.27 ? 52% perf-profile.self.cycles-pp.get_cpu_device
0.28 ? 27% perf-profile.self.cycles-pp.check_cpu_stall
0.28 ? 40% perf-profile.self.cycles-pp.perf_rotate_context
0.28 ? 64% perf-profile.self.cycles-pp.rcu_pending
0.30 ? 50% perf-profile.self.cycles-pp.irqtime_account_process_tick
0.31 ? 27% perf-profile.self.cycles-pp.__intel_pmu_enable_all
0.31 ? 37% perf-profile.self.cycles-pp.__update_blocked_fair
0.34 ? 26% perf-profile.self.cycles-pp.sysvec_apic_timer_interrupt
0.34 ? 6% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.35 ? 39% perf-profile.self.cycles-pp.scheduler_tick
0.35 ? 37% perf-profile.self.cycles-pp.update_process_times
0.36 ? 44% perf-profile.self.cycles-pp.native_apic_mem_write
0.38 ? 20% perf-profile.self.cycles-pp.nr_iowait_cpu
0.39 ? 27% perf-profile.self.cycles-pp.cpuidle_idle_call
0.40 ? 23% perf-profile.self.cycles-pp.rb_erase
0.41 ? 30% perf-profile.self.cycles-pp.tick_check_oneshot_broadcast_this_cpu
0.44 ? 12% perf-profile.self.cycles-pp.irqtime_account_irq
0.46 ? 31% perf-profile.self.cycles-pp.timerqueue_add
0.46 ? 21% perf-profile.self.cycles-pp.run_posix_cpu_timers
0.51 ? 22% perf-profile.self.cycles-pp.lapic_next_deadline
0.52 ? 29% perf-profile.self.cycles-pp.update_irq_load_avg
0.54 ? 44% perf-profile.self.cycles-pp.__hrtimer_run_queues
0.55 ? 11% perf-profile.self.cycles-pp.arch_scale_freq_tick
0.55 ? 21% perf-profile.self.cycles-pp.read_tsc
0.56 ? 12% perf-profile.self.cycles-pp.__hrtimer_next_event_base
0.58 ? 19% perf-profile.self.cycles-pp.native_sched_clock
0.60 ? 25% perf-profile.self.cycles-pp.ktime_get
1.57 ? 29% perf-profile.self.cycles-pp.menu_select
2.02 ? 18% perf-profile.self.cycles-pp.cpuidle_enter_state
61.40 ? 2% perf-profile.self.cycles-pp.intel_idle
*************************************************************************
1 sockets Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz (Coffee Lake) with 32G memory
===============================================================
kconfig/need_x/option_a/rootfs/tbox_group/test/testcase:
4-rhel-8.3/true/500px Copy From Window To Window/debian-x86_64-phoronix/lkp-cfl-d2/x11perf-1.1.1/phoronix-test-suite
Remove fast_switch_possible flag from active driver")
Return failure if fast_switch is not set and fast_switch_possible is set")
278dfbd17c5d0767834471d7fea
---------------------------
fail:runs
|
6:6 kmsg.cpufreq:fast_switch_possible_is_enabled_but_fast_switch_callback_is_not_set
%stddev
\
6093 ? 45% phoronix-test-suite.time.involuntary_context_switches
-4.0% 120523 phoronix-test-suite.time.maximum_resident_set_size
-23.6% 147783 phoronix-test-suite.time.minor_page_faults
83.00 phoronix-test-suite.time.percent_of_cpu_this_job_got
99.90 phoronix-test-suite.time.user_time
-79.4% 290969 ? 10% phoronix-test-suite.time.voluntary_context_switches
-7.3% 10294 phoronix-test-suite.x11perf.500pxCopyFromWindowToWindow.operations___second
-10.0% 1.436e+09 cpuidle..time
-31.3% 2612320 cpuidle..usage
0.32 mpstat.cpu.all.iowait%
3.70 mpstat.cpu.all.irq%
0.26 mpstat.cpu.all.soft%
2.23 mpstat.cpu.all.sys%
12.59 mpstat.cpu.all.usr%
-9.1% 80.00 vmstat.cpu.id
12.00 vmstat.cpu.us
2045 vmstat.io.bi
-76.8% 2519224 vmstat.memory.cache
+40.1% 29541029 vmstat.memory.free
1.00 vmstat.procs.r
15452 ? 2% vmstat.system.cs
-7.5% 16870 vmstat.system.in
-9.1% 417876 meminfo.AnonPages
-77.5% 2446909 meminfo.Cached
-78.3% 2346035 meminfo.Committed_AS
-7.4% 1121115 meminfo.Inactive(anon)
75030 meminfo.KReclaimable
-23.1% 115779 meminfo.Mapped
+38.6% 30641210 meminfo.MemAvailable
+40.7% 29538155 meminfo.MemFree
-73.2% 3125406 meminfo.Memused
6547 meminfo.PageTables
75030 meminfo.SReclaimable
61884 meminfo.SUnreclaim
-89.2% 1015386 meminfo.Shmem
-30.1% 136915 meminfo.Slab
-96.4% 308501 meminfo.Unevictable
-75.7% 3154603 meminfo.max_used_kB
0.75 ? 21% sched_debug.cfs_rq:/.h_nr_running.avg
31685 ? 12% sched_debug.cfs_rq:/.min_vruntime.avg
68497 ? 16% sched_debug.cfs_rq:/.min_vruntime.max
17233 ? 12% sched_debug.cfs_rq:/.min_vruntime.min
15442 ? 18% sched_debug.cfs_rq:/.min_vruntime.stddev
0.60 ? 11% sched_debug.cfs_rq:/.nr_running.avg
+45.2% 537.98 ? 11% sched_debug.cfs_rq:/.runnable_avg.avg
1397 ? 6% sched_debug.cfs_rq:/.runnable_avg.max
+43.6% 373.22 ? 12% sched_debug.cfs_rq:/.runnable_avg.stddev
15448 ? 18% sched_debug.cfs_rq:/.spread0.stddev
+32.7% 467.44 ? 11% sched_debug.cfs_rq:/.util_avg.avg
1238 ? 6% sched_debug.cfs_rq:/.util_avg.max
+34.4% 345.04 ? 10% sched_debug.cfs_rq:/.util_avg.stddev
+78.9% 136.54 ? 21% sched_debug.cfs_rq:/.util_est_enqueued.avg
+18.8% 894877 ? 10% sched_debug.cpu.avg_idle.avg
1.85 ? 13% sched_debug.cpu.clock.stddev
0.00 ? 13% sched_debug.cpu.next_balance.stddev
0.75 ? 18% sched_debug.cpu.nr_running.avg
92716 ? 19% sched_debug.cpu.nr_switches.avg
-53.4% 209625 ? 18% sched_debug.cpu.nr_switches.max
16484 ? 29% sched_debug.cpu.nr_switches.min
57142 ? 25% sched_debug.cpu.nr_switches.stddev
-9.1% 104454 proc-vmstat.nr_anon_pages
+38.5% 766481 proc-vmstat.nr_dirty_background_threshold
+38.5% 1534837 proc-vmstat.nr_dirty_threshold
-77.4% 612214 proc-vmstat.nr_file_pages
+40.7% 7384644 proc-vmstat.nr_free_pages
-7.4% 280420 proc-vmstat.nr_inactive_anon
6236 proc-vmstat.nr_kernel_stack
28939 proc-vmstat.nr_mapped
1638 proc-vmstat.nr_page_table_pages
-89.2% 253805 proc-vmstat.nr_shmem
18757 proc-vmstat.nr_slab_reclaimable
15470 proc-vmstat.nr_slab_unreclaimable
76929 proc-vmstat.nr_unevictable
-7.4% 280420 proc-vmstat.nr_zone_inactive_anon
76929 proc-vmstat.nr_zone_unevictable
-92.9% 765514 proc-vmstat.numa_hit
-92.9% 765422 proc-vmstat.numa_local
-91.8% 902283 proc-vmstat.pgalloc_normal
+6.5% 534236 ? 2% proc-vmstat.pgfault
-92.9% 775875 proc-vmstat.pgfree
70928 ? 2% proc-vmstat.pgreuse
-97.0% 305188 proc-vmstat.unevictable_pgs_culled
-97.0% 305188 proc-vmstat.unevictable_pgs_rescued
-97.0% 305217 proc-vmstat.unevictable_pgs_scanned
96.32 turbostat.Any%C0
-51.4% 165.67 turbostat.Avg_MHz
20.71 turbostat.Busy%
-81.0% 800.00 turbostat.Bzy_MHz
-12.5% 743127 turbostat.C1E
2.48 turbostat.C1E%
24104 ? 2% turbostat.C3
0.31 ? 2% turbostat.C3%
+14.3% 827187 turbostat.C6
-20.8% 671679 turbostat.C8
35.46 turbostat.C8%
24.45 turbostat.CPU%c1
0.40 ? 4% turbostat.CPU%c3
13.58 turbostat.CPU%c7
64.90 turbostat.CPUGFX%
4.28 turbostat.CorWatt
38.83 turbostat.CoreTmp
64.94 turbostat.GFX%C0
34.42 turbostat.GFX%rc6
9.48 turbostat.GFXWatt
5641 ? 53% turbostat.POLL
0.01 ? 35% turbostat.POLL%
39.17 turbostat.PkgTmp
14.65 turbostat.PkgWatt
2.59 turbostat.RAMWatt
227.52 turbostat.Totl%C0
78.69 perf-stat.i.MPKI
-21.1% 3.333e+08 perf-stat.i.branch-instructions
3.93 perf-stat.i.branch-miss-rate%
+10.0% 13851976 perf-stat.i.branch-misses
-12.8% 18457066 ? 2% perf-stat.i.cache-misses
-21.3% 1.006e+08 perf-stat.i.cache-references
15699 ? 2% perf-stat.i.context-switches
1.27 perf-stat.i.cpi
-47.9% 1.823e+09 perf-stat.i.cpu-cycles
+7.8% 29.10 ? 2% perf-stat.i.cpu-migrations
-67.8% 208.52 ? 6% perf-stat.i.cycles-between-cache-misses
0.17 ? 3% perf-stat.i.dTLB-load-miss-rate%
-22.4% 477839 ? 4% perf-stat.i.dTLB-load-misses
-26.4% 3.616e+08 perf-stat.i.dTLB-loads
-33.0% 1.714e+08 perf-stat.i.dTLB-stores
-40.5% 272629 ? 14% perf-stat.i.iTLB-load-misses
-16.5% 3554676 ? 2% perf-stat.i.iTLB-loads
-22.3% 1.584e+09 perf-stat.i.instructions
0.83 perf-stat.i.ipc
2.08 ? 10% perf-stat.i.major-faults
0.15 perf-stat.i.metric.GHz
-21.7% 516.58 perf-stat.i.metric.K/sec
80.56 perf-stat.i.metric.M/sec
3211 ? 2% perf-stat.i.minor-faults
0.00 ? 12% perf-stat.i.node-load-miss-rate%
-6.0% 1363538 perf-stat.i.node-loads
0.00 ? 19% perf-stat.i.node-store-miss-rate%
-79.8% 1225203 ? 2% perf-stat.i.node-stores
3213 ? 2% perf-stat.i.page-faults
4.15 perf-stat.overall.branch-miss-rate%
18.35 ? 2% perf-stat.overall.cache-miss-rate%
1.15 perf-stat.overall.cpi
98.75 perf-stat.overall.cycles-between-cache-misses
0.87 perf-stat.overall.ipc
0.00 ? 9% perf-stat.overall.node-store-miss-rate%
-21.1% 3.31e+08 perf-stat.ps.branch-instructions
+9.9% 13751528 perf-stat.ps.branch-misses
-12.7% 18338973 ? 2% perf-stat.ps.cache-misses
-21.2% 99965770 perf-stat.ps.cache-references
15579 ? 2% perf-stat.ps.context-switches
-47.9% 1.81e+09 perf-stat.ps.cpu-cycles
+7.8% 28.90 ? 2% perf-stat.ps.cpu-migrations
-22.4% 474612 ? 4% perf-stat.ps.dTLB-load-misses
-26.4% 3.591e+08 perf-stat.ps.dTLB-loads
-33.0% 1.702e+08 perf-stat.ps.dTLB-stores
-40.5% 270782 ? 14% perf-stat.ps.iTLB-load-misses
-16.5% 3531717 ? 2% perf-stat.ps.iTLB-loads
-22.3% 1.573e+09 perf-stat.ps.instructions
2.06 ? 10% perf-stat.ps.major-faults
3187 ? 2% perf-stat.ps.minor-faults
-5.9% 1354982 perf-stat.ps.node-loads
-79.8% 1217164 ? 2% perf-stat.ps.node-stores
3189 ? 2% perf-stat.ps.page-faults
-19.6% 2.338e+11 perf-stat.total.instructions
19.38 ? 4% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
17.90 ? 5% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
3.44 ? 12% perf-profile.calltrace.cycles-pp.drm_ioctl_kernel.drm_ioctl.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe
4.07 ? 9% perf-profile.calltrace.cycles-pp.drm_ioctl.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe
4.44 ? 8% perf-profile.calltrace.cycles-pp.__x64_sys_ioctl.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.30 ?101% perf-profile.calltrace.cycles-pp.evict.__dentry_kill.dentry_kill.dput.__fput
0.30 ?101% perf-profile.calltrace.cycles-pp.shmem_evict_inode.evict.__dentry_kill.dentry_kill.dput
0.30 ?101% perf-profile.calltrace.cycles-pp.shmem_undo_range.shmem_evict_inode.evict.__dentry_kill.dentry_kill
3.05 ? 15% perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.60 ? 9% perf-profile.calltrace.cycles-pp.do_writev.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.47 ? 10% perf-profile.calltrace.cycles-pp.vfs_writev.do_writev.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.37 ? 10% perf-profile.calltrace.cycles-pp.do_iter_write.vfs_writev.do_writev.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.30 ? 10% perf-profile.calltrace.cycles-pp.do_iter_readv_writev.do_iter_write.vfs_writev.do_writev.do_syscall_64
2.27 ? 11% perf-profile.calltrace.cycles-pp.sock_write_iter.do_iter_readv_writev.do_iter_write.vfs_writev.do_writev
2.14 ? 10% perf-profile.calltrace.cycles-pp.unix_stream_sendmsg.sock_write_iter.do_iter_readv_writev.do_iter_write.vfs_writev
1.48 ? 13% perf-profile.calltrace.cycles-pp.__sys_recvmsg.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.72 ? 9% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret
1.37 ? 12% perf-profile.calltrace.cycles-pp.___sys_recvmsg.__sys_recvmsg.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.90 ? 26% perf-profile.calltrace.cycles-pp.__x64_sys_epoll_wait.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.24 ? 11% perf-profile.calltrace.cycles-pp.____sys_recvmsg.___sys_recvmsg.__sys_recvmsg.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.33 ?101% perf-profile.calltrace.cycles-pp.schedule_hrtimeout_range_clock.ep_poll.do_epoll_wait.__x64_sys_epoll_wait.do_syscall_64
0.86 ? 25% perf-profile.calltrace.cycles-pp.do_epoll_wait.__x64_sys_epoll_wait.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.19 ? 14% perf-profile.calltrace.cycles-pp.sock_recvmsg.____sys_recvmsg.___sys_recvmsg.__sys_recvmsg.do_syscall_64
0.80 ? 27% perf-profile.calltrace.cycles-pp.ep_poll.do_epoll_wait.__x64_sys_epoll_wait.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.09 ? 17% perf-profile.calltrace.cycles-pp.unix_stream_recvmsg.sock_recvmsg.____sys_recvmsg.___sys_recvmsg.__sys_recvmsg
1.05 ? 16% perf-profile.calltrace.cycles-pp.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.____sys_recvmsg.___sys_recvmsg
0.44 ? 72% perf-profile.calltrace.cycles-pp.__alloc_skb.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter
0.53 ? 47% perf-profile.calltrace.cycles-pp.alloc_skb_with_frags.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter.do_iter_readv_writev
0.72 ? 16% perf-profile.calltrace.cycles-pp.sock_alloc_send_pskb.unix_stream_sendmsg.sock_write_iter.do_iter_readv_writev.do_iter_write
0.28 ?100% perf-profile.calltrace.cycles-pp.consume_skb.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.____sys_recvmsg
0.69 ? 49% perf-profile.calltrace.cycles-pp.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.53 ? 48% perf-profile.calltrace.cycles-pp.unix_stream_recvmsg.sock_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64
0.59 ? 48% perf-profile.calltrace.cycles-pp.sock_recvmsg.__sys_recvfrom.__x64_sys_recvfrom.do_syscall_64.entry_SYSCALL_64_after_hwframe
0.52 ? 47% perf-profile.calltrace.cycles-pp.unix_stream_read_generic.unix_stream_recvmsg.sock_recvmsg.__sys_recvfrom.__x64_sys_recvfrom
0.95 ? 20% perf-profile.calltrace.cycles-pp.__entry_text_start
1.44 ? 15% perf-profile.calltrace.cycles-pp.__do_softirq.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
1.22 ? 12% perf-profile.calltrace.cycles-pp.scheduler_tick.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues
1.64 ? 12% perf-profile.calltrace.cycles-pp.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
0.65 ? 11% perf-profile.calltrace.cycles-pp.asm_exc_page_fault
0.64 ? 18% perf-profile.calltrace.cycles-pp.tick_irq_enter.irq_enter_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
0.66 ? 19% perf-profile.calltrace.cycles-pp.irq_enter_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
0.73 ? 16% perf-profile.calltrace.cycles-pp.perf_mux_hrtimer_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
2.22 ? 9% perf-profile.calltrace.cycles-pp.update_process_times.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt
0.94 ? 23% perf-profile.calltrace.cycles-pp.tick_nohz_get_sleep_length.menu_select.cpuidle_idle_call.do_idle.cpu_startup_entry
2.58 ? 8% perf-profile.calltrace.cycles-pp.tick_sched_handle.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt
2.40 ? 14% perf-profile.calltrace.cycles-pp.menu_select.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
2.16 ? 10% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
2.92 ? 10% perf-profile.calltrace.cycles-pp.tick_sched_timer.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
4.86 ? 9% perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
5.95 ? 7% perf-profile.calltrace.cycles-pp.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state
6.25 ? 7% perf-profile.calltrace.cycles-pp.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter
9.46 ? 9% perf-profile.calltrace.cycles-pp.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call
10.46 ? 11% perf-profile.calltrace.cycles-pp.asm_sysvec_apic_timer_interrupt.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
59.02 ? 8% perf-profile.calltrace.cycles-pp.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary
64.61 ? 8% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
64.67 ? 8% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64_no_verify
64.66 ? 8% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
62.12 ? 8% perf-profile.calltrace.cycles-pp.cpuidle_idle_call.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64_no_verify
50.99 ? 2% perf-profile.calltrace.cycles-pp.intel_idle_ibrs.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle
65.93 ? 2% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.cpuidle_enter.cpuidle_idle_call.do_idle.cpu_startup_entry
72.75 ? 2% perf-profile.calltrace.cycles-pp.secondary_startup_64_no_verify
19.40 ? 4% perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
17.93 ? 5% perf-profile.children.cycles-pp.do_syscall_64
3.45 ? 12% perf-profile.children.cycles-pp.drm_ioctl_kernel
4.08 ? 10% perf-profile.children.cycles-pp.drm_ioctl
4.44 ? 8% perf-profile.children.cycles-pp.__x64_sys_ioctl
1.71 ? 15% perf-profile.children.cycles-pp.shmem_read_mapping_page_gfp
1.68 ? 17% perf-profile.children.cycles-pp.shmem_get_folio_gfp
0.78 ? 14% perf-profile.children.cycles-pp.dentry_kill
0.80 ? 13% perf-profile.children.cycles-pp.__fput
0.78 ? 14% perf-profile.children.cycles-pp.__dentry_kill
0.77 ? 14% perf-profile.children.cycles-pp.evict
0.77 ? 14% perf-profile.children.cycles-pp.shmem_evict_inode
0.77 ? 14% perf-profile.children.cycles-pp.shmem_undo_range
0.80 ? 12% perf-profile.children.cycles-pp.dput
3.08 ? 14% perf-profile.children.cycles-pp.syscall_exit_to_user_mode
0.09 ? 43% perf-profile.children.cycles-pp.poll_idle
0.45 ? 15% perf-profile.children.cycles-pp.shmem_add_to_page_cache
0.32 ? 50% perf-profile.children.cycles-pp.clear_page_erms
0.47 ? 30% perf-profile.children.cycles-pp.shmem_alloc_and_acct_folio
0.41 ? 9% perf-profile.children.cycles-pp.release_pages
0.34 ? 12% perf-profile.children.cycles-pp.__pagevec_release
0.42 ? 35% perf-profile.children.cycles-pp.check_move_unevictable_pages
0.41 ? 39% perf-profile.children.cycles-pp.check_move_unevictable_folios
0.36 ? 15% perf-profile.children.cycles-pp.truncate_inode_folio
0.42 ? 35% perf-profile.children.cycles-pp.shmem_alloc_folio
0.33 ? 16% perf-profile.children.cycles-pp.filemap_remove_folio
0.38 ? 33% perf-profile.children.cycles-pp.vma_alloc_folio
0.30 ? 11% perf-profile.children.cycles-pp.__filemap_remove_folio
2.60 ? 9% perf-profile.children.cycles-pp.do_writev
2.47 ? 10% perf-profile.children.cycles-pp.vfs_writev
0.32 ? 33% perf-profile.children.cycles-pp.__folio_alloc
0.36 ? 31% perf-profile.children.cycles-pp.__alloc_pages
2.37 ? 10% perf-profile.children.cycles-pp.do_iter_write
2.32 ? 9% perf-profile.children.cycles-pp.do_iter_readv_writev
2.27 ? 11% perf-profile.children.cycles-pp.sock_write_iter
0.15 ? 19% perf-profile.children.cycles-pp.free_unref_page_list
2.14 ? 10% perf-profile.children.cycles-pp.unix_stream_sendmsg
0.29 ? 36% perf-profile.children.cycles-pp.get_page_from_freelist
0.16 ? 30% perf-profile.children.cycles-pp.xas_store
1.85 ? 14% perf-profile.children.cycles-pp.sock_recvmsg
0.22 ? 21% perf-profile.children.cycles-pp.__mod_lruvec_page_state
1.68 ? 17% perf-profile.children.cycles-pp.unix_stream_recvmsg
1.64 ? 17% perf-profile.children.cycles-pp.unix_stream_read_generic
1.49 ? 13% perf-profile.children.cycles-pp.__sys_recvmsg
0.26 ? 49% perf-profile.children.cycles-pp.folio_add_lru
2.76 ? 8% perf-profile.children.cycles-pp.syscall_return_via_sysret
0.23 ? 41% perf-profile.children.cycles-pp.rmqueue
1.37 ? 12% perf-profile.children.cycles-pp.___sys_recvmsg
0.24 ? 50% perf-profile.children.cycles-pp.folio_batch_move_lru
0.33 ? 25% perf-profile.children.cycles-pp._raw_spin_lock_irq
0.90 ? 26% perf-profile.children.cycles-pp.__x64_sys_epoll_wait
1.24 ? 11% perf-profile.children.cycles-pp.____sys_recvmsg
0.12 ? 50% perf-profile.children.cycles-pp.__mem_cgroup_charge
0.86 ? 25% perf-profile.children.cycles-pp.do_epoll_wait
0.17 ? 31% perf-profile.children.cycles-pp.__rmqueue_pcplist
0.80 ? 27% perf-profile.children.cycles-pp.ep_poll
0.12 ? 21% perf-profile.children.cycles-pp.__mod_lruvec_state
0.10 ? 22% perf-profile.children.cycles-pp.free_pcppages_bulk
0.10 ? 50% perf-profile.children.cycles-pp.lru_add_fn
0.12 ? 60% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state
0.10 ? 58% perf-profile.children.cycles-pp.rmqueue_bulk
0.72 ? 16% perf-profile.children.cycles-pp.sock_alloc_send_pskb
0.61 ? 18% perf-profile.children.cycles-pp.alloc_skb_with_frags
0.08 ? 21% perf-profile.children.cycles-pp.__mod_node_page_state
0.60 ? 19% perf-profile.children.cycles-pp.__alloc_skb
0.08 ? 81% perf-profile.children.cycles-pp.charge_memcg
0.69 ? 13% perf-profile.children.cycles-pp.consume_skb
0.12 ? 21% perf-profile.children.cycles-pp.filemap_unaccount_folio
0.70 ? 14% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.08 ? 51% perf-profile.children.cycles-pp.__free_one_page
0.04 ?101% perf-profile.children.cycles-pp.find_lock_entries
0.05 ? 81% perf-profile.children.cycles-pp.xas_load
0.76 ? 26% perf-profile.children.cycles-pp.__sys_recvfrom
0.22 ? 30% perf-profile.children.cycles-pp.vm_mmap_pgoff
0.10 ? 53% perf-profile.children.cycles-pp.__list_del_entry_valid
0.20 ? 25% perf-profile.children.cycles-pp.do_mmap
0.29 ? 22% perf-profile.children.cycles-pp.__kmem_cache_alloc_node
0.45 ? 26% perf-profile.children.cycles-pp.ep_poll_callback
0.04 ?104% perf-profile.children.cycles-pp.xas_clear_mark
0.34 ? 19% perf-profile.children.cycles-pp.unix_destruct_scm
0.34 ? 18% perf-profile.children.cycles-pp.sock_wfree
0.37 ? 27% perf-profile.children.cycles-pp.skb_release_head_state
0.27 ? 33% perf-profile.children.cycles-pp.kmalloc_reserve
0.05 ? 72% perf-profile.children.cycles-pp.xas_find_conflict
0.06 ? 80% perf-profile.children.cycles-pp.rcu_do_batch
0.05 ?107% perf-profile.children.cycles-pp.cgroup_rstat_updated
0.29 ? 17% perf-profile.children.cycles-pp.skb_release_data
0.18 ? 34% perf-profile.children.cycles-pp.mmap_region
0.25 ? 33% perf-profile.children.cycles-pp.__kmalloc_node_track_caller
0.57 ? 18% perf-profile.children.cycles-pp._raw_spin_lock
0.03 ?101% perf-profile.children.cycles-pp.folio_unlock
0.14 ? 22% perf-profile.children.cycles-pp.rcu_core
0.30 ? 21% perf-profile.children.cycles-pp.kmem_cache_alloc_node
0.24 ? 21% perf-profile.children.cycles-pp.unix_write_space
0.21 ? 34% perf-profile.children.cycles-pp.ep_autoremove_wake_function
0.31 ? 28% perf-profile.children.cycles-pp.memcg_slab_post_alloc_hook
0.05 ? 72% perf-profile.children.cycles-pp.__list_add_valid
0.04 ?105% perf-profile.children.cycles-pp.try_charge_memcg
0.06 ? 58% perf-profile.children.cycles-pp.__folio_throttle_swaprate
0.27 ? 22% perf-profile.children.cycles-pp.sock_poll
0.06 ?110% perf-profile.children.cycles-pp.do_vmi_munmap
0.17 ? 33% perf-profile.children.cycles-pp.__kmem_cache_free
0.04 ?115% perf-profile.children.cycles-pp.blk_cgroup_congested
0.13 ? 17% perf-profile.children.cycles-pp.unix_poll
0.21 ? 30% perf-profile.children.cycles-pp.skb_copy_datagram_from_iter
0.04 ?110% perf-profile.children.cycles-pp.memset_orig
0.22 ? 39% perf-profile.children.cycles-pp.rep_movs_alternative
0.18 ? 27% perf-profile.children.cycles-pp.__slab_free
0.11 ? 36% perf-profile.children.cycles-pp.copy_msghdr_from_user
0.05 ? 77% perf-profile.children.cycles-pp._copy_from_iter
0.10 ? 67% perf-profile.children.cycles-pp.mod_objcg_state
0.08 ? 65% perf-profile.children.cycles-pp.import_iovec
0.07 ? 80% perf-profile.children.cycles-pp.__import_iovec
0.19 ? 33% perf-profile.children.cycles-pp.kmem_cache_free
0.06 ? 64% perf-profile.children.cycles-pp.mas_store_prealloc
0.08 ? 30% perf-profile.children.cycles-pp.dma_resv_iter_walk_unlocked
0.09 ? 37% perf-profile.children.cycles-pp.obj_cgroup_charge
0.12 ? 16% perf-profile.children.cycles-pp.cpuidle_not_available
0.08 ? 27% perf-profile.children.cycles-pp.account_process_tick
0.10 ? 15% perf-profile.children.cycles-pp.link_path_walk
0.14 ? 20% perf-profile.children.cycles-pp.rb_insert_color
0.10 ? 46% perf-profile.children.cycles-pp.tick_nohz_tick_stopped
0.11 ? 30% perf-profile.children.cycles-pp.cpuidle_reflect
0.10 ? 36% perf-profile.children.cycles-pp.pipe_write
0.11 ? 28% perf-profile.children.cycles-pp.task_tick_idle
0.15 ? 24% perf-profile.children.cycles-pp.dup_mmap
0.30 ? 24% perf-profile.children.cycles-pp.ct_idle_exit
0.10 ? 36% perf-profile.children.cycles-pp.show
0.10 ? 36% perf-profile.children.cycles-pp.show_time_in_state
0.19 ? 18% perf-profile.children.cycles-pp.copy_process
0.17 ? 21% perf-profile.children.cycles-pp.dup_mm
0.11 ? 34% perf-profile.children.cycles-pp.sysfs_kf_seq_show
0.13 ? 28% perf-profile.children.cycles-pp.do_futex
0.12 ? 30% perf-profile.children.cycles-pp.futex_wait
0.16 ? 44% perf-profile.children.cycles-pp.ct_kernel_exit_state
0.14 ? 35% perf-profile.children.cycles-pp.ksys_mmap_pgoff
0.16 ? 20% perf-profile.children.cycles-pp.error_entry
0.13 ? 34% perf-profile.children.cycles-pp.ct_kernel_exit
0.31 ? 24% perf-profile.children.cycles-pp.irqtime_account_irq
0.14 ? 31% perf-profile.children.cycles-pp.__x64_sys_futex
0.21 ? 22% perf-profile.children.cycles-pp.__do_sys_clone
0.21 ? 22% perf-profile.children.cycles-pp.kernel_clone
0.13 ? 58% perf-profile.children.cycles-pp.__irqentry_text_end
0.16 ? 41% perf-profile.children.cycles-pp.irqentry_exit_to_user_mode
0.12 ? 45% perf-profile.children.cycles-pp.traverse
0.15 ? 45% perf-profile.children.cycles-pp.vfs_write
0.24 ? 17% perf-profile.children.cycles-pp.__intel_pmu_enable_all
0.31 ? 26% perf-profile.children.cycles-pp.find_busiest_group
0.17 ? 45% perf-profile.children.cycles-pp.ksys_write
0.13 ? 41% perf-profile.children.cycles-pp.vsnprintf
0.14 ? 27% perf-profile.children.cycles-pp.pipe_read
0.19 ? 18% perf-profile.children.cycles-pp.hrtimer_update_next_event
0.58 ? 13% perf-profile.children.cycles-pp.perf_rotate_context
0.34 ? 18% perf-profile.children.cycles-pp.sched_clock_cpu
0.20 ? 27% perf-profile.children.cycles-pp.do_filp_open
0.20 ? 27% perf-profile.children.cycles-pp.path_openat
0.22 ? 50% perf-profile.children.cycles-pp.do_read_fault
0.28 ? 45% perf-profile.children.cycles-pp.lapic_next_deadline
0.24 ? 22% perf-profile.children.cycles-pp.native_apic_msr_eoi_write
0.21 ? 56% perf-profile.children.cycles-pp.filemap_map_pages
0.33 ? 11% perf-profile.children.cycles-pp.native_irq_return_iret
0.23 ? 18% perf-profile.children.cycles-pp.__x64_sys_openat
0.31 ? 33% perf-profile.children.cycles-pp.__hrtimer_next_event_base
0.23 ? 18% perf-profile.children.cycles-pp.do_sys_openat2
0.26 ? 26% perf-profile.children.cycles-pp.__mmput
0.26 ? 26% perf-profile.children.cycles-pp.exit_mmap
0.46 ? 23% perf-profile.children.cycles-pp.__handle_mm_fault
0.34 ? 29% perf-profile.children.cycles-pp.hrtimer_next_event_without
0.25 ? 15% perf-profile.children.cycles-pp.wait_for_xmitr
0.48 ? 25% perf-profile.children.cycles-pp.handle_mm_fault
0.31 ? 39% perf-profile.children.cycles-pp.arch_scale_freq_tick
0.31 ? 31% perf-profile.children.cycles-pp.load_elf_binary
0.57 ? 20% perf-profile.children.cycles-pp.exc_page_fault
0.57 ? 19% perf-profile.children.cycles-pp.do_user_addr_fault
0.62 ? 34% perf-profile.children.cycles-pp.rebalance_domains
0.32 ? 33% perf-profile.children.cycles-pp.search_binary_handler
0.33 ? 31% perf-profile.children.cycles-pp.bprm_execve
0.33 ? 32% perf-profile.children.cycles-pp.exec_binprm
0.26 ? 49% perf-profile.children.cycles-pp.seq_read_iter
0.72 ? 14% perf-profile.children.cycles-pp.irq_enter_rcu
0.71 ? 14% perf-profile.children.cycles-pp.tick_irq_enter
0.51 ? 15% perf-profile.children.cycles-pp.read_tsc
0.41 ? 27% perf-profile.children.cycles-pp.__x64_sys_execve
0.41 ? 27% perf-profile.children.cycles-pp.do_execveat_common
0.70 ? 11% perf-profile.children.cycles-pp.asm_exc_page_fault
0.44 ? 30% perf-profile.children.cycles-pp.clockevents_program_event
0.83 ? 18% perf-profile.children.cycles-pp.perf_mux_hrtimer_handler
0.48 ? 35% perf-profile.children.cycles-pp.vfs_read
0.51 ? 32% perf-profile.children.cycles-pp.ksys_read
0.82 ? 15% perf-profile.children.cycles-pp.ktime_get
2.00 ? 10% perf-profile.children.cycles-pp.__irq_exit_rcu
1.47 ? 11% perf-profile.children.cycles-pp.scheduler_tick
1.03 ? 27% perf-profile.children.cycles-pp.tick_nohz_get_sleep_length
2.69 ? 10% perf-profile.children.cycles-pp.update_process_times
3.03 ? 9% perf-profile.children.cycles-pp.tick_sched_handle
2.22 ? 10% perf-profile.children.cycles-pp.intel_idle
2.56 ? 17% perf-profile.children.cycles-pp.menu_select
3.43 ? 7% perf-profile.children.cycles-pp.tick_sched_timer
5.56 ? 7% perf-profile.children.cycles-pp.__hrtimer_run_queues
6.63 ? 6% perf-profile.children.cycles-pp.hrtimer_interrupt
7.00 ? 6% perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
10.44 ? 7% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
11.31 ? 7% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
64.67 ? 8% perf-profile.children.cycles-pp.start_secondary
51.00 ? 2% perf-profile.children.cycles-pp.intel_idle_ibrs
66.55 ? 2% perf-profile.children.cycles-pp.cpuidle_enter_state
66.63 ? 2% perf-profile.children.cycles-pp.cpuidle_enter
72.75 ? 2% perf-profile.children.cycles-pp.secondary_startup_64_no_verify
72.75 ? 2% perf-profile.children.cycles-pp.cpu_startup_entry
72.73 ? 2% perf-profile.children.cycles-pp.do_idle
70.00 ? 2% perf-profile.children.cycles-pp.cpuidle_idle_call
0.08 ? 59% perf-profile.self.cycles-pp.poll_idle
0.15 ? 37% perf-profile.self.cycles-pp.shmem_get_folio_gfp
0.28 ? 40% perf-profile.self.cycles-pp.clear_page_erms
2.49 ? 15% perf-profile.self.cycles-pp.syscall_exit_to_user_mode
2.76 ? 8% perf-profile.self.cycles-pp.syscall_return_via_sysret
0.15 ? 38% perf-profile.self.cycles-pp.check_move_unevictable_folios
0.07 ? 71% perf-profile.self.cycles-pp.xas_store
0.09 ? 62% perf-profile.self.cycles-pp.release_pages
1.50 ? 3% perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
0.09 ? 83% perf-profile.self.cycles-pp.shmem_add_to_page_cache
0.08 ? 21% perf-profile.self.cycles-pp.__mod_node_page_state
0.09 ? 44% perf-profile.self.cycles-pp.lru_add_fn
0.08 ? 82% perf-profile.self.cycles-pp.rmqueue_bulk
0.09 ? 66% perf-profile.self.cycles-pp.__mod_memcg_lruvec_state
0.06 ? 83% perf-profile.self.cycles-pp.__mod_lruvec_page_state
0.06 ? 48% perf-profile.self.cycles-pp.__free_one_page
0.10 ? 50% perf-profile.self.cycles-pp.__list_del_entry_valid
0.04 ?105% perf-profile.self.cycles-pp.shmem_alloc_folio
0.03 ?100% perf-profile.self.cycles-pp.__filemap_remove_folio
0.03 ?101% perf-profile.self.cycles-pp.vma_alloc_folio
0.04 ?112% perf-profile.self.cycles-pp.__alloc_pages
0.16 ? 35% perf-profile.self.cycles-pp._raw_spin_lock_irq
0.64 ? 13% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.05 ?107% perf-profile.self.cycles-pp.cgroup_rstat_updated
0.56 ? 19% perf-profile.self.cycles-pp._raw_spin_lock
0.05 ? 72% perf-profile.self.cycles-pp.__list_add_valid
0.26 ? 29% perf-profile.self.cycles-pp.memcg_slab_post_alloc_hook
0.04 ?110% perf-profile.self.cycles-pp.memset_orig
0.16 ? 20% perf-profile.self.cycles-pp.__slab_free
0.03 ?101% perf-profile.self.cycles-pp.try_charge_memcg
0.10 ? 50% perf-profile.self.cycles-pp.kmem_cache_free
0.04 ?107% perf-profile.self.cycles-pp.__import_iovec
0.10 ? 30% perf-profile.self.cycles-pp.sock_wfree
0.04 ?112% perf-profile.self.cycles-pp.tick_nohz_stop_idle
0.03 ?100% perf-profile.self.cycles-pp.filemap_unaccount_folio
0.08 ? 27% perf-profile.self.cycles-pp.account_process_tick
0.12 ? 16% perf-profile.self.cycles-pp.cpuidle_not_available
0.22 ? 15% perf-profile.self.cycles-pp.sysvec_apic_timer_interrupt
0.09 ? 48% perf-profile.self.cycles-pp.tick_nohz_tick_stopped
0.10 ? 36% perf-profile.self.cycles-pp.cpuidle_reflect
0.15 ? 29% perf-profile.self.cycles-pp.update_sg_lb_stats
0.14 ? 20% perf-profile.self.cycles-pp.rb_insert_color
0.11 ? 28% perf-profile.self.cycles-pp.task_tick_idle
0.10 ? 20% perf-profile.self.cycles-pp.tick_sched_timer
0.16 ? 45% perf-profile.self.cycles-pp.ct_kernel_exit_state
0.16 ? 20% perf-profile.self.cycles-pp.error_entry
0.13 ? 58% perf-profile.self.cycles-pp.__irqentry_text_end
0.24 ? 17% perf-profile.self.cycles-pp.__intel_pmu_enable_all
0.18 ? 47% perf-profile.self.cycles-pp.tick_nohz_next_event
0.28 ? 38% perf-profile.self.cycles-pp.__hrtimer_next_event_base
0.23 ? 21% perf-profile.self.cycles-pp.native_apic_msr_eoi_write
0.28 ? 45% perf-profile.self.cycles-pp.lapic_next_deadline
0.40 ? 33% perf-profile.self.cycles-pp.ktime_get
0.33 ? 11% perf-profile.self.cycles-pp.native_irq_return_iret
0.31 ? 39% perf-profile.self.cycles-pp.arch_scale_freq_tick
0.50 ? 16% perf-profile.self.cycles-pp.read_tsc
1.03 ? 14% perf-profile.self.cycles-pp.cpuidle_enter_state
2.22 ? 10% perf-profile.self.cycles-pp.intel_idle
50.98 ? 2% perf-profile.self.cycles-pp.intel_idle_ibrs
estimated based on internal Intel analysis and are provided
purposes only. Any difference in system hardware or software
may affect actual performance.
om/intel/lkp-tests/wiki">https://github.com/intel/lkp-tests/wiki

Attachments:

(No filename) (240.61 kB)
config-6.4.0-rc2-00007-g278dfbd17c5d (160.86 kB)
job-script (7.98 kB)
job.yaml (5.37 kB)
reproduce (313.00 B)
Download all attachments

2023-05-24 17:56:38

by Rafael J. Wysocki

[permalink] [raw]

Subject: Re: [PATCH v4 1/3] amd_pstate: Add ->fast_switch() callback

On Wed, May 17, 2023 at 6:29 PM Wyes Karny <[email protected]> wrote:
>
> From: "Gautham R. Shenoy" <[email protected]>
>
> Schedutil normally calls the adjust_perf callback for drivers with
> adjust_perf callback available and fast_switch_possible flag set.
> However, when frequency invariance is disabled and schedutil tries to
> invoke fast_switch. So, there is a chance of kernel crash if this
> function pointer is not set. To protect against this scenario add
> fast_switch callback to amd_pstate driver.
>
> Fixes: 1d215f0319c2 ("cpufreq: amd-pstate: Add fast switch function for AMD P-State")
>
> Signed-off-by: Wyes Karny <[email protected]>
> Signed-off-by: Gautham R. Shenoy <[email protected]>
> ---
> drivers/cpufreq/amd-pstate.c | 36 ++++++++++++++++++++++++++++++------
> 1 file changed, 30 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
> index 5a3d4aa0f45a..45711fc0a856 100644
> --- a/drivers/cpufreq/amd-pstate.c
> +++ b/drivers/cpufreq/amd-pstate.c
> @@ -444,9 +444,8 @@ static int amd_pstate_verify(struct cpufreq_policy_data *policy)
> return 0;
> }
>
> -static int amd_pstate_target(struct cpufreq_policy *policy,
> - unsigned int target_freq,
> - unsigned int relation)
> +static int amd_pstate_update_freq(struct cpufreq_policy *policy,
> + unsigned int target_freq, bool fast_switch)
> {
> struct cpufreq_freqs freqs;
> struct amd_cpudata *cpudata = policy->driver_data;
> @@ -465,14 +464,37 @@ static int amd_pstate_target(struct cpufreq_policy *policy,
> des_perf = DIV_ROUND_CLOSEST(target_freq * cap_perf,
> cpudata->max_freq);
>
> - cpufreq_freq_transition_begin(policy, &freqs);
> + WARN_ON(fast_switch && !policy->fast_switch_enabled);
> + /*
> + * If fast_switch is desired, then there aren't any registered
> + * transition notifiers. See comment for
> + * cpufreq_enable_fast_switch().
> + */
> + if (!fast_switch)
> + cpufreq_freq_transition_begin(policy, &freqs);
> +
> amd_pstate_update(cpudata, min_perf, des_perf,
> - max_perf, false, policy->governor->flags);
> - cpufreq_freq_transition_end(policy, &freqs, false);
> + max_perf, fast_switch, policy->governor->flags);
> +
> + if (!fast_switch)
> + cpufreq_freq_transition_end(policy, &freqs, false);
>
> return 0;
> }
>
> +static int amd_pstate_target(struct cpufreq_policy *policy,
> + unsigned int target_freq,
> + unsigned int relation)
> +{
> + return amd_pstate_update_freq(policy, target_freq, false);
> +}
> +
> +static unsigned int amd_pstate_fast_switch(struct cpufreq_policy *policy,
> + unsigned int target_freq)
> +{
> + return amd_pstate_update_freq(policy, target_freq, true);
> +}
> +
> static void amd_pstate_adjust_perf(unsigned int cpu,
> unsigned long _min_perf,
> unsigned long target_perf,
> @@ -715,6 +737,7 @@ static int amd_pstate_cpu_exit(struct cpufreq_policy *policy)
>
> freq_qos_remove_request(&cpudata->req[1]);
> freq_qos_remove_request(&cpudata->req[0]);
> + policy->fast_switch_possible = false;
> kfree(cpudata);
>
> return 0;
> @@ -1309,6 +1332,7 @@ static struct cpufreq_driver amd_pstate_driver = {
> .flags = CPUFREQ_CONST_LOOPS | CPUFREQ_NEED_UPDATE_LIMITS,
> .verify = amd_pstate_verify,
> .target = amd_pstate_target,
> + .fast_switch = amd_pstate_fast_switch,
> .init = amd_pstate_cpu_init,
> .exit = amd_pstate_cpu_exit,
> .suspend = amd_pstate_cpu_suspend,
> --

Applied along with the [2/3], thanks!

Do you need them in 6.4 or would 6.5 be sufficient? Also do you need
them to go into "stable"?

2023-05-24 18:18:48

by Rafael J. Wysocki

[permalink] [raw]

Subject: Re: [PATCH v4 3/3] cpufreq: Return failure if fast_switch is not set and fast_switch_possible is set

On Wed, May 17, 2023 at 6:30 PM Wyes Karny <[email protected]> wrote:
>
> If fast_switch_possible flag is set by the scaling driver, the governor
> is free to select fast_switch function even if adjust_perf is set. Some
> scaling drivers which use adjust_perf don't set fast_switch thinking
> that the governor would never fall back to fast_switch. But the governor
> can fall back to fast_switch even in runtime if frequency invariance is
> disabled due to some reason. This could crash the kernel if the driver
> didn't set the fast_switch function pointer.
>
> Therefore, return failure in cpufreq_online function if fast_switch is
> not set and fast_switch_possible is set.
>
> Signed-off-by: Wyes Karny <[email protected]>
> ---
> drivers/cpufreq/cpufreq.c | 5 +++++
> include/linux/cpufreq.h | 4 +++-
> 2 files changed, 8 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index 6b52ebe5a890..7835ba4fa34c 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -1376,6 +1376,11 @@ static int cpufreq_online(unsigned int cpu)
> goto out_free_policy;
> }
>
> + if (policy->fast_switch_possible && !cpufreq_driver->fast_switch) {
> + pr_err("fast_switch_possible is enabled but fast_switch callback is not set\n");
> + ret = -EINVAL;
> + goto out_destroy_policy;
> + }

The driver registration can fail if the driver has ->adjust_perf
without ->fast_switch. Then the check above would not be necessary
any more.

> /*
> * The initialization has succeeded and the policy is online.
> * If there is a problem with its frequency table, take it
> diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> index 26e2eb399484..8cdf77bb3bc1 100644
> --- a/include/linux/cpufreq.h
> +++ b/include/linux/cpufreq.h
> @@ -340,7 +340,9 @@ struct cpufreq_driver {
> /*
> * ->fast_switch() replacement for drivers that use an internal
> * representation of performance levels and can pass hints other than
> - * the target performance level to the hardware.
> + * the target performance level to the hardware. If driver is setting this,
> + * then it needs to set fast_switch also. Because in certain scenario scale
> + * invariance could be disabled and governor can switch back to fast_switch.

I would say something like "This can only be set if ->fast_switch is
set too, because in those cases (under specific conditions) scale
invariance can be disabled, which causes the schedutil governor to
fall back to the latter."

> */
> void (*adjust_perf)(unsigned int cpu,
> unsigned long min_perf,
> --
> 2.34.1
>

2023-05-29 07:03:51

by Wyes Karny

[permalink] [raw]

Subject: Re: [PATCH v4 1/3] amd_pstate: Add ->fast_switch() callback

Hi Rafael,

On 24 May 19:47, Rafael J. Wysocki wrote:
> On Wed, May 17, 2023 at 6:29 PM Wyes Karny <[email protected]> wrote:
> >
> > From: "Gautham R. Shenoy" <[email protected]>
> >
> > Schedutil normally calls the adjust_perf callback for drivers with
> > adjust_perf callback available and fast_switch_possible flag set.
> > However, when frequency invariance is disabled and schedutil tries to
> > invoke fast_switch. So, there is a chance of kernel crash if this
> > function pointer is not set. To protect against this scenario add
> > fast_switch callback to amd_pstate driver.
> >
> > Fixes: 1d215f0319c2 ("cpufreq: amd-pstate: Add fast switch function for AMD P-State")
> >
> > Signed-off-by: Wyes Karny <[email protected]>
> > Signed-off-by: Gautham R. Shenoy <[email protected]>
> > ---
> > drivers/cpufreq/amd-pstate.c | 36 ++++++++++++++++++++++++++++++------
> > 1 file changed, 30 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
> > index 5a3d4aa0f45a..45711fc0a856 100644
> > --- a/drivers/cpufreq/amd-pstate.c
> > +++ b/drivers/cpufreq/amd-pstate.c
> > @@ -444,9 +444,8 @@ static int amd_pstate_verify(struct cpufreq_policy_data *policy)
> > return 0;
> > }
> >
> > -static int amd_pstate_target(struct cpufreq_policy *policy,
> > - unsigned int target_freq,
> > - unsigned int relation)
> > +static int amd_pstate_update_freq(struct cpufreq_policy *policy,
> > + unsigned int target_freq, bool fast_switch)
> > {
> > struct cpufreq_freqs freqs;
> > struct amd_cpudata *cpudata = policy->driver_data;
> > @@ -465,14 +464,37 @@ static int amd_pstate_target(struct cpufreq_policy *policy,
> > des_perf = DIV_ROUND_CLOSEST(target_freq * cap_perf,
> > cpudata->max_freq);
> >
> > - cpufreq_freq_transition_begin(policy, &freqs);
> > + WARN_ON(fast_switch && !policy->fast_switch_enabled);
> > + /*
> > + * If fast_switch is desired, then there aren't any registered
> > + * transition notifiers. See comment for
> > + * cpufreq_enable_fast_switch().
> > + */
> > + if (!fast_switch)
> > + cpufreq_freq_transition_begin(policy, &freqs);
> > +
> > amd_pstate_update(cpudata, min_perf, des_perf,
> > - max_perf, false, policy->governor->flags);
> > - cpufreq_freq_transition_end(policy, &freqs, false);
> > + max_perf, fast_switch, policy->governor->flags);
> > +
> > + if (!fast_switch)
> > + cpufreq_freq_transition_end(policy, &freqs, false);
> >
> > return 0;
> > }
> >
> > +static int amd_pstate_target(struct cpufreq_policy *policy,
> > + unsigned int target_freq,
> > + unsigned int relation)
> > +{
> > + return amd_pstate_update_freq(policy, target_freq, false);
> > +}
> > +
> > +static unsigned int amd_pstate_fast_switch(struct cpufreq_policy *policy,
> > + unsigned int target_freq)
> > +{
> > + return amd_pstate_update_freq(policy, target_freq, true);
> > +}
> > +
> > static void amd_pstate_adjust_perf(unsigned int cpu,
> > unsigned long _min_perf,
> > unsigned long target_perf,
> > @@ -715,6 +737,7 @@ static int amd_pstate_cpu_exit(struct cpufreq_policy *policy)
> >
> > freq_qos_remove_request(&cpudata->req[1]);
> > freq_qos_remove_request(&cpudata->req[0]);
> > + policy->fast_switch_possible = false;
> > kfree(cpudata);
> >
> > return 0;
> > @@ -1309,6 +1332,7 @@ static struct cpufreq_driver amd_pstate_driver = {
> > .flags = CPUFREQ_CONST_LOOPS | CPUFREQ_NEED_UPDATE_LIMITS,
> > .verify = amd_pstate_verify,
> > .target = amd_pstate_target,
> > + .fast_switch = amd_pstate_fast_switch,
> > .init = amd_pstate_cpu_init,
> > .exit = amd_pstate_cpu_exit,
> > .suspend = amd_pstate_cpu_suspend,
> > --
>
> Applied along with the [2/3], thanks!
>
> Do you need them in 6.4 or would 6.5 be sufficient? Also do you need
> them to go into "stable"?

Sorry for late reply.
Thanks for picking this for 6.4 and stable.
I see the patch is not applying to the 6.3-stable tree.
I'll check that.

Thanks & Regards,
Wyes

2023-05-29 14:26:48

by Wyes Karny

[permalink] [raw]

Subject: Re: [PATCH v4 3/3] cpufreq: Return failure if fast_switch is not set and fast_switch_possible is set

Hi Rafael,

On 24 May 19:44, Rafael J. Wysocki wrote:
> On Wed, May 17, 2023 at 6:30 PM Wyes Karny <[email protected]> wrote:
> >
> > If fast_switch_possible flag is set by the scaling driver, the governor
> > is free to select fast_switch function even if adjust_perf is set. Some
> > scaling drivers which use adjust_perf don't set fast_switch thinking
> > that the governor would never fall back to fast_switch. But the governor
> > can fall back to fast_switch even in runtime if frequency invariance is
> > disabled due to some reason. This could crash the kernel if the driver
> > didn't set the fast_switch function pointer.
> >
> > Therefore, return failure in cpufreq_online function if fast_switch is
> > not set and fast_switch_possible is set.
> >
> > Signed-off-by: Wyes Karny <[email protected]>
> > ---
> > drivers/cpufreq/cpufreq.c | 5 +++++
> > include/linux/cpufreq.h | 4 +++-
> > 2 files changed, 8 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> > index 6b52ebe5a890..7835ba4fa34c 100644
> > --- a/drivers/cpufreq/cpufreq.c
> > +++ b/drivers/cpufreq/cpufreq.c
> > @@ -1376,6 +1376,11 @@ static int cpufreq_online(unsigned int cpu)
> > goto out_free_policy;
> > }
> >
> > + if (policy->fast_switch_possible && !cpufreq_driver->fast_switch) {
> > + pr_err("fast_switch_possible is enabled but fast_switch callback is not set\n");
> > + ret = -EINVAL;
> > + goto out_destroy_policy;
> > + }
>
> The driver registration can fail if the driver has ->adjust_perf
> without ->fast_switch. Then the check above would not be necessary
> any more.

Sure. Will do that.

>
> > /*
> > * The initialization has succeeded and the policy is online.
> > * If there is a problem with its frequency table, take it
> > diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> > index 26e2eb399484..8cdf77bb3bc1 100644
> > --- a/include/linux/cpufreq.h
> > +++ b/include/linux/cpufreq.h
> > @@ -340,7 +340,9 @@ struct cpufreq_driver {
> > /*
> > * ->fast_switch() replacement for drivers that use an internal
> > * representation of performance levels and can pass hints other than
> > - * the target performance level to the hardware.
> > + * the target performance level to the hardware. If driver is setting this,
> > + * then it needs to set fast_switch also. Because in certain scenario scale
> > + * invariance could be disabled and governor can switch back to fast_switch.
>
> I would say something like "This can only be set if ->fast_switch is
> set too, because in those cases (under specific conditions) scale
> invariance can be disabled, which causes the schedutil governor to
> fall back to the latter."

Sure. Will update and send the updated patch in-reply-to this.

Thanks & Regards,
Wyes

>
> > */
> > void (*adjust_perf)(unsigned int cpu,
> > unsigned long min_perf,
> > --
> > 2.34.1
> >

2023-05-29 14:35:21

by Wyes Karny

[permalink] [raw]

Subject: [PATCH v4.1] cpufreq: Fail driver register if it has adjust_perf without fast_switch

If fast_switch_possible flag is set by the scaling driver, the governor
is free to select fast_switch function even if adjust_perf is set. Some
scaling drivers which use adjust_perf don't set fast_switch thinking
that the governor would never fall back to fast_switch. But the governor
can fall back to fast_switch even in runtime if frequency invariance is
disabled due to some reason. This could crash the kernel if the driver
didn't set the fast_switch function pointer.

Therefore, fail driver registration if it has adjust_perf without
fast_switch.

Suggested-by: Rafael J. Wysocki <[email protected]>
Suggested-by: Viresh Kumar <[email protected]>
Signed-off-by: Wyes Karny <[email protected]>
---
drivers/cpufreq/cpufreq.c | 3 ++-
include/linux/cpufreq.h | 5 ++++-
2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 6b52ebe5a890..50bbc969ffe5 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -2828,7 +2828,8 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
(driver_data->setpolicy && (driver_data->target_index ||
driver_data->target)) ||
(!driver_data->get_intermediate != !driver_data->target_intermediate) ||
- (!driver_data->online != !driver_data->offline))
+ (!driver_data->online != !driver_data->offline) ||
+ (driver_data->adjust_perf && !driver_data->fast_switch))
return -EINVAL;

pr_debug("trying to register driver %s\n", driver_data->name);
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 26e2eb399484..172ff51c1b2a 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -340,7 +340,10 @@ struct cpufreq_driver {
/*
* ->fast_switch() replacement for drivers that use an internal
* representation of performance levels and can pass hints other than
- * the target performance level to the hardware.
+ * the target performance level to the hardware. This can only be set
+ * if ->fast_switch is set too, because in those cases (under specific
+ * conditions) scale invariance can be disabled, which causes the
+ * schedutil governor to fall back to the latter.
*/
void (*adjust_perf)(unsigned int cpu,
unsigned long min_perf,
--
2.34.1

2023-06-15 16:44:39

by Rafael J. Wysocki

[permalink] [raw]

Subject: Re: [PATCH v4.1] cpufreq: Fail driver register if it has adjust_perf without fast_switch

On Mon, May 29, 2023 at 4:26 PM Wyes Karny <[email protected]> wrote:
>
> If fast_switch_possible flag is set by the scaling driver, the governor
> is free to select fast_switch function even if adjust_perf is set. Some
> scaling drivers which use adjust_perf don't set fast_switch thinking
> that the governor would never fall back to fast_switch. But the governor
> can fall back to fast_switch even in runtime if frequency invariance is
> disabled due to some reason. This could crash the kernel if the driver
> didn't set the fast_switch function pointer.
>
> Therefore, fail driver registration if it has adjust_perf without
> fast_switch.
>
> Suggested-by: Rafael J. Wysocki <[email protected]>
> Suggested-by: Viresh Kumar <[email protected]>
> Signed-off-by: Wyes Karny <[email protected]>
> ---
> drivers/cpufreq/cpufreq.c | 3 ++-
> include/linux/cpufreq.h | 5 ++++-
> 2 files changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index 6b52ebe5a890..50bbc969ffe5 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -2828,7 +2828,8 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
> (driver_data->setpolicy && (driver_data->target_index ||
> driver_data->target)) ||
> (!driver_data->get_intermediate != !driver_data->target_intermediate) ||
> - (!driver_data->online != !driver_data->offline))
> + (!driver_data->online != !driver_data->offline) ||
> + (driver_data->adjust_perf && !driver_data->fast_switch))
> return -EINVAL;
>
> pr_debug("trying to register driver %s\n", driver_data->name);
> diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> index 26e2eb399484..172ff51c1b2a 100644
> --- a/include/linux/cpufreq.h
> +++ b/include/linux/cpufreq.h
> @@ -340,7 +340,10 @@ struct cpufreq_driver {
> /*
> * ->fast_switch() replacement for drivers that use an internal
> * representation of performance levels and can pass hints other than
> - * the target performance level to the hardware.
> + * the target performance level to the hardware. This can only be set
> + * if ->fast_switch is set too, because in those cases (under specific
> + * conditions) scale invariance can be disabled, which causes the
> + * schedutil governor to fall back to the latter.
> */
> void (*adjust_perf)(unsigned int cpu,
> unsigned long min_perf,
> --

Applied as 6.5 material, thanks!