lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e4730b8c68bea16ac9135293b82828012d2206a7.camel@linux.intel.com>
Date:   Thu, 27 Aug 2020 12:52:31 -0700
From:   Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
To:     "Rafael J. Wysocki" <rjw@...ysocki.net>,
        Linux PM <linux-pm@...r.kernel.org>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Doug Smythies <dsmythies@...us.net>,
        Artem Bityutskiy <artem.bityutskiy@...ux.intel.com>
Subject: Re: [PATCH v3 0/5] cpufreq: intel_pstate: Address some HWP-related
 oddities

On Thu, 2020-08-27 at 17:08 +0200, Rafael J. Wysocki wrote:
> Hi All,
> 
> The v3 is here to address feedback from Srinivas and Doug, and Artem
> concerns
> regarding the EPP sysfs interface changes.
> 
> The purpose of this series is to address some peculiarities related
> to
> taking CPUs offline/online and switching between different operation
> modes with HWP enabled that have become visible after allowing the
> driver to work in the passive mode with HWP enabled in 5.9-rc1 (and
> one that was there earlier, but can be addressed easily after the
> changes made in 5.9-rc1).
> 
> Please refer to the patch changelogs for details.
> 
> For easier testing/review, the series is available from the git
> branch at:
> 
>  git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git \
>  intel_pstate-testing
> 
> I've done my best to address all of the possible corner cases, but
> the test
> matrix is quite extensive and I may have missed something, so go
> ahead
> and test.

Unchecked MSR access (attached dmesg) then all HWP settings goes to 0s
except CPU 0 after resume after S3.

Full log of my console (full copy paste)

[labuser@...pl-perf-test-skx-i9 ~]$ sudo -s
[root@...pl-perf-test-skx-i9 labuser]# cd
/sys/devices/system/cpu/intel_pstate/
[root@...pl-perf-test-skx-i9 intel_pstate]# grep . *
hwp_dynamic_boost:0
max_perf_pct:100
min_perf_pct:27
no_turbo:0
num_pstates:32
status:active
turbo_pct:32
[root@...pl-perf-test-skx-i9 intel_pstate]# cd ../cpu0/cpufreq/
[root@...pl-perf-test-skx-i9 cpufreq]# grep . *
affected_cpus:0
base_frequency:3300000
cpuinfo_max_freq:4300000
cpuinfo_min_freq:1200000
cpuinfo_transition_latency:0
energy_performance_available_preferences:default performance
balance_performance balance_power power 
energy_performance_preference:balance_performance
related_cpus:0
scaling_available_governors:performance powersave
scaling_cur_freq:1199695
scaling_driver:intel_pstate
scaling_governor:powersave
scaling_max_freq:4300000
scaling_min_freq:1200000
scaling_setspeed:<unsupported>
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -a 0x771
70c212b
70c212b
70c212b
70c212d
70c212b
70c212b
70c212b
70c212b
70c212b
70c212d
70c212b
70c212b
70c212b
70c212d
70c212b
70c212b
70c212b
70c212b
70c212b
70c212d
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -a 0x774
80002b0c
80002b0c
80002b0c
80002d0c
80002b0c
80002b0c
80002b0c
80002b0c
80002b0c
80002d0c
80002b0c
80002b0c
80002b0c
80002d0c
80002b0c
80002b0c
80002b0c
80002b0c
80002b0c
80002d0d
[root@...pl-perf-test-skx-i9 cpufreq]# echo 4000000 > scaling_max_freq 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 9x774
rdmsr: CPU 0 cannot read MSR 0x00000009
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
8000280c
[root@...pl-perf-test-skx-i9 cpufreq]# echo 4200000 > scaling_max_freq 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
80002a0c
[root@...pl-perf-test-skx-i9 cpufreq]# echo 4300000 > scaling_max_freq 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
80002b0c
[root@...pl-perf-test-skx-i9 cpufreq]# echo 4000000 > scaling_max_freq 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
8000280c
[root@...pl-perf-test-skx-i9 cpufreq]# echo 127 >
energy_performance_preference 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
7f00280c
[root@...pl-perf-test-skx-i9 cpufreq]# echo 0 > ../../
cpu0/            cpu13/           cpu18/           cpu5/            cpu
freq/         kernel_max       online           uevent
cpu1/            cpu14/           cpu19/           cpu6/            cpu
idle/         microcode/       possible         vulnerabilities/
cpu10/           cpu15/           cpu2/            cpu7/            hot
plug/         modalias         power/           
cpu11/           cpu16/           cpu3/            cpu8/            int
el_pstate/    nohz_full        present          
cpu12/           cpu17/           cpu4/            cpu9/            iso
lated         offline          smt/             
[root@...pl-perf-test-skx-i9 cpufreq]# echo 0 > ../
acpi_cppc/        cpuidle/          driver/           microcode/       
 subsystem/        uevent            
cache/            crash_notes       firmware_node/    node0/           
 thermal_throttle/ 
cpufreq/          crash_notes_size  hotplug/          power/           
 topology/         
[root@...pl-perf-test-skx-i9 cpufreq]# echo 128 >
energy_performance_preference 
[root@...pl-perf-test-skx-i9 cpufreq]# echo 4300000 > scaling_max_freq 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
80002b0c
[root@...pl-perf-test-skx-i9 cpufreq]# cd ../../cpu1
cpu1/  cpu10/ cpu11/ cpu12/ cpu13/ cpu14/ cpu15/ cpu16/ cpu17/ cpu18/
cpu19/ 
[root@...pl-perf-test-skx-i9 cpufreq]# cd ../../cpu1
[root@...pl-perf-test-skx-i9 cpu1]# cd cpufreq/
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
80002b0c
[root@...pl-perf-test-skx-i9 cpufreq]# echo 127 >
energy_performance_preference 
[root@...pl-perf-test-skx-i9 cpufreq]# echo 4000000 > scaling_max_freq 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 0 0x774
80002b0c
[root@...pl-perf-test-skx-i9 cpufreq]# cat scaling_max_freq 
4000000
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 1 0x774
7f00280c
[root@...pl-perf-test-skx-i9 cpufreq]# echo 0 > ../online 
bash: ../online: Permission denied
[root@...pl-perf-test-skx-i9 cpufreq]# echo 0 > ../online 
affected_cpus                             energy_performance_available_
preferences  scaling_driver
base_frequency                            energy_performance_preference
             scaling_governor
cpuinfo_max_freq                          related_cpus                 
             scaling_max_freq
cpuinfo_min_freq                          scaling_available_governors  
             scaling_min_freq
cpuinfo_transition_latency                scaling_cur_freq             
             scaling_setspeed
[root@...pl-perf-test-skx-i9 cpufreq]# cat ../
cat: ../: Is a directory
[root@...pl-perf-test-skx-i9 cpufreq]# cat ../online 
cat: ../online: No such file or directory
[root@...pl-perf-test-skx-i9 cpufreq]# ls ../../
cpu0   cpu11  cpu14  cpu17  cpu2  cpu5  cpu8     cpuidle       isolated
    modalias   online    present  vulnerabilities
cpu1   cpu12  cpu15  cpu18  cpu3  cpu6  cpu9     hotplug       kernel_m
ax  nohz_full  possible  smt
cpu10  cpu13  cpu16  cpu19  cpu4  cpu7  cpufreq  intel_pstate  microcod
e   offline    power     uevent
[root@...pl-perf-test-skx-i9 cpufreq]# echo 0 > ../../cpu1
cpu1/  cpu10/ cpu11/ cpu12/ cpu13/ cpu14/ cpu15/ cpu16/ cpu17/ cpu18/
cpu19/ 
[root@...pl-perf-test-skx-i9 cpufreq]# echo 0 > ../../cpu1/online 
[root@...pl-perf-test-skx-i9 cpufreq]# echo 1 > ../../cpu1/online 
[root@...pl-perf-test-skx-i9 cpufreq]# dmesg | tail
[  626.188805] intel_pstate: CPU 1 stopping
[  626.188808] intel_pstate: CPU 1 going offline
[  626.205589] smpboot: CPU 1 is now offline
[  633.671812] x86: Booting SMP configuration:
[  633.671815] smpboot: Booting Node 0 Processor 1 APIC 0x2
[  633.677208] intel_pstate: CPU 1 going online
[  633.677253] intel_pstate: set_policy cpuinfo.max 4300000 policy->max 
4000000
[  633.677259] intel_pstate: cpu:1 max_state 43 min_policy_perf:12
max_policy_perf:40
[  633.677260] intel_pstate: cpu:1 global_min:12 global_max:43
[  633.677262] intel_pstate: cpu:1 max_perf_ratio:40 min_perf_ratio:12
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 1 0x774
7f00280c
[root@...pl-perf-test-skx-i9 cpufreq]# rtcwake -m mem -s 10
rtcwake: wakeup from "mem" using /dev/rtc0 at Thu Aug 27 19:42:51 2020
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -p 1 0x774
0
[root@...pl-perf-test-skx-i9 cpufreq]# grep . *
affected_cpus:1
base_frequency:3300000
cpuinfo_max_freq:4300000
cpuinfo_min_freq:1200000
cpuinfo_transition_latency:0
energy_performance_available_preferences:default performance
balance_performance balance_power power 
energy_performance_preference:performance
related_cpus:1
scaling_available_governors:performance powersave
scaling_cur_freq:699896
scaling_driver:intel_pstate
scaling_governor:powersave
scaling_max_freq:4000000
scaling_min_freq:1200000
scaling_setspeed:<unsupported>
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -a 0x774
80002b0c
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
[root@...pl-perf-test-skx-i9 cpufreq]# dmesg > /tmp/dmesg.txt
[root@...pl-perf-test-skx-i9 cpufreq]# vi /tmp/dmesg.txt 
[root@...pl-perf-test-skx-i9 cpufreq]# rdmsr -a 0x774
80002b0c
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
[root@...pl-perf-test-skx-i9 cpufreq]# 

Thanks,
Srinivas

> 
> Thanks,
> Rafael
> 
> 
> 

View attachment "dmesg_08_27.txt" of type "text/plain" (118159 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ