Sem eventos de hardware na lista Perf no PC W / Intel Core i7-4770 CPU executando o servidor Ubuntu? -- drivers campo com server campo com kernel campo com performance camp askubuntu Relacionado O problema

No hardware events in perf list on PC w/ Intel Core i7-4770 CPU running Ubuntu Server?


1
vote

problema

português

Eu tenho Dell Desktop com Intel Core i7-4770 @ 3.4GHz CPU.

Lá, instalei o Ubuntu Server 15.04. Depois de instalar ABCDEFGHIJKLMNABCDEFGHIJKLMN7 e ABCDEFGHIJKLMNABCDEFGHIJKLMN8 para instalar Ferramentas Perf. No entanto, quando executo ABCDEFGHIJKLMNABCDEFGHIJKLMN9 A seção de hardware não está lá, nem mesmo "não suportada". A única outra coisa é que eu fiz antes é instalar o Xen, e o Ubuntu está atualmente em execução como Dom0. Não tenho certeza se isso pode ter qualquer efeito no meu problema.

Aqui está o que eu recebo do comando plasmashell0 :

  plasmashell1   

Eu gostaria de saber qual é o meu problema? É o próprio CPU que não tem suporte para eventos de hardware? Ou é o kernel do repositório que não tem eventos de hardware ativados? Se apenas kernel, acho que vou ter que

english

I have dell desktop with Intel Core i7-4770 @ 3.4GHz CPU.

There, I have installed Ubuntu server 15.04. Afterwards I have install linux-tools-generic and linux-tools-common to install perf tools. However, when I run perf list the hardware section is not there, not even "not supported". The only other thing is that I have done before is to install xen, and Ubuntu is currently running as Dom0. I am not sure if this can have any effect on my problem.

Here is what I get from command perf list:

ubuntuuser@ubuntu:~/work/linux$ perf list  List of pre-defined events (to be used in -e):   cpu-clock                                          [Software event]   task-clock                                         [Software event]   page-faults OR faults                              [Software event]   context-switches OR cs                             [Software event]   cpu-migrations OR migrations                       [Software event]   minor-faults                                       [Software event]   major-faults                                       [Software event]   alignment-faults                                   [Software event]   emulation-faults                                   [Software event]   dummy                                              [Software event]    power/energy-cores/                                [Kernel PMU event]   power/energy-gpu/                                  [Kernel PMU event]   power/energy-pkg/                                  [Kernel PMU event]   power/energy-ram/                                  [Kernel PMU event]    rNNN                                               [Raw hardware event descriptor]   cpu/t1=v1[,t2=v2,t3 ...]/modifier                  [Raw hardware event descriptor]    (see 'man perf-list' on how to encode it)    mem:<addr>[:access]                                [Hardware breakpoint]    [ Tracepoints not available: Permission denied ] 

I would like to know what is my problem? Is it the CPU itself that doesn't have support for hardware events? Or is it kernel from repository that doesn't have hardware events enabled? If it just kernel, I guess I will just have to

           
   
   

Lista de respostas

1
 
vote
vote
Melhor resposta
 

Eu não recebi nenhum evento de hardware, até que eu usei alguns. Agora, mesmo depois de uma nova reabacção, alguns estão listados:

  doug@s15:~$ perf list hw  List of pre-defined events (to be used in -e):    branch-instructions OR branches                    [Hardware event]   branch-misses                                      [Hardware event]   bus-cycles                                         [Hardware event]   cache-misses                                       [Hardware event]   cache-references                                   [Hardware event]   cpu-cycles OR cycles                               [Hardware event]   instructions                                       [Hardware event]   ref-cycles                                         [Hardware event]   stalled-cycles-backend OR idle-cycles-backend      [Hardware event]   stalled-cycles-frontend OR idle-cycles-frontend    [Hardware event]  doug@s15:~$ perf list pmu  List of pre-defined events (to be used in -e):    branch-instructions OR cpu/branch-instructions/    [Kernel PMU event]   branch-misses OR cpu/branch-misses/                [Kernel PMU event]   bus-cycles OR cpu/bus-cycles/                      [Kernel PMU event]   cache-misses OR cpu/cache-misses/                  [Kernel PMU event]   cache-references OR cpu/cache-references/          [Kernel PMU event]   cpu-cycles OR cpu/cpu-cycles/                      [Kernel PMU event]   instructions OR cpu/instructions/                  [Kernel PMU event]   mem-loads OR cpu/mem-loads/                        [Kernel PMU event]   mem-stores OR cpu/mem-stores/                      [Kernel PMU event]   power/energy-cores/                                [Kernel PMU event]   power/energy-gpu/                                  [Kernel PMU event]   power/energy-pkg/                                  [Kernel PMU event]   ref-cycles OR cpu/ref-cycles/                      [Kernel PMU event]   stalled-cycles-backend OR cpu/stalled-cycles-backend/ [Kernel PMU event]   stalled-cycles-frontend OR cpu/stalled-cycles-frontend/ [Kernel PMU event]   uncore_cbox_0/clockticks/                          [Kernel PMU event]   uncore_cbox_1/clockticks/                          [Kernel PMU event]   uncore_cbox_2/clockticks/                          [Kernel PMU event]   uncore_cbox_3/clockticks/                          [Kernel PMU event]   uncore_imc/data_reads/                             [Kernel PMU event]   uncore_imc/data_writes/                            [Kernel PMU event]    

anteriormente eu tinha feito isso:

  doug@s15:~$ sudo ~/bin/perf stat -a sleep 5   Performance counter stats for 'system wide':        39997.230896      task-clock (msec)         #    7.998 CPUs utilized            (100.00%)                396      context-switches          #    0.010 K/sec                    (100.00%)                  8      cpu-migrations            #    0.000 K/sec                    (100.00%)                 82      page-faults               #    0.002 K/sec        359,328,178      cycles                    #    0.009 GHz                      (83.29%)        827,261,275      stalled-cycles-frontend   #  230.22% frontend cycles idle     (83.33%)        797,161,705      stalled-cycles-backend    #  221.85% backend  cycles idle     (66.74%)         49,220,050      instructions              #    0.14  insns per cycle                                                   #   16.81  stalled cycles per insn  (83.37%)         10,011,396      branches                  #    0.250 M/sec                    (83.37%)          1,800,790      branch-misses             #   17.99% of all branches          (83.29%)         5.000925953 seconds time elapsed  doug@s15:~$ sudo ~/bin/perf stat -e cycles,instructions,cache-references,cache-misses,bus-cycles -a sleep 5   Performance counter stats for 'system wide':          24,268,338      cycles                                                        (100.00%)          8,505,561      instructions              #    0.35  insns per cycle          (100.00%)            657,099      cache-references                                              (100.00%)             73,174      cache-misses              #   11.136 % of all cache refs      (100.00%)          1,507,017      bus-cycles         5.001004407 seconds time elapsed  doug@s15:~$ sudo ~/bin/perf stat -e L1-dcache-loads,L1-dcache-load-misses,L1-dcache-stores sleep 5   Performance counter stats for 'sleep 5':             170,208      L1-dcache-loads             14,139      L1-dcache-load-misses     #    8.31% of all L1-dcache hits      <not counted>      L1-dcache-stores           (0.00%)         5.000957676 seconds time elapsed    

Eu não sei como minha lista está completa, nem como obter a lista completa se não for.

 

I didn't get any hardware events listed either, until after I used some. Now, even after a fresh re-boot, some are listed:

doug@s15:~$ perf list hw  List of pre-defined events (to be used in -e):    branch-instructions OR branches                    [Hardware event]   branch-misses                                      [Hardware event]   bus-cycles                                         [Hardware event]   cache-misses                                       [Hardware event]   cache-references                                   [Hardware event]   cpu-cycles OR cycles                               [Hardware event]   instructions                                       [Hardware event]   ref-cycles                                         [Hardware event]   stalled-cycles-backend OR idle-cycles-backend      [Hardware event]   stalled-cycles-frontend OR idle-cycles-frontend    [Hardware event]  doug@s15:~$ perf list pmu  List of pre-defined events (to be used in -e):    branch-instructions OR cpu/branch-instructions/    [Kernel PMU event]   branch-misses OR cpu/branch-misses/                [Kernel PMU event]   bus-cycles OR cpu/bus-cycles/                      [Kernel PMU event]   cache-misses OR cpu/cache-misses/                  [Kernel PMU event]   cache-references OR cpu/cache-references/          [Kernel PMU event]   cpu-cycles OR cpu/cpu-cycles/                      [Kernel PMU event]   instructions OR cpu/instructions/                  [Kernel PMU event]   mem-loads OR cpu/mem-loads/                        [Kernel PMU event]   mem-stores OR cpu/mem-stores/                      [Kernel PMU event]   power/energy-cores/                                [Kernel PMU event]   power/energy-gpu/                                  [Kernel PMU event]   power/energy-pkg/                                  [Kernel PMU event]   ref-cycles OR cpu/ref-cycles/                      [Kernel PMU event]   stalled-cycles-backend OR cpu/stalled-cycles-backend/ [Kernel PMU event]   stalled-cycles-frontend OR cpu/stalled-cycles-frontend/ [Kernel PMU event]   uncore_cbox_0/clockticks/                          [Kernel PMU event]   uncore_cbox_1/clockticks/                          [Kernel PMU event]   uncore_cbox_2/clockticks/                          [Kernel PMU event]   uncore_cbox_3/clockticks/                          [Kernel PMU event]   uncore_imc/data_reads/                             [Kernel PMU event]   uncore_imc/data_writes/                            [Kernel PMU event] 

Earlier I had done these:

doug@s15:~$ sudo ~/bin/perf stat -a sleep 5   Performance counter stats for 'system wide':        39997.230896      task-clock (msec)         #    7.998 CPUs utilized            (100.00%)                396      context-switches          #    0.010 K/sec                    (100.00%)                  8      cpu-migrations            #    0.000 K/sec                    (100.00%)                 82      page-faults               #    0.002 K/sec        359,328,178      cycles                    #    0.009 GHz                      (83.29%)        827,261,275      stalled-cycles-frontend   #  230.22% frontend cycles idle     (83.33%)        797,161,705      stalled-cycles-backend    #  221.85% backend  cycles idle     (66.74%)         49,220,050      instructions              #    0.14  insns per cycle                                                   #   16.81  stalled cycles per insn  (83.37%)         10,011,396      branches                  #    0.250 M/sec                    (83.37%)          1,800,790      branch-misses             #   17.99% of all branches          (83.29%)         5.000925953 seconds time elapsed  doug@s15:~$ sudo ~/bin/perf stat -e cycles,instructions,cache-references,cache-misses,bus-cycles -a sleep 5   Performance counter stats for 'system wide':          24,268,338      cycles                                                        (100.00%)          8,505,561      instructions              #    0.35  insns per cycle          (100.00%)            657,099      cache-references                                              (100.00%)             73,174      cache-misses              #   11.136 % of all cache refs      (100.00%)          1,507,017      bus-cycles         5.001004407 seconds time elapsed  doug@s15:~$ sudo ~/bin/perf stat -e L1-dcache-loads,L1-dcache-load-misses,L1-dcache-stores sleep 5   Performance counter stats for 'sleep 5':             170,208      L1-dcache-loads             14,139      L1-dcache-load-misses     #    8.31% of all L1-dcache hits      <not counted>      L1-dcache-stores           (0.00%)         5.000957676 seconds time elapsed 

I do not how know if my list is now complete, nor how to get the complete list if it isn't.

 
 

Perguntas relacionadas

1  Ubuntu 12,04 lts versus Ubuntu 13.10 [Duplicado]  ( Ubuntu 12 04 lts versus ubuntu 13 10 ) 
. Esta pergunta já tem respostas aqui : Qual é a diferença entre um suporte a longo prazo Liberação e...

0  Ubuntu 18.04.4 lts lento em AMD 4300 8GB de RAM  ( Ubuntu 18 04 4 lts slow on amd 4300 8gb ram ) 
Eu instalei o Ubuntu 18.04.4 Lts Versão em um PC de desktop com o processador AMD 4300 Quad Core, 8GB RAM e Hardware HDD de 160GB. leva mais de dois minutos...

3  Software Anti Virus / Anti Malware [duplicado]  ( Anti virus anti malware software ) 
. Esta pergunta já tem respostas aqui : fechado 10 anos atrás . . Duplicado possível: Digital...

347  Como posso melhorar o desempenho geral do sistema do Ubuntu?  ( How can i improve ubuntu overall system performance ) 
Quais são as suas dicas para melhorar o desempenho geral do sistema no Ubuntu? Inspirado por este pergunta Eu percebi que algumas configurações padrão podem...

5  Netbook Edition Battery Life - Ubuntu vs Kubuntu?  ( Netbook edition battery life ubuntu vs kubuntu ) 
Eu tenho executado em meu eeee 1000 praticamente desde que saiu há mais de um ano (eu estava usando-se amasia antes disso), mas estou curioso sobre a nova edi...

1  Cache SSD (por exemplo, Asus P9X79 WS) - possível no Ubuntu?  ( Ssd caching e g asus p9x79 ws possible on ubuntu ) 
Algumas placas-mãe como ASUS P9X79 WS Suporte algo chamado "Asus SSD Caching" . Você pode conectar 1 HDD e 1 SSD a 2 portas SATA designadas e IUUC, você pod...

10  Intel_PowerClamp: Iniciar / Parar Injeção Idle Forçada  ( Intel powerclamp start stop forced idle injection ) 
Eu realmente não sei quando começou, acabei de atualizar de 13,10 a 14.04 e estava prestes a fazer algo sobre performances gráficas (Unity sentiu lento). Essa...

19  Qual é a perda de desempenho se você executar o Ubuntu Desktop Edition em uma máquina de servidor?  ( What is the performance loss if you run ubuntu desktop edition on a server machi ) 
Nós instalamos o Ubuntu Desktop Edition em nosso servidor dev. Eu queria saber se há alguma perda de desempenho perceptível em comparação com a edição do se...

1  Animações lentas usando o Ubuntu 13.04  ( Slow animations using ubuntu 13 04 ) 
Eu decidi agora trocar de ganhar para o Ubuntu. Agora, após a instalação do mais recente Ubuntu (13.04), as animações da área de trabalho (assim como abrir o ...

6  Como posso fazer o Perf Script -g Python funciona?  ( How can i make perf script g python work ) 
Após instalar sudo apt-get install pbuilder sudo pbuilder create # create isolated environment for building packages 3 com o pacote sudo apt-get ...




© 2022 pergunte.org All Rights Reserved. Casa de perguntas e respostas todos os direitos reservados


Licensed under cc by-sa 3.0 with attribution required.