linux-kernel - Re: [PATCH V2 0/5] Support metric group constraint

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]

Date:   Wed, 26 Feb 2020 16:35:49 +0100
From:   Jiri Olsa <jolsa@...hat.com>
To:     kan.liang@...ux.intel.com
Cc:     acme@...nel.org, mingo@...hat.com, peterz@...radead.org,
        linux-kernel@...r.kernel.org, mark.rutland@....com,
        namhyung@...nel.org, ravi.bangoria@...ux.ibm.com,
        yao.jin@...ux.intel.com, ak@...ux.intel.com
Subject: Re: [PATCH V2 0/5] Support metric group constraint

On Mon, Feb 24, 2020 at 01:59:19PM -0800, kan.liang@...ux.intel.com wrote:
> From: Kan Liang <kan.liang@...ux.intel.com>
> 
> Changes since V1:
> - Remove global static flag violate_nmi_constraint, and add a new
>   function metricgroup___watchdog_constraint_hint() for all
>   watchdog constraint hints in patch 4.
>   The rest of the patches are not changed.

Acked-by: Jiri Olsa <jolsa@...hat.com>

thanks,
jirka

> 
> Some metric groups, e.g. Page_Walks_Utilization, will never count when
> NMI watchdog is enabled.
> 
>  $echo 1 > /proc/sys/kernel/nmi_watchdog
>  $perf stat -M Page_Walks_Utilization
> 
>  Performance counter stats for 'system wide':
> 
>  <not counted>      itlb_misses.walk_pending       (0.00%)
>  <not counted>      dtlb_load_misses.walk_pending  (0.00%)
>  <not counted>      dtlb_store_misses.walk_pending (0.00%)
>  <not counted>      ept.walk_pending               (0.00%)
>  <not counted>      cycles                         (0.00%)
> 
>        2.343460588 seconds time elapsed
> 
>  Some events weren't counted. Try disabling the NMI watchdog:
>         echo 0 > /proc/sys/kernel/nmi_watchdog
>         perf stat ...
>         echo 1 > /proc/sys/kernel/nmi_watchdog
>  The events in group usually have to be from the same PMU. Try
>  reorganizing the group.
> 
> A metric group is a weak group, which relies on group validation
> code in the kernel to determine whether to be opened as a group or
> a non-group. However, group validation code may return false-positives,
> especially when NMI watchdog is enabled. (The metric group is allowed
> as a group but will never be scheduled.)
> 
> The attempt to fix the group validation code has been rejected.
> https://lore.kernel.org/lkml/20200117091341.GX2827@hirez.programming.kicks-ass.net/
> Because we cannot accurately predict whether the group can be scheduled
> as a group, only by checking current status.
> 
> This patch set provides another solution to mitigate the issue.
> Add "MetricConstraint" in event list, which provides a hint for perf tool,
> e.g. "MetricConstraint": "NO_NMI_WATCHDOG". Perf tool can change the
> metric group to non-group (standalone metrics) if NMI watchdog is enabled.
> 
> After applying the patch,
> 
>  $echo 1 > /proc/sys/kernel/nmi_watchdog
>  $perf stat -M Page_Walks_Utilization
>   Splitting metric group Page_Walks_Utilization into standalone metrics.
>   Try disabling the NMI watchdog to comply NO_NMI_WATCHDOG metric constraint:
>         echo 0 > /proc/sys/kernel/nmi_watchdog
>         perf stat ...
>         echo 1 > /proc/sys/kernel/nmi_watchdog
> 
>  Performance counter stats for 'system wide':
> 
>         18,253,454      itlb_misses.walk_pending  #      0.0
>                               Page_Walks_Utilization   (50.55%)
>         78,051,525      dtlb_load_misses.walk_pending  (50.55%)
>         29,213,063      dtlb_store_misses.walk_pending (50.55%)
>                  0      ept.walk_pending               (50.55%)
>      2,542,132,364      cycles                         (49.92%)
> 
>        1.037095993 seconds time elapsed
> 
> Kan Liang (5):
>   perf jevents: Support metric constraint
>   perf metricgroup: Factor out metricgroup__add_metric_weak_group()
>   perf util: Factor out sysctl__nmi_watchdog_enabled()
>   perf metricgroup: Support metric constraint
>   perf vendor events: Add NO_NMI_WATCHDOG metric constraint
> 
>  .../arch/x86/cascadelakex/clx-metrics.json         |   3 +-
>  .../pmu-events/arch/x86/skylake/skl-metrics.json   |   3 +-
>  .../pmu-events/arch/x86/skylakex/skx-metrics.json  |   3 +-
>  tools/perf/pmu-events/jevents.c                    |  19 ++--
>  tools/perf/pmu-events/jevents.h                    |   2 +-
>  tools/perf/pmu-events/pmu-events.h                 |   1 +
>  tools/perf/util/metricgroup.c                      | 109 ++++++++++++++++-----
>  tools/perf/util/stat-display.c                     |   6 +-
>  tools/perf/util/util.c                             |  18 ++++
>  tools/perf/util/util.h                             |   2 +
>  10 files changed, 128 insertions(+), 38 deletions(-)
> 
> -- 
> 2.7.4
>