Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1715926 > unrolled thread

[PATCH] tools/perf: fix documentation for sysctls perf_event_paranoid and perf_event_mlock_kb

Started byKonstantin Khlebnikov <khlebnikov@yandex-team.ru>
First post2017-08-20 13:50 +0200
Last post2017-08-24 10:40 +0200
Articles 3 — 3 participants

Back to article view | Back to linux.kernel


Contents

  [PATCH] tools/perf: fix documentation for sysctls  perf_event_paranoid and perf_event_mlock_kb Konstantin Khlebnikov <khlebnikov@yandex-team.ru> - 2017-08-20 13:50 +0200
    Re: [PATCH] tools/perf: fix documentation for sysctls  perf_event_paranoid and perf_event_mlock_kb Arnaldo Carvalho de Melo <acme@kernel.org> - 2017-08-22 18:00 +0200
    [tip:perf/core] perf: Fix documentation for sysctls  perf_event_paranoid and perf_event_mlock_kb tip-bot for Konstantin Khlebnikov <tipbot@zytor.com> - 2017-08-24 10:40 +0200

#1715926 — [PATCH] tools/perf: fix documentation for sysctls perf_event_paranoid and perf_event_mlock_kb

FromKonstantin Khlebnikov <khlebnikov@yandex-team.ru>
Date2017-08-20 13:50 +0200
Subject[PATCH] tools/perf: fix documentation for sysctls perf_event_paranoid and perf_event_mlock_kb
Message-ID<ugwTU-6pr-1@gated-at.bofh.it>
Fix misprint CAP_IOC_LOCK -> CAP_IPC_LOCK. This capability have nothing
to do with raw tracepoints. This part is about bypassing mlock limits.

Sysctl kernel.perf_event_paranoid = -1 allows raw and ftrace function
tracepoints without CAP_SYS_ADMIN.

Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
---
 Documentation/sysctl/kernel.txt |   13 ++++++++++++-
 tools/perf/util/evsel.c         |    4 +++-
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
index bac23c198360..ce61d1fe08ca 100644
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -61,6 +61,7 @@ show up in /proc/sys/kernel:
 - perf_cpu_time_max_percent
 - perf_event_paranoid
 - perf_event_max_stack
+- perf_event_mlock_kb
 - perf_event_max_contexts_per_stack
 - pid_max
 - powersave-nap               [ PPC only ]
@@ -654,7 +655,9 @@ Controls use of the performance events system by unprivileged
 users (without CAP_SYS_ADMIN).  The default value is 2.
 
  -1: Allow use of (almost) all events by all users
->=0: Disallow raw tracepoint access by users without CAP_IOC_LOCK
+     Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK
+>=0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN
+     Disallow raw tracepoint access by users without CAP_SYS_ADMIN
 >=1: Disallow CPU event access by users without CAP_SYS_ADMIN
 >=2: Disallow kernel profiling by users without CAP_SYS_ADMIN
 
@@ -673,6 +676,14 @@ The default value is 127.
 
 ==============================================================
 
+perf_event_mlock_kb:
+
+Control size of per-cpu ring buffer not counted agains mlock limit.
+
+The default value is 512 + 1 page
+
+==============================================================
+
 perf_event_max_contexts_per_stack:
 
 Controls maximum number of stack frame context entries for
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 413f74df08de..5ed7056d1fd0 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -2535,7 +2535,9 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
 		 "unprivileged users (without CAP_SYS_ADMIN).\n\n"
 		 "The current value is %d:\n\n"
 		 "  -1: Allow use of (almost) all events by all users\n"
-		 ">= 0: Disallow raw tracepoint access by users without CAP_IOC_LOCK\n"
+		 "      Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK\n"
+		 ">= 0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN\n"
+		 "      Disallow raw tracepoint access by users without CAP_SYS_ADMIN\n"
 		 ">= 1: Disallow CPU event access by users without CAP_SYS_ADMIN\n"
 		 ">= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN\n\n"
 		 "To make this setting permanent, edit /etc/sysctl.conf too, e.g.:\n\n"

[toc] | [next] | [standalone]


#1717524

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2017-08-22 18:00 +0200
Message-ID<uhjKV-4r8-7@gated-at.bofh.it>
In reply to#1715926
Em Sun, Aug 20, 2017 at 02:39:20PM +0300, Konstantin Khlebnikov escreveu:
> Fix misprint CAP_IOC_LOCK -> CAP_IPC_LOCK. This capability have nothing
> to do with raw tracepoints. This part is about bypassing mlock limits.
> 
> Sysctl kernel.perf_event_paranoid = -1 allows raw and ftrace function
> tracepoints without CAP_SYS_ADMIN.

Thanks, looks sane, applied.

- Arnaldo
 
> Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
> ---
>  Documentation/sysctl/kernel.txt |   13 ++++++++++++-
>  tools/perf/util/evsel.c         |    4 +++-
>  2 files changed, 15 insertions(+), 2 deletions(-)
> 
> diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
> index bac23c198360..ce61d1fe08ca 100644
> --- a/Documentation/sysctl/kernel.txt
> +++ b/Documentation/sysctl/kernel.txt
> @@ -61,6 +61,7 @@ show up in /proc/sys/kernel:
>  - perf_cpu_time_max_percent
>  - perf_event_paranoid
>  - perf_event_max_stack
> +- perf_event_mlock_kb
>  - perf_event_max_contexts_per_stack
>  - pid_max
>  - powersave-nap               [ PPC only ]
> @@ -654,7 +655,9 @@ Controls use of the performance events system by unprivileged
>  users (without CAP_SYS_ADMIN).  The default value is 2.
>  
>   -1: Allow use of (almost) all events by all users
> ->=0: Disallow raw tracepoint access by users without CAP_IOC_LOCK
> +     Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK
> +>=0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN
> +     Disallow raw tracepoint access by users without CAP_SYS_ADMIN
>  >=1: Disallow CPU event access by users without CAP_SYS_ADMIN
>  >=2: Disallow kernel profiling by users without CAP_SYS_ADMIN
>  
> @@ -673,6 +676,14 @@ The default value is 127.
>  
>  ==============================================================
>  
> +perf_event_mlock_kb:
> +
> +Control size of per-cpu ring buffer not counted agains mlock limit.
> +
> +The default value is 512 + 1 page
> +
> +==============================================================
> +
>  perf_event_max_contexts_per_stack:
>  
>  Controls maximum number of stack frame context entries for
> diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
> index 413f74df08de..5ed7056d1fd0 100644
> --- a/tools/perf/util/evsel.c
> +++ b/tools/perf/util/evsel.c
> @@ -2535,7 +2535,9 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
>  		 "unprivileged users (without CAP_SYS_ADMIN).\n\n"
>  		 "The current value is %d:\n\n"
>  		 "  -1: Allow use of (almost) all events by all users\n"
> -		 ">= 0: Disallow raw tracepoint access by users without CAP_IOC_LOCK\n"
> +		 "      Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK\n"
> +		 ">= 0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN\n"
> +		 "      Disallow raw tracepoint access by users without CAP_SYS_ADMIN\n"
>  		 ">= 1: Disallow CPU event access by users without CAP_SYS_ADMIN\n"
>  		 ">= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN\n\n"
>  		 "To make this setting permanent, edit /etc/sysctl.conf too, e.g.:\n\n"

[toc] | [prev] | [next] | [standalone]


#1719003 — [tip:perf/core] perf: Fix documentation for sysctls perf_event_paranoid and perf_event_mlock_kb

Fromtip-bot for Konstantin Khlebnikov <tipbot@zytor.com>
Date2017-08-24 10:40 +0200
Subject[tip:perf/core] perf: Fix documentation for sysctls perf_event_paranoid and perf_event_mlock_kb
Message-ID<uhVQg-43p-53@gated-at.bofh.it>
In reply to#1715926
Commit-ID:  ac0bb6b72f4bbab08f270a919406d971e73698b5
Gitweb:     http://git.kernel.org/tip/ac0bb6b72f4bbab08f270a919406d971e73698b5
Author:     Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
AuthorDate: Sun, 20 Aug 2017 14:39:20 +0300
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Tue, 22 Aug 2017 13:24:54 -0300

perf: Fix documentation for sysctls perf_event_paranoid and perf_event_mlock_kb

Fix misprint CAP_IOC_LOCK -> CAP_IPC_LOCK. This capability have nothing
to do with raw tracepoints. This part is about bypassing mlock limits.

Sysctl kernel.perf_event_paranoid = -1 allows raw and ftrace function
tracepoints without CAP_SYS_ADMIN.

Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/150322916080.129746.11285255474738558340.stgit@buzz
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 Documentation/sysctl/kernel.txt | 13 ++++++++++++-
 tools/perf/util/evsel.c         |  4 +++-
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
index bac23c1..ce61d1f 100644
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -61,6 +61,7 @@ show up in /proc/sys/kernel:
 - perf_cpu_time_max_percent
 - perf_event_paranoid
 - perf_event_max_stack
+- perf_event_mlock_kb
 - perf_event_max_contexts_per_stack
 - pid_max
 - powersave-nap               [ PPC only ]
@@ -654,7 +655,9 @@ Controls use of the performance events system by unprivileged
 users (without CAP_SYS_ADMIN).  The default value is 2.
 
  -1: Allow use of (almost) all events by all users
->=0: Disallow raw tracepoint access by users without CAP_IOC_LOCK
+     Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK
+>=0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN
+     Disallow raw tracepoint access by users without CAP_SYS_ADMIN
 >=1: Disallow CPU event access by users without CAP_SYS_ADMIN
 >=2: Disallow kernel profiling by users without CAP_SYS_ADMIN
 
@@ -673,6 +676,14 @@ The default value is 127.
 
 ==============================================================
 
+perf_event_mlock_kb:
+
+Control size of per-cpu ring buffer not counted agains mlock limit.
+
+The default value is 512 + 1 page
+
+==============================================================
+
 perf_event_max_contexts_per_stack:
 
 Controls maximum number of stack frame context entries for
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 5dfb8bc..a5888c7 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -2674,7 +2674,9 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
 		 "unprivileged users (without CAP_SYS_ADMIN).\n\n"
 		 "The current value is %d:\n\n"
 		 "  -1: Allow use of (almost) all events by all users\n"
-		 ">= 0: Disallow raw tracepoint access by users without CAP_IOC_LOCK\n"
+		 "      Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK\n"
+		 ">= 0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN\n"
+		 "      Disallow raw tracepoint access by users without CAP_SYS_ADMIN\n"
 		 ">= 1: Disallow CPU event access by users without CAP_SYS_ADMIN\n"
 		 ">= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN\n\n"
 		 "To make this setting permanent, edit /etc/sysctl.conf too, e.g.:\n\n"

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web